Arimaa Forum (http://arimaa.com/arimaa/forum/cgi/YaBB.cgi)
Team Games >> 2007 One vs TheMob >> Move 12
(Message started by: RonWeasley on Aug 22nd, 2007, 2:24pm)

Title: Move 12
Post by RonWeasley on Aug 22nd, 2007, 2:24pm
Well it looks like I survived by trip with Harry, so I'm back to coordinate TheMob.  Things are a bit dicey for the Slytherins these days, which may explain Fritzlein buckling under the pressure.  Thanx to Fritzlein for covering for me in his snake-like way.  At least it kept him away from Hogwarts.

We'll see if TheMob has taken the initiative with that horse move or simply given chessandgo a fat, juicy target.

Title: Re: Move 12
Post by The_Jeh on Aug 23rd, 2007, 5:27pm
Chessandgo has chosen 12w hb6w Eb5n Eb6s rb7s.

Apparently, he thinks we will lose more positionally from him pulling our b-file rabbit than he will from us pulling his h-file rabbit.

Title: Re: Move 12
Post by Fritzlein on Aug 23rd, 2007, 6:14pm
12b hh3n Rh2n hh4n Rh3n

This seems idiot simple.  If we race straight up, both sides will capture on move 14, leaving chessandgo with move 15w to take the initiative, which would mean we were slightly losing.  But we don't have to race straight up, we can complicate on 13b.  The fact that both elephants are near his rabbit pull seems to favor us.

My clinching argument for the straightforward move is not so much that it is a good move as that I don't see anything else for us.  Trying to complicate on this move seems to waste us more time than it gains us.  Is this one of those rare moves on which a snap vote is appropriate?  Maybe the sentiment won't be as unanimous as I think.

Now that the game is more unsettled, it is harder for me to evaluate.  Does anyone have an opinion as to who is winning?

Title: Re: Move 12
Post by The_Jeh on Aug 23rd, 2007, 8:38pm
Maybe 12b ed4w hh3n Rh2n xxxx (ha6n?).

My only reasoning would be that we could complicate things by threatening to flip his horse or pull his dog. Also, it puts our elephant beside the c3 trap early. But I actually like your idea just as well, Fritzlein.

Title: Re: Move 12
Post by RonWeasley on Aug 24th, 2007, 6:32am
I would do both pulls on this move.  After gold flips our r we get another pull and two steps, presumeably ed4w rb4w.  If we don't do both pulls on this move, we end up in a similar position after 13b but with one less pull of gold's rabbit.

If there is only sparse discussion, I will start the vote on Monday morning (EDT) ending Tuesday morning if we have 10 votes and Tuesday evening otherwise.  Remember, if this is an easy decision then voting early will put more time back into our reserve.

Title: Re: Move 12
Post by UruramTururam on Aug 24th, 2007, 7:20am
C&G decided to ignore our charge...
I suggest playing:

12b md6e ed4w ec4n Hc3n

Reasoning:

md6e:
  • moves our m farther from opponent's E
  • secures f6
  • allows m-h attack on f3

    ed4w ec4n:
  • seals E - now it needs quite a long time to switch wings

    Hc3n
  • makes the seal even stronger.

    Now gold cannot afford to flip our r (H is lost).

    The answer could possibly be
    13w Eb5s rb4s Hc4s xxx

    if xxx is Eb4e then our r is safe for a while and we may double pull the left wing R.

    if xxx is hc3e then our answer should contain moving our e southwards blocking golden E.

  • Title: Re: Move 12
    Post by 99of9 on Aug 24th, 2007, 7:24am

    on 08/24/07 at 06:32:11, RonWeasley wrote:
    After gold flips our r we get another pull and two steps, presumeably ed4w rb4w.

    Whoah...  why will he flip?  The drag leaves it much harder for us to compicate with 2 steps:
    13w Eb5s rb6s Eb4e rb5s

    Jeh's move means we get the prominent c4 outpost.  An alternative final step is dd8s.

    I'm not sure which is better, but I'd like to hear some better analysis before fasttracking the double pull.

    Title: Re: Move 12
    Post by Fritzlein on Aug 24th, 2007, 8:37am

    on 08/24/07 at 07:24:00, 99of9 wrote:
    Whoah...  why will he flip?  The drag leaves it much harder for us to compicate with 2 steps:
    13w Eb5s rb6s Eb4e rb5s

    Oops, I didn't see the drag.  I was assuming he would flip.  As usual the position isn't as easy as it first appears.

    If we pull once and occupy c4 with our elephant, we need to be more aware of chessandgo pulling once and centralizing his elephant to go after our camel, or directly going after our camel, since we've lost a step decentralizing our elephant.

    Now that our double-pull looks inferior, I will venture to guess that we are slightly losing, although we need more lines to be confident of anything in a race-like position.

    Title: Re: Move 12
    Post by jdb on Aug 24th, 2007, 9:11am
    12b hh3n Rh2n ed4w rb6e

    I think silver needs to drag the rabbit one step on this move. The reason being, in some lines, the horse can flip the rabbit instead of dragging it. After flipping, the horse on h4 has the possibility of sneaking a second rabbit pull from h1. If the horse is on h5, this is not possible.

    Putting the elephant on c4 is important. It threatens gold's c3 horse. This means gold can at best drag the rabbit one step since he needs to spend a step to tend to the horse.

    Moving the rabbit to c6 makes it harder for gold to continue dragging the rabbit. If gold drags it to c5, he has to decide which side of the c5/c4 wall to put his elephant on. He has weaknesses on either side of the wall (a2 rabbit, h3 rabbit), so either way we should get something out of it.

    Some possible lines:

    A)
    12b hh3n Rh2n ed4w rb6e
    13w Eb5n Eb6n Eb7s rb8s
    13b ec4e Hc3n Hc4n ed4w

    B)
    12b hh3n Rh2n ed4w rb6e
    13w Eb5e Ec5w rc6s Hc3e
    13b hh4w Rh3n Rh4n hg4e

    C)
    12b hh3n Rh2n ed4w rb6e
    13w Eb5e Ec5e rc6s Ed5e
    13b hh4n Rh3n hh5w hg5n









    Title: Re: Move 12
    Post by arimaa_master on Aug 24th, 2007, 12:31pm
    12b hh3n Rh2n hh4n Rh3n - at first glance it seemed to me like the only one possible good move but after reading through the suggested moves now I changed my mind too :) (mainly due to that devilish drag of our rabbit ending with c4 phant for Chessandgo)

    12b md6e ed4w ec4n Hc3n - at first sight it looks good but then I changed my mind because of 13w Eb5s rb6s Hc4s Eb4e but now I changed my mind again coz after 13b hh3n Rh2n hh4n Rh3n it looks playable.


    12b hh3n Rh2n ed4w rb6e  - this is very interesting move though after

        13w Eb5e Ec5e rc6s ra2e
        13b hh4n Rh3n hh5n Rh4n
        14w Hg3n Hg4n Hg5n De2

    there are some sharp lines where I am not quite sure who could be better.


    Thus  12b md6e ed4w ec4n Hc3n is my favourite so far.

    Title: Re: Move 12
    Post by jdb on Aug 24th, 2007, 1:26pm

    on 08/24/07 at 12:31:11, arimaa_master wrote:
     12b hh3n Rh2n ed4w rb6e  - this is very interesting move though after

        13w Eb5e Ec5e rc6s ra2e
        13b hh4n Rh3n hh5n Rh4n
        14w Hg3n Hg4n Hg5n De2

    there are some sharp lines where I am not quite sure who could be better.


    12b hh3n Rh2n ed4w rb6e
    13w Eb5e Ec5e rc6s ra2e
    13b ha6s ec4w eb4e Db3n

    Gold's dog is in danger, so gold can either safe it, or find a bigger threat somewhere else. Either way I think silver is ok, but again the position is getting sharp and some checking would be required.

    Title: Re: Move 12
    Post by UruramTururam on Aug 24th, 2007, 5:57pm
    For hh3n Rh2n ed4w rb6e we risk:
    13w Eb5e Ec5e rc6s Hc3e
    and we have golden E in the center freezing our camel and charging towards our horse, not to mention the rabbit in trouble. I think we should keep the golden Elephant out of the center now.

    Title: Re: Move 12
    Post by jdb on Aug 24th, 2007, 7:14pm

    on 08/24/07 at 17:57:26, UruramTururam wrote:
    For hh3n Rh2n ed4w rb6e we risk:
    13w Eb5e Ec5e rc6s Hc3e
    and we have golden E in the center freezing our camel and charging towards our horse, not to mention the rabbit in trouble. I think we should keep the golden Elephant out of the center now.


    12b hh3n Rh2n ed4w rb6e
    13w Eb5e Ec5e rc6s Hc3e
    13b ha6s ec4w eb4e Db3n

    I think silver is ok here, but maybe gold still has something.

    Title: Re: Move 12
    Post by Fritzlein on Aug 24th, 2007, 8:37pm

    on 08/24/07 at 07:20:06, UruramTururam wrote:
    C&G decided to ignore our charge...
    I suggest playing:

    12b md6e ed4w ec4n Hc3n

    Now that I've done a little more analysis rather than just spouting off my instincts, I am starting to like UruramTururam's move.  If Chessandgo doesn't pull our rabbit once, he isn't making any progress, but he can't pull it twice, so next move it is pre-unfrozen and able to move to a5, for a gain in the race large enough to spend the whole present move on.

    Furthermore, if the Gold elephant doesn't swing to c4, our elephant can take one step to restore the hindrance.

    12b md6e ed4w ec4n Hc3n
    13w Eb5s rb6s Hc4s Rh2w
    13b ec5s hh3s hh2n Rh1n

    and I think we are ahead.  So it seems that Gold must let us retreat our rabbit with

    12b md6e ed4w ec4n Hc3n
    13w Eb5s rb6s Hc4s Eb4e
    13b rb5w hh3n Rh2n xxxx

    Where we are slightly behind in the race, but this is somewhat compensated by the awkwardness of Gold's piece placement.

    Just one point of disagreement:


    Quote:
    md6e:
  • moves our m farther from opponent's E
  • secures f6
  • allows m-h attack on f3

  • I do not think we should contemplate an MH attack any time in the foreseeable future.  It's too slow and can't accomplish anything.  An EMH attack is more within the realm of possibility, but that doesn't combine well with a rabbit pull, i.e. once we pull the h-rabbit, he may have an easier time defending against an EMH attack, so we should choose only the one that looks better (the rabbit pull IMHO).

    Title: Re: Move 12
    Post by jdb on Aug 24th, 2007, 9:57pm

    on 08/24/07 at 20:37:29, Fritzlein wrote:
    Now that I've done a little more analysis rather than just spouting off my instincts, I am starting to like UruramTururam's move.  If Chessandgo doesn't pull our rabbit once, he isn't making any progress, but he can't pull it twice, so next move it is pre-unfrozen and able to move to a5, for a gain in the race large enough to spend the whole present move on.

    Furthermore, if the Gold elephant doesn't swing to c4, our elephant can take one step to restore the hindrance.

    12b md6e ed4w ec4n Hc3n
    13w Eb5s rb6s Hc4s Rh2w
    13b ec5s hh3s hh2n Rh1n


    12b md6e ed4w ec4n Hc3n
    13w Eb5s rb6s Hc4s Hc3e

    How does silver maintain the advantage now? Gold's dog and horse are not on the usual squares. Is there a way to take advantage of this?



    Title: Re: Move 12
    Post by The_Jeh on Aug 24th, 2007, 10:14pm

    on 08/24/07 at 20:37:29, Fritzlein wrote:
    So it seems that Gold must let us retreat our rabbit with

    12b md6e ed4w ec4n Hc3n
    13w Eb5s rb6s Hc4s Eb4e
    13b rb5w hh3n Rh2n xxxx

    Where we are slightly behind in the race, but this is somewhat compensated by the awkwardness of Gold's piece placement.


    But then 14w Ec4e Ed4e Ee4e Ef4e.

    Does this position favor us, too? We could move our elephant to f4, but then I think the c6 trap looks awkward; maybe it's an illusion. Or would you retreat the horse with step 4 of 13b?

    Title: Re: Move 12
    Post by mdk on Aug 25th, 2007, 12:14am
    how about 12b ed4n ed5w rb6e ha6e ?

    Title: Re: Move 12
    Post by UruramTururam on Aug 25th, 2007, 5:16pm

    on 08/24/07 at 20:37:29, Fritzlein wrote:
    I do not think we should contemplate an MH attack any time in the foreseeable future.  It's too slow and can't accomplish anything.  An EMH attack is more within the realm of possibility, but that doesn't combine well with a rabbit pull, i.e. once we pull the h-rabbit, he may have an easier time defending against an EMH attack, so we should choose only the one that looks better (the rabbit pull IMHO).


    Not that we have to perform the MH attack.  ;) I think more important is we could easily perform it if the golden elephant was too busy elsewhere. If both elephants are to far to count, MH attack when the defender has a camel is as strong as EH one when the defender has an elephant. This limits possible Chessandgo's strategies.

    Title: Re: Move 12
    Post by Fritzlein on Aug 25th, 2007, 6:23pm

    on 08/25/07 at 17:16:49, UruramTururam wrote:
    If both elephants are to far to count, MH attack when the defender has a camel is as strong as EH one when the defender has an elephant. This limits possible Chessandgo's strategies.

    I should analyze some of the lines people have proposed, but I am first drawn to respond to this comment.  There are some similarities about an EH attack when an E defends and an MH attack when an M defends.  For example, neither attack can hope to capture a piece in the attacked trap, because the strongest defender is the same strength as the strongest attacker.  Also both attacks have a danger of the horse (the weaker attacker) being taken hostage by the strong defender.

    But the differences are critical.  If the EH attack gives up the horse hostage to the defending elephant, it can make the defending elephant less mobile.  The reduced mobility of the defending elephant creates an increased mobility for the attacker's camel to operate elsewhere.  The shape of the EH vs. Exxxx fight can make one camel stronger than the other on the rest of the board, because that camel is less in danger of an elephant switching quadrants.

    In contrast, if the defending M takes the attacking H hostage in an MH attack, this has no effect whatsoever on the mobility of the two elephants.  The elephants weren't afraid of the camels in the first place.  What's more, a camel in the opening has no higher aspiration than holding a horse hostage, so it wouldn't want to cross sides, so it doesn't matter if it has lost mobility.

    If the defending M takes the attacking H hostage, it forces the attacking M to stay exposed forward to prevent the capture of the horse.  If the two elephants arrive two moves later (or whenever) the erstwhile MH attacker will have the worst of it because of the exposed M and the hostage H.

    The only way an MH attack makes sense against a trap defended by M is if our elephant is going to arrive before our H gets taken hostage, i.e. only if it is really an EMH attack.

    Title: Re: Move 12
    Post by mdk on Aug 25th, 2007, 6:41pm

    on 08/25/07 at 17:16:49, UruramTururam wrote:
    Not that we have to perform the MH attack.  ;) I think more important is we could easily perform it if the golden elephant was too busy elsewhere. If both elephants are to far to count, MH attack when the defender has a camel is as strong as EH one when the defender has an elephant. This limits possible Chessandgo's strategies.


    The problem with MH attacks is that they readily allow for camel hostages. if we advance our camel then chessandgo is certainly going to attempt to take it hostage and unless you have some way to tie down the gold phant or blockade it (at least partially) MH attacks simply aren't particularly viable. An EMH attack, one of chessandgo's favorite is much more viable but they are also rather commital and based on the voting of the mob up to this point i doubt such an option would be voted for.

    Just an overall note to the mob, all the strategic things that have been discussed in the first 11 moves are nice but i really feel that we need to spend greater time looking at immediate tactics. I feel like the possibilty of chessandgo pulling our rabbit was basically completely ignored in our discussion for the last move. I am equally at fault with everyone else. I spent the vast majority of my forum posts arguing for my own move and only alluded to this possibility which i was concerned about rather than clearly pointing this out although. I hope we have all learned something about the importance of everyone contributing to the discussion as much as possible.

    That being said I do not feel that our position is a bad one. This move was second on my ballot. I'd say we are about even or slightly behind but behind no more than the advantage i believe gold has by moving first. Perhaps further analysis by the mob will convince me that we are better off than i currently feel we are.

    [edit] i guess much of what i said at the beginning was posted by fritz except more precisely while i was typing all of this up

    Title: Re: Move 12
    Post by RonWeasley on Aug 25th, 2007, 6:42pm
    In light of all this discussion, don't expect a vote Monday morning.  Maybe as late as Wednesday morning.

    Title: Re: Move 12
    Post by mdk on Aug 25th, 2007, 8:25pm

    on 08/24/07 at 08:37:34, Fritzlein wrote:
    Oops, I didn't see the drag.  I was assuming he would flip.  As usual the position isn't as easy as it first appears.

    If we pull once and occupy c4 with our elephant, we need to be more aware of chessandgo pulling once and centralizing his elephant to go after our camel, or directly going after our camel, since we've lost a step decentralizing our elephant.

    Now that our double-pull looks inferior, I will venture to guess that we are slightly losing, although we need more lines to be confident of anything in a race-like position.


    I feel like we have completely ignored the option of pulling the rabbit twice after Fritz said that he feels the double pull looks inferior. We were worried about the following line where chessandgo drags the rabbit rather than flipping it.

    12b hh3n Rh2n hh4n Rh3n
    13w Eb5s rb6s Eb4e rb5s

    Note however that we can continue:

    13b  ha6s ha5s ha4s ed4s !!!

    Now we are threatening a second rabbit pull and chessandgo's camel has no real hope of taking a horse hostage. After this 13b I don't feel that the immediate double pull looks inferior. Instead I would say that chessandgo pulling the rabbit twice is inferior!

    I don't know how other people feel but I now feel that pulling the rabbit twice is clearly best and I would be happy to vote early in order to add to our reserve.

    Title: Re: Move 12
    Post by Fritzlein on Aug 25th, 2007, 8:44pm

    on 08/25/07 at 18:41:47, mdk wrote:
    Just an overall note to the mob, all the strategic things that have been discussed in the first 11 moves are nice but i really feel that we need to spend greater time looking at immediate tactics.

    You are right, we tipped too much into fuzzy impressions.  I plead guilty to just reacting on first impressions rather than playing out lines.  We need a good mix of concrete variations as well as strategy.  In fact, the tactics should be our advantage, because we will be at best equal with chessandgo on strategy, but with more eyeballs on particular lines we should be better able to avoid not only blunders but also deep tactical traps.

    Fortunately there have been a lot of good concrete suggestions so far this move.  I have put them all into a tree for easy review.  I have added some variations and comments.  Right now I think that UT's move leaves us a little behind, so I've added a slight variation for discussion: dd8s instead of md6e.

    Mdk's continuation for the double-pull can get super-sharp in a way that seems to favor Gold, so I don't like it yet, but definitely we need more analysis.  This is the critical line so far.  I'm curious to see mdk's refutation for my continuation for Gold.

    I've included all posted variations of jdb's move and others, but haven't had time to review them yet.

    The analysis tree:
    12b md6e ed4w ec4n Hc3n (block Gold elephant from center)
    .    13w Eb5s rb6s Hc4s Eb4e (keep elephant mobility)
    .    .    13b hh3n Rh2n rb5w ha6e (protect rabbit instead f racing?)
    .    .    .    14w Db3w Da3n Da4s ra5s (now a straight race ends tied)
    .    .    .    14w Ec4e Ed4e Ee4e Ef4e (uh oh)
    .    .    13b hh3n Rh2n hh4n Rh3n (race instead of protecting rabbit?)
    .    .    .    14w Db3n Db4s rb5s Db3e (now Gold wins a straight race!)
    .    .    .    .    14b ha6s ha5s ha4n Da3n (complicating?)
    .    .    .    .    .    15w rb4s Ec4w Da4s Hc3e (Gold is OK?)
    .    13w Eb5s rb6s Hc4s Rh2w (safety h2-rabbit)
    .    .    13b hh3s hh2n Rh1n ec5s (still keep elephant from center)
    .    .    .    14w Eb4w rb5s Ea4n Ea5e (regain mobility)
    .    .    .    .    14b hh3n Rh2n rb4w Ha6e (seems about even?)
    .    13w Eb5s rb6s Hc4s Hc3e (safety c3-horse)
    .    .    13b hh3n Rh2n ec5s ha6e (still keep elephant from center)
    .    .    .    14w Eb4w rb5s Db3w rb4s (still threaten rabbit)
    .    .    .    .   14b hb6s hb5s rb3s hb4s (exploiting b3 hole)
    .    .    .    .    .    15w Ea4e hb3e Eb4s De2n (wrong side frame)
    .    .    .    .    .    .    15b me6s me5s de3s me4s (crazy but Silver's OK)  
    12b hh3n Rh2n hh4n Rh3n (simple: just race)
    .    13w Eb5s rb6s Eb4e rb5s (blocking off c4)
    .    .    13b ha6s ha5s ha4s ed4s (each Silver horse pulls a rabbit)
    .    .    .    14w Eb5e Ec5e md6e Ed5n (counter-attack)
    .    .    .    .    14b Db3s ha3e rb8s rb7s (very sharp but favors Gold?)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    .    .    13b hh5n Rh4n ed4w rb4w (we win race unless Gold complicates)
    .    .    .    14w Eb5e Ec5e md6e Ed5n (complicating?)
    .    .    .    .    14b ha6e me6e Rh5w hh6n (he's in more trouble than us)
    12b hh3n Rh2n ed4w rb6e
    .    13w Eb5n Eb6n Eb7s rb8s
    .    .    13b ec4e Hc3n Hc4n ed4w
    .    13w Eb5e Ec5w rc6s Hc3e
    .    .    13b hh4w Rh3n Rh4n hg4e
    .    13w Eb5e Ec5e rc6s Ed5e
    .    .    13b hh4n Rh3n hh5w hg5n
    .    13w Eb5e Ec5e rc6s ra2e
    .    .    13b ha6s ec4w eb4e Db3n
    .    .    13b hh4n Rh3n hh5n Rh4n
    .    .    .    14w Hg3n Hg4n Hg5n De2
    12b ed4n ed5w rb6e ha6e
    12b ed4w hh3n Rh2n dd8s (occupy c4, advance dog)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    12b ed4w hh3n Rh2n ha6n (occupy c4, safety horse)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    12b dd8s ed4w ec4n Hc3n (block Gold elephant from center)


    Title: Re: Move 12
    Post by 99of9 on Aug 26th, 2007, 2:47am
    The tree is great - thanks Fritz.  If anyone is keen, they could put it on the wiki, so we can all shoulder the responsibility of keeping it up to date.

    Title: Re: Move 12
    Post by UruramTururam on Aug 26th, 2007, 8:03am

    on 08/25/07 at 18:23:40, Fritzlein wrote:
    The only way an MH attack makes sense against a trap defended by M is if our elephant is going to arrive before our H gets taken hostage, i.e. only if it is really an EMH attack.


    Yes. Thus it's better to keep our elephant in the center while trying to keep C&G's one out of it. As for now the situation looks OK. Trying to keep golden elephant right-wing-sealed leaves us much freedom on the left (including EMH if necessary).

    Title: Re: Move 12
    Post by jdb on Aug 26th, 2007, 8:52am
    Nice tree fritz. Anything I added is in bold

    The analysis tree:
    12b md6e ed4w ec4n Hc3n (block Gold elephant from center)
    .    13w Eb5s rb6s Hc4s Eb4e (keep elephant mobility)
    .    .    13b hh3n Rh2n rb5w ha6e (protect rabbit instead f racing?)
    .    .    .    14w Db3w Da3n Da4s ra5s (now a straight race ends tied)
    .    .    .    14w Ec4e Ed4e Ee4e Ef4e (uh oh)
    .    .    13b hh3n Rh2n hh4n Rh3n (race instead of protecting rabbit?)
    .    .    .    14w Db3n Db4s rb5s Db3e (now Gold wins a straight race!)
    .    .    .    .    14b ha6s ha5s ha4n Da3n (complicating?)
    .    .    .    .    .    15w rb4s Ec4w Da4s Hc3e (Gold is OK?)
    .    13w Eb5s rb6s Hc4s Rh2w (safety h2-rabbit)
    .    .    13b hh3s hh2n Rh1n ec5s (still keep elephant from center)
    .    .    .    14w Eb4w rb5s Ea4n Ea5e (regain mobility)
    .    .    .    .    14b hh3n Rh2n rb4w Ha6e (seems about even?)
    .    13w Eb5s rb6s Hc4s Hc3e (safety c3-horse)
    .    .    13b hh3n Rh2n ec5s ha6e (still keep elephant from center)
    .    .    .    14w Eb4w Ea5n rb5s Ea6e
    .    .    .    14w Eb4w rb5s Db3w rb4s (still threaten rabbit)
    .    .    .    .   14b hb6s hb5s rb3s hb4s (exploiting b3 hole)
    .    .    .    .    .    15w Ea4e hb3e Eb4s De2n (wrong side frame)
    .    .    .    .    .    .    15b me6s me5s de3s me4s (crazy but Silver's OK)  
    12b hh3n Rh2n hh4n Rh3n (simple: just race)
    .    13w Eb5s rb6s Eb4e rb5s (blocking off c4)
    .    .    13b ha6s ha5s ha4s ed4s (each Silver horse pulls a rabbit)
    .    .    .    14w Eb5e Ec5e md6e Ed5n (counter-attack)
    .    .    .    .    14b Db3s ha3e rb8s rb7s (very sharp but favors Gold?)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    .    .    13b hh5n Rh4n ed4w rb4w (we win race unless Gold complicates)
    .    .    .    14w Eb5e Ec5e md6e Ed5n (complicating?)
    .    .    .    .    14b ha6e me6e Rh5w hh6n (he's in more trouble than us)
    12b hh3n Rh2n ed4w rb6e
    .    13w Eb5n Eb6n Eb7s rb8s
    .    .    13b ec4e Hc3n Hc4n ed4w
    .    13w Eb5e Ec5w rc6s Hc3e
    .    .    13b hh4w Rh3n Rh4n hg4e
    .    13w Eb5e Ec5e rc6s Ed5e
    .    .    13b hh4n Rh3n hh5w hg5n
    .    13w Eb5e Ec5e rc6s ra2e
    .    .    13b ha6s ec4w eb4e Db3n
    .    .    13b hh4n Rh3n hh5n Rh4n
    .    .    .    14w Hg3n Hg4n Hg5n De2
    12b ed4n ed5w rb6e ha6e
    12b ed4w hh3n Rh2n dd8s (occupy c4, advance dog)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    12b ed4w hh3n Rh2n ha6n (occupy c4, safety horse)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    12b dd8s ed4w ec4n Hc3n (block Gold elephant from center)


    Title: Re: Move 12
    Post by mdk on Aug 26th, 2007, 3:14pm
    12b hh3n Rh2n hh4n Rh3n (simple: just race)
    .    13w Eb5s rb6s Eb4e rb5s (blocking off c4)
    .    .    13b ha6s ha5s ha4s ed4s (each Silver horse pulls a rabbit)
    .    .    .    14w Eb5e Ec5e md6e Ed5n (counter-attack)

    I was thinking 14b cc7w cb7s md6e dd8s which leaves the position extremely sharp.

    However another possibility is 13b ha6s ha5s md6w ed4s.

    Title: Re: Move 12
    Post by Fritzlein on Aug 26th, 2007, 4:14pm
    99of9, it would be great to have a communally edited analysis tree on the Wiki.  For now, though, we can cut and paste all or part of it in different posts as needed.

    I have updated the tree with possible replies to mdk's and jdb's suggestions.  Whenever the tree branches, I have put the better move first, according to my guess as to which move is better.  My current assessment is that UT's move leaves us about even, while both the double pull and jdb's move leave us slightly behind with best play.  That leaves the bottom four moves which have gotten little attention so far.  One of them (or something new!) may be better than the three heavily-analyzed moves.  I will analyze more as I have time.

    The analysis tree:
    12b md6e ed4w ec4n Hc3n (block Gold elephant from center)
    .    13w Eb5s rb6s Hc4s Eb4e (keep elephant mobility)
    .    .    13b hh3n Rh2n rb5w ha6e (protect rabbit instead of racing?)
    .    .    .    14w Db3w Da3n Da4s ra5s (straight race ends tied)
    .    .    .    .    14b hh4n Rh3n hh5n Rh4n (rabbit trade =)
    .    .    .    14w Ec4e Ed4e Ee4e Ef4e (freeze horse)
    .    .    .    .    14b ec5s ec4w hc3n hb6s (race better than defense)
    .    .    .    .    .    15w Eg4w hh4w Ef4n hg4w (horse trade =)
    .    .    13b hh3n Rh2n hh4n Rh3n (race instead of protecting rabbit?)
    .    .    .    14w Db3n Db4s rb5s Db3e (now Gold wins a straight race!)
    .    .    .    .    14b ha6s ha5s ha4n Da3n (complicating?)
    .    .    .    .    .    15w rb4s Ec4w Da4s Hc3e (Gold is OK?)
    .    13w Eb5s rb6s Hc4s Rh2w (safety h2-rabbit)
    .    .    13b hh3s hh2n Rh1n ec5s (still keep elephant from center)
    .    .    .    14w Eb4w rb5s Ea4n Ea5e (regain mobility)
    .    .    .    .    14b hh3n Rh2n rb4w Ha6e (seems about even?)
    .    13w Eb5s rb6s Hc4s Hc3e (safety c3-horse)
    .    .    13b hh3n Rh2n ec5s ha6e (still keep elephant from center)
    .    .    .    14w Eb4w Ea4n rb5s Ea5e (regain mobility)
    .    .    .    .    14b rb4w hh4n Rh3n dd8s (Silver slightly ahead?)
    .    .    .    14w Eb4w rb5s Db3w rb4s (still threaten rabbit)
    .    .    .    .   14b hb6s hb5s rb3s hb4s (exploiting b3 hole)
    .    .    .    .    .    15w Ea4e hb3e Eb4s De2n (wrong side frame)
    .    .    .    .    .    .    15b me6s me5s de3s me4s (crazy but Silver's OK)  
    12b hh3n Rh2n hh4n Rh3n (simple: just race)
    .    13w Eb5s rb6s Eb4e rb5s (blocking off c4)
    .    .    13b ha6s ha5s md6w ed4s
    .    .    .    14w Db3s Hc3w Hb3w rb4s (Gold slightly ahead in race)
    .    .    13b ha6s ha5s ha4s ed4s (each Silver horse pulls a rabbit)
    .    .    .    14w Eb5e Ec5e md6e Ed5n (counter-attack)
    .    .    .    .    14b cc7w cb7s md6e dd8s
    .    .    .    .    .    15w Hc3n Hc4n Cc2w Md2w (Gold's attack is better)
    .    .    .    .    14b Db3s ha3e rb8s rb7s (very sharp but favors Gold?)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    .    .    13b hh5n Rh4n ed4w rb4w (we win race unless Gold complicates)
    .    .    .    14w Eb5e Ec5e md6e Ed5n (complicating?)
    .    .    .    .    14b ha6e me6e Rh5w hh6n (he's in more trouble than us)  
    12b hh3n Rh2n ed4w rb6e (rabbit into trap)
    .    13w Eb5e Ec5e rc6s ra2e (pull rabbit + tuck rabbit)
    .    .    13b ha6s ec4w eb4e Db3n (pull dog)
    .    .    .    14w Ed5e md6s md5s Ee5w (flip camel)
    .    .    .    .    14b md4e me4e mf4n mf5n (run away)
    .    .    .    .    .    15w Ed5e Ee5e Ef5e mf6s
    .    .    .    .    .    .    15b ce7s ce6s mf5n ce5n
    .    .    .    .    .    .    .    16w Eg5s Eg4n hh4w Hf3w (Gold +=)  
    .    .    13b hh4n Rh3n hh5n Rh4n (pull rabbit)
    .    .    .    14w Hg3n Hg4n Hg5n De2n (attack f6, Gold initiative)
    .    13w Eb5e Ec5w rc6s Hc3e (pull rabbit + safety horse)
    .    .    13b hh4w Rh3n Rh4n hg4e (flip rabbit)
    .    .    .    14w Ed5e md6s md5s Ee5w (flip camel)
    .    .    .    .    14b md4e me4e mf4n mf5n (run away)
    .    .    .    .    .    15w Ed5e Ee5e Ef5e mf6s (Gold has initiative)
    .    13w Eb5e Ec5e rc6s Ed5e (pull rabbit + threaten f6)
    .    .    13b hh4n Rh3n hh5w hg5n (pull rabbit and block horse)
    .    13w Eb5n Eb6n Eb7s rb8s (pull second rabbit)
    .    .    13b ec4e Hc3n Hc4n ed4w (horse flip, Silver winning)
    12b ed4n ed5w rb6e ha6e
    12b ed4w hh3n Rh2n dd8s (occupy c4, advance dog)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    12b ed4w hh3n Rh2n ha6n (occupy c4, safety horse)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    12b dd8s ed4w ec4n Hc3n (block Gold elephant from center)

    Title: Re: Move 12
    Post by jdb on Aug 26th, 2007, 6:02pm
    .    13w Eb5s rb6s Hc4s Hc3e (safety c3-horse)
    .    .    13b hh3n Rh2n ec5s ha6e (still keep elephant from center)
    .    .    .    14w Eb4w Ea4n rb5s Ea5e (regain mobility)
    .    .    .    .    14b rb4w hh4n Rh3n dd8s (Silver slightly ahead?)

    15w Hd3w Db3s Hc3w Eb5e

    Looks about equal.  Gold's horses are back on the proper squares and the rabbit dragging is close too. If anything silver's camel is a bit of a target, but nothing serious.

    Title: Re: Move 12
    Post by mdk on Aug 26th, 2007, 6:02pm

    on 08/26/07 at 16:14:51, Fritzlein wrote:
    99of9,
    [font=courier]The analysis tree:
    [edit]  
    12b hh3n Rh2n hh4n Rh3n (simple: just race)
    .    13w Eb5s rb6s Eb4e rb5s (blocking off c4)
    .    .    13b ha6s ha5s md6w ed4s
    .    .    .    14w Db3s Hc3w Hb3w rb4s (Gold slightly ahead in race)


    14b hh5n Rh4n hh6w Rh5n

    in this case arent we the ones to capture first? the phant cant save the rabbit because then we can capture at the c3 trap.

    so

    12b hh3n Rh2n hh4n Rh3n (simple: just race)
    .    13w Eb5s rb6s Eb4e rb5s (blocking off c4)
    .    .    13b ha6s ha5s md6w ed4s
    .    .    .    14w Db3s Hc3w Hb3w rb4s (Gold NOT slightly ahead in race)
    .    .    .    .    14b hh5n Rh4n hh6w Rh5n
    .    .    .    .    .    15w Ec4n Ec5e Ed5e Ee5n (Ee5e)
    .    .    .    .    .    .    15b mc6w ed3w ec3e Cc2n Cc3x (clearly bad for gold)
    .    .    .    .    .    15w Hg3n Hg4n  Hg5w xxxx
    .    .    .    .    .    .    15b mc6e md6e me6s rg7e (we are clearly better)
    .    .    .    .    .    15w Hg3n Hg4n Hg5e Rh6n
    .    .    .    .    .    .    15b hg6e hh6w Rh7s rg7e (and we will again capture the rabbit first)

    i hope this bit of analysis changes your evaluation of the position fritz. or is there some other response you were thinking of for 15w? if any of you guys have responses i didnt mention that you want me to respond to i would be happy to do so.

    Title: Re: Move 12
    Post by jdb on Aug 26th, 2007, 6:20pm
    12b hh3n Rh2n ed4w rb6e (rabbit into trap)
    .    13w Eb5e Ec5e rc6s ra2e (pull rabbit + tuck rabbit)
    .    .    13b ha6s ec4w eb4e Db3n (pull dog)
    .    .    .    14w Ed5e md6s md5s Ee5w (flip camel) This looks strong
    .    .    .    .    14b md4e me4e mf4n mf5n (run away)
    .    .    .    .    .    15w Ed5e Ee5e Ef5e mf6s
    .    .    .    .    .    .    15b ce7s ce6s mf5n ce5n
    .    .    .    .    .    .    .    16w Eg5s Eg4n hh4w Hf3w (Gold +=)  
    .    .    13b hh4n Rh3n hh5n Rh4n (pull rabbit)
    .    .    .    14w Hg3n Hg4n Hg5n De2n (attack f6, Gold initiative)
    .    13w Eb5e Ec5w rc6s Hc3e (pull rabbit + safety horse)
    .    .    13b hh4w Rh3n Rh4n hg4e (flip rabbit)
    .    .    .    14w Ed5e md6s md5s Ee5w (flip camel)    This move is illegal!
    .    .    .    .    14b md4e me4e mf4n mf5n (run away)
    .    .    .    .    .    15w Ed5e Ee5e Ef5e mf6s (Gold has initiative)
    .    13w Eb5e Ec5e rc6s Ed5e (pull rabbit + threaten f6)
    .    .    13b hh4n Rh3n hh5w hg5n (pull rabbit and block horse)
    .    13w Eb5n Eb6n Eb7s rb8s (pull second rabbit)
    .    .    13b ec4e Hc3n Hc4n ed4w (horse flip, Silver winning)

    Title: Re: Move 12
    Post by jdb on Aug 26th, 2007, 7:30pm
    Analysis by mdk:


    12b hh3n Rh2n hh4n Rh3n (simple: just race)
    .    13w Eb5s rb6s Eb4e rb5s (blocking off c4)
    .    .    13b ha6s ha5s md6w ed4s

    .    .    .      14w Db3s rb4s rb3w Hc3w
    .    .    .    .    .   14b hh5n Rh4n hh6w Rh5n

    Can gold do something with this? Gold has alot fewer weaknesses now, but is there something concrete he can do to save his rabbit?


    .    .    .    14w Db3s Hc3w Hb3w rb4s (Gold NOT slightly ahead in race)
    .    .    .    .    14b hh5n Rh4n hh6w Rh5n
    .    .    .    .    .    15w Ec4n Ec5e Ed5e Ee5n (Ee5e)
    .    .    .    .    .    .    15b mc6w ed3w ec3e Cc2n Cc3x (clearly bad for gold)
    .    .    .    .    .    15w Hg3n Hg4n  Hg5w xxxx
    .    .    .    .    .    .    15b mc6e md6e me6s rg7e (we are clearly better)
    .    .    .    .    .    15w Hg3n Hg4n Hg5e Rh6n
    .    .    .    .    .    .    15b hg6e hh6w Rh7s rg7e (and we will again capture the rabbit first)

    Title: Re: Move 12
    Post by Fritzlein on Aug 26th, 2007, 8:26pm

    on 08/26/07 at 18:02:36, mdk wrote:
    12b hh3n Rh2n hh4n Rh3n (simple: just race)
    .    13w Eb5s rb6s Eb4e rb5s (blocking off c4)
    .    .    13b ha6s ha5s md6w ed4s
    .    .    .    14w Db3s Hc3w Hb3w rb4s (Gold NOT slightly ahead in race)
    .    .    .    .    14b hh5n Rh4n hh6w Rh5n

    i hope this bit of analysis changes your evaluation of the position fritz. or is there some other response you were thinking of for 15w?

    Yes, I was thinking of a different 15w, namely 15w rb3e Ha3e Ec4n Ec5e.  Silver can't capture the rabbit on 15b because the Silver camel is hanging.  That will give Gold time to save the h6 rabbit.

    The evaluation of the resulting position is complex.  Who is ahead if each side has a threatened and/or framed rabbit?  To my mind it seemed that Gold would be ahead because the silver camel must head west, after which Gold can defend f6 with a horse, and have a free elephant.  I'm not sure how the silver camel can linger near enough to scare away the gold horse without itself becoming a target of the Gold elephant.

    However, the position remains loose and double-edged.  It is quite possible you have a refutation for my 15w, and I'm eager to hash out any lines.  Perhaps instead of saying "Gold is slightly ahead in the race" I should say, "Gold's position appears more flexible to me, because the silver rabbit is already framed, tying down the silver elephant while the gold elephant is still mobile.  The tactics remain complex, but my expectation is that the variations will tend to favor Gold."

    I hope it doesn't seem that the analysis tree is an attempt to lay down the law and stifle dissent with off-hand comments like "Gold is ahead".  Your challenge of the evaluation and presentation of possible lines is precisely the way the discussion should proceed.

    That said, you haven't changed my mind (yet).  The Gold position still looks better to me in this line, and I would still rather play UruramTururam's move, or perhaps one of the less heavily analyzed moves like 99of9's suggestion.

    Title: Re: Move 12
    Post by Fritzlein on Aug 26th, 2007, 8:38pm

    on 08/26/07 at 18:20:35, jdb wrote:
    12b hh3n Rh2n ed4w rb6e (rabbit into trap)
    .    13w Eb5e Ec5w rc6s Hc3e (pull rabbit + safety horse)
    .    .    13b hh4w Rh3n Rh4n hg4e (flip rabbit)
    .    .    .    14w Ed5e md6s md5s Ee5w (flip camel)    This move is illegal!

    Whoops, sorry about that.  I didn't notice the subtle change of direction there.  To my mind it seems mandatory for Gold to respond to 12b hh3n Rh2n ed4w rb6e by pulling the rabbit to c5 and leaving the elephant on d5, threatening to flip the camel.  Therefore I didn't read carefully and thought the variation was merely in the last step.

    But actually, if the threat to flip the camel is strong, then we don't need to figure out exactly which fourth step is best for Gold.  We can chalk up the line as slightly bad for us, spend more time on lines that seem OK for us (if there are any) and revive this line only if everything else seems bad too, or if we can find a way to do better against the current best for Gold.

    Title: Re: Move 12
    Post by mdk on Aug 26th, 2007, 9:59pm

    on 08/26/07 at 20:26:11, Fritzlein wrote:
    Yes, I was thinking of a different 15w, namely 15w rb3e Ha3e Ec4n Ec5e.  Silver can't capture the rabbit on 15b because the Silver camel is hanging.  That will give Gold time to save the h6 rabbit.

    The evaluation of the resulting position is complex.  Who is ahead if each side has a threatened and/or framed rabbit?  To my mind it seemed that Gold would be ahead because the silver camel must head west, after which Gold can defend f6 with a horse, and have a free elephant.  I'm not sure how the silver camel can linger near enough to scare away the gold horse without itself becoming a target of the Gold elephant.

    However, the position remains loose and double-edged.  It is quite possible you have a refutation for my 15w, and I'm eager to hash out any lines.  Perhaps instead of saying "Gold is slightly ahead in the race" I should say, "Gold's position appears more flexible to me, because the silver rabbit is already framed, tying down the silver elephant while the gold elephant is still mobile.  The tactics remain complex, but my expectation is that the variations will tend to favor Gold."

    I hope it doesn't seem that the analysis tree is an attempt to lay down the law and stifle dissent with off-hand comments like "Gold is ahead".  Your challenge of the evaluation and presentation of possible lines is precisely the way the discussion should proceed.

    That said, you haven't changed my mind (yet).  The Gold position still looks better to me in this line, and I would still rather play UruramTururam's move, or perhaps one of the less heavily analyzed moves like 99of9's suggestion.


    15b ha4s mc6w dd8s rg7e

    yes our phant may be pinned down but we will gain rabbit pulls on both wings. If chessandgo's phant crosses the we can use our camel to pull the horse on b3 and move our own horse there while pulling the a2 rabbit, thereby freeing our phant. If chessandgo's phant doesnt cross then we will end up with a gold rabbit framed on our f6 trap protected by gold's horse. we will then gain a second pulled rabbit or partial trap control of the c3 trap or both.

    personally i prefer our position at this point although i may be wandering back to the realm of murky strategy. yes things may be less flexible for us but i dont really see how gold can make any improvements to his position. perhaps there are some lines some of you would like to play out?

    [edit] also dont forget that chessandgo has to be careful with regards to where he places his phant because we always have the ability to give up the rabbit in order to attack his horse once it moves to protect the rabbit. this limits where on the western part of the board chessandgo can place his phant which i believe will prevent him from being able to take advantage of any additional "flexibility" he may appear to have.

    Title: Re: Move 12
    Post by Fritzlein on Aug 26th, 2007, 11:26pm

    on 08/26/07 at 21:59:30, mdk wrote:
    15b ha4s mc6w dd8s rg7e
    [...]
    also dont forget that chessandgo has to be careful with regards to where he places his phant because we always have the ability to give up the rabbit in order to attack his horse once it moves to protect the rabbit. this limits where on the western part of the board chessandgo can place his phant which i believe will prevent him from being able to take advantage of any additional "flexibility" he may appear to have.

    I have played a few lines, and it seems you are right.  The Gold elephant has to stay in the west to keep watch on the camel, but also has to stay in the center to make the Gold horse less vulnerable.  This makes the extra flexibility  that I thought Gold had rather illusory.  Your "murky strategy" has convinced me.  ;-)

    But before I embrace your line, let me back up a turn and wonder whether exposing the camel on 13b might have been a problem:

    12b hh3n Rh2n hh4n Rh3n (simple: just race)
    .    13w Eb5s rb6s Eb4e rb5s (blocking off c4)
    .    .    13b ha6s ha5s md6w ed4s
    .    .    .    14w Ec4n Ec5w mc4s xxxx

    It seems that moving the camel west on 13b to indirectly menace c3 is a good idea.  What is conceived of as a rescue operation for the exposed rabbit can turn into a full-blown EMHR attack in some case.  But what if Gold anticipates and resists this?

    Title: Re: Move 12
    Post by mdk on Aug 27th, 2007, 12:20am

    on 08/26/07 at 23:26:53, Fritzlein wrote:
    I have played a few lines, and it seems you are right.  The Gold elephant has to stay in the west to keep watch on the camel, but also has to stay in the center to make the Gold horse less vulnerable.  This makes the extra flexibility  that I thought Gold had rather illusory.  Your "murky strategy" has convinced me.  ;-)

    But before I embrace your line, let me back up a turn and wonder whether exposing the camel on 13b might have been a problem:

    12b hh3n Rh2n hh4n Rh3n (simple: just race)
    .    13w Eb5s rb6s Eb4e rb5s (blocking off c4)
    .    .    13b ha6s ha5s md6w ed4s
    .    .    .    14w Ec4n Ec5w mc4s xxxx

    It seems that moving the camel west on 13b to indirectly menace c3 is a good idea.  What is conceived of as a rescue operation for the exposed rabbit can turn into a full-blown EMHR attack in some case.  But what if Gold anticipates and resists this?


    Well I'm glad to hear that all the time I have spent analyzing the position hasn't been in vain. I have a couple of possible ideas that I will throw out there for those of you in different time zones although I haven't fully analyzed them. I don't know how much longer I will be up as it is currently 1am here.

    One possibility is 14b rb4e ha4s Db3n ha3e. This looks pretty strong to me although it is possible there are some tactics I am missing.

    Another possibility is 14b rb4e mc5n mc6w ha4s. I actually saw this possibility first but I'm not sure how much I like it. It creates a position similar to that in the other line Fritz and I have been discussing.

    Of course these options are really based on Fritz's 3 step move because I'm really not sure how/if chessandgo would use the 4th step.

    I'm looking foward to hearing what Fritz and hopefully some other mobsters think.

    Title: Re: Move 12
    Post by Fritzlein on Aug 27th, 2007, 2:12am
    Since both of your response involve advancing the horse, consider the move to be 14w Ec4n Ec5w mc4s Rc1e, which makes it easier for the camel to cross behind the trap after a later Cc2s.

    Title: Re: Move 12
    Post by mdk on Aug 27th, 2007, 7:42pm

    on 08/27/07 at 02:12:35, Fritzlein wrote:
    Since both of your response involve advancing the horse, consider the move to be 14w Ec4n Ec5w mc4s Rc1e, which makes it easier for the camel to cross behind the trap after a later Cc2s.


    In this case the best 14b that I see is ed3n ed4w mc5n mc6w.  I still prefer our position at this point. Perhaps I am missing some better response for 14b?

    Or perhaps on 13b we should respond edd4s md6w mc6w ha6s. This seems to have most of the benefits of 13b ha6s ha5s md6w ed4s and avoids our camel being threatened.

    Title: Re: Move 12
    Post by Fritzlein on Aug 27th, 2007, 9:54pm
    I looked a bit at 99of9's suggestion just now,  12b ed4w hh3n Rh2n dd8s.  I thought the obvious Gold response was flipping the rabbit, but it seems to me that after

    12b ed4w hh3n Rh2n dd8s
    13w Eb5w rb6s rb5s Ea5e
    13b rb4w hh4n Rh3n ha6e

    we are quite simply a little bit ahead.  Does Gold have a good continuation, or a better 13w?

    It seems that it would be good for the team to have at least looked at all the suggestions a little bit, so we're not in the position of possibly selecting a move for which not even one line has been posted.  (like last move ;-))  If no one else comments on the last three ballot options, I'll try to give at least brief impressions before the voting begins.

    Title: Re: Move 12
    Post by jdb on Aug 28th, 2007, 8:09am

    on 08/27/07 at 21:54:22, Fritzlein wrote:
    I looked a bit at 99of9's suggestion just now,  12b ed4w hh3n Rh2n dd8s.  I thought the obvious Gold response was flipping the rabbit, but it seems to me that after

    12b ed4w hh3n Rh2n dd8s
    13w Eb5w rb6s rb5s Ea5e
    13b rb4w hh4n Rh3n ha6e

    we are quite simply a little bit ahead.  Does Gold have a good continuation, or a better 13w?


    12b ed4w hh3n Rh2n dd8s
    13w Eb5e rb6s Db3s Hc3w
    13b ha6e rb5w ha4n Ra3n

    To me, this looks playable for either side

    Title: Re: Move 12
    Post by Fritzlein on Aug 28th, 2007, 9:04am
    Yeah, JDB, the more I play through lines suggested by mdk, the more I realize that it is truly a nuisance for chessandgo to have a dog on b3 instead of a horse, and the more it seems worthwhile for chessandgo to spend two steps fixing that.  Still, if chessandgo takes that time, doesn't it mean we are a bit ahead in the race?  I am starting to think maybe our position is better than I first thought, which is to say that maybe we are slightly ahead after all, and the horse charge last move turns out to have been a good idea.

    Your 13w is perhaps worth considering against the double pull as well, although in the case of the double pull we are one square closer to having his rabbit in the bag than in this line.  I guess that means if fixing the misplaced horse is the best chessandgo can do against the double-pull then (A)  he's flat out admitting we are winning and (B) the double pull is the best line for us.  That in turn would mean the critical line to understand is still our double-pull followed by his double-pull.  I'll try to get back to that after a quick look at the move suggestions with no lines so far.

    Title: Re: Move 12
    Post by RonWeasley on Aug 28th, 2007, 11:23am
    My schedule is a bit odd this week.  Sorry I haven't contributed to this, but it looks like the best week of analysis so far.

    I still want to start the vote tomorrow at about this time tomorrow, as analysis seems to be stabilizing, but I would be able to extend the voting window until later Thursday evening.  Please let me know if TheMob needs more time.

    Title: Re: Move 12
    Post by jdb on Aug 28th, 2007, 12:16pm
    I think in any of these variations, our dream position is to capture gold's rabbit and end up protecting our own rabbit for as long as possible. Eventually we will likely have to abandon it, but by then we hopefully will have been able to drag another rabbit.

    It looks like there are ways to end up with a decent position without getting into complicated tactics that could lead to unexpected things.

    Title: Re: Move 12
    Post by Fritzlein on Aug 28th, 2007, 2:16pm
    Nice summary JDB.  If we can capture a rabbit in f6 without immediately having to give up ours in c3, we are ahead.  However, it seems that even in Gold's best lines, we can return the c3 rabbit for an equal position and no risk.

    I have updated the analysis tree with everyone's suggestions, plus a few new lines, plus at least a little analysis of each of the moves.  I have ordered the moves and branches approximately best to worst, according to minimax on the tree and my guesstimate of the nodes.  I think we are a tiny bit ahead ahead with either the double pull or 99of9's move, and I'm not sure which I prefer.  We need to look more at the current top variation to see if Gold has a refutation.

    Gold does seem to have an outright refutation for safetying our rabbit with 12b ed4n ed5w rb6e ha6e.  Between the similar moves 12b ed4w hh3n Rh2n dd8s and 12b ed4w hh3n Rh2n ha6n, the former clearly has the better fourth step, so we needn't pay too much attention to the latter.  In contrast, it is an interesting choice between the similar moves 12b ed4w ec4n Hc3n md6e and 12b ed4w ec4n Hc3n dd8s.  Advancing the dog for the fourth step gives us the possibility to play a very sharp horse flip in one of the main lines.

    The analysis tree:
    12b ed4w hh3n Rh2n dd8s (occupy c4, pull rabbit, advance dog)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b ha4n Ra3n ha5n Ra4n
    .    .    .    14w Ec5e Ed5e md4s Hg3n
    .    .    .    .    14b ha6e Ra5n rb5e md5n (Silver seems better)
    .    .    13b ha6e rb5w ha4n Ra3n
    .    .    .    14w Ec5e Ed5e md4s Hg3n
    .    .    .    .    14b ha5e Ra4n ec4e md5n
    .    .    .    .    .    15w Ee5e Ef5w hg5w Hg4n
    .    .    .    .    .    .    15b md6e me6e hf5s hf4e (unclear)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    .    .    13b rb4w hh4n Rh3n ha6e
    12b hh3n Rh2n hh4n Rh3n (pull rabbit twice)
    .    13w Eb5s rb6s Eb4e rb5s (pull rabbit twice, blocking off c4)
    .    .    13b ha6s ha5s md6w ed4s
    .    .    .    14w Ec4n Ec5w mc4s Rc1e
    .    .    .    .    14b ed3n ed4w mc5n mc6w
    .    .    .    .    .    15w Db3s Hc3w Hg3n Hg4n
    .    .    .    .    .    .    15b hh5n Rh4n hh6w Rh5n
    .    .    .    .    .    .    .    16w Hg5w Rh6n mb6w Eb5n (unclear)
    .    .    .    .    .    .    15b ec4e ed4e ee4e ef4e
    .    .    .    .    .    .    .    16w Eb5e Ec5e Ed5e Ee5n (unclear)
    .    .    .    .    14b rb4e ha4s Db3n ha3e
    .    .    .    .    .    15w rc4e Db4e Cc2s Md2w
    .    .    .    .    .    .    15b hb3w ha3n Ra2n ha4n
    .    .    .    .    .    .    .    16w Dc4w rd4w Hc3w rc4s (unclear)
    .    .    .    .    14b rb4e mc5n mc6w ha4s
    .    .    .    .    .    15w Db3s Hc3w rc4s Hg3n
    .    .    .    .    .    .    16b hh5w Ra4n ha3n Ra2n
    .    .    .    .    .    .    .    16w Eb5w mb6s mb5e Ea5e (unclear)
    .    .    .    14w Db3s Hc3w Hb3w rb4s
    .    .    .    .    14b hh5n Rh4n hh6w Rh5n
    .    .    .    .    .    15w rb3e Ha3e Ec4n Ec5e (only rabbit save)
    .    .    .    .    .    .    15b ha4s mc6w dd8s rg7e (Silver +=)
    .    .    13b edd4s md6w mc6w ha6s
    .    .    .    14w Db3s Hc3w Hb3w rb4s
    .    .    .    .    14b hh5n Rh3n hh6w Ra5n
    .    .    .    .    .    15w rb3e Ha3e Ec4n Ec5w (likely rabbit trade =?)
    .    .    13b ha6s ha5s ha4s ed4s (each Silver horse pulls a rabbit)
    .    .    .    14w Eb5e Ec5e md6e Ed5n (counter-attack)
    .    .    .    .    14b cc7w cb7s md6e dd8s
    .    .    .    .    .    15w Hc3n Hc4n Cc2w Md2w (Gold's attack is better)
    .    .    .    .    14b Db3s ha3e rb8s rb7s (very sharp but favors Gold?)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    .    .    13b hh5n Rh4n ed4w rb4w (we win race unless Gold complicates)
    .    .    .    14w Eb5e Ec5e md6e Ed5n (complicating?)
    .    .    .    .    14b ha6e me6e Rh5w hh6n (he's in more trouble than us)  
    12b ed4w ec4n Hc3n dd8s (block Gold elephant from center, D south)
    .    13w Eb5s rb6s Hc4s Rh2w (safety h2-rabbit)
    .    .    13b hh3s hh2n Rh1n ec5s (still keep elephant from center)
    .    .    .    14w Eb4w rb5s Ea4n Ea5e (regain mobility)
    .    .    .    .    14b ec4e Hc3n Hc4n ed4w (flip horse !?)
    .    .    .    .    .    15w Eb5n Eb6e md6s Ec6e
    .    .    .    .    .    .    15b ec4e Hc5w md5w ec5n (unclear)
    .    .    .    .    .    15w Hc5e Eb5e Hd5s Ec5e
    .    .    .    .    .    .    15b Hd4e ec4e He4n ed4e
    .    .    .    .    .    .    .    16w Ed5w md6s md5s Ec5e
    .    .    .    .    .    .    .    .    16b md4w rb4w mc4w ha6e (unclear)
    .    .    .    .    14b hh3n Rh2n rb4w Ha6e
    .    .    .    .    .    15w Db3s Hc3w Eb5e Ec5e (seems about even?)
    12b ed4w ec4n Hc3n md6e (block Gold elephant from center, M east)
    .    13w Eb5s rb6s Hc4s Rh2w (safety h2-rabbit)
    .    .    13b hh3s hh2n Rh1n ec5s (still keep elephant from center)
    .    .    .    14w Eb4w rb5s Ea4n Ea5e (regain mobility)
    .    .    .    .    14b hh3n Rh2n rb4w Ha6e (seems about even?)
    .    13w Eb5s rb6s Hc4s Eb4e (keep elephant mobility)
    .    .    13b hh3n Rh2n hh4n Rh3n (race instead of protecting rabbit?)
    .    .    .    14w Db3n Db4s rb5s Db3e (now Gold wins a straight race!)
    .    .    .    .    14b ha6s ha5s ha4n Da3n (complicating)
    .    .    .    .    .    15w rb4s Ec4w Da4s Eb4n
    .    .    .    .    .    .    15b ec4s me6w hh5n Rh4n (Silver +=?)
    .    .    13b hh3n Rh2n rb5w ha6e (protect rabbit instead of racing?)
    .    .    .    14w Db3w Da3n Da4s ra5s (straight race ends tied)
    .    .    .    .    14b hh4n Rh3n hh5n Rh4n (rabbit trade =)
    .    .    .    14w Ec4e Ed4e Ee4e Ef4e (freeze horse)
    .    .    .    .    14b ec5s ec4w hc3n hb6s (race better than defense)
    .    .    .    .    .    15w Eg4w hh4w Ef4n hg4w (horse trade =)
    .    13w Eb5s rb6s Hc4s Hc3e (safety c3-horse)
    .    .    13b hh3n Rh2n ec5s ha6e (still keep elephant from center)
    .    .    .    14w Eb4w Ea4n rb5s Ea5e (regain mobility)
    .    .    .    .    14b rb4w hh4n Rh3n dd8s (Silver slightly ahead?)
    .    .    .    .    .    15w Hd3w Db3s Hc3w Eb5e
    .    .    .    14w Eb4w rb5s Db3w rb4s (still threaten rabbit)
    .    .    .    .   14b hb6s hb5s rb3s hb4s (exploiting b3 hole)
    .    .    .    .    .    15w Ea4e hb3e Eb4s De2n (wrong side frame)
    .    .    .    .    .    .    15b me6s me5s de3s me4s (crazy but Silver's OK)  
    12b hh3n Rh2n ed4w rb6e (occupy c4, pull rabbit, rabbit into trap)
    .    13w Eb5e Ec5e rc6s ra2e (pull rabbit + tuck rabbit)
    .    .    13b ha6s ec4w eb4e Db3n (pull dog)
    .    .    .    14w Ed5e md6s md5s Ee5w (flip camel)
    .    .    .    .    14b md4e me4e mf4n mf5n (run away)
    .    .    .    .    .    15w Ed5e Ee5e Ef5e mf6s
    .    .    .    .    .    .    15b ce7s ce6s mf5n ce5n
    .    .    .    .    .    .    .    16w Eg5s Eg4n hh4w Hf3w (Gold +=)  
    .    .    13b hh4n Rh3n hh5n Rh4n (pull rabbit)
    .    .    .    14w Hg3n Hg4n Hg5n De2n (attack f6, Gold initiative)
    .    13w Eb5e Ec5w rc6s Hc3e (pull rabbit + safety horse)
    .    .    13b hh4w Rh3n Rh4n hg4e (flip rabbit, Silver winning)
    .    13w Eb5e Ec5e rc6s Ed5e (pull rabbit + threaten f6)
    .    .    13b hh4n Rh3n hh5w hg5n (pull rabbit and block horse)
    .    13w Eb5n Eb6n Eb7s rb8s (pull second rabbit)
    .    .    13b ec4e Hc3n Hc4n ed4w (horse flip, Silver winning)
    12b ed4w hh3n Rh2n ha6n (occupy c4, pull rabbit, retreat horse)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b hh4n Rh3n hh5n Rh4n
    .    .    .    14w Ec5e md6e Ed5n Hb3n (unclear)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    .    .    13b hh4n Rh3n hh5n Rh4n
    .    .    .    14w Db3s Hc3w Eb5e Ec5e
    .    .    .    .    14b rb4w ec4e hh6w rh6n (equal)
    12b ed4n ed5w rb6e ha6e (safety rabbit)
    .    13w Rh2w Md2n Md3e Me3e
    .    .    13b hh3n hh4n hh5n hh6w
    .    .    .    14w Eb5s Eb4e Db3s Hc3w (Gold +=)
    .    .    13b hh3s hh2n Rh1n hh3n
    .    .    .    14w Hg3n Mf3e Mg3e Eb5s (Gold +)

    Title: Re: Move 12
    Post by mdk on Aug 28th, 2007, 5:22pm

    on 08/28/07 at 14:16:20, Fritzlein wrote:
    Nice summary JDB.  If we can capture a rabbit in f6 without immediately having to give up ours in c3, we are ahead.  However, it seems that even in Gold's best lines, we can return the c3 rabbit for an equal position and no risk.

    I have updated the analysis tree with everyone's suggestions, plus a few new lines, plus at least a little analysis of each of the moves.  I have ordered the moves and branches approximately best to worst, according to minimax on the tree and my guesstimate of the nodes.  I think we are a tiny bit ahead ahead with either the double pull or 99of9's move, and I'm not sure which I prefer.  We need to look more at the current top variation to see if Gold has a refutation.

    Gold does seem to have an outright refutation for safetying our rabbit with 12b ed4n ed5w rb6e ha6e.  Between the similar moves 12b ed4w hh3n Rh2n dd8s and 12b ed4w hh3n Rh2n ha6n, the former clearly has the better fourth step, so we needn't pay too much attention to the latter.  In contrast, it is an interesting choice between the similar moves 12b ed4w ec4n Hc3n md6e and 12b ed4w ec4n Hc3n dd8s.  Advancing the dog for the fourth step gives us the possibility to play a very sharp horse flip in one of the main lines.

    The analysis tree:
    [edit]...
    12b hh3n Rh2n hh4n Rh3n (pull rabbit twice)
    .    13w Eb5s rb6s Eb4e rb5s (pull rabbit twice, blocking off c4)
    .    .    13b ha6s ha5s md6w ed4s
    .    .    .    14w Ec4n Ec5w mc4s Rc1e
    .    .    .    .    14b ed3n ed4w mc5n mc6w
    .    .    .    .    .    15w Db3s Hc3w Hg3n Hg4n
    .    .    .    .    .    .    15b hh5n Rh4n hh6w Rh5n
    .    .    .    .    .    .    .    16w Hg5w Rh6n mb6w Eb5n (unclear)
    .    .    .    .    .    .    15b ec4e ed4e ee4e ef4e
    .    .    .    .    .    .    .    16w Eb5e Ec5e Ed5e Ee5n (unclear)
    .    .    .    .    14b rb4e ha4s Db3n ha3e
    .    .    .    .    .    15w rc4e Db4e Cc2s Md2w
    .    .    .    .    .    .    15b hb3w ha3n Ra2n ha4n
    .    .    .    .    .    .    .    16w Dc4w rd4w Hc3w rc4s (unclear)
    .    .    .    .    14b rb4e mc5n mc6w ha4s
    .    .    .    .    .    15w Db3s Hc3w rc4s Hg3n
    .    .    .    .    .    .    16b hh5w Ra4n ha3n Ra2n
    .    .    .    .    .    .    .    16w Eb5w mb6s mb5e Ea5e (unclear)
    .    .    .    14w Db3s Hc3w Hb3w rb4s
    .    .    .    .    14b hh5n Rh4n hh6w Rh5n
    .    .    .    .    .    15w rb3e Ha3e Ec4n Ec5e (only rabbit save)
    .    .    .    .    .    .    15b ha4s mc6w dd8s rg7e (Silver +=)
    .    .    13b edd4s md6w mc6w ha6s
    .    .    .    14w Db3s Hc3w Hb3w rb4s
    .    .    .    .    14b hh5n Rh3n hh6w Ra5n
    .    .    .    .    .    15w rb3e Ha3e Ec4n Ec5w (likely rabbit trade =?)
    .    .    13b ha6s ha5s ha4s ed4s (each Silver horse pulls a rabbit)
    .    .    .    14w Eb5e Ec5e md6e Ed5n (counter-attack)
    .    .    .    .    14b cc7w cb7s md6e dd8s
    .    .    .    .    .    15w Hc3n Hc4n Cc2w Md2w (Gold's attack is better)
    .    .    .    .    14b Db3s ha3e rb8s rb7s (very sharp but favors Gold?)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    .    .    13b hh5n Rh4n ed4w rb4w (we win race unless Gold complicates)
    .    .    .    14w Eb5e Ec5e md6e Ed5n (complicating?)
    .    .    .    .    14b ha6e me6e Rh5w hh6n (he's in more trouble than us)  
    ...[edit]


    regarding the line

    12b hh3n Rh2n hh4n Rh3n (pull rabbit twice)
    .    13w Eb5s rb6s Eb4e rb5s (pull rabbit twice, blocking off c4)
    .    .    13b edd4s md6w mc6w ha6s
    .    .    .    14w Db3s Hc3w Hb3w rb4s

    i would continue
    .    .    .    .    14b ha5s hh5n Rh4n dd8s

    and i think we are better. this seems similar to me to the lines that fritz and i were discussing where silver advanced the horse twice on 13b and only moved the camel once. i think we are better here for the same reasons we are in the other lines. i'm curious to here what fritz and others think. i do not, however, think that this line will lead to a simple rabbit trade where the position is left about equal.

    Title: Re: Move 12
    Post by 99of9 on Aug 28th, 2007, 5:31pm

    Quote:
    12b ed4w hh3n Rh2n ha6n (occupy c4, pull rabbit, retreat horse)


    I don't have the board in front of me, but I just realized I misread the original suggestion of this move.  I thought it meant advance the a-file horse.  Which I quite liked actually.  So there's a new option:

    12b ed4w hh3n Rh2n ha6s

    Again I don't have time, but let me know if your refutations still apply.

    Title: Re: Move 12
    Post by Fritzlein on Aug 28th, 2007, 8:08pm

    on 08/28/07 at 17:22:48, mdk wrote:
    regarding the line

    12b hh3n Rh2n hh4n Rh3n (pull rabbit twice)
    .    13w Eb5s rb6s Eb4e rb5s (pull rabbit twice, blocking off c4)
    .    .    13b edd4s md6w mc6w ha6s
    .    .    .    14w Db3s Hc3w Hb3w rb4s
    .    .    .    .    14b ha5s hh5n Rh4n dd8s

    Wow, that 14b really shuts down Gold's ability to force a rabbit trade.  I think these lines are teaching me that even in a "race" I am far too eager to make racing moves instead of positional moves.  Not only was I too eager to pull the rabbit on 14b, but I was too eager to pull the rabbit on 14w.

    How would you continue for Silver if Gold preemptively freezes the horse and camel?

    12b hh3n Rh2n hh4n Rh3n (pull rabbit twice)
    .    13w Eb5s rb6s Eb4e rb5s (pull rabbit twice, blocking off c4)
    .    .    13b edd4s md6w mc6w ha6s
    .    .    .    14w Db3s Hc3w Ec4n Ec5w

    With the silver camel frozen, Gold is threatening to advance his g3-horse to g6.  Also, more subtly, Gold is threatening 15w rb4e Hb3n rc4e Hb4e, with camel flip to follow.

    Title: Re: Move 12
    Post by Fritzlein on Aug 28th, 2007, 9:02pm

    on 08/28/07 at 17:31:38, 99of9 wrote:
    12b ed4w hh3n Rh2n ha6s

    Again I don't have time, but let me know if your refutations still apply.

    Revisiting the line, I'm not sure my "refutation" applies even to the horse retreat, never mind the horse advance.  This is why we need people double-checking any lines I give!

    Anyway, on second inspection I'd give a different line, but I still think Gold is better after the horse retreat.

    12b ed4w hh3n Rh2n ha6n (occupy c4, pull rabbit, retreat horse)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b hh4n Rh3n hh5n Rh4n
    .    .    .    14w Ec5e Ed5w md6s Hb3n
    .    .    .    .    14b ec4e md5n hh6w Rh5n
    .    .    .    .    .    15w Hb4w rb5s Db2n Ec5e

    The horse advance is trickier to meet, and it foils my main line.  For one thing, the horse could dart in and get a second rabbit pull, although I'm not sure that's so good:

    12b ed4w hh3n Rh2n ha6s (occupy c4, pull rabbit, advance horse)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b ha5s ha4s ha3n Ra2n
    .    .    .    14w rb5s Ec5w Eb5e Ec5e

    But at the end of the first line I gave the horse can definitely dive in to good effect:

    12b ed4w hh3n Rh2n ha6s (occupy c4, pull rabbit, advance horse)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b hh4n Rh3n hh5n Rh4n
    .    .    .    14w Ec5e Ed5w md6s Hb3n
    .    .    .    .    14b ec4e md5n hh6w Rh5n
    .    .    .    .    .    15w Hb4w rb5s Db2n Ec5e
    .    .    .    .    .    .    15b ha5e rb4e hb5s ed4s

    so Gold needs a new plan.  Maybe

    12b ed4w hh3n Rh2n ha6s (occupy c4, pull rabbit, advance horse)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b hh4n Rh3n hh5n Rh4n
    .    .    .    14w Hb3n Db2n Hb4w rb5s

    But Silver is at worst equal there, and probably a bit ahead.

    In short, after a cursory glance, the horse advance looks fine.

    Title: Re: Move 12
    Post by mdk on Aug 28th, 2007, 9:06pm

    on 08/28/07 at 20:08:17, Fritzlein wrote:
    Wow, that 14b really shuts down Gold's ability to force a rabbit trade.  I think these lines are teaching me that even in a "race" I am far too eager to make racing moves instead of positional moves.  Not only was I too eager to pull the rabbit on 14b, but I was too eager to pull the rabbit on 14w.

    How would you continue for Silver if Gold preemptively freezes the horse and camel?

    12b hh3n Rh2n hh4n Rh3n (pull rabbit twice)
    .    13w Eb5s rb6s Eb4e rb5s (pull rabbit twice, blocking off c4)
    .    .    13b edd4s md6w mc6w ha6s
    .    .    .    14w Db3s Hc3w Ec4n Ec5w

    With the silver camel frozen, Gold is threatening to advance his g3-horse to g6.  Also, more subtly, Gold is threatening 15w rb4e Hb3n rc4e Hb4e, with camel flip to follow.


    I'm glad to hear that you are learning as much as the rest of us from this game Fritz. I guess this just proves what a great idea this game was.

    Ok i have a couple of ideas. First:
    .    .    .    .    14b cc7w dd8s dd7w xxxx.

    This allows the camel to avoid being flipped and also allows it to quickly cross, avoiding any possibility of the g3 horse advancing, the two things you mentioned fritz.
    I'm not sure what would be the best way to use the fourth step yet. any ideas?

    The other is:
    .    .    .    .    14b ed3n Md2n ed4w Md3n.

    This is certainly different than the other moves we have been looking at in this line but i believe that with this move we can take advantage of the position of the gold phant.

    As usual I'm looking foward to hearing what everyone thinks.

    Title: Re: Move 12
    Post by RonWeasley on Aug 29th, 2007, 1:54pm
    I find I don't have time to weigh in on this.  I think I want to see a response to mdk and then I'm going to start the vote.  The top two moves are close and I'd like to see mdk's suggestion commented.

    Title: Re: Move 12
    Post by mdk on Aug 29th, 2007, 2:38pm

    on 08/28/07 at 21:02:00, Fritzlein wrote:
    Revisiting the line, I'm not sure my "refutation" applies even to the horse retreat, never mind the horse advance.  This is why we need people double-checking any lines I give!

    Anyway, on second inspection I'd give a different line, but I still think Gold is better after the horse retreat.

    12b ed4w hh3n Rh2n ha6n (occupy c4, pull rabbit, retreat horse)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b hh4n Rh3n hh5n Rh4n
    .    .    .    14w Ec5e Ed5w md6s Hb3n
    .    .    .    .    14b ec4e md5n hh6w Rh5n
    .    .    .    .    .    15w Hb4w rb5s Db2n Ec5e

    The horse advance is trickier to meet, and it foils my main line.  For one thing, the horse could dart in and get a second rabbit pull, although I'm not sure that's so good:

    12b ed4w hh3n Rh2n ha6s (occupy c4, pull rabbit, advance horse)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b ha5s ha4s ha3n Ra2n
    .    .    .    14w rb5s Ec5w Eb5e Ec5e

    But at the end of the first line I gave the horse can definitely dive in to good effect:

    12b ed4w hh3n Rh2n ha6s (occupy c4, pull rabbit, advance horse)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b hh4n Rh3n hh5n Rh4n
    .    .    .    14w Ec5e Ed5w md6s Hb3n
    .    .    .    .    14b ec4e md5n hh6w Rh5n
    .    .    .    .    .    15w Hb4w rb5s Db2n Ec5e
    .    .    .    .    .    .    15b ha5e rb4e hb5s ed4s

    so Gold needs a new plan.  Maybe

    12b ed4w hh3n Rh2n ha6s (occupy c4, pull rabbit, advance horse)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b hh4n Rh3n hh5n Rh4n
    .    .    .    14w Hb3n Db2n Hb4w rb5s

    But Silver is at worst equal there, and probably a bit ahead.

    In short, after a cursory glance, the horse advance looks fine.


    Fritz... did you consider any other moves on 13w for gold?

    For example:

    12b ed4w hh3n Rh2n ha6s (occupy c4, pull rabbit, advance horse)
    .    13w Eb5e rb6s rb5s Eb5w
    .    .    13b hh4n Rh3n hh5n Rh4n
    .    .    .    14w Eb5e Ec5e Ed5e md6s

    and suddenly our position doesn't seem so favorable as our camel is threatened and gold is threatening an EH attack. So how should we continue on 13b as clearly the move i suggest here is bad for us?

    Also, I, like Ron, would like to hear some comments on my suggestion.

    [edit]
    actually

    12b ed4w hh3n Rh2n ha6s (occupy c4, pull rabbit, advance horse)
    .    13w Eb5e Ec5e Ed5e md6s

    may be a better line for gold so id like to know how we would respond to both potential moves.

    Title: Re: Move 12
    Post by Fritzlein on Aug 29th, 2007, 4:46pm

    on 08/29/07 at 13:54:06, RonWeasley wrote:
    I find I don't have time to weigh in on this.  I think I want to see a response to mdk and then I'm going to start the vote.  The top two moves are close and I'd like to see mdk's suggestion commented.


    I'll comment on mdk's line if I have time tonight, but if I don't, I think we can trust his analysis, or vote on our own instincts.  We should be less worried that no one has responded to a sub-variation of a heavily analyzed move than that nobody has responded to the top line of my analysis tree:

    12b ed4w hh3n Rh2n dd8s (occupy c4, pull rabbit, advance dog)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b ha4n Ra3n ha5n Ra4n
    .    .    .    14w Ec5e Ed5e md4s Hg3n
    .    .    .    .    14b ha6e Ra5n rb5e md5n (Silver seems better)

    Is everyone just going to take my word that we are better off if we play 99of9's suggestion?

    On the one hand, it is better to extend an active discussion than to extend a discussion where nothing is being said, but even so, we don't collectively have time to examine everything, and we ultimately will have to vote with some questions unanswered.

    Title: Re: Move 12
    Post by mdk on Aug 29th, 2007, 5:09pm

    on 08/29/07 at 16:46:34, Fritzlein wrote:
    I'll comment on mdk's line if I have time tonight, but if I don't, I think we can trust his analysis, or vote on our own instincts.  We should be less worried that no one has responded to a sub-variation of a heavily analyzed move than that nobody has responded to the top line of my analysis tree:

    12b ed4w hh3n Rh2n dd8s (occupy c4, pull rabbit, advance dog)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b ha4n Ra3n ha5n Ra4n
    .    .    .    14w Ec5e Ed5e md4s Hg3n
    .    .    .    .    14b ha6e Ra5n rb5e md5n (Silver seems better)

    Is everyone just going to take my word that we are better off if we play 99of9's suggestion?

    On the one hand, it is better to extend an active discussion than to extend a discussion where nothing is being said, but even so, we don't collectively have time to examine everything, and we ultimately will have to vote with some questions unanswered.


    Ok ill question your line Fritz... your notation must be wrong as this line isn't even possible!!!!!

    i'll add some analysis in a bit but i thought i should throw this out there

    [edit]
    12b ed4w hh3n Rh2n dd8s (occupy c4, pull rabbit, advance dog)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b hh4n Rh3n hh5n Rh4n
    .    .    .    14w Ec5e Ed5e md4s Hg3n (i believe this is what you intended)
    .    .    .    .    14b ha6e Ra5n rb5e md5n (this move isn't legal)

    Title: Re: Move 12
    Post by 99of9 on Aug 29th, 2007, 5:11pm

    on 08/23/07 at 18:14:38, Fritzlein wrote:
    This seems idiot simple.

    Is this one of those rare moves on which a snap vote is appropriate?

    Yep, I think you're right Fritz, let's vote ;-).  Gee I love that you're bold enough to risk statements like that.

    Title: Re: Move 12
    Post by mdk on Aug 29th, 2007, 5:23pm
    12b ed4w hh3n Rh2n dd8s (occupy c4, pull rabbit, advance dog)
    .    13w Eb5e Ec5e Ed5e md6s

    this is a very similar position to
    12b ed4w hh3n Rh2n ha6n
    .    13w Eb5e Ec5e Ed5e md6s

    perhaps there is some refutation to both of these lines, but no one has offered one yet and until someone does I do not feel comfortable playing either of these moves.

    I think as a general note we need to be more aware of the potential for chessandgo to attack our camel. Collectively we have been so focused on the rabbit pulling race that we have been ignoring larger threats that either side can potentially make.

    Title: Re: Move 12
    Post by mdk on Aug 29th, 2007, 5:54pm
    also i don't have time now but could someone put together an updated analysis tree?

    Title: Re: Move 12
    Post by mdk on Aug 29th, 2007, 5:59pm

    on 08/28/07 at 14:16:20, Fritzlein wrote:
    Nice summary JDB.  If we can capture a rabbit in f6 without immediately having to give up ours in c3, we are ahead.  However, it seems that even in Gold's best lines, we can return the c3 rabbit for an equal position and no risk.

    I have updated the analysis tree with everyone's suggestions, plus a few new lines, plus at least a little analysis of each of the moves.  I have ordered the moves and branches approximately best to worst, according to minimax on the tree and my guesstimate of the nodes.  I think we are a tiny bit ahead ahead with either the double pull or 99of9's move, and I'm not sure which I prefer.  We need to look more at the current top variation to see if Gold has a refutation.

    Gold does seem to have an outright refutation for safetying our rabbit with 12b ed4n ed5w rb6e ha6e.  Between the similar moves 12b ed4w hh3n Rh2n dd8s and 12b ed4w hh3n Rh2n ha6n, the former clearly has the better fourth step, so we needn't pay too much attention to the latter.  In contrast, it is an interesting choice between the similar moves 12b ed4w ec4n Hc3n md6e and 12b ed4w ec4n Hc3n dd8s.  Advancing the dog for the fourth step gives us the possibility to play a very sharp horse flip in one of the main lines.

    The analysis tree:
    12b ed4w hh3n Rh2n dd8s (occupy c4, pull rabbit, advance dog)
    .    13w Eb5e rb6s Db3s Hc3w
    .    .    13b ha4n Ra3n ha5n Ra4n
    .    .    .    14w Ec5e Ed5e md4s Hg3n
    .    .    .    .    14b ha6e Ra5n rb5e md5n (Silver seems better)
    .    .    13b ha6e rb5w ha4n Ra3n
    .    .    .    14w Ec5e Ed5e md4s Hg3n
    .    .    .    .    14b ha5e Ra4n ec4e md5n
    .    .    .    .    .    15w Ee5e Ef5w hg5w Hg4n
    .    .    .    .    .    .    15b md6e me6e hf5s hf4e (unclear)
    .    13w Eb5e rb6s rb5s Ec5w (flip rabbit)
    .    .    13b rb4w hh4n Rh3n ha6e
    ...[edit]



    i don't know if this has been edited elsewhere or changed but this part of the analysis tree is riddled with unplayable moves

    Title: Re: Move 12
    Post by Fritzlein on Aug 29th, 2007, 10:13pm

    on 08/28/07 at 21:06:12, mdk wrote:
    I'm glad to hear that you are learning as much as the rest of us from this game Fritz.

    Oh, yes, I am learning a ton.  Indeed, the fact that I think I know something is often the biggest obstacle.  For example, when I try to apply my intuitions from games with pure dual-lone-elephant attacks to this game with horses charging forward as well, it seems I usually get the wrong answer.

    In the line
    12b hh3n Rh2n hh4n Rh3n (pull rabbit twice)
    .    13w Eb5s rb6s Eb4e rb5s (pull rabbit twice, blocking off c4)
    .    .    13b edd4s md6w mc6w ha6s
    .    .    .    14w Db3s Hc3w Ec4n Ec5w
    .    .    .    .    14b cc7w dd8s dd7w xxxx

    I would look first at 15w rb4e Hb3n rc4e Hb4e.  The camel is unfrozen, true, but it doesn't really have a better place to be.  If it heads back to d6, then the gold elephant can push it to e6 and invite dueling EH attacks.  The critical line  is probably 15b rd4e ed3n Hc4n ed4w, but after 16w Hc5e Hd5e re4s He5s Gold looks fine.

    In the line
    12b hh3n Rh2n hh4n Rh3n (pull rabbit twice)
    .    13w Eb5s rb6s Eb4e rb5s (pull rabbit twice, blocking off c4)
    .    .    13b edd4s md6w mc6w ha6s
    .    .    .    14w Db3s Hc3w Ec4n Ec5w
    .    .    .    .    14b ed3n Md2n ed4w Md3n.

    The gold camel is in danger, but I think this is compensated by its greater ability to access the east.  In some lines it could come into play that we can threaten to take a horse hostage with our camel.  15w De2n De3n Md4s De4s, and Gold has very active pieces.  One point is that now if Gold can frame the silver rabbit, Gold can later pull it to d2, forking and winning it, so just racing and capturing the h-rabbit is less likely to produce an advantage for Silver.  My best guess at how Silver could play for advantage is to attack the exposed camel and dog on d3 and e3, but the timing is very tricky because Gold pieces are threatened as well.  For example Silver could err with 15b ec4e ed4e Md3n dd8s 16w Hb3w rb4s rb3e Ha3s 16b Md4n ee4w Md5n ed4n 17w Eb5e mb6s mb5s Ec5w and Silver has lost a camel and a rabbit for a camel.


    Title: Re: Move 12
    Post by Fritzlein on Aug 29th, 2007, 10:38pm
    This isn't the type of position I understand very well, but I will venture a summary of the opposing merits of the double rabbit pull and occupying c4.

    If Silver pulls the rabbit twice, then chessandgo can pull in response, blocking us from c4, and forcing our elephant to defend from d3 if we want to play for advantage.  This is not a bad square for our elephant, but to make it potent it seems we need to threaten some sort of action with our horse and/or camel on the same wing.  All of mdk's proposed advances allow some sort of counter and complication, to wit:

    13b ed4s ha6s ha5s ha4s  Charging three with the horse allows our camel to be pushed east in an invitation to dueling EH attacks.

    13b ed4s ha6s ha5s md6w Trading one horse step for moving our camel west allows the camel to be pulled south, which sort of forces the silver elephant back to c4, releasing the gold camel that was stuck on d2.

    13b edd4s md6w mc6w ha6s Trading two horse steps for moving our camel west two allows Gold to freeze both at once with his elephant.

    These lines all got more complex than simple racing rabbit pulls, and I have trouble evaluating them.  Also, there may be yet other good tries for 13b.  All in all, however, I fail to find a way for Silver to force the dream scenario outlined by JDB, where we delay capture of our exposed rabbit while capturing Gold's exposed rabbit for a clear advantage.  Rather it seems Silver's edge is more subtle, and requires that superior piece play to realize.

    In contrast,  12b ed4w hh3n Rh2n dd8s is less ambitious and less wild.  Placing our elephant on c4 is purely an attempt to temporarily delay capture of our exposed rabbit, not a threat to mount a rescue mission that Gold must guard against if our elephant is on d3.  If our elephant on c4 is eventually pinned to our rabbit on c3, we will surely lose the rabbit when it is dragged to d3 and forked.

    However, the delay should be long enough to capture chessandgo's rabbit, and perhaps get some play while he is taking the time to finish off ours.  If he doesn't want to accept this slight disadvantage, then he is the one who is forced to seek complications, perhaps to his disadvantage.

    As you may guess, I favor the calmer move, not because it is necessarily objectively better, but because it doesn't require us to charge forward in order to play for advantage.  This is just a stylistic preference, and those who don't mind mixing it up by advancing pieces behind our exposed rabbit may like the greater tension inherent in pulling twice and eventually putting our elephant on d3.


    Title: Re: Move 12
    Post by Fritzlein on Aug 29th, 2007, 10:40pm

    on 08/29/07 at 17:59:07, mdk wrote:
    i don't know if this has been edited elsewhere or changed but this part of the analysis tree is riddled with unplayable moves

    If you look on my last full post of the analysis tree, I edited yesterday evening to change the a's to h's, so the rabbit pulls should now be legal.

    Title: Re: Move 12
    Post by mdk on Aug 29th, 2007, 10:48pm

    on 08/29/07 at 22:13:11, Fritzlein wrote:
    Oh, yes, I am learning a ton.  Indeed, the fact that I think I know something is often the biggest obstacle.  For example, when I try to apply my intuitions from games with pure dual-lone-elephant attacks to this game with horses charging forward as well, it seems I usually get the wrong answer.

    In the line
    12b hh3n Rh2n hh4n Rh3n (pull rabbit twice)
    .    13w Eb5s rb6s Eb4e rb5s (pull rabbit twice, blocking off c4)
    .    .    13b edd4s md6w mc6w ha6s
    .    .    .    14w Db3s Hc3w Ec4n Ec5w
    .    .    .    .    14b cc7w dd8s dd7w xxxx

    I would look first at 15w rb4e Hb3n rc4e Hb4e.  The camel is unfrozen, true, but it doesn't really have a better place to be.  If it heads back to d6, then the gold elephant can push it to e6 and invite dueling EH attacks.  The critical line  is probably 15b rd4e ed3n Hc4n ed4w, but after 16w Hc5e Hd5e re4s He5s Gold looks fine.

    In the line
    12b hh3n Rh2n hh4n Rh3n (pull rabbit twice)
    .    13w Eb5s rb6s Eb4e rb5s (pull rabbit twice, blocking off c4)
    .    .    13b edd4s md6w mc6w ha6s
    .    .    .    14w Db3s Hc3w Ec4n Ec5w
    .    .    .    .    14b ed3n Md2n ed4w Md3n.

    The gold camel is in danger, but I think this is compensated by its greater ability to access the east.  In some lines it could come into play that we can threaten to take a horse hostage with our camel.  15w De2n De3n Md4s De4s, and Gold has very active pieces.  One point is that now if Gold can frame the silver rabbit, Gold can later pull it to d2, forking and winning it, so just racing and capturing the h-rabbit is less likely to produce an advantage for Silver.  My best guess at how Silver could play for advantage is to attack the exposed camel and dog on d3 and e3, but the timing is very tricky because Gold pieces are threatened as well.  For example Silver could err with 15b ec4e ed4e Md3n dd8s 16w Hb3w rb4s rb3e Ha3s 16b Md4n ee4w Md5n ed4n 17w Eb5e mb6s mb5s Ec5w and Silver has lost a camel and a rabbit for a camel.


    12b hh3n Rh2n hh4n Rh3n (pull rabbit twice)
    .    13w Eb5s rb6s Eb4e rb5s (pull rabbit twice, blocking off c4)
    .    .    13b edd4s md6w mc6w ha6s
    .    .    .    14w Db3s Hc3w Ec4n Ec5w
    .    .    .    .    14b ed3n Md2n ed4w Md3n
    .    .    .    .    .    15w De2n De3n Md4s De4s
    .    .    .    .    .    .    15b ec4e ed4e Md3n dd8s
    .    .    .    .    .    .    .    16w Hb3w rb4s rb3e Ha3s
    .    .    .    .    .    .    .    .    16b Md4n ee4w cc7w mb6w

    and as far as i can tell silver is guaranteed a camel hostage. so i would say that this 15b isn't a mistake rather a reasonable continuation and that the 16w you propose is a mistake. looking foward to hearing some comments.

    Title: Re: Move 12
    Post by mdk on Aug 29th, 2007, 11:00pm

    on 08/29/07 at 22:38:24, Fritzlein wrote:
    This isn't the type of position I understand very well, but I will venture a summary of the opposing merits of the double rabbit pull and occupying c4.

    If Silver pulls the rabbit twice, then chessandgo can pull in response, blocking us from c4, and forcing our elephant to defend from d3 if we want to play for advantage.  This is not a bad square for our elephant, but to make it potent it seems we need to threaten some sort of action with our horse and/or camel on the same wing.  All of mdk's proposed advances allow some sort of counter and complication, to wit:

    13b ed4s ha6s ha5s ha4s  Charging three with the horse allows our camel to be pushed east in an invitation to dueling EH attacks.

    13b ed4s ha6s ha5s md6w Trading one horse step for moving our camel west allows the camel to be pulled south, which sort of forces the silver elephant back to c4, releasing the gold camel that was stuck on d2.

    13b edd4s md6w mc6w ha6s Trading two horse steps for moving our camel west two allows Gold to freeze both at once with his elephant.

    These lines all got more complex than simple racing rabbit pulls, and I have trouble evaluating them.  Also, there may be yet other good tries for 13b.  All in all, however, I fail to find a way for Silver to force the dream scenario outlined by JDB, where we delay capture of our exposed rabbit while capturing Gold's exposed rabbit for a clear advantage.  Rather it seems Silver's edge is more subtle, and requires that superior piece play to realize.

    In contrast,  12b ed4w hh3n Rh2n dd8s is less ambitious and less wild.  Placing our elephant on c4 is purely an attempt to temporarily delay capture of our exposed rabbit, not a threat to mount a rescue mission that Gold must guard against if our elephant is on d3.  If our elephant on c4 is eventually pinned to our rabbit on c3, we will surely lose the rabbit when it is dragged to d3 and forked.

    However, the delay should be long enough to capture chessandgo's rabbit, and perhaps get some play while he is taking the time to finish off ours.  If he doesn't want to accept this slight disadvantage, then he is the one who is forced to seek complications, perhaps to his disadvantage.

    As you may guess, I favor the calmer move, not because it is necessarily objectively better, but because it doesn't require us to charge forward in order to play for advantage.  This is just a stylistic preference, and those who don't mind mixing it up by advancing pieces behind our exposed rabbit may like the greater tension inherent in pulling twice and eventually putting our elephant on d3.


    well obviously based on all my arguing for the double rabbit pull i respectfully disagree with fritz. i would also suggest that the complex variations that fritz and i have been discussing over the past few days are part of what make the double rabbit pull a better option for us the mob. With all the potential lines that chessandgo has to look at he is more likely to make a mistake and i think that the number of times fritz has had to change his responses to my suggestions shows just how likely this is. it is in lines such as these that we will be able to demonstrate the advantage of having a greater number of people collaborating to examine lines. also, through the detailed analysis that has already been done we have quite a good idea of any complications chessandgo can throw and we have solid responses to all of them. this is more than can be said for any other line.

    alas we already have charged foward with our horse on one wing and i believe it is chessandgo who has shown how effective charging foward with horses on both wings can be. in any case in none of the lines has it been demonstrated that any of the pieces which "charge" foward are in any real danger of being captured or even hostaged other than of course the rabbit which is inevitable.

    Title: Re: Move 12
    Post by Fritzlein on Aug 29th, 2007, 11:03pm

    on 08/29/07 at 17:23:12, mdk wrote:
    12b ed4w hh3n Rh2n dd8s (occupy c4, pull rabbit, advance dog)
    .    13w Eb5e Ec5e Ed5e md6s

    this is a very similar position to
    12b ed4w hh3n Rh2n ha6n
    .    13w Eb5e Ec5e Ed5e md6s

    perhaps there is some refutation to both of these lines, but no one has offered one yet and until someone does I do not feel comfortable playing either of these moves.

    I believe that 13b ec4e Hc3n hh4n Rh3n is playable in either case, and indeed probably to Silver's advantage.  Gold's threatened EH attack is worth less than a whole horse, so Gold probably can't switch wings quite so easily.  This is what I meant in my vague summary that Gold is the one forced to seek complications "perhaps to his disadvantage".  Obviously it is difficult to tell who is winning in any sharp line, and we'll have to vote without the detailed analysis we had for the sharp lines in the double rabbit pull variant, but my gut impression (for what it's worth) is that Gold has no forcing tactical way to hurt us in that line, and indeed may worsen his position by trying to complicate.

    Title: Re: Move 12
    Post by mdk on Aug 29th, 2007, 11:07pm
    also why doesnt the ballot include the move 12b ed4w hh3n Rh2n ed4w ha6s which was one of the moves proposed in the thread?

    im not sure if this is a significant enough option that the balloting should be restarted

    Title: Re: Move 12
    Post by Fritzlein on Aug 29th, 2007, 11:29pm

    on 08/29/07 at 23:00:35, mdk wrote:
    With all the potential lines that chessandgo has to look at he is more likely to make a mistake and i think that the number of times fritz has had to change his responses to my suggestions shows just how likely this is.

    And don't forget to mention that you had to change your mind about 13b twice after I eventually refuted your first two suggestions by changing my mind.  We have proved that it is a complicated position, not that Silver has a big advantage, and we have proved that we can also make mistakes.  (Yes, I also made a mistake in 16w in the last line I proposed, and I could easily be totally wrong about my overall evaluation of the double pull.)


    on 08/29/07 at 17:54:05, mdk wrote:
    also i don't have time now but could someone put together an updated analysis tree?

    Yes, the one thing you always have time for is arguing for your own line and arguing against other lines.  Meanwhile maintaining a tree and wondering if there is merit in some other branch is someone else's job.


    on 08/29/07 at 17:09:17, mdk wrote:
    Ok ill question your line Fritz... your notation must be wrong as this line isn't even possible!!!!!

    I have exhausted myself this move maintaining the analysis tree.  I apologize for all the errors it contains in notation and judgment.  You or anyone is welcome to take over.


    Title: Re: Move 12
    Post by Fritzlein on Aug 30th, 2007, 12:03am
    Sorry, mdk, my frustration bubbled over there.  I could delete the post, but let me instead try to rephrase my frustration more constructively.

    There are two ways to contribute to the Mob playing well, namely discovering which move is best and advocating for that move so it gets the most votes.  Either by itself is futile.  If you discover the best move and nobody believes you that it is best, the Mob will play an inferior move.  If you fail to discover the best move and advocate for an inferior one, the Mob will be more like to vote incorrectly as well.

    It has been my perception, perhaps erroneously, that you have been focused on advocacy at the expense of discovery.  It seems that when you even looked at a move other than the double pull, you didn't look to see if that move kicked butt for Silver too, you looked only for a way to prove that the move would hand the advantage to Gold.  Meanwhile when you looked at the double pull, you didn't rack your brain for a way to refute it, only for a way to make it work to Silver's advantage.

    Maybe this isn't fair to you.  Certainly I sometimes fall into the trap myself, and try to prove what I already believe rather than trying to see how I might be wrong.  However, from all the times I have changed my mind in this thread, I hope it comes through that I am at least trying to be impartial.  When I advocate for or against a move, I hope I have at least tried to see both advantages and disadvantages to that move.  Yet it is difficult to stay in this impartial mode of discussion when I perceive (correctly or not) that you don't share the same aim, and I find myself biased to want to contradict you.

    Does this sound at all fair to you?  I shouldn't say that I don't want to maintain the analysis tree any more, because actually I enjoyed it.  The less enjoyable part, which has led me to suspend discussing moves to raise issues of process, is feeling that we aren't quite working with the same objective in contributing to the tree.

    Putting that to one side, I have very much enjoyed your voluminous and insightful contributions of analysis on this move.  I think we go into the voting this time better prepared than on any other move, and this difference has been your participation.  I hope that my candor here doesn't discourage you from participating actively in the future as well, because your analysis has been fantastic.

    Title: Re: Move 12
    Post by mdk on Aug 30th, 2007, 12:10am

    on 08/29/07 at 23:29:40, Fritzlein wrote:
    And don't forget to mention that you had to change your mind about 13b twice after I eventually refuted your first two suggestions by changing my mind.  We have proved that it is a complicated position, not that Silver has a big advantage, and we have proved that we can also make mistakes.

    Yes, the one thing you always have time for is arguing for your own line and arguing against other lines.  Meanwhile maintaining a tree and wondering if there is merit in some other branch is someone else's job.

    I have exhausted myself this move maintaining the analysis tree.  I apologize for all the errors it contains in notation and judgment.  You or anyone is welcome to take over.

    I don't claim to know which move is best.  I did make a mistake in 16w in the last line I proposed.  Perhaps I am capable of correcting that and reaching a position in which Silver is not obviously winning.  However I am not capable of getting in the last word, so I'll let it pass, turn in my vote, and go to bed.


    Fritz I certainly respect your opinion and I understand that we all make mistakes. My goal isn't to try and get the last word in, simply to be able to express my opinion and generate as much discussion as possible. I think we all should be glad that there has been a much higher level of forum activity this move than there has been since the beginning of this game. I am sorry that this is not how you are seeing things.

    I would also like to apologize if any of my posts seemed arrogant at any time. Regarding the analysis tree I was not trying to suggest that it was your responsibility Fritz, or anyone elses to put together such a tree. It was a great service to the mob that you put it together in the first place and I hope my post is seen as expressing what a great idea and how helpful the tree was in my analysis and that of others. I was actually working on putting together an updated analysis tree myself this evening when you posted that you had updated the one in your previous post.

    And yes I did change my proposed 13b twice because the lines that you were willing to work out with me allowed me to hold a far better understanding of the position than that which I held when I first proposed these moves and I hope you came to understand the position better yourself because of these lines. As I said before we all make mistakes and I am glad that I have been able to learn so much this week from mine. I think I have learned more this week about arimaa than any previous one except for the week when I first joined the site. I only suggest that while we are all prone to making mistakes by having all of us examine a position rather than just one of us, as a group we are far less likely than chessandgo to make a blunder. I think this weeks analysis has shown this as well.

    Finally I admit that I have not spent my time analyzing all of the moves equally as perhaps I should. I did, however, spend time looking at all of the moves proposed by the mob. I would like you to remember, however, that before coming to support the double rabbit pull I proposed a move of my own which I came to feel was inferior and so chose to support this one which based on the lines that have been examined and the style of play that I prefer I like best.

    Well it is time for me to cast my ballot as well and go to bed. I am sure that whatever decision the mob makes it will be a good one.

    Title: Re: Move 12
    Post by Fritzlein on Aug 30th, 2007, 12:17am
    Fair enough.  I do recall that you proposed a different move before coming to like the double pull.  Come to think of it, I was the one that proposed the double pull before 99of9 made me suspicious of it.  It is ironic that I would criticize you for advocating your move too forcefully when in fact you were advocating my move!

    Also it is fair to say that a generally risky position may be more risky to chessandgo than to us, because we can help each other avoid blunders.  Sorry I focused only on the "look how many mistakes Fritz made" part of the sentence.  Objectively it seem reasonable to steer for complications in the hope that they advantage the team over the individual.

    I see you read and quoted my "get in the last word" comment before I thought the better of it and edited it out.  Let me try to get in the last word after all by apologizing for that.  There is nothing wrong with responding last, in the spirit of helping the team, to whatever analysis someone else has posted.  There's no denying how helpful your analysis has been, whether first, middle, or last.

    Title: Re: Move 12
    Post by mdk on Aug 30th, 2007, 12:22am

    on 08/30/07 at 00:03:01, Fritzlein wrote:
    Sorry, mdk, my frustration bubbled over there.  I could delete the post, but let me instead try to rephrase my frustration more constructively.

    There are two ways to contribute to the Mob playing well, namely discovering which move is best and advocating for that move so it gets the most votes.  Either by itself is futile.  If you discover the best move and nobody believes you that it is best, the Mob will play an inferior move.  If you fail to discover the best move and advocate for an inferior one, the Mob will be more like to vote incorrectly as well.

    It has been my perception, perhaps erroneously, that you have been focused on advocacy at the expense of discovery.  It seems that when you even looked at a move other than the double pull, you didn't look to see if that move kicked butt for Silver too, you looked only for a way to prove that the move would hand the advantage to Gold.  Meanwhile when you looked at the double pull, you didn't rack your brain for a way to refute it, only for a way to make it work to Silver's advantage.

    Maybe this isn't fair to you.  Certainly I sometimes fall into the trap myself, and try to prove what I already believe rather than trying to see how I might be wrong.  However, from all the times I have changed my mind in this thread, I hope it comes through that I am at least trying to be impartial.  When I advocate for or against a move, I hope I have at least tried to see both advantages and disadvantages to that move.  Yet it is difficult to stay in this impartial mode of discussion when I perceive (correctly or not) that you don't share the same aim, and I find myself biased to want to contradict you.

    Does this sound at all fair to you?  I shouldn't say that I don't want to maintain the analysis tree any more, because actually I enjoyed it.  The less enjoyable part, which has led me to suspend discussing moves to raise issues of process, is feeling that we aren't quite working with the same objective in contributing to the tree.

    Putting that to one side, I have very much enjoyed your voluminous and insightful contributions of analysis on this move.  I think we go into the voting this time better prepared than on any other move, and this difference has been your participation.  I hope that my candor here doesn't discourage you from participating actively in the future as well, because your analysis has been fantastic.


    Well your candor won't discourage me from participating but school starting this coming Wednesday and college applications (to a ridiculous number of schools... 15 at last count) and scholarship applications might.

    And honestly I find it more interesting when you choose to contradict my lines as no one else in the mob has. It forces me to take a second (or third or fourth) look at them and see if there is a mistake and if not how i need to continue and it benefits the mob as a whole. Again I would say that during the course of the week I have changed my mind about which move to advocate although I guess it isn't quite so apparent in the forum.

    Title: Re: Move 12
    Post by mdk on Aug 30th, 2007, 12:39am
    finally i should mention as i think i forgot to do so that i only was asking for the analysis tree as i thought it would be helpful to the mob if one complete with everything mentioned in the forum was compiled into a single post prior to the voting to help all the mob members make the most informed vote possible. i would have done it myself... and started to as i said previously had i had time at that moment you just beat me to it. perhaps in the future it would be best to put the tree in the wiki so that everyone can easily edit it without making the thread extraordinarily long. of course we would  have to ask chessandgo not to look at the tree in the wiki.

    Title: Re: Move 12
    Post by UruramTururam on Aug 30th, 2007, 5:05am
    Uff... The analysis is so deep that I can only wish C&G would replay to our move with something completely unpredicted.  ;)

    Title: Re: Move 12
    Post by RonWeasley on Aug 30th, 2007, 11:48am
    Thank you to those who noticed that I forgot to include the credible ed4w ha3n Ra2n ha6s horse advance.  It got a little bit of analysis and was considered quite playable.  So I'm starting the election over with this move included.  7 voted in the aborted election and the top two moves were at 4-3, so this is a close one.  Sorry about that.

    Let me state that I think the game tree is very valuable in that one can follow the rationale for liking, or not, various move proposals.  The tree can be reused in future moves if gold chooses one of the analyzed responses.  It seems worth maintaining even though it requires some work.

    On advocating or discrediting a particular move, I am in favor of both practices.  Either way, the reasoning exposes information about the position and I have benefitted from this discussion.  Disagreeing about a move is makes TheMob more dangerous, hopefully to our opponent.  Thank you for disagreeing!

    Title: Re: Move 12
    Post by 99of9 on Aug 30th, 2007, 7:35pm
    Thanks for putting that one back in Ron... it turned out to be my top vote!  I think the horse advance is marginally better than the defensive dog advance because it starts threatening new things if his elephant were ever to leave, whereas the dog advance is only to protect our camel (which I don't think is in actual danger).

    Title: Re: Move 12
    Post by 99of9 on Aug 30th, 2007, 7:51pm

    on 08/28/07 at 14:16:20, Fritzlein wrote:

    12b hh3n Rh2n hh4n Rh3n (pull rabbit twice)
    .    13w Eb5s rb6s Eb4e rb5s (pull rabbit twice, blocking off c4)
    .    .    13b ha6s ha5s md6w ed4s
    .    .    .    14w Db3s rb4s rb3w Hc3w
    .    .    .    .    14b hh5n Rh4n hh6w Rh5n
    .    .    .    .    .    15w Ec4e Ed4e Ee4n Ee5e


    When playing through the double-pull line you guys have done to great depth... this was my continuation.  To me it seems hard for silver to gain the advantage, and if we try, it looks like we might end up in an EMH attack.

    Title: Re: Move 12
    Post by mdk on Aug 30th, 2007, 11:36pm

    on 08/30/07 at 19:51:07, 99of9 wrote:
    When playing through the double-pull line you guys have done to great depth... this was my continuation.  To me it seems hard for silver to gain the advantage, and if we try, it looks like we might end up in an EMH attack.


    I think that this 14w is actually far better for silver than at first glance because silver's elephant doesnt have the mobility it appears to have and ill try to add a bit more analysis but im pretty tired right now and may just go to sleep.

    [edit]alright here is a bit
    12b hh3n Rh2n hh4n Rh3n (pull rabbit twice)
    .    13w Eb5s rb6s Eb4e rb5s (pull rabbit twice, blocking off c4)
    .    .    13b ha6s ha5s md6w ed4s
    .    .    .    14w Db3s rb4s rb3w Hc3w
    .    .    .    .    14b hh5n Rh4n hh6w Rh5n
    .    .    .    .    .    15w Ec4e Ed4e Ee4n Ee5e
    .    .    .    .    .    .    15b mc6w mb6s dd8s rg7e

    and gold has a problem as if gold's phant moves west then silver captures the rabbit and if it doesnt then the camel can drag the gold horse leaving gold with other issues that if ignored will lead to problems greater than the loss of a rabbit. there are other lines involving gold attacking the silver camel earlier on that make this potential response for 13b look nonoptimal rather than lines such as these. there are other moves for gold for 14w that worry me far more than this one does, as i believe this line leads us to a clear positional advantage for gold, although others may disagree.

    Title: Re: Move 12
    Post by RonWeasley on Aug 31st, 2007, 8:39pm
    Some Death Eater put a jinx on the power to the Hogwarts server.  My e-mail containing the voting administrator's page has been made unavailable.  As soon as I learn that power has been restored, I will get the results of the vote and make the move.  Until then, don't panic and don't say his name aloud.

    Hopefully tomorrow morning.  It takes about 30 minutes for me to apparate from The Burrow to Hogwarts.

    Title: Re: Move 12
    Post by mdk on Aug 31st, 2007, 8:59pm

    on 08/31/07 at 20:39:58, RonWeasley wrote:
    Some Death Eater put a jinx on the power to the Hogwarts server.  My e-mail containing the voting administrator's page has been made unavailable.  As soon as I learn that power has been restored, I will get the results of the vote and make the move.  Until then, don't panic and don't say his name aloud.

    Hopefully tomorrow morning.  It takes about 30 minutes for me to apparate from The Burrow to Hogwarts.


    Are you sure its not flying your broom that takes that long? It's rather difficult to apparate into Hogwarts ;)

    Title: Re: Move 12
    Post by Fritzlein on Sep 1st, 2007, 12:13am
    In addition to which apparition is instantaneous.  Are you sure that it's the real Ron Weasley moderating this forum, not someone transmogrified by Polyjuice Potion?

    Title: Re: Move 12
    Post by RonWeasley on Sep 4th, 2007, 7:50am
    The curse has been lifted.  Ministry people had to fix our power.  Then we needed house elves to boot the servers and we needed Filch to boot the house elves.  The hit on the time reserve is unfortunate, but that's one reason why the reserve is there.

    TheMob chooses the double R pull.  We are idiot simple!  With all that time only 11 votes, so the other six mobsters must be playing with their flobberworms.

    Title: Re: Move 12
    Post by Fritzlein on Sep 4th, 2007, 8:25am

    on 08/29/07 at 17:11:06, 99of9 wrote:
    Yep, I think you're right Fritz, let's vote ;-).  Gee I love that you're bold enough to risk statements like that.

    I told you so!  If we had just had a snap vote for the rabbit pull, we would have saved a week of reserve.  ;)

    Title: Re: Move 12
    Post by Fritzlein on Sep 4th, 2007, 8:32am
        .    .    .    .    .    .    .    .    .    .    1.  2.  3.  4.  5.  6.  7.  8.
    1. hh3n Rh2n hh4n Rh3n (pull R twice)  .    .    .     -   6   7   8   8  10  11  11
    2. hh3n Rh2n ed4w dd8s (pull R, e to c4, d to d7).     5   -   5   7   8   9  10   9
    3. hh3n Rh2n ed4w ha6s (pull R, e to c4, h to a5).     4   6   -   4   5   6   8   7
    4. ed4w ec4n Hc3n md6e (block E from center, m to e6)  3   4   6   -   4   7   8   7
    5. ed4w ec4n Hc3n dd8s (block E from center, d to d7)  3   3   5   3   -   6   7   6
    6. hh3n Rh2n ed4w rb6e (pull R, e to c4, r to c6).     1   1   5   4   4   -  10   9
    7. hh3n Rh2n ed4w ha6n (pull R, e to c4, h to a7).     0   1   3   3   4   1   -   8
    8. ed4n ed5w rb6e ha6e (e to c5, r to c6, h to b6)     0   2   3   2   2   2   3   -


    This is our closest vote yet, not only with a one-vote margin at the top, but with a three-way circular preference for second through fourth.  Hopefully that is because all of the top four moves would have given us a slight advantage.

    Title: Re: Move 12
    Post by UruramTururam on Sep 4th, 2007, 8:42am
    When looking at the matrix above I have an impression that most of the voters thought like this:
    "I prefer moves X,Y... I don't like many others. So I place the silly 'Pull-The-Wabbit' move just between the ones I like and the ones I don't.".

    Title: Re: Move 12
    Post by Fritzlein on Sep 4th, 2007, 3:53pm
    On the contrary, UT, if you were looking at my ballot (or at my ordering of the analysis tree), you would have seen that I put the "block the E from the center" moves in between the moves that I liked and the moves that I didn't like.

    More generally, I expect everyone put the moves that they felt neutral about in between the moves they favored and the moves they disfavored.

    More generally still, I expect everyone ranked the moves from best to worst.  That's a good thing, isn't it?

    Even if pulling the rabbit was the #1 choice of nobody (which you seem to imply, and which I expect is not true), even then we know one thing for sure: if pulling the rabbit had been in a two-way vote with any other move the majority would have wanted to pull the rabbit.  That's the beauty of Condorcet voting.

    Title: Re: Move 12
    Post by UruramTururam on Sep 4th, 2007, 4:50pm
    I don't think that absolutely no-one put the double pull first. But it is possible. Yet surely no-one put it near the bottom.

    The two general thoughts you present are of course true, but this case where there is no "absolute" winner the preferences on the lower locations became more and more important. By the way I think that sometimes analyzing specific cases could probably be better than generalizing up to obviousness... Of course I am not a mathematician!  

    And - yes - the Condorcet method is something like selecting "the smallest evil", looking for the compromise. What is really counterintuitive here, and worth our investigation  - the safe solutions not always win! This is the case of safe solution winning, but for the previous move or move 9 aggressive moves won. However we should take into account that gamers usually like taking risks.



    Arimaa Forum » Powered by YaBB 1 Gold - SP 1.3.1!
    YaBB © 2000-2003. All Rights Reserved.