Arimaa Forum
        (http://arimaa.com/arimaa/forum/cgi/YaBB.cgi)
      

        Arimaa >> General Discussion >> Opening Strategy 1
        
(Message started by: myopicblur on Jul 24th, 2009, 2:51pm)
Title: Opening Strategy 1
Post by myopicblur on Jul 24th, 2009, 2:51pm
For the following openings:
Silver
r r r  d  d  r r r
r h c e m c h r

------------------

RHRMERHR
RCRDDRCR
Gold

Suggest the best opening responses (8 turns) for both sides

for

a. exactly the same as the above initial position
b. as the above initial positions, but the colours are interchanged.
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 24th, 2009, 3:00pm
and discuss which strategies are preferable for both sides.
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 25th, 2009, 12:35am
Perhaps I have not stated clearly what I intended to discuss in this thread. The title of this thread is "Opening Strategy", thus I suppose fellow players should suggest moves by explaining its intention of each moves.

Let's discuss move 2 first, for gold.
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 25th, 2009, 12:55am
i.e. Move 2g.

I thought of advancing the e-side H to b3, E to d5. But is that a feasible choice for move 2g?

Not sure about the intention of this move... is that a move intended for launching EH attack?

For this particular move 2g, what would be the appropriate move 2s then? Feel free to discuss.
Title: Re: Opening Strategy 1
Post by Fritzlein on Jul 25th, 2009, 6:38am
What do you think of the strategies discussed in the Arimaa Wikibook (http://en.wikibooks.org/wiki/Arimaa)?
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 25th, 2009, 9:23am

on 07/25/09 at 06:38:02, Fritzlein wrote:
What do you think of the strategies discussed in the Arimaa Wikibook (http://en.wikibooks.org/wiki/Arimaa)?


Good for an introduction--- but here I would like to discuss some details--- especially for the later stage of the openings.

I consider Move 2s: ed7s ed6e hg7s hb7s, to counter the EH attack.
Is that a suitable response for the aforementioned move 2g?
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 25th, 2009, 9:28am
or move 2s: hg7s ed7s ed6s ed5s. Initiating an attack.

Seemingly a better move than the one mentioned in the above post.

then move 3g: Md2e Me2n De1n Dd1n.

Please comment on my choices.
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 25th, 2009, 9:39am
By move 3g following the above line, 3s has an option to initiate a EH attack, by hb6s hb5s ed4s md7s.

then 4g: Me3e Hb2n Cb1n Cg1n. This move 4g intends to slow down the attack by silver, on the other hand trying to develop attack. At this point it often gets me confusing (if I am the gold player)

For this 4g, 4s may push the c2 rabbit to north, by Dd2s ed3s Rc2n ed2w.

I am not sure how 5g should respond... let experienced players comment on all the above suggestions.

I'll now go for a sleep.
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 25th, 2009, 9:50am
Just read the following, feel free to comment on the following quoted comment:


on 11/18/08 at 21:06:00, 99of9 wrote:
Off-balance is much riskier for gold than it is for silver, because the silver player gets a chance to start exploiting our imbalance during their setup.  For example, if we put the camel on b2, Fritz can put his elephant directly opposite on b7 (and his horses far away).

If we are playing gold, I favour a roughly symmetrical setup of some kind.
Title: Re: Opening Strategy 1
Post by Fritzlein on Jul 25th, 2009, 11:20am
OK, I understand now.  I was going to say that beginners should not worry about specific opening moves.  It is much better to first study basic capture and goal tactics, and next to study the elements of strategy such as camel hostages, elephant blockades, frames, etc.  It is no good to talk about specific moves until you have learned enough to understand why those specific moves might be good or bad.

I see, however, that you are not a beginner.  You don't play like a beginner, and you don't talk like a beginner.  You don't make common beginner blunders against CluelessP1 and BombP1, and also you are clearly familiar with all the basic strategic elements.  If there were Arimaa clubs all over the world, I might suspect you were a club player joining arimaa.com for the first time, but there are no clubs.  Clearly you have played on arimaa.com before under a duplicate account, but now wish to remain anonymous.  Let me just say that I would have appreciated if you would have been open about wanting to remain anonymous, like aaaa has done, instead of feeding us a fake identity.

Since you have good grasp of the game already, you would be ready for a more advanced discussion.


on 07/25/09 at 00:55:00, myopicblur wrote:
i.e. Move 2g.

I thought of advancing the e-side H to b3, E to d5. But is that a feasible choice for move 2g?

Not sure about the intention of this move... is that a move intended for launching EH attack?

For this particular move 2g, what would be the appropriate move 2s then? Feel free to discuss.

I think that you must mean 2g elephant to e5, not to d5, because later you suggest a 2s for Silver that moves her elephant through d5.  When gold advances his elephant three steps, it is more common to advance the horse on the same flank as the elephant, in order to be threatening the opposite trap in four steps.  After 2g Ee2n Ee3n Ee4n Hb2n, Gold would need five steps to get both elephant and horse next to the c6-trap, and five steps to get both elephant and horse next to the f6-trap.  So instead of one threat in one turn, Gold has two threats in two turns, which is less forcing.  It would be much more critical to play  2g Ee2n Ee3n Ee4n Hg2n, with an immediate threat to attack the f6-trap in only four steps, which is why this opening move 2g is the most popular among experts.

The long line you have analyzed for Silver's response begins with the counter-attacking move 2s hg7s ed7s ed6s ed5s, i.e. a 180 degree symmetrical rotation of Gold's 2g Ee2n Ee3n Ee4n Hb2n.  This is not bad.  Lots of moves for Silver are not bad in that situation because Gold's 2g didn't put on any pressure.

If, however, Gold opens with the more critical move 2g Ee2n Ee3n Ee4n Hg2n, then Silver should definitely not open with the 180 degree symmetrical move of 2s hb7s ed7s ed6s ed5s.  Gold will simply complete the attack on the f6-trap with 3g Ee5n Hg3n Hg4n Hg5n.  Silver can't respond by attacking the c3-trap, because she will only win a rabbit from c2 while Gold is winning a cat from f7.

Indeed, one of the reasons for Gold to set up with rabbits behind his traps is to prevent Silver from making a rotationally symmetrical attack.  If Silver responds to the Gold setup by placing cats behind her traps, it is not a mistake, but it means she is giving up the option of immediate counter-attack, and must defensively answer 2g Ee2n Ee3n Ee4n Hg2n with something like 2s hg7s ed7s ed6s ed5s or even 2s hg7s hb7s ed7s ed6s


on 07/25/09 at 09:50:54, myopicblur wrote:
Just read the following, feel free to comment on the following quoted comment:

Oops, I clicked on the link to see where 99of9 had made this comment, since I didn't remember hearing it before.  I didn't realize that it was part of the Mob 2009 discussion until after I opened the thread.  My apologies to the Mob; I trust this accidental glance at an old thread won't give me an unfair advantage.


Quote:
Off-balance is much riskier for gold than it is for silver, because the silver player gets a chance to start exploiting our imbalance during their setup.  For example, if we put the camel on b2, Fritz can put his elephant directly opposite on b7 (and his horses far away).

I agree with 99of9's analysis.  If he is correct, it might someday prove that Gold does not have an advantage in Arimaa because Gold can't afford an unbalanced setup while Silver can afford it.

Here is the funny thing: When I am Gold I have started experimenting with setting up a flank camel in spite of the risk 99of9 identifies, but nobody as Silver seems willing to set up the Silver elephant directly opposite!  We are so entrenched in believing that the elephant must be set up in the center that nobody will set up a flank elephant even when I strategically invite it.  Therefore I have to contradict 99of9's theoretical analysis with a practical truth.  In reality there is no risk for Gold to set up with a flank camel because Silver will never respond with an elephant on the same flank.  ::)
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 25th, 2009, 8:13pm

on 07/25/09 at 11:20:46, Fritzlein wrote:
I see, however, that you are not a beginner.  You don't play like a beginner, and you don't talk like a beginner.  You don't make common beginner blunders against CluelessP1 and BombP1, and also you are clearly familiar with all the basic strategic elements.  If there were Arimaa clubs all over the world, I might suspect you were a club player joining arimaa.com for the first time, but there are no clubs.
 

There is a small but private hobby group (i.e. not a club) of around 10 individuals in HK, which play and discuss abstract games frequently. I suspect the same goes for Omar as well.

Arimaa was never a popular game among us, we treated Arimaa as a "chess variant".

The popularity peaked on 2008, when Karl's image from Dallas was posted posted onto the wall of common room. That particular image of Karl was removed after 2 months.


Quote:
Clearly you have played on arimaa.com before under a duplicate account, but now wish to remain anonymous.  Let me just say that I would have appreciated if you would have been open about wanting to remain anonymous, like aaaa has done, instead of feeding us a fake identity.


Not really, but I would like to remain anonymous here.


Quote:
I agree with 99of9's analysis.  If he is correct, it might someday prove that Gold does not have an advantage in Arimaa because Gold can't afford an unbalanced setup while Silver can afford it.

Here is the funny thing: When I am Gold I have started experimenting with setting up a flank camel in spite of the risk 99of9 identifies, but nobody as Silver seems willing to set up the Silver elephant directly opposite!  We are so entrenched in believing that the elephant must be set up in the center that nobody will set up a flank elephant even when I strategically invite it.  Therefore I have to contradict 99of9's theoretical analysis with a practical truth.  In reality there is no risk for Gold to set up with a flank camel because Silver will never respond with an elephant on the same flank.  ::)


You may have to demonstrate that by real game examples.
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 25th, 2009, 8:16pm

on 07/25/09 at 11:20:46, Fritzlein wrote:
I think that you must mean 2g elephant to e5, not to d5, because later you suggest a 2s for Silver that moves her elephant through d5.  When gold advances his elephant three steps, it is more common to advance the horse on the same flank as the elephant, in order to be threatening the opposite trap in four steps.


Thanks for correcting the mistake.


Quote:
After 2g Ee2n Ee3n Ee4n Hb2n, Gold would need five steps to get both elephant and horse next to the c6-trap, and five steps to get both elephant and horse next to the f6-trap.  So instead of one threat in one turn, Gold has two threats in two turns, which is less forcing.

It would be much more critical to play  2g Ee2n Ee3n Ee4n Hg2n, with an immediate threat to attack the f6-trap in only four steps, which is why this opening move 2g is the most popular among experts.


Move 2g should be Ee2n Ee3n Ee4n Hg2n. My original post was unclear, which caused your confusion.


Quote:
The long line you have analyzed for Silver's response begins with the counter-attacking move 2s hg7s ed7s ed6s ed5s, i.e. a 180 degree symmetrical rotation of Gold's 2g Ee2n Ee3n Ee4n Hb2n.  This is not bad.  Lots of moves for Silver are not bad in that situation because Gold's 2g didn't put on any pressure.

If, however, Gold opens with the more critical move 2g Ee2n Ee3n Ee4n Hg2n, then Silver should definitely not open with the 180 degree symmetrical move of 2s hb7s ed7s ed6s ed5s.  Gold will simply complete the attack on the f6-trap with 3g Ee5n Hg3n Hg4n Hg5n.  Silver can't respond by attacking the c3-trap, because she will only win a rabbit from c2 while Gold is winning a cat from f7.


Again, not hb7s, but hg7s on my post.


Quote:
Indeed, one of the reasons for Gold to set up with rabbits behind his traps is to prevent Silver from making a rotationally symmetrical attack.  If Silver responds to the Gold setup by placing cats behind her traps, it is not a mistake, but it means she is giving up the option of immediate counter-attack, and must defensively answer 2g Ee2n Ee3n Ee4n Hg2n with something like 2s hg7s ed7s ed6s ed5s or even 2s hg7s hb7s ed7s ed6s


read.
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 25th, 2009, 8:30pm

on 07/25/09 at 09:39:29, myopicblur wrote:
By move 3g following the above line, 3s has an option to initiate a EH attack, by hb6s hb5s ed4s md7s.

then 4g: Me3e Hb2n Cb1n Cg1n. This move 4g intends to slow down the attack by silver, on the other hand trying to develop attack. At this point it often gets me confusing (if I am the gold player)

For this 4g, 4s may push the c2 rabbit to north, by Dd2s ed3s Rc2n ed2w.

I am not sure how 5g should respond... let experienced players comment on all the above suggestions.


How about this line? I am on a plan window to see how 5g should work.

The following: Silver to move. 4s.

http://farm3.static.flickr.com/2548/3756451073_0692fa931d_o.jpg
Title: Re: Opening Strategy 1
Post by Fritzlein on Jul 25th, 2009, 10:02pm

on 07/25/09 at 20:13:48, myopicblur wrote:
There is a small but private hobby group (i.e. not a club) of around 10 individuals in HK, which play and discuss abstract games frequently. [...] I would like to remain anonymous here.

Thank you for admitting you are not a young Japanese woman.  With that I will respect your anonymity.


on 07/25/09 at 20:30:13, myopicblur wrote:
How about this line? I am on a plan window to see how 5g should work.

Has this position ever occurred in one of your games?  If it has occurred once, it seems very unlikely to occur again.  If it has not yet happened, it seems very unlikely to ever occur for the first time.  Unless you are playing a bot that always does the same thing, one human or the other will have deviated before reaching this position.

The trouble with Arimaa opening theory is that the branching factor is too high.  You can study a certain position and memorize the perfect move, but what good is that knowledge if the position never arises?

In the position you give, Silver has already misplayed.  Her attack is premature and toothless.  The elephant and horse together can't force a capture in the c3-trap for several more turns.  Since Silver can't capture in c3, what is she really threatening?  Her strongest immediate threat is to flip the gold dog from d2 to d4, taking it back to Silver's side as a hostage.  But for this threat the advanced silver horse is unnecessary, and would in fact be better placed on b6, defending the c6-trap.

Because Silver's attack is so slow, Gold could choose ignore it completely with 5g Ee5n Ee6w Ed6n Rh2n.  However, this seems to allow Silver to retreat without disadvantage (5s hb4n hb6n cf7e me7e).  Therefore a more critical line is probably threatening Silver's exposed horse with 5g Ee5w Ed5w Ec5s Ra2n.  Silver would not like to prevent the horse frame with a rabbit advance, since the advanced rabbit would itself become a target, and preventing the horse frame with 5s ed3e Dd2n Dd3n ee3w fails to 6g Dd4e De4s De2w Ra3n, so only 5s cc7w cb7s cb6s hb4w remains.  But this exposes the silver cat, which can be persecuted with 6g Ec4n cb5s Ec5w Dd2s.  Gold has the upper hand, and Silver would be happy to escape with winning a rabbit in exchange for her forlorn cat.
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 25th, 2009, 10:36pm

on 07/25/09 at 22:02:56, Fritzlein wrote:
The trouble with Arimaa opening theory is that the branching factor is too high.  You can study a certain position and memorize the perfect move, but what good is that knowledge if the position never arises?


You may have misinterpreted my intention here. I expected the branching factor would be too high that we need several threads to discuss some cases that most Arimaa players may not realise at the moment. Not intend for one to memorise the move--- but to realise the strategy behind whilst understand the possibility.

There may be some interesting line added. Perfect moves was intended as an introduction to this (expected) heated discussion of strategy.


Quote:
In the position you give, Silver has already misplayed.  Her attack is premature and toothless.  The elephant and horse together can't force a capture in the c3-trap for several more turns.  Since Silver can't capture in c3, what is she really threatening?  Her strongest immediate threat is to flip the gold dog from d2 to d4, taking it back to Silver's side as a hostage.  But for this threat the advanced silver horse is unnecessary, and would in fact be better placed on b6, defending the c6-trap.

Because Silver's attack is so slow, Gold could choose ignore it completely with 5g Ee5n Ee6w Ed6n Rh2n.  However, this seems to allow Silver to retreat without disadvantage (5s hb4n hb6n cf7e me7e).  Therefore a more critical line is probably threatening Silver's exposed horse with 5g Ee5w Ed5w Ec5s Ra2n.  Silver would not like to prevent the horse frame with a rabbit advance, since the advanced rabbit would itself become a target, and preventing the horse frame with 5s ed3e Dd2n Dd3n ee3w fails to 6g Dd4e De4s De2w Ra3n, so only 5s cc7w cb7s cb6s hb4w remains.  But this exposes the silver cat, which can be persecuted with 6g Ec4n cb5s Ec5w Dd2s.  Gold has the upper hand, and Silver would be happy to escape with winning a rabbit in exchange for her forlorn cat.


The position I have posted shows the position before move 4s. Silver can threaten to pull the c2 rabbit to c3, disrupting the position. Are you sure silver's attack "so slow" as you have interpreted on your last post? I wonder 4s Dd2s ed3s Rc2n ed2w would get gold into trouble. This results in a centralized R for gold.

http://farm4.static.flickr.com/3617/3756660211_e66ef0961f_o.jpg
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 25th, 2009, 11:01pm
At the above position, it seems to me if silver is "misplayed", then gold is more "misplayed" at that awkward position with R being flipped that way.
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 25th, 2009, 11:13pm
(Not the same line as above)

See game 111916 for an attempt by silver against bot.
Title: Re: Opening Strategy 1
Post by Fritzlein on Jul 26th, 2009, 7:36am

on 07/25/09 at 23:01:56, myopicblur wrote:
At the above position, it seems to me if silver is "misplayed", then gold is more "misplayed" at that awkward position with R being flipped that way.

I beg your pardon, I thought it was Gold to move, not Silver to move.  I didn't see the diagram caption.  I agree that with Silver to move in the diagram, it is probably Silver who is winning.

Perhaps my confusion about the diagram is excusable given that it doesn't follow from the moves you give in text.  You clarified that you intended the opening moves

2g Ee2n Ee3n Ee4n Hg2n
2s ed7s ed6s ed5s hg7
3g Md2e Me2n De1n Dd1n

But then the only move you give for Silver is

3s hb6s hb5s ed4s md7s

and you quote this move from yourself as if to repeat it leading in to the diagram you pasted, but in fact Silver has no camel on d7 and no horse on b6.  Even if we suppose that you meant to use some different move 2s for Silver, it would be impossible for Silver to have a horse on b6, an elephant on d4, and a camel on d7.  This is why I skipped to the diagram without understanding who had made what moves, and whose turn it was.  Again, I am sorry I didn't read the caption correctly.


Quote:
You may have misinterpreted my intention here. I expected the branching factor would be too high that we need several threads to discuss some cases that most Arimaa players may not realise at the moment. Not intend for one to memorise the move--- but to realise the strategy behind whilst understand the possibility.

You may have misinterpreted my objection.  I agree that studying any Arimaa position, whether it has ever happened or ever will, can teach something.  I could randomly select five positions from the game database, post them in the Forum, and if everyone discussed them we would learn.

You seem to be saying that the positions you propose are of particular interest, i.e. that they are not random positions, but my point is that the multitude of options makes the position you illustrated, in fact, essentially random.  We can't cover the possibilities up to move 4s in "several threads".  Thousands of threads, perhaps, or millions, but not "several".

So while the diagrammed position may be useful to discuss in the way that any position is useful to discuss, I doubt that it has any special status.  Before the diagram, Gold's move 3g Md2e Me2n De1n Dd1n is far too passive to deserve consideration, to say nothing of the explosion of alternatives on 3s and 4g.

In the past it was popular for Gold to use 3g to try to get either a rabbit or horse pulled out with his elephant, for example 3g Ee5e Ef5e Eg5e hg6s, basically insuring a slight lead in the rabbit-pulling race.  Yet if Gold intended to enter a rabbit-pulling race, why did he open with four rabbits forward?  Perhaps Gold should have deviated on move 1g, not on 3g.

An alternative idea is to try to force the Silver camel to commit to one wing before launching an elephant-horse attack on the other wing.  3g Ee5n puts pressure on the silver camel and allows Gold three other developing steps.  But apart from Hb2n it is hard to know how to use those steps before the camel has committed, so there is some inefficiency in that plan.  If it was Gold's intention to force the Silver camel to commit and then launch an EH attack on the other wing, then Gold should probably have deviated with 2g Ee2n Ee3n Ee4n Ee5n.

A forcing alternative is 3g Ee5n Ee6w me7s Hb2n, choosing the east for the Silver camel with the intention of an EH attack in the west.  But Silver has many ways to play a safe defense, and also the ambitious 3s Hb7s Dd8s De8s Me6w.  We really don't know whether the camel is better placed in the f6-trap than on e7.

I could go on suggesting moves, but the strong probability is that anything I suggest for 3g at present will be in the rubbish bin of history a year from now, even if we have the same 1g, 1s, 2g, 2s leading to this position, which we probably won't.  Talking about 4g and 4s with the expectation that those positions will occur in practice is a wild leap of the imagination.

The problem is that we understand neither the value (positive or negative) of an elephant holding a horse hostage, nor the value (positive or negative) of advanced rabbits, nor the value (positive or negative) of an unbalanced setup with both horses on one flank and the camel on the other.  The fact that a flank camel might be good when the horses are unbalanced raises the possibility that even when horses are balanced it is perhaps better to have a flank camel than a centralized camel as conventional wisdom prefers.  If our understanding of these four factors changes, it will completely re-write any opening theory we discuss now.  (Furthermore, those aren't the only questions.  None of your five threads even has the opening position I am currently intrigued by, namely with two rabbits forward on one flank and none on the other.  If you play against me in next year's Postal Mixer, you will have to throw away all these threads you started, and think about move 1s from scratch.  The following year I will probably use yet another new opening.)

Let us suppose that I not only suggest a specific move, but also, exactly as you say, "realise the strategy behind".  Let's say a certain move by Gold has the strategy to pull out a rabbit.  But what if next year we realize that pulling this rabbit is a bad idea in that exact situation?  We can spin our wheels talking about how Gold can try to force a rabbit pull and how Silver can prevent that rabbit pull, but that will be missing the point because the real question is when the rabbit pull is a good idea and when it isn't.  Without the deeper understanding that none of us have yet, our discussion of specific moves will be misguided.  The same applies to elephant-horse attacks, unbalancing forces, and centralizing/decentralizing the camel, etc.

There are myriad answers to the question, "What can we do in the opening?"  I can give you as many as you like.  On the other hand, we don't know the answers to the question, "What should you do in the opening?"  Yes, there are some moves that I feel are clearly bad, so I don't think we lose much by narrowing down the discussion to discard those.  But we can by no means narrow the discussion enough to get the branching factor under control and feel confident that the specific positions we arrive at are the relevant ones.  Trying to map out opening moves is relatively uninteresting and futile compared to trying to answer the unknowns that would motivate those opening moves.

May I suggest that you participate in the Mob 2009 game?  The current board position is admittedly essentially random, and it will never occur again, but it has the virtue that a dozen people are motivated to discuss it with you.  The position is relevant because it is the one the Mob will have to understand in order to beat me.  The discussion there will not just be about specific moves, but also about the ideas behind the moves.  In terms of illuminating general principles of Arimaa strategy, it will probably be more useful to you than any position on 2g you could propose.
Title: Re: Opening Strategy 1
Post by omar on Jul 28th, 2009, 9:42am
Trying to propose the best moves even from a fixed initial setup is a really difficult task. I would be afraid to propose anything specific. When I reflect on how I play openings, I guess I am thinking more about what I am trying to achieve; do I want to chase the opponent's camel and try to take it hostage, do I want to try an elephant horse attack on the east wing, do I want to pull out a rabbit, or maybe just play passive and wait for the opponent to make a mistake. At present I don't know what would be a good way to formally study opening strategies.
Title: Re: Opening Strategy 1
Post by myopicblur on Jul 28th, 2009, 10:48am

on 07/28/09 at 09:42:41, omar wrote:
Trying to propose the best moves even from a fixed initial setup is a really difficult task. I would be afraid to propose anything specific. When I reflect on how I play openings, I guess I am thinking more about what I am trying to achieve; do I want to chase the opponent's camel and try to take it hostage, do I want to try an elephant horse attack on the east wing, do I want to pull out a rabbit, or maybe just play passive and wait for the opponent to make a mistake. At present I don't know what would be a good way to formally study opening strategies.


(Currently I'm busy at work, therefore just a brief reply for Omar)

Actually my proposed threads for opening strategy stemmed from the one from chess. Chess openings are categorised under codes A00-E99, and they are intensively studied by chess players. I suppose a systematic methodology of studying strategy would be beneficial to Arimaa Community as a whole. Arimaa is a new game, with a much narrower player spectrum. As Arimaa Community grows, I expect the motivation of our community for such study would be larger. Starting from one thread, to multiple threads, then to a subforum.
Title: Re: Opening Strategy 1
Post by Fritzlein on Jul 28th, 2009, 2:51pm

on 07/28/09 at 10:48:46, myopicblur wrote:
Chess openings are categorised under codes A00-E99, and they are intensively studied by chess players. I suppose a systematic methodology of studying strategy would be beneficial to Arimaa Community as a whole.

I look forward to seeing what classification system is produced by you and everyone who joins you in pursuing this methodology.

Some previous discussion on the topic:

http://arimaa.com/arimaa/forum/cgi/YaBB.cgi?board=talk;action=display;num=1106048769

http://arimaa.com/arimaa/forum/cgi/YaBB.cgi?board=talk;action=display;num=1204896182

http://arimaa.com/arimaa/forum/cgi/YaBB.cgi?board=talk;action=display;num=1202228829
Title: Re: Opening Strategy 1
Post by chessandgo on Aug 3rd, 2009, 12:35pm
Hi myopicblur,

Just a quick reply to agree with my fellow "old" (;)) arimaa players. It's great to discuss any arimaa position, so I encourage you to continue, and agree that the best opportunity right now is the Mob game.

The opening position are not that good to analyse extensively as has been said. I've played chess very seriously for about half of my (short) life, and have studied extensively openings as everyone, but as others have pointed out it's not doable in arimaa. At chess there are, say, about 3-4 candidate moves for every decision, while at arimaa there might be 20-30 very serious candidates, even much more sometimes taking into account the combinatorics of steps which can be played one move or the next indifferently.

So I would advise you as well to put more thoughts into midgame turning points of actual games, which is very far more rewarding than an extensive analysis of openings I think.

This said of course very respecfully, in any case have fun with arimaa the way(s) you prefer, and see you sometimes over the board I hope!

Jean
Title: Re: Opening Strategy 1
Post by myopicblur on Aug 8th, 2009, 1:34am

on 07/26/09 at 07:36:51, Fritzlein wrote:
You may have misinterpreted my objection.  I agree that studying any Arimaa position, whether it has ever happened or ever will, can teach something.  I could randomly select five positions from the game database, post them in the Forum, and if everyone discussed them we would learn.


Quoted. I suppose the top players should take the lead on that. Perhaps you may consider starting your thread--- one position a day/ a week/ a month, etc.


(I am still replying earlier posts at the moment)
Title: Re: Opening Strategy 1
Post by myopicblur on Aug 8th, 2009, 1:43am

on 07/26/09 at 07:36:51, Fritzlein wrote:
A forcing alternative is 3g Ee5n Ee6w me7s Hb2n, choosing the east for the Silver camel with the intention of an EH attack in the west.  But Silver has many ways to play a safe defense, and also the ambitious 3s Hb7s Dd8s De8s Me6w.  We really don't know whether the camel is better placed in the f6-trap than on e7.


So we really don't know whether the camel is better placed in the f6-trap than on e7, but what would you suggest for evaluating such kind of positions, as you are top human players ? Perhaps you may suggest your own interested opening positions in the following of this thread.
Title: Re: Opening Strategy 1
Post by myopicblur on Aug 8th, 2009, 1:48am

on 07/28/09 at 14:51:22, Fritzlein wrote:
I look forward to seeing what classification system is produced by you and everyone who joins you in pursuing this methodology.

Some previous discussion on the topic:

http://arimaa.com/arimaa/forum/cgi/YaBB.cgi?board=talk;action=display;num=1106048769

http://arimaa.com/arimaa/forum/cgi/YaBB.cgi?board=talk;action=display;num=1204896182

http://arimaa.com/arimaa/forum/cgi/YaBB.cgi?board=talk;action=display;num=1202228829


Thanks for the link. My reply about those links will be given soon.

.
Title: Re: Opening Strategy 1
Post by myopicblur on Aug 8th, 2009, 1:53am

on 02/05/05 at 20:03:52, Fritzlein wrote:
The postal tournament can provide us some statistics on the opening theory as it stood in late January of 2005.  Of course I had to lump things into broad categories to do statistics, but it still gives us some idea. Of the 160 opening setups:

  • 126 had six rabbits back

    • 89 of these had the forward rabbits on the flanks
    • 37 of these had the forward rabbits behind the traps

  • 23 had all eight rabbits on the back rank
  • 9 had seven rabbits back

    • 4 of these had the forward rabbit on a flank
    • 5 of these had the forward rabbit behind a trap

  • 2 had four rabbits back

To judge our penchant for symmetry, I considered the elephant and camel to be identical.  If the two sides of the midline were mirror images, I called it a symmetrical setup.  If the two halves had equal forces but not in mirror image I called it a balanced setup.  If the two halves didn't have equal forces (e.g. both horses on one side, or camel and elephant on the same side) I called it an unbalanced setup. Of the 160 opening setups:

  • 91 were symmetrical
  • 36 were balanced
  • 33 were unbalanced

I also tried to categorize the opening moves, but it quickly gets hard to define irregular openings. Of the 80 first moves by Gold:

  • 55 moved the elephant up 4 steps
  • 9 moved the elephant up 3 steps and over 1
  • 5 moved the elephant up 3 steps and another piece up 1
  • 3 moved the elephant up 2 steps and two pieces up 1
  • 3 moved each of four pieces up one step
  • 5 were in categories of two or fewer occurances

Beyond the first move it totally gets out of control in trying to make categories.  I'm not sure opening theory in Arimaa will ever resemble opening theory in chess.


An attempt to categorise the first move in 2005.


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