Welcome, Guest. Please Login or Register.
Apr 28th, 2024, 5:39pm

Home Home Help Help Search Search Members Members Login Login Register Register
Arimaa Forum « Fixing the rules / eliminating draws »


   Arimaa Forum
   Arimaa
   General Discussion
(Moderator: supersamu)
   Fixing the rules / eliminating draws
« Previous topic | Next topic »
Pages: 1 2 3  Reply Reply Notify of replies Notify of replies Send Topic Send Topic Print Print
   Author  Topic: Fixing the rules / eliminating draws  (Read 12391 times)
Fritzlein
Forum Guru
*****



Arimaa player #706

   
Email

Gender: male
Posts: 5928
Re: Fixing the rules / eliminating draws
« Reply #30 on: Jul 9th, 2005, 8:31am »
Quote Quote Modify Modify

I haven't been following the games closely for a few weeks, but I'm guessing that if there had been a draw, someone would have mentioned it.  Thus we have again gone several thousand games with no draws.
 
When was the last near-draw?  The closest I can think of is Belbo vs. Adanac postal, and that wasn't actually very close to drawing.
 
My guess is that draws are most likely when the players start exchanging rabbits before pieces.  If there are lots of pieces left at the end and only a few rabbits, then there may be more possibility to sacrifice in exchange for the opponent's last rabbit or two.  On the other hand, if games start out with piece exchanges, which empties the board but leaves lots of rabbits on, then the game becomes even less drawish than at the start.
 
About a year ago 99of9 was prediciting a draw soon between top-level players as knowledge of strategy got more refined.  It seemed reasonable at the time, because top-level games were dominated by lone-elephant openings resulting in rabbit pulls.  Nowadays, not only is it more common to see pieces in addition to the elephant committed to an attack, but also it seems that lone-elephant attacks are more focused on the opposing heavy pieces than on rabbits.
 
Therefore it seems that, unless the style of game involving opposing rabbit pulls returns to prevalence, we may not see a top-level draw (or any draw at all) for years to come.
IP Logged

99of9
Forum Guru
*****




Gnobby's creator (player #314)

  toby_hudson  


Gender: male
Posts: 1413
Re: Fixing the rules / eliminating draws
« Reply #31 on: Jul 10th, 2005, 7:48am »
Quote Quote Modify Modify

Yes, I agree with that analysis Fritz.  The chance of a draw has definitely decresed since the era of rabbit-pulling has gone out of favour.  I of course am still of that mindset - so as soon as someone similar comes along we can have that draw!
IP Logged
Fritzlein
Forum Guru
*****



Arimaa player #706

   
Email

Gender: male
Posts: 5928
Re: Fixing the rules / eliminating draws
« Reply #32 on: Jul 10th, 2005, 2:22pm »
Quote Quote Modify Modify

on Jul 10th, 2005, 7:48am, 99of9 wrote:
he chance of a draw has definitely decresed since the era of rabbit-pulling has gone out of favour.  I of course am still of that mindset -

 
I have seen you punish the opening elephant-horse attack mercilessly on defense.  Somehow you always get the attacking horse framed (which shouldn't be as easy as you make it look) and somehow you convert the framed horse into a captured horse (which also shouldn't be as easy as you make it look).  Because of watching your games, I'm not comfortable with anything other than lone-elephant attacks in the opening, except against bots, of course.  The more multi-piece attacks you refute, the more converts you will make to your lone-elephant-rabbit-pulling style.
 
Interestingly, though, I noticed in my current postal game against RonWeasley that as soon as we traded a pair of horses, I was suddenly eager to launch an E+H attack and afraid he would be able to stage an E+H attack on me first.  I conclude that the balance between offense and defense in the opening position is so fine that  a single trade can violently tip the scales.  You have opined that a camel trade stabilizes the opening position, which seems reasonable in my judgement, but given this it is counter-intuitive how much a horse trade destablizes the opening position.
 
To stray off-topic, let me add that the more I learn about Arimaa strategy the more mystical it becomes to me.  I keep groping for easy rules of thumb, but the divergent effect between a camel trade and a horse trade as the first trade of the game doesn't yet fit into a larger picture for me.  I know what I feel about a position before I can explain why, which didn't happen to me much in chess.  Maybe that's just the difference between learning chess strategy from books and learning Arimaa strategy by playing, but I don't think that's the whole picture.  I think Arimaa is simply a more strategic game than chess.
 
Back to the topic of draws, I think that sentiment might well swing back in the direction of lone-elephant rabbit-pulling contests, because I've seen nothing yet to discredit this strategy.  Another possibility is that opening theory might lead to early camel trades, which would then force the game into rabbit-pulling mode.
 
Moreover, I'm not even convinced that any progress at all can be made against the perfect blend of defense and counter attack.  Perhaps all multi-piece attacks from the opening position are unsound.  Perhaps in dual-lone-elephant openings piece trades can always be avoided, and futher it might be that for one player to guarantee himself a rabbit pull, he has to give up a horse hostage, which isn't worth it.  Arimaa might then devolve into endless thrust and parry, a practical draw albeit not a rulebook draw.
 
On the other hand, the recent awareness of more attacking possibilities makes the game seem so rich and complicated that I doubt we can "solve" Arimaa any time soon.  For example, it might be that when a rabbit is being pulled on one wing, the E+H (or even E+M!) attack gains strength on that wing and can become sound.  Recall how aggressively Omar used his camel in the postal tournament to good effect, not to mention his recently beating BombP3 with a cat and two rabbits for a lost camel.  Our ideas of "lone-elephant rabbit pulling opening" and "multi-piece attack opening" don't stay conveniently separate from one another in practice.  In fact, I don't yet discount the E+M+H opening attack: BlackKnight almost beat me with such an attack in last year's World Championship match, and I still don't know what I should have done differently.
 
Arimaa is so far too mind-boggling to guess what will happen next in opening theory.  Rather than predicting the frequency/infrequency of draws in the future, I should instead predict only this: Arimaa theory will remain unpredictable for a long time.
IP Logged

Pages: 1 2 3  Reply Reply Notify of replies Notify of replies Send Topic Send Topic Print Print

« Previous topic | Next topic »

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