lycier Posted September 10, 2016 Report Share Posted September 10, 2016 This is a interesting hand. [hv=http://www.bridgebase.com/tools/handviewer.html?sn=Robot&s=S85HAKQ3D85CAT932&wn=Robot&w=SJ96HT74DQJT43C74&nn=lycier&n=SAKT4HJ9865DA97C8&en=Robot&e=SQ732H2DK62CKQJ65&d=n&v=o&b=1&a=1H(Major%20suit%20opening%20--%205+%20%21H%3B%2011-21%20HCP%3B)D(Takeout%20double%20--%203-5%20%21C%3B%203-5%20%21D%3B%202-%20%21H%3B%203-4%20%21S%3B%2012+%20total%20points)2N!(Truscott%20%5BJordan%5D%20--%203+%20%21H%3B%2011+%20total%20points)P4H(5+%20%21H%3B%2013+%20HCP%3B%2014-19%20total%20points)PPP&p=CKCAC7C8D5DTDAD2D7D6D8DJS9SAS7S5D9DKH3DQS8S6SKS3S4S2HQSJC2C4H5C6STSQHKH4HAH7H6H2C3D4H8C5HJCJC9HTH9CQCTD3]400|300[/hv] Today I play against Gibs, see 9th round, Gib W inexplicably discards trump ♥4, my result is 4♥+2. So I am gonna think I can make 6♥ contract on a whim. I would better try it again. [hv=http://www.bridgebase.com/tools/handviewer.html?sn=Robot&s=S85HAKQ3D85CAT932&wn=Robot&w=SJ96HT74DQJT43C74&nn=lycier&n=SAKT4HJ9865DA97C8&en=Robot&e=SQ732H2DK62CKQJ65&d=n&v=o&b=1&a=1H(Major%20suit%20opening%20--%205+%20%21H%3B%2011-21%20HCP%3B%2012-22%20total%20points)D(Takeout%20double%20--%203-5%20%21C%3B%203-5%20%21D%3B%202-%20%21H%3B%203-4%20%21S%3B%2012+%20total%20points)2N!(Truscott%20%5BJordan%5D%20--%203+%20%21H%3B%2011+%20total%20points)P6H(5+%20%21H%3B%2021-%20HCP%3B%2021-22%20total%20points)PPP&p=CKCAC7C8D8DTD7D2S9SKS7S5DAD6D5D4D9DKH3DQS8S6SAS3S4S2HQSJC2C4H5C6STSQHAH4HKH7H6H2CTHTHJC5H9CJC3DJH8CQC9D3]400|300[/hv] Similarly, see 9th round, Gib W still does a same thing - discarding trump ♥4, even my playing is not best, my result is 6♥=. Quote Link to comment Share on other sites More sharing options...
Stephen Tu Posted September 10, 2016 Report Share Posted September 10, 2016 Double dummy it doesn't matter, and presumably GIB is able to see that. If it discards, you can ruff club and lead back to HA and lead club for a coup. Now arguably discard is better in case human opp not able to figure that out, but at this point GIB can't really distinguish between what is an easy play or hard play from a human standpoint. Against decent humans it really didn't cost anything, the takeout double should make the coup rather obvious. Quote Link to comment Share on other sites More sharing options...
lycier Posted September 10, 2016 Author Report Share Posted September 10, 2016 I played against Gibs normally in the first hand, not double dummy. Quote Link to comment Share on other sites More sharing options...
smerriman Posted September 10, 2016 Report Share Posted September 10, 2016 Stephen Tu is talking about Gib, not you. Gib defends based on double dummy simulations; all simulations prove you can make all remaining tricks regardless of the card Gib plays, so it doesn't see one as better than another. Quote Link to comment Share on other sites More sharing options...
Stephen Tu Posted September 10, 2016 Report Share Posted September 10, 2016 Yeah I know. My point is that at the point in play you are questioning, GIB can see that all plays are equivalent, the hand is essentially face up for everyone at that point. You are always taking all the tricks if you play correctly. So underruffing doesn't cost. It only costs against a bad player who can't see the line. The software isn't really able to distinguish easy plays vs moderately difficult plays vs hard plays for humans. To computer correct line for declarer is trivially easy, so it thinks underruff is equivalent play so it will go ahead and randomly play it since it makes no difference. It can't comprehend that a bad human might cash the high trump in dummy first instead of ruffing club. It would take a lot more computation to distinguish plays based on possible bad plays from opps. Current is assuming best play from opps. Quote Link to comment Share on other sites More sharing options...
jdonn Posted September 10, 2016 Report Share Posted September 10, 2016 Stephen is right. In fact, a human "should" make the rest of the tricks, since I think it can't cost to ruff a club back to your hand, after which a trump to dummy reveals the exact distribution. Obviously a lot of humans would miss that play, but GIB doesn't know that. Let me edit to say, it is probably better to ruff a club in hand at trick 2 and simply play to ruff all the clubs in your hand and/or cross ruff. But I am talking about what a human should do at the point you found yourself near the end. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.