Hands played at the Palmerston North Bridge Club published here for discussion and question and answer.
Any bidding shown is an actual auction and not necessarily recommended.
Please feel free to participate and make suggestions.
Tough hand to play. Some would have bid 3NT with east but even 2NT requires an endplay to make.
It was a much easier hand to defend. After the dQ lead won by the king in dummy and then a heart to ace and queen, west's hand is an open book. West is marked with the dA from the play to trick one and with the hAQ that makes 10 hcp. Given the pass of 2NT that means that south must have the sA. Otherwise west would have fourteen points and would have accepted the invite.
I think some players would baulk at leading low from QJxx(x) at no trumps, but Mike Lawrence's view is that if you must lead that suit, then lead low. You are hoping your partner has at least one of A,K or T. If they don't then you are probably stuffed anyway.
However, with the 9,8 there, you have some insurance, and I should imagine most defenders would lead the queen.
Bird/Anthias profile this auction. EW average 6.88 cards in the majors and 6.12 cards in the minors, so a major suit bias exists. They give a hand not too dissimilar to the above of QJ9x,K9xx,QJ9x, x. QS was best followed closely by QD (ie queens better than low - at both IMPs and MPs).
In playing the hand I could see 7 tricks (after JH appears) and a 50% 2 way shot for an extra trick in clubs, neither way working as it happens. I also had some idea of throwing west in given the obvious diamond situation, but little idea of how to execute that given I was reluctant to play on spades. Suggestions?
I think you basically have to pick the club position to make. In practice that is hard. You have no bidding to go on and only limited information from the passes and play. On top of that you have fairly reasonable plays in clubs.
You could exit spades and hope the defense can only get three tricks in that suit and will eventually be endplayed to lead clubs and you can guess well.
Also at the table I thought which way I played on clubs was a 50/50 given no useful other info. This is true is the honours are split, but if they are not split then there is a distinction. If LHO has both u get 2 tricks either way, whereas if RHO has both u get 2 tricks by playing low to the queen and only 1 trick by finessing the T. Therefore the first play should be play low to Q and not finesse the T.
Re leading low or the Q, while it is correct u are hoping for pard to have A,K, or T, the probability is reasonably high that he has none of those. The T being random is 2/3rds, while the prob of a single honour with oppo is ~23/32 (ie oppo have fallen short of game thus have ~23 and pard thus must have ~9). The conditional prob of oppo having both honours is I think 23/32 * 19/28, so together with the T it is ~a 34% chance. If that is the case then it matters a lot whether u lead low or high. Leading low will almost always cost a trick, whereas leading high will only cost a trick where dummy has the T and has at least one honour, so 50% *(50% split hons+25% both honours) or 37.5%, so a net 62.5% (of 34%) in favour of leading high
You don't calculate probabilities of high card points like this. It is quite a bit more complex. Aside from everything else you have to make some assumptions on what shape hands the opponents will bid like this.
In addition your numbers seem to be slightly wrong. On the assumption that they have 23 (in fact the range is probably (22)23-24(25) 11-11, 12-11, 11-12, 12-12, 13-11 and 13-12 are all possible and that is given partner never invites with 10) then we have 9 (sK dQJ cK) so partner has 8.
Empirically I get the probability of the opponents having the dA to be 0.81927 (cf 23/31 = 0.74194). The discrepancy is because there is a tendency for aces to follow the high cards. That is a 12 hcp hand is disproportionately more likely to have any given ace than 8 hcp hand. There is also an affect here in favour of west having the ace (king or ten) because west has denied a major. There are other minor affects going on too.
For a ten there is another affect - with the opponents having more high cards then it is correspondingly slightly more likely that partner with fewer high cards will have any particular spot card. Again these calculations are complex. Empirically I get 0.30403 for partner having the dT. This is below the 1/3 chance and is the result of the complex mix of partner having more spot cards and likely to have fewer diamonds (minor cards) on this bidding. By contrast empirically I get partner to be 0.36512 for partner to have the hT, which is more than the a priori 1/3.
Even my empirical results are not completely accurate. I did 100000 simulations but I had to constrain the hands on which in particular east would bid Stayman and invite. This is difficult when there are some distributional hands say 4=1=2=6 with 10 hcp that different players might handle differently.
Overall I got, again empirically, that there was a 0.57925 chance that partner had at least one diamond honour.
Also the issue of whether you cost a trick has a bigger impact if it is the trick that gives the contract when we are playing IMPs.
There is another hand in the set where the leader has KQxxx and some led the king. Bob lead low at our table. The king might allow the contract to make for example when partner has Ax and the suit breaks 3=3. Of course leading low might give away a trick unnecessarily too. This illustrates the complexity of these decisions.
Tough hand to play. Some would have bid 3NT with east but even 2NT requires an endplay to make.
ReplyDeleteIt was a much easier hand to defend. After the dQ lead won by the king in dummy and then a heart to ace and queen, west's hand is an open book. West is marked with the dA from the play to trick one and with the hAQ that makes 10 hcp. Given the pass of 2NT that means that south must have the sA. Otherwise west would have fourteen points and would have accepted the invite.
Not so tough when North leads their fourth highest diamond ...
ReplyDeleteA similar lead happened another night recently - maybe even the same suit QJ98.
ReplyDeleteSome advocate to lead low from QJxx(x) and the like. However I think the 98 sway heavily back in favour of the top honour lead.
I think some players would baulk at leading low from QJxx(x) at no trumps, but Mike Lawrence's view is that if you must lead that suit, then lead low. You are hoping your partner has at least one of A,K or T. If they don't then you are probably stuffed anyway.
ReplyDeleteHowever, with the 9,8 there, you have some insurance, and I should imagine most defenders would lead the queen.
Bird/Anthias profile this auction. EW average 6.88 cards in the majors and 6.12 cards in the minors, so a major suit bias exists. They give a hand not too dissimilar to the above of QJ9x,K9xx,QJ9x, x. QS was best followed closely by QD (ie queens better than low - at both IMPs and MPs).
ReplyDeleteIn playing the hand I could see 7 tricks (after JH appears) and a 50% 2 way shot for an extra trick in clubs, neither way working as it happens. I also had some idea of throwing west in given the obvious diamond situation, but little idea of how to execute that given I was reluctant to play on spades. Suggestions?
ReplyDeleteI think you basically have to pick the club position to make. In practice that is hard. You have no bidding to go on and only limited information from the passes and play. On top of that you have fairly reasonable plays in clubs.
ReplyDeleteYou could exit spades and hope the defense can only get three tricks in that suit and will eventually be endplayed to lead clubs and you can guess well.
Also at the table I thought which way I played on clubs was a 50/50 given no useful other info. This is true is the honours are split, but if they are not split then there is a distinction. If LHO has both u get 2 tricks either way, whereas if RHO has both u get 2 tricks by playing low to the queen and only 1 trick by finessing the T. Therefore the first play should be play low to Q and not finesse the T.
ReplyDeleteRe leading low or the Q, while it is correct u are hoping for pard to have A,K, or T, the probability is reasonably high that he has none of those. The T being random is 2/3rds, while the prob of a single honour with oppo is ~23/32 (ie oppo have fallen short of game thus have ~23 and pard thus must have ~9). The conditional prob of oppo having both honours is I think 23/32 * 19/28, so together with the T it is ~a 34% chance. If that is the case then it matters a lot whether u lead low or high. Leading low will almost always cost a trick, whereas leading high will only cost a trick where dummy has the T and has at least one honour, so 50% *(50% split hons+25% both honours) or 37.5%, so a net 62.5% (of 34%) in favour of leading high
ReplyDeleteYou don't calculate probabilities of high card points like this. It is quite a bit more complex. Aside from everything else you have to make some assumptions on what shape hands the opponents will bid like this.
ReplyDeleteIn addition your numbers seem to be slightly wrong. On the assumption that they have 23 (in fact the range is probably (22)23-24(25) 11-11, 12-11, 11-12, 12-12, 13-11 and 13-12 are all possible and that is given partner never invites with 10) then we have 9 (sK dQJ cK) so partner has 8.
Empirically I get the probability of the opponents having the dA to be 0.81927 (cf 23/31 = 0.74194). The discrepancy is because there is a tendency for aces to follow the high cards. That is a 12 hcp hand is disproportionately more likely to have any given ace than 8 hcp hand. There is also an affect here in favour of west having the ace (king or ten) because west has denied a major. There are other minor affects going on too.
For a ten there is another affect - with the opponents having more high cards then it is correspondingly slightly more likely that partner with fewer high cards will have any particular spot card. Again these calculations are complex. Empirically I get 0.30403 for partner having the dT. This is below the 1/3 chance and is the result of the complex mix of partner having more spot cards and likely to have fewer diamonds (minor cards) on this bidding. By contrast empirically I get partner to be 0.36512 for partner to have the hT, which is more than the a priori 1/3.
Even my empirical results are not completely accurate. I did 100000 simulations but I had to constrain the hands on which in particular east would bid Stayman and invite. This is difficult when there are some distributional hands say 4=1=2=6 with 10 hcp that different players might handle differently.
Overall I got, again empirically, that there was a 0.57925 chance that partner had at least one diamond honour.
Also the issue of whether you cost a trick has a bigger impact if it is the trick that gives the contract when we are playing IMPs.
There is another hand in the set where the leader has KQxxx and some led the king. Bob lead low at our table. The king might allow the contract to make for example when partner has Ax and the suit breaks 3=3. Of course leading low might give away a trick unnecessarily too. This illustrates the complexity of these decisions.
Toes were stuck together. HCP adds to 41!
ReplyDelete