xcurt

XIMPS, unfavorable, 2♠-Pass-4♠-? ♠--, ♥QT762, ♦T2, ♣KQJT96 Is bidding here just sick? If you act, what do you choose? If you pass, how much better does your hand need to be to change your decision?

Lemme guess, 3♣ was on 2434 8 count. I guess 3♣ was the winning action in that thread. I'll head right over and post that suggestion now. Where do I go to collect my prize. Sorry about the snark but these two hands just illustrate that if you randomize your results a lot, you'll get a few nice stories out of it. That's why bridge tournaments are not 1-board events.

IMO, he played you for a real 3♥ call. Passing over 3♥ is chicken bridge. I saw someone do something a little more egregious the other night in a pair game at a local sectional. I opened 5♦ in first seat white against red and my LHO overcalled 5♥. My RHO passed holding AKQxx Jx Ax KJxx. Declarer could not quite handle the hand, and did lost a trick. 680 was not a great success for their side. I know 3♥ here shows more playing strength than a minimum opener with a decent suit. But responder knows his side's fit is bad. Bidding 3NT, then, is betting on a make on power, which is equivalent to playing partner for something like 16++ here. Most of the time, partner's overcall will be near his minimum. I agree completely that passing because we don't have full limit values with at least three trumps is an error in these auctions -- but that applies when we're coming in partner's suit, which we aren't. Now the guy that passed in your example, that's just sick. I would be worrying about missing grand.

If declarer has 5 hearts, then declarer has either AQ, Jxxxx, AKQ, KQJ or AKQJ, xxxxx, AQ, KQ, or some minor variation on one of those hand. In any event, we can't bet 4♥ if he has one of those.

Diamond now, give partner his presumed spade ruff after winning the other trump honor. If partner has SKJx we're not beating this. If partner has a club card we're not beating this. If we give the ruff now, the other diamond goes on declarer's fourth spade. Edit, edit -- I guess this is a problem. If declarer has AKQJ, Jxxx, AQx, KQ we need to defend as given above. If declarer has AKQJ, Jxxx, AQ, KQx, then we need to defend as given above. If declarer has AKQJ, Jxxx, AKQ, Qx then I need to play a black suit. If I play a diamond, then declarer can play 3 rounds of diamonds before the last trump and then if we take our spade ruff partner has --, --, x, KJxxx and he's endplayed. I'll play for partner to hold one specific card (the ♦K) instead of two specific cards and fire back the ♦9.

He played for hearts 3-3 rather than playing on diamonds. He therefore lost 2 hearts, 4 clubs and 2 diamonds. Isn't there a commuications issue with this line? On a club lead you don't have the entries. On a high spade lead lead you have to unguard a black suit. On a heart lead you might be able to read the position.

1100? What line did your partner take? Even if the opponents led clubs (best) it only looks like 500 to me. Your partner's hand isn't much better than average for the auction and position. I'm not a big fan of the 3NT call. He played you for ♥AQJxxx, the ♦A, and some black stuff too. If you had a hand that wanted to catch a raise to 3NT on a random 9-count, you would have overcalled 2NT.

The point is.....wait for it.....that you're maybe a fraction of a trick better, if at all. And you might be worse. Here's another seeming paradox. It might be better to have ♥Axxxxx than ♥AJxxxx. Why? Because if the opponents have 5 hearts including the K, Q, J, and T, they're making the same number of heart tricks on play in spades as they are on defense against hearts.* If they are missing some of the interior spots, they might not be able to play the suit for tricks on play, but they'll make the same number of heart tricks on defense -- the number of trumps in their longer hand, less one. So you get better Law protection with the emptier suit, since it's less likely to be a Law failure hand. I agree with other posters that passing could miss game. I don't agree with other posters that we're getting crushed as often as they think we are. So by arguments others made previously we should be in this auction. And if you read anything else I posted, you know prefer what you called the "thinking" option. * -- unless there is a cut or the defense can cut off the dummy against spades when the long heart is there

In hearts? THIS ONE! Meaning, your example suit. Give partner ♥xx. Let's analyze a 4-1 break with ♥AQ offside (you can put the ♥T onside stiff if you like). The bad guys are going to make two trump tricks on power and probably will tap you out to make a long one. If the hearts split 3-2 with length over you, you need two honors tight onside to do better than I will with ace-empty-sixth. Plus, having the ace is just huge on a hand like this since we can control the number of rounds of trumps taken out. Often that's worth at least one trick in the play. You can go through a bunch of these analyses, or you can just consider that the longer the suit, the more the relative rank of the ace versus the king, and the less the relative value of the secondary honors and high spots versus the low spots. I still like overcalling.

What happens when you play against such a pair, and you have an auction where the meaning of the opponents calls is dependent on the meaning of your calls? There is going to be a UI problem around questions asked or not asked, or tempo, or both.

I consider strong NT and four-card majors to be much better at MP than IMPs - very important to be able to find your 4-4 and 4-3 major suit part-scores at MPs, where +110 can be much better than +90. For game and slam bidding, on the other hand, you are better off playing five-card majors, which is more important at IMPs. Weak NT and four-card majors, on the other hand, has little to commend it at either form of scoring. I agree with this statement, but I think the reasoning is wrong. 4cm/strongNT has a relative advantage at matchpoints because you can conduct rapid and efficient auctions on partscore hands but slower and more informative auctions on game and slam hands, which are upweighted at IMPs. You also give the opponents nastier low-level bidding problems but get nastier high-level competitive problems. I have a little experience playing these methods in national pair games and they work just fine as long as you put in some artificial methods to fill some of the holes in the system (and all systems have some holes).

This means you give up on the "I've got a secret" double. I haven't thought through the implications of this, but I guess you can survive transferring to NT and bidding 2NT, invitational, or passing and bidding 2NT, invitational (and implying you would have penalty doubled at least one of the other two suits) since you're probably picking up a trick by putting doubler on lead on a hand where advancer is bankrupt or nearly so.

Why is everyone hating the heart suit so much? Which has a greater trick expectation integrated over partner's possible hands -- the hand we hold, or Axx, KJ976x, xx, Ax? I think they are about the same in the heart suit, and having the trump ace gives us more possible lines of play in how we handle the offsuits. Also, the argument that the opponents can double us more readily with the 2♠ method they used cuts both ways. Partner also has to be cautious balancing -- and he has zero or one aces and therefore much more reason to fear conceding a big penalty.

Who says we don't have a good suit? I'd rather have this hand for 1♠ than KT432, xx, x, A8xxx. Anyway, I agree totally with Josh on both of Tim's hands.

I don't think this is a good way to ask the question. I would say that typical doubles will respect 3NT from partner, and will respect 4m from partner unless offshape, advancer hit doublers doubleton, and having 5-card hearts or some other prepared response. Since I would say that less than 10% of takeout doubles will fulfill all of these conditions, you are looking at almost the same probability distribution as for the original double. So, in other words, responder shouldn't count on opener to recover the heart suit very often, and should strain to bid hearts when could be right. But we already knew that.

I generally agree with the sentiment that not making unforced errors is the most important single factor. Let's think about this quantitatively. I would guess (somewhat from experience*) that the best players in the world are about 1 IMP per board better than players who regularly win open regional events, and place in NABCs with some frequency. One way to think about this is to ask "what's the average winning margin for the top 10 seeds in the Spingold in the round of 64 or round of 32." One imp per board is about one game swing and one partscore swing per 16-board segment. That's not that much. I know that when I was playing regularly -- and I did try to review every session objectively -- I would make 2-3 unforced errors per session, on average. Just really silly things that I wouldn't do if I focused entirely on bridge. Of course not all of those errors would cost or would be game or slam swings, but that's probably half of the gap between me and someone in the late rounds of a National KO.

Clearly penalty of course. But this is also a brutal lead problem, no?

2♠ is just obvious regardless of what it shows by agreement. Your side has 8 spades to (at least) the AQ. One of the bad guys has a 6-card diamond suit and it's breaking onside. If your partner is burying you for taking calls like this you need to discuss this stuff (and that Law thing) with him.

Hmm I was never aware that the ACBL published clarification on this point. I'm glad to read it. I was once on a committee where delcarer was in a slam with something like 9 trumps to the AKQ, and not needing any ruffs to get up to 12 tricks. Declarer cashed a high trump with everyone following, cashed a second one with one opponent following, faced her hand, and (as far as I could tell from the testimony) the opponents immediately screamed for the director before she could state a line and wanted to make their baby trump. I argued and we ruled that even if declarer didn't have an exact count on the trump suit and hadn't intended to state a line involving drawing the last trump it was illogical to cash running side suits before cashing the remaining high trump. I discussed this one later with someone who is a pretty good player and he wanted strongly to award the opponents a trick with their small trump. Your point C seems to imply we made the correct ruling. It sounds like this stuff is only available if you have ACBLScore. How come this material isn't published to a wider audience?

I'm a little late to this thread so rather than opine on your auction I'll just say 3NT was a poor call since it's going down often when clubs aren't running (and some other cases too).

My first reaction was easy pass, but I started giving partner some hands and they're strong favorites to be making. Of course no action comes with a guarantee and the more creatively partner preempts in this position the less likely I would be to bid here. Let's assume we're leading the ♥K. I don't claim this is best, but it's something to work with: * HQJ8th and out -- they make 6♠, we're -500 in 5♥x. * HA8th and out -- they make 5♠ or 6♠, we're -300 in 5♥x. * HAQ7th and another card -- we're -300 and they're making anywhere from 9-11 tricks at spades. Add another card to partner's hand and in the first two cases they're still making 4♠ and we're probably -100/-300. So when we're right we're gaining 4-8-13 IMPs and when we're wrong we're losing 9 IMPs (-300 and -100 from teammates). I'll bid 5♥. I'm pushed to bid by having an ace -- it protects me against them making a slam, and provides more chances we can beat them if we can get the up one more level.

If east has ♣A9 and we misguess by playing a low spade, how about ♠Q, ♣Q, ♠A, ♦A, club. If east began with 2xx2 shape he's now endplayed. I don't think this is better than flying ♠J though.

Sorry, I should have added, this was BBOFlash. Full stack was Linux (centos 5/i386/latest kernel) Ffox 3.01 Flash 9.latest BBOflash latest Full disclosure -- it also doesn't include the ad. Curt

Nope, actual numbers for 45 minutes of typical activity (including 9 hands of pretty random MBC bridge) was about 2.5 MB down and 0.5 MB up.

Aaah this number again. Well you're getting 10:6 odds assuming your only possible results are 9 or 10 tricks undoubled. So you need 37.5% given that: you're playing an infinitely long match you don't assign any value to giving your opponents the cube Personally, in the absence of any information about partner or the opponents, I think this is very close and I don't have strong feelings about it. I guess that means I disagree with jdonn that there's a clearly right action.