TimG

No finer gradations will do any good so long as people are self-rating. If someone who is advanced misses a splinter, how will it help to add low/middle/high intermediate?

Yes, good things can happen when you lurk in the weeds -- sort of like on the auction 1x-DBL, if you pass you may get a chance to penalize at a higher level than you would had you warned them with a redouble -- but that doesn't mean that when you pass initially with four trumps and none of those good things happen, that you will always want to convert opener's reopening double. The reopening double may itself warn you against defending because of opener's now announced shortness. Anyway, I wouldn't feel compelled to take action after 1D-(1H) with a balanced invitation (or less) with length in their suit. Passing and awaiting developments might be a good course of action. But, balanced or unbalanced, you wouldn't trap pass with a GF hand and four good trumps, would you? Because, as you say, when opener has a minimum with some length in their suit, you may end up playing for 100 per undertrick. I'm not trying to argue with you, just talk it through so I better understand.

This question is prompted by a thread in rgb. A poster there asserted that it would be routine to pass after 1D-(1H)-P-(P); DBL-(p) with heart holdings such as AQTx or AQJ8 (and an otherwise suitable values). The conditions were MP, red v red. While I can probably construct hands were passing with only four trumps is indicated, I would think it far from routine. And, if I were to set up a simulation, I would start with the premise that the penalty passer held five (or six) hearts. Am I out in left field here? I tend to be more inclined than most to try for a penalty, but defending their likely 8-card fit seems far from routine to me.

Is that the same Nate Silver who developed the PECOTA system for Baseball Prospectus? Interesting that he has taken his projections from sports to politics.

Is there a practical difference?

We won't really know whether their analysis is any good until we see the actual results, will we?

That looks consistent with my results. But, I think you want to avoid the trap of thinking that because game makes 50% of the time, you want to bid it. Here an invitation will get you to game when opener has the top end of the range (and game will make more than 50%) and will stop short when he has the bottom end of the range (and game will make less than 50%). That is why I focused on opener's minimum HCP holdings.

Just curious, I hope...

FWIW, some simple simulation results. Give opener a 4333 or 4432 hand with 4 hearts and 15 HCP (to match a 1N-2C-2H start), and game makes ~32% of the time. Even giving declarer some advantage vs double dummy, those who think this is an invite are correct. Give opener 16 and game makes 48%; 17 and game makes 67%. Seems to me like a straightforward quantitative invitation will work well. Give opener 4=4=3=2, 3=4=4=2 or 2=4=4=3 and 15 HCP (to match a 1D-1H-2H start), and game makes ~28% of the time. Yes, less than when we don't have any inference about the diamond semi-fit. So, I was wrong when I suggested that the extra bit of information might make bidding game more attractive -- we clearly still want to invite. Give opener 1453, 2452 or 3451 and 14 HCP and game makes about 36% of the time. Still not worth blasting to game. Increase to 15 HCP and game makes about 54% of the time, so here again a simple quantitative invitation would seem to do well. Of course, if you have more sophisticated invitational methods, they may work even better. Thanks for the responses. I guess I could have answered the questions myself with the use of simulation, but I wanted to get a better idea how the field viewed this.

From the KS summary provided at the Bridge World site: "Usually 15-17, 4-card fit. Could be unbalanced equivalent. Rarely, 3-card support, too concentrated for 1 NT rebid: ♠AQx ♥xx ♦xxx ♣AKQxx." If the 1=4=5=3 12 HCP hand does not qualify as 15-17 in support of hearts, it is probably right not to open (in the context of KS).

Doubt all you want, but I would surely not raise to 3H with 4333 and 17 HCP (even considering that we would open that 1C, not 1D). As for the 12 HCP 2452, I would not open 1D if the hand did not evaluate to 15+ in support of responder's hearts. (I think that within the KS framework we're dealing with, the options would be Pass, 1H and 1N if the hand was not good enough to raise 1H to 2H.)

Thanks for all the responses. Recall that the hand is ♠QJT5 ♥KQ43 ♦J94 ♣75. We were playing weak NTs at the time, so the auction we had actually started 1♦-1♥-2♥, where 2♥ shows 15-17 balanced with 4 hearts (or an unbalanced equivalent). Two questions: 1) Is the semi-fit in diamonds enough to tip the scales in favor of a game bid rather than an invite? (Yes, partner could be 4=4=3=2, but we open 1♣ with 33 in the minors.) 2) In a MP event, does it matter to you that many pairs will start 1N-2♣-2♥? That is, do you want to duplicate their invitation/force choice even though you may have a tad more information? (Let's assume we consider ourselves to be as good as the field and we're not going out of our way to create anti-field swings or the like, the weak notrump not withstanding.)

MP, red v white, you hold ♠QJT5 ♥KQ43 ♦J94 ♣75. Partner opens 1N (15-17)and replies 2H to your Stayman bid. Is this hand good enough to insist on game?

No, the overcall or lack of an overcall made little or no difference to the final contract or play of the hand. I posted the had because I would have expected a near unanimous forum answer of "1♠, wtp", so I was a little surprised to see 2/5 forum regulars in the individual pass with this hand. Due to the suit quality the answer "1S, wtp" will certainly never be unanimous. I said "near unanimous".

No, the overcall or lack of an overcall made little or no difference to the final contract or play of the hand. I posted the had because I would have expected a near unanimous forum answer of "1♠, wtp", so I was a little surprised to see 2/5 forum regulars in the individual pass with this hand.

♠AJ765 ♥T72 ♦A9763 ♣ -- MP, white v red, 1st seat. 3/5 passed; 2/5 opened 1♠. Your thoughts? And, does it matter whether you are playing 2/1 or not?

♠JT973 ♥T6 ♦7 ♣AK832 MP, red v white, dealer on your right opens 1♦. 3/5 overcalled 1♠; 2/5 passed. Your thoughts?

I find it incomprehensible that there were 14 spades in the deck. I also find it annoying that the point of the post seems to have been to tell us about how good you are and how bad your partner was.

How does a direct 5♥ differ from 4NT followed by pulling 5m to 5♥?

Wouldn't 1♠-P-1N-3♣; DBL be for penalties, or at least strongly suggest penalties?

What do you all think of responder starting with 1D? When NT is right, it is probably best played from opener's side (we have no holdings that need protection).

That does not follow logically. Some of the negative equity situations are likely only small negative positions -- $200,000 owed on a house that would sell for $199,000, for example. The amount could actually be pretty small.

Edit: I guess you do want to be in 7♦.

You know what my 12 year-old daughter knows -- honestly, she looked at my 10 year-old son like he was crazy when he said it had something to do with math. Said 10 year-old son was told by his 5th grade teacher that the median is determined by averaging the middle two elements when there are an even number of elements. Nothing really wrong with him following the prescribed procedure, but I want to know if it would be wrong to call one of the two middle elements a median (or any other value between the two middle elements).

I had to try, too. When I eliminated a lot of the hands that EW would intervene with, I got 2H making 43%, 2D making 64%, and 3D making 37%. A couple of further items for comparison: Average HCP for opener: 15.4 Average hearts for opener: 1.09 And, even though talking about spades got me into trouble earlier, the result of the simulation was that the opponents could make 2S 76% of the time. The usual simulation disclaimers apply even more than usual: I'm sure EW hands that would intervene slipped through, some they wouldn't intervene with were discarded, and that both those items depend a lot on the opponents' approach to intervention; double-dummy tends to favor the defenders relative to table results, but this effect decreases as the auction reveals more; yada yada; the exact percentages are probably off, but they are probably not bad in relative terms (2D making 50% more often than 2H and 2H making about as often as 3D).