Siegmund

I would be very surprised if the relationship between total trumps and expected tricks were fit well by a parabola. The relationship between (for instance) your side's HCP and your side's expected number of tricks in notrump is extremely close to linear, all the way up to 11.5 tricks at 31 HCP, and then rather abruptly flattens out. A parabola captures neither the long straight portion nor the sharp elbow well. It would not surprise me at all if the LoTT relationship were linear up to about 21 or 22 total trumps and then rather abruptly flattened out.

Had a funny thing happen yesterday. My non-bridge-playing girlfriend was in the mood to ask me a bunch of questions about how tournaments (and rubber bridge games) are scored, and at some point I grabbed the nearest hand record from a recent sectional, and my convention card, and I talked her through a few of the hands: "On board 1, you see my side has all the hearts, and the analysis box says -450 for hearts making 5 is par... when I actually played it, my opponents misjudged to bid 5C, and we doubled them and got -500... and got 19 matchpoints out of 20." (Yes, a 21-table pairs game with a web movement.) We got off on a tangent about whether people were scored by comparing their results to double-dummy par rather than to the results at other tables .... I pointed out a couple other hands where reasonable bidding never reached the par result (e.g. a 1NT-3NT hand where the other side has a lucky paying sacrifice in a 4-4 club fit) ... then looked for a simple hand where most people achieved double dummy par to see how close to average I scored on it. Twenty-six boards, and not a single one where the result at my table was the double-dummy par score. Yes, it was a night when I had had a big game and ruined it with four disasters in the last six boards ... but wow. If you had asked me two days ago, I would have guessed that in a typical session, double-dummy par is achieved at the table six or eight times.

It is a little odd that he would suggest equal level correction and two-suited overcalls with the lower suit stronger or longer. I am a fan of having a method to show the 4-5s that tend to have the 4-card suit get lost in standard methods, even at the expense of having to bid 5-5s naturally. My preferred adaptation of Roman jump overcalls, to squeeze in most of the hand types and remain GCC legal, has been floating around on the web for several years -- briefly, using 1C-2C to show 4 spades and 5+ of a red suit, and 1C-2D to show 4 hearts and 5+ diamonds, but leaving 1C-2H and 1C-2S as natural, instead of playing all three jump overcalls as Roman. I have talked several of my partners into trying it -- the ones who turned me down counter-offered Equal Level Correction, and encouraged me to double-and-correct with at least some of those same 4-5 hands.

Yes, should always be at least 6 spades.

I seem to remember there being either a Law or an ACBL regulation that said something like "no result achieved by a player who has already seen the cards may stand." In 2006 I had a fight with a unit board after cancelling all 27 results achieved by the club manager's wife and her partner, because she had made the boards for a pre-duplicated event at the club and then played in that event. I remember having a very clear rule to cite by which I removed all of her scores. But looking today at both the current and the 1997 law books, I can only find the one that says that if a player played a board twice, the second result may not stand. I thought that law used to be worded more generally, so that it applied equally to boards played twice because of mismovement, boards not shuffled from a previous session, and exposed hand records. Where did it go?

It's not my style, even at favorable in 3rd, but I give you credit for being within one trick of your bid, and for picking the right time to swing from the chandeliers.

In the meantime, you will do well to seek out the partners who self-rate as "advanced" :)

Well, that's one more thing I can add to my list of reasons not to like the book.

You've not provided enough information about the system to properly constrain either opener's or responder's hand. For instance: Does responder have any other ways to show a weakish hand with a 6- or 7-card minor (intermediate jump shifts, and 3-level weak jump shifts, both intrude on "6-10".) Does responder ever have 3 spades on this auction? Does opener open 1S or 1NT on 5(332) hands? If 1NT, what notrump range? You say "may contain a 4-card minor" ... but I am sure you don't mean all hands with a 4-card minor. Only 5224 and 5242? Only if the 4-card suit is weak? I did take a first stab at it using simple (too simple) conditions. 1000 hands, which makes the percentages good to ±3%. Responder: 6-10, 0-2 spades, 0-6 hearts, 0-6 diamonds, 0-6 clubs. Opener: 12-16, 5(332), 5242, or 5224. Lead, P(par possible after this lead), P(1NT can be set after this lead), avg. # tricks worst than best double-dummy lead. Didn't report average IMP loss because I accidentally left my script calculating IMPs in 3NT rather than 1NT. ♠A 20% 18% 1.30 ♠J 52% 31% 0.64 ♥K 50% 29% 0.83 ♥T 55% 33% 0.60 ♥4 57% 34% 0.56 ♦K 45% 28% 0.90 ♦9 60% 35% 0.50 ♦7 63% 36% 0.46 ♣6 67% 35% 0.42 Repeated with opener always 5(332), but still allowing 12-16. We expect this situation to show a stronger contrast, because it magnifies the importance of our fourth diamond -- a diamond lead never is into a 4-card suit -- and a "passive" club never runs into a running 4-card side suit. ♠A 18% 18% 1.32 ♠J 50$ 33% 0.66 ♥K 43% 25% 0.96 ♥T 51% 29% 0.69 ♥4 54% 32% 0.62 ♦K 47% 27% 0.90 ♦9 62% 36% 0.50 ♦7 66% 37% 0.45 ♣6 72% 38% 0.36 Sorry, but the club wins, at least under these conditions. I would have led a diamond at the table, too. Apparently I am strong enough that I should expect my partner not to be much help and I need to seek a passive lead.

To test different leads you can have a series of calls to the solver like set t [deal::tricks south notrump] set t_after_as_led [dds -reuse -leader west -trick as south notrump] set t_after_2s_led [dds -reuse -leader west -trick 2s south notrump] set t_after_7c_led [dds -reuse -leader west -trick 7c south notrump] set t_after_2c_led [dds -reuse -leader west -trick 2c south notrump] ... and then count how many times $t == $t_after_as_led, etc, to find out how often leading the ♠A blows a trick. I have a script to automatically find all possible leads and test them, but it is a little opaque, and requires a couple of custom libraries to help me count all the things I want to count.

How complicated of a method you use, depends how much TCL you feel like learning, vs. importing into another system. If I am going to use an outside program for analysis, I usually customize the output format rather than using ddline (and it saves time, to not do all 20 analyses if you dont care about all 20 of them.) You can edit the format definitions, but I usually use source format/none to suppress the automatic output and then write one PUTS statement inside your main loop to (for instance) create a .csv file that has data on each generated hand in a format that is easy to import into a spreadsheet. For simple questions like what percentage of the time a contract makes, you can do that inside your deal script. If you use Thomas's 'sdev' structure, you define statistics accumulators in the preamble, you submit data to them in the main loop, and you view the results afterward: source format/none #if you dont want to see a thousand hands spewed on the screen sdev avgtricks sdev chanceofmaking main { # put your restrictions on hands here set t [deal::tricks south notrump] avgtricks add $t if {$t>=9} {chanceofmaking add 1} {chanceofmaking add 0} accept } deal_finished { puts "Average tricks: [avgtricks average]" puts "Probability of making game: [chanceofmaking average]" }

3♣ has the merit of preventing partner from bidding again later and getting us much too high... :\

You can put me down as a second protester. If the goal is to make the symbols "stand out better" or be more visible to people who can't distinguish the color well, add an outline to the outside of the red symbols, perhaps. Even as a "normal-sighted" person, I prefer to see hollow hearts and diamonds (or, rarely, red interiors and black edges) over all-red ones when I am reading a book. And yes, if we were taking polls, I would prefer the forum and the bidding boxes use red diamonds and black clubs. The 5-color bidboxes are usable, but to me, add very little to the game. (Here again, as size is no issue, I wish that a black outline of the suit symbol was standard on the bidding box symbols.) Could be worse, though. At least we aren't talking about those decks that are still 2 colors of ink but use faded-out diamonds and clubs. Recipe for confusion and revokes even when the cards are new.

You want a sim, OK, here's a sim. 1000 deals, opener 15-17 balanced or semibalanced, responder 9-13 no 4CM unless 4333. (Yes, this leaves in a few 3163s etc that would have done something else.) With or without a sim, any non-jack lead here is one part psychic and two parts psychotic. It is not remotely close, whichever measure you use to decide what lead is best: Card , #deals on which par is still possible after that lead, #deals on which 3NT can be set after that lead, #tricks lost vs. best lead, IMPs lost vs. best lead: ♠J: 670 150 -516 -1566 ♠4: 637 138 -581 -1728 ♥J: 714 182 -449 -1224 ♥8: 633 164 -562 -1499 ♥2: 630 164 -568 -1504 ♦Q: 406 76 -1053 -2660 ♦9: 463 91 -927 -2428 ♦4: 489 95 -887 -2352 ♣4: 551 110 -702 -2087 With 1000 replications, margin of error (95% confidence interval) is about ±30 on the first column, ±20 on the second, ±50 on the third, ±100 on the fourth. More than enough to resolve the H > S > C > D pattern. Edit: I see 1eyedjack got a head start on me. I am also happy to see we are reporting the same qualitative results despite small differences in our constraints and in what we chose to count.I report more frequent sets, of course, since my responder was 9-13 rather than 10-15.

I have the methods to get started down the right track -- in my preferred method it will start 1NT - 3C asks for 5CM, usually (31)(54) or (30)55 but occasionally 33(52) 3D - 3H no 5CM / singleton or void But I think that as South I am going to try to sign off now, and as North I don't think I have quite enough to try again.

My area has been stricken by a disease of 10AM-and-3PM start times, with long alcohol-fueled discussions over a late dinner afterward. On the traditional 1-and-7 schedule, the afternoon hands got discussed at dinner still though.

They do that. You can get your matchpoint result on all 26 boards, from any idle bridgemate in the room, along with your rank. It doesn't display the hand records -- the screen is too small for that right now, and we dont really want people to stay in the playing area all night to do that. (It doesn't send everyone's results to every bridgemate: you type in your pair number, the client sends a request, and the server responds with that pairs results) But the "in" thing now is to use the Fast Results service, and get a text with a link to a webpage with everyone's scores, hand records, and double-dummy analysis sent to you. Of course this does exacerbate the tendency for people to eat dinner with their noses buried in their smartphones.

I've been exploring the same question for awhile. My comment in your other thread about transfers was because of a time when I tried 1C = weak with 4 spades, or any strong hand; 1D = 11-18ish with 4 hearts; 1M = 5+; 1NT = 11-14 bal no 4CM 2m = 6+ instead of a more simple fine-tuning of Polish. I think it is hard to find a 1C/1D combination that is actually better than a Polish style club and a natural (5+ or 4441, or something like that) 1D. Having run into the ruling that GCC doesn't allow 1D-promising-4-hearts, I was forced to try something more nebulous: 1C = 11-14 balanced with 4CM, or 15+ bal or minor 1-suiter, or 19+ any; 1D = 11-18 unbalanced with either or both 4CMs 1M = 5+ 1NT = 11-14 no 4CM 2m = (9-10-)11-14(-15), 6+, no 4CM This system actually feels more comfortable than the C=S, D=H version did in competition; opposite the balanced hand, responder can compete to 2 of any 5-card suit or 3 of any 6-card suit fairly safely, opposite the unbalanced hand, you can be confident opener has 2 or 3 places to play, so responder can compete easily with most 2-suited hands. It does, however, make the frequency of the 1D opening very low - something like 4-5% vs. the 6-7% of the 1M and 1NT openings. That "feels wrong" -- but ruining 1D by making it so nebulous that the first round of the auction is wasted "feels even wronger" to me. Of course I am not sure that it's better than a super-natural Polish diamond; it's just more fun, if you like to play with systems.

I would have bid 5C without a second thought. Interesting thread highlighting bad-breaks-in-both-majors as a reason not to push so hard.

No option in the poll for "there's enough blame to go around for both North and South"? I think 3S over 3D by North, and 3H over 3D-X by South, are both quite clearcut.

One more vote for "you needed to bid 4C last time, but since you didn't, 5S now."

1♦-1♠, 2♣-2♥, 3♥-4♥ is sufficiently straightforward that I am surprised to see it mentioned less often than several quite strange alternatives. Admittedly it is more appealing if you play FSF-round (as I do) than if you play FSF-game. If you made opener's minor strength wildly lopsided, like x AKQx xxxx AKQx, I would be very sympathetic to a reverse without quite having 4-5 shape.

It's a tendency I'd like to see disclosed, even in the US. I don't know what the best method for disclosing it is; alerting doesn't feel like a solution.

You've been going to the wrong theaters if you've never heard anyone shout "sic semper tyrannis!" (In the nitpicks department: I wasn't saying it wasn't a long i, just that long-i doesn't sound like an English long i, in just about every other language I can pronounce. And it is veni -- so sounds like waynie rather then weenie :) )

I confess I gave up on the entire thread, for being unable to decipher the auction. Could we please not do the "one bid in parentheses" thing, as in, never for the rest of our lives?