Join Bridge Winners
All comments by Lawrence Diamond
1 2 3
You are ignoring the author of this comment. Click to temporarily show the comment.
Items have been sold.
Nov. 20, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
Everything has been sold. Thank you to all who were interested.
April 29, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
Bridge World Magazines have been sold. Thank you everyone for your response.
March 28, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
I devoted a significant part of a chapter on this exact issue with statistics from many thousands of hands in “Mastering Hand Evaluation” using modified Kleinman points to determine when to invite. Inviting with 8+ Kleinman points, assuming that partner will accept with any 16 HCP, is OK at IMPs, where less than a 50% success rate at 3NT shows a small profit. My double-dummy results on 10000 hands did not show that it was reasonable to invite with 8 HCP at MPs (44.6% chance of 9+ tricks with 8 HCP opposite 16-17 HCP).
Jan. 31, 2017
You are ignoring the author of this comment. Click to temporarily show the comment.
In my opinion, the problems you are experiencing are not due to system, they result from the fact that the partnership doesn't understand the hand evaluation implications of the “conversation.” A 15-17 NT shows, on average, about 6.75 losers using the new losing trick count (see the book “Mastering Hand Evaluation”).

Once a fit has been established, NLTC should be used by both partners and an invitational sequence after a NT opener shows that responder has 6.5 NLTC losers. Here responder has 7 raw NLTC losers minus 0.5 for the doubleton with a 4-card fit, so 6.5. Therefore, he has an invitation towards slam (not a jump to 4NT with a worthless doubleton, followed by a guess).

Opener cooperates with 6.5 or fewer losers. On this occasion, opener has a great hand. Only 6 losers with at least 2nd round control in all suits and a couple of good spot cards. Therefore, if opener knows that responder wouldn't make a slam invitation unless he had 6.5 losers, opener has plenty to accept. It is not “shooting the slam.” The combined NLTC of the two hands ensures that the slam has a decent play.

In your case, you arranged for responder to guess whether opener had a better than average NT (your actual hand) or a worse than average NT (Eric's example hand). With a proper invitational sequence, opener (not responder) would make the decision and get to the slam knowing that he wasn't “shooting.”
Nov. 25, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
Even if that was not the best example because other bids might be available after 3, the point I am trying to make is that “generically” opener's sequence has to be in the slam try range (and therefore a hand that is worth more than 10 tricks opposite a 9 NLTC loser hand). If opener was only worth game, he would just bid 4 the first time. Responder's first responsibility is to evaluate his hand for game. It is possible for responder's cards to be poor for game (opposite a presumed “help-suit” try), but excellent for slam opposite what is now known to be a diamond cue bid (and no club control).
Oct. 15, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
Does a player evaluate his hand in a vacuum (“I paid my entry fee and therefore I am entitled to make my best guess”) or should it be done with repect to the partnership.

Unfortunately, I have rarely found partners who are willing to embrace the partnership aspects of the game by trying to understand my preferred methods of hand evaluation. If players do not make the effort to understand how partner is thinking, then they are shortchanging themselves on one of the better aspects of bridge (IMHO).

Are all working cards in the combined partnership holdings held in the same suits (similarly, combined holdings with no values but shortness in one hand are very powerful)?

To me (despite the fact that all bridge players evaluate their hand independently using whatever methods they choose) the purpose of hand evaluation in general, and specifically the reason for using NLTC once a fit has been established, is to recognize the trigger areas (part-score only vs game-try vs game only vs slam try vs slam force) as a partnership. If my partner understands my hand evaluation methods, then after a fit has been established he/she will be able to construct possible hands for me (important) knowing how many losers I have shown.

For me, “help-suit” game tries are holdings where you want partner to upgrade a king or queen in that suit (or possibly shortness with extra trumps). It is true that many good players think that natural “help suit” game tries by opener give away too much information to the defense. Those players may choose to disguise the game tries allowing responder to describe something that doesn't give up as much info to the defence) or choose a pass/bash approach (which has some potential for game bidding, since defenders often give up a trick, but may lead to very poor slam bidding). Remember, what starts out as a “game try” may actually be the start of a slam try. The auction 1-2-3 (help suit GT)-3 (no help)- 4 doesn't mean “I am borderline between a game try and game so let's try it anyway,” it means “the combined hands have more than game and I just made a cue bid; please re-evaluate your known limited hand with respect to a possible slam.”

A lot of people are turned off by LTC because they don't understand that a really good LTC takes into consideration upgrades for holdings such as AKJ, AQJ, KJ10, and downgrades for duplication of values (a king or queen opposite shortness). It is true that we cannot always tell when there is duplication of values (unless a splinter has been made). That problem is not specific to losing trick counts, however.

Agreeing on partnership hand evaluation methods really helps in getting to the lay-down 19-21 HCP major suit games, avoiding the bad 23-24 games, and improving slam bidding (just go through the records of top events to see how poor slam bidding still is).

The subtitle of my book “Mastering Hand Evaluation” is “Understanding the Principles of Partnership Bidding.” To me, using an advanced losing trick count in fit situations really improves partnership communication.
Oct. 15, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
As others have mentioned: (1) losing trick count is only useful when a fit has been discovered, and (2) the “basic” LTC which counts an ace and queen equally, is not accurate.

New losing trick count (NLTC) with additonal modifications for distribution, honor combinations, and duplication of values is highly accurate when an 8-card or longer fit has been found. See “Mastering Hand Evaluation.”
Oct. 14, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
New Losing Trick Count works beautifully with 6-card suits opposite NT openers. Table 5.6 in “Mastering Hand Evaluation” show that a 20-21 HCP 2NT opener has 5.09 losers on average. Therefore, an 8-loser responding hand puts you “in the slam zone” opposite an average or better hand. That means a slam that you would want to be in providing that there are sufficient controls for you not to have 2 losers off the top.

Here, the responding hand is slightly better than 8 “raw” losers because NLTC doesn't count any value to the singleton K. The 2NT opener is also slightly better than its raw 5 losers because the Q is not being fully counted and the !J is in combination with higher honors.

Both players have enough information to know that slam is worth trying for opposite an average or better hand for the sequence 2NT-3-3-4.
Aug. 10, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
OK, you are talking about AT9 T98 T98 KJT9. The hand that was given by the original poster is 6.45 and 7-.

An A-KJ combination is worth an invite in any hand that is not 4333. In Aleksis' example hand, the 10s with higher honors are very useful.
July 9, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
The K&R evaluation looks like 6.45 to me. On the Kleinman scale, it is a bad 7 which is not nearly enough to invite. According to my data, you should invite with 8 HCP that are also 8 Kleinman points.
July 9, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
In covering many possible point count systems for NT evaluation in my book “Mastering Hand Evaluation,” I extensively cover this exact issue, i.e. when to invite with 8 HCP opposite a 15-17 NT (or 9 HCP opposite 14-16, etc.) if partner accepts with anything but the minimum HCP in the range.

There is a lot of data (both at-the-table and double dummy deals) backing up the recommendations in the book.

The 4333 distribution is a negative (balanced by the 10s somewhat but not a “good enough” 8 HCP to invite at matchpoints. Move the 5 to another suit and the value of the hand would go up significantly.
July 9, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
Message sent regarding 1973 and 1983.
July 2, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
I play that a super-accept shows 4 trumps and 6.5 or fewer losers (New Losing Trick Count), a better than average hand. You have reasonable protection at the 3-level when partner has a weak hand, but this method is also superb for evaluating slam potential. Since partner knows that you have a balanced hand, there is less chance for duplication of values and if (s)he also has 6.5 or fewer losers, partner will know that slam is a good contract as long as you have a sufficient number of keycards and appropriate control of all suits.
Feb. 15, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
If you want to quantitate what a “good 8” looks like, a modified “Kleinman count” (13-9-5-2) works nicely. The modification I suggest is to add 2 raw points for a good 5-card suit (Q10xxx or better). After adding for honors in combination, and 10s and/or 9s with other honors, as well as subtracting for 4-3-3-3 distribution or honors in short suits, and dividing by 3, if the hand is 8 points, it is worth a raise (and partner accepts with all 16s). Based on a study of 10000 hands, this works out to about a 45% game, which according to popular wisdom, is worth bidding at IMPs, even NV, but is probably not worth bidding at matchpoints. Danny Kleinman told me that he doesn't invite with 8 (he probably plays mostly matchpoints) and I am told that Kenny Gee (the Canadian pro) has a “rule” never to invite with 8 HCP, as well.

With 4-4-3-2 distribution it works out to honor combinations like, two Aces, A-KJ, 8 points including a 4-card suit head by AQJ and the other Jack in a 4-card suit, or K-KQ (with a KQ combination, not K-K-Q).
Jan. 24, 2016
Lawrence Diamond edited this comment Jan. 24, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
The Law of Total Tricks (LOTT) is excellent for determining when to compete at the 2- and 3-levels in a deal where the points are more evenly divided between the two sides. In my opinion, it is not particularly useful for making a decision about when to sacrifice.

The New Losing Trick Count (NLTC) with adjustments, however, will help you determine how many tricks you can expect in any fit auction.

In NLTC, missing an ace is 1.5 losers, missing a king is 1, missing a queen is 0.5. You can have more “losers” in a suit than cards.

So, West has 11.5 “raw” losers: 3 each in spades, hearts, and clubs and 2.5 in diamonds. Table 5.3 in my book, “Mastering Hand Evaluation” shows the adjustment for distribution. The person who first learns about the 8+ card fit (in this case West) can deduct 1/2 loser when holding 4 trumps and a doubleton. That brings West down to 11 losers. Furthermore, Jack-Ten combinations are undervalued when a higher honor is held. As Belladonna and Garozzo suggested, we can adjust for combinations such as AJ10 and KJ10. In this case, the King is in partners hand. However, we should recognize the potential for the J109 and consider that the West hand is a fraction better than 11 losers.

When West knows that NS have at least an 8-card spade fit, he/she knows that partner has at most one spade and is unbalanced. East's most likely distribution in that case is some 5431 such as 1=5=3=4 or 1=5=4=3. The maximum number of losers with 5431 distribution is 10.5 losers (Table 5.5 in my book, but easily calculated at the table). Give East 12 HCP for a “sound” opening bid, like AKKQ and that amounts to 4 cover cards which we deduct from East's maximum of 10.5 losers to estimate that East has 6.5 losers.

Larry Cohen talked about positive adjustments for “purity” in the LOTT. Vacant suits with shortness (in this deal, spades from the point of view of EW) are good things. Here, EW are playing in a 34-point deck and whatever assets they do have are working better in combination.

The NLTC equation is 25-(sum of both partners losers)=Expected number of tricks. West can calculate 25-(11+6.5)=7.5. This tells us that we are probably on a finesse for 8 tricks. If partner has either the K or A, then our J-10-9 combination has extra value. If the spade suit is vacant (i.e., partner does not hold the stiff Q) then that is a slight positive adjustment.

No guarantees, but if you judge partner's hand correctly, you can tell that you are slightly better than 50% to take 8 tricks, making for a successful sacrifice at this vulnerability.
Jan. 13, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
When I lived in England, I got a chance to play the MIDMAC bidding system with its inventor, Jon Drabble. If you open 1 in that system, it promises a 4-card major. In principle, responder wasn't supposed to bid 1M unless he/she held a 5-card suit or 4 with two top honors. While it was not always possible to stick to those criteria, we did acknowledge that KJ109 was good enough. We once reached 4 on exactly this kind of hand knowing that it was a 4-2 fit (AQ opposite KJ109) and that it was also the best contract.
Jan. 4, 2016
You are ignoring the author of this comment. Click to temporarily show the comment.
It is possible to express in cover cards. The example hand (Kxx Kxx Jxxx Kxx) has 3 NLTC cover cards and is not a LR by my calculation (9 losers). Upgrade any of the honors by one and it would contain 3.5 cover cards (8.5 losers) and qualify as a LR. If you hold 4 trumps, you can count a doubleton as 1/2 cover card (see my book “Mastering Hand Evaluation” for a complete explanation).
Dec. 26, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Fit auctions shouldn't be about HCP. My LR show 8-8.5 new losing trick count losers, whether or not I am a passed hand. That way, partner knows the offensive potential of the hand. I see no reason to bid a 9-loser hand like an 8.5-loser hand.
Dec. 25, 2015
You are ignoring the author of this comment. Click to temporarily show the comment.
Once an 8-card or greater fit has been established, an advanced losing trick count, such as New Losing Trick count (NLTC), is more accurate than HCP if you take into consideration the negative effect of duplication of values (kings and/or queens opposite shortness).

For details in how to apply the NLTC, see (shameless plug follows) my book, “Mastering Hand Evaluation” available from the Bridge Winners e-book shop or as a regular book from Amazon.
Dec. 18, 2015
1 2 3
.

Bottom Home Top