[bluechip10]

I am a Bergen disciple. That is why I like Bergen raises, support dbl/redbl so we can follow the LOTT (Law of total tricks). Take the overcall example (first hand). First, I would dbl the 1 dia bid and then bid spades to show a 16+ hand. Even though this hand is "only" 15hcp, the texture gives it the extra value. Anyway, given the 1 spade overcall, I would bid 3sp as the responder to show 4 spades in support. The only forcing, strength showing bid would be a cuebid. The other hands that the partnership opened, Bergen raises would help. You would know you have 10 trumps between you. So, bidding 4 in most cases (avoiding hindsight) satisfies me. Bottom line: The Law of Total Tricks works.

[cherdano]

Hand 1: West has a beautiful 15 hcp hand: a 6th trump, three aces, heart and spade tenaces over opener, all cards working. He should at least invite (and bidding 4♠ is better than bidding 3♠).
East would accept the invite with his 8 hcp despite the possible wasted king, because of his 4th trump.

Hand 2: West has 15 well-working hcp, a 6th trump, and a singleton in their suit. He should at least invite - if in doubt, overbid with shortness in their suit. North should accept with all 6 hcp working nicely (even a J is a decent card when in a KJ combination), and with his 4th trump.

Hand 3: West has a very suitable 6-4 (good suit and aces plus intermediates in the trump suit). Assuming 1♠ shows 5, the crucial thing to realise is that this hand becomes much better opposite a 5-card spade suit - this makes it much less likely that dummy can be forced successfully, or that our ruffs (in red suits in dummy, and clubs in hand) will promote a trump trick for opponents. It would already be tempting to bid 3♠ if 1♠ showed 4+ (I wouldn't, because partner won't be able to evaluate well since we have to bid the same way with a balanced 18 count), but I think opposite 5+ you have to. We have play for 4♠ opposite ♠Axxxx and out.


In my view, life is much easier if you start with a simple and decent hand evaluation method, i.e. hcp, instead of a flawed one like losing trick count. How many adjusted versions of losing trick count are there?

[Winstonm]

 cherdano, on 2019-January-21, 23:57, said:

Hand 1: West has a beautiful 15 hcp hand: a 6th trump, three aces, heart and spade tenaces over opener, all cards working. He should at least invite (and bidding 4♠ is better than bidding 3♠).
East would accept the invite with his 8 hcp despite the possible wasted king, because of his 4th trump.

Hand 2: West has 15 well-working hcp, a 6th trump, and a singleton in their suit. He should at least invite - if in doubt, overbid with shortness in their suit. North should accept with all 6 hcp working nicely (even a J is a decent card when in a KJ combination), and with his 4th trump.

Hand 3: West has a very suitable 6-4 (good suit and aces plus intermediates in the trump suit). Assuming 1♠ shows 5, the crucial thing to realise is that this hand becomes much better opposite a 5-card spade suit - this makes it much less likely that dummy can be forced successfully, or that our ruffs (in red suits in dummy, and clubs in hand) will promote a trump trick for opponents. It would already be tempting to bid 3♠ if 1♠ showed 4+ (I wouldn't, because partner won't be able to evaluate well since we have to bid the same way with a balanced 18 count), but I think opposite 5+ you have to. We have play for 4♠ opposite ♠Axxxx and out.


In my view, life is much easier if you start with a simple and decent hand evaluation method, i.e. hcp, instead of a flawed one like losing trick count. How many adjusted versions of losing trick count are there?

Extending this thinking a little further, all methods of evaluation, i.e., losing trick count, law of total tricks, point count, are all simply tools to be used rather than principles to which we must adhere. Although, AQJ would be 1 loser in LTC evaulation, when RHO bids or overcalls this suit LTC probabilities change. Length, fit with partner, double fits, interior spots, honor placement - all of these things and more go into the evaluation of a hand. It's good to have a general idea of the value of a hand - but generalizations are falsehoods concerning the specific.

[phoenixmj]

I just read an article by Kevin Wilson on Modified LTC. Using this method - you actually subtract from 25 instead of 24. There are also some slight modifications in calculation of the LTC. In this case if I apply it - On hand number 1 and 2, I get that we should be in game. On number 3, I get that we should stop at 3.

All three hands made 4 in "real life" - but double dummy says that hand 1 and 2 should make 4 and number 3 (on best defense) should make 3. Now, we were not the only ones to make 4 on all three. I think that number 3 made 4 every time played that day - but it is interesting that the modified ltc does say to bid 4 on 1 and 2 but not 3.

Has anyone used the mltc and used 25 instead 24 as the total losers? If so, what has been your experience with it?

[billyjef]

 phoenixmj, on 2019-January-22, 10:50, said:

I just read an article by Kevin Wilson on Modified LTC. Using this method - you actually subtract from 25 instead of 24. There are also some slight modifications in calculation of the LTC. In this case if I apply it - On hand number 1 and 2, I get that we should be in game. On number 3, I get that we should stop at 3.

All three hands made 4 in "real life" - but double dummy says that hand 1 and 2 should make 4 and number 3 (on best defense) should make 3. Now, we were not the only ones to make 4 on all three. I think that number 3 made 4 every time played that day - but it is interesting that the modified ltc does say to bid 4 on 1 and 2 but not 3.

Has anyone used the mltc and used 25 instead 24 as the total losers? If so, what has been your experience with it?

It is what I use and I am also a student of Kevin's

[phoenixmj]

 bluechip10, on 2019-January-21, 09:42, said:

I am a Bergen disciple. That is why I like Bergen raises, support dbl/redbl so we can follow the LOTT (Law of total tricks). Take the overcall example (first hand). First, I would dbl the 1 dia bid and then bid spades to show a 16+ hand. Even though this hand is "only" 15hcp, the texture gives it the extra value. Anyway, given the 1 spade overcall, I would bid 3sp as the responder to show 4 spades in support. The only forcing, strength showing bid would be a cuebid. The other hands that the partnership opened, Bergen raises would help. You would know you have 10 trumps between you. So, bidding 4 in most cases (avoiding hindsight) satisfies me. Bottom line: The Law of Total Tricks works.

Without the intervening bids, we would bid Bergen raise to show 4 card support.

With the bidding for number 1 (the 1 spade bid was an overcall) - I think a 3 spade bid would show 4 cards support but also a weaker hand. In this case, west has a 9 LTC and 8 HCP - we would think the hand is too strong to jump to 3.

I will read up on it - but I think we play that Bergen is off the table with an intervening bid. Else, in hand number 2, we would bid 3 clubs to show a 9 ltc and 4 card support. If I bid 3 clubs for number 2 having had them overcall 2 clubs, it would be a cue bid showing a limit raise - which south does not have.

In hand number 3, south opens a club and after an intervening bid, north bids a spade. the 2 spade bid says south has 4 spades. That totals 9 so LOTT would imply 9 tricks.

[mikeh]

I used to use LTC as one of the several metrics we applied in the most complex partnership method I have ever played, but it was the least important and used only after we looked at hcp and controls, as well as suit texture, location of honours, and the like. As with virtually all valuation methods of any utility, there is overlap between methods.

I don't consciously use LTC anymore, although I do find myself, from time to time, doing the arithmetic, probably more from habit than anything else. I would never recommend using it as the main metric, let alone the only one.

LTC was invented early in the history of contract bridge: I have an edition of what I think was the original LTC book. It is important to understand the state of bidding theory at the time. Players were generally hung up on honour tricks aka quick tricks, which was a holdover from earlier forms of the game, where bidding to one's contract was meaningless. Play 1D and make 12 tricks...you got the slam bonus!

If one had the balance of power, one tried to win the auction (in 'auction bridge', the immediate forerunner of contract) as cheaply as possible. One lost points for going down but making 12 tricks got you slam unless you had bid 7. So bidding accuracy was far less relevant than it became when one had to make a 'contract' to take a certain number of tricks, and one didn't score game or slam bonuses unless one bid to the appropriate level.

Honour tricks held on for many years, even in the face of the Work Point Count, which was originally limited to and used only in the context of notrump bidding. Many other methods were tried, mostly variants of point count, including fractional points.

LTC was simple. It was almost as simple as quick tricks...indeed, take a look at quick trick tables from bridge texts from the early 1930's and one could well argue that LTC was simpler even than quick tricks. It also afforded a very simple way of evaluating how high to bid, which had gained importance with contract.

So those who use the LTC, even modified versions, are using an outdated methodology that was better, in its day, than almost all of the competing methods, but which is no longer an important part of the valuation methods of most experts today. That isn't to say that it is utterly irrelevant, but it should be at most a minor factor in what should be a multi-faceted valuation approach.

I've often said that one should look at hcp, modified by number of controls (Aces are worth more, in most contexts, than the 4 hcp in the 4321 method, as one example, and one needs 10 controls, or corresponding shortness, to bid slams, etc), shape (and I mean not merely length but degree of fit), location of hcp (in long suits better, in short suits not so much, in partner's suits excellent), combination of values (QJx xxx is better, initially, than Qxx Jxx), possession or lack of suit texture (gotta love 10's, and 9's are nice too), and LTC.

I know of no good player who would ever describe any hand by reference to a number, whether it be 'LTC of 6', or '14 hcp' or '4 controls', etc except when addressing a specific aspect in light of the auction. Focusing on one metric to the large exclusion of others prevents one from learning that hand evaluation is an holistic endeavor, and the weight to be given to individual aspects varies with every round of the auction.

I'd suggest starting with the 4321 count, learn to add for length, learn to adjust for shortness only after the auction tells you that it may be valuable, and be aware that one needs to re-evaluate with every call at the table. Once one has that reasonably learned, start looking at how things like LTC can cause one to perhaps upgrade some hands and downgrade others.

[phoenixmj]

 mikeh, on 2019-January-22, 15:24, said:

I used to use LTC as one of the several metrics we applied in the most complex partnership method I have ever played, but it was the least important and used only after we looked at hcp and controls, as well as suit texture, location of honours, and the like. As with virtually all valuation methods of any utility, there is overlap between methods.

I don't consciously use LTC anymore, although I do find myself, from time to time, doing the arithmetic, probably more from habit than anything else. I would never recommend using it as the main metric, let alone the only one.

LTC was invented early in the history of contract bridge: I have an edition of what I think was the original LTC book. It is important to understand the state of bidding theory at the time. Players were generally hung up on honour tricks aka quick tricks, which was a holdover from earlier forms of the game, where bidding to one's contract was meaningless. Play 1D and make 12 tricks...you got the slam bonus!

If one had the balance of power, one tried to win the auction (in 'auction bridge', the immediate forerunner of contract) as cheaply as possible. One lost points for going down but making 12 tricks got you slam unless you had bid 7. So bidding accuracy was far less relevant than it became when one had to make a 'contract' to take a certain number of tricks, and one didn't score game or slam bonuses unless one bid to the appropriate level.

Honour tricks held on for many years, even in the face of the Work Point Count, which was originally limited to and used only in the context of notrump bidding. Many other methods were tried, mostly variants of point count, including fractional points.

LTC was simple. It was almost as simple as quick tricks...indeed, take a look at quick trick tables from bridge texts from the early 1930's and one could well argue that LTC was simpler even than quick tricks. It also afforded a very simple way of evaluating how high to bid, which had gained importance with contract.

So those who use the LTC, even modified versions, are using an outdated methodology that was better, in its day, than almost all of the competing methods, but which is no longer an important part of the valuation methods of most experts today. That isn't to say that it is utterly irrelevant, but it should be at most a minor factor in what should be a multi-faceted valuation approach.

I've often said that one should look at hcp, modified by number of controls (Aces are worth more, in most contexts, than the 4 hcp in the 4321 method, as one example, and one needs 10 controls, or corresponding shortness, to bid slams, etc), shape (and I mean not merely length but degree of fit), location of hcp (in long suits better, in short suits not so much, in partner's suits excellent), combination of values (QJx xxx is better, initially, than Qxx Jxx), possession or lack of suit texture (gotta love 10's, and 9's are nice too), and LTC.

I know of no good player who would ever describe any hand by reference to a number, whether it be 'LTC of 6', or '14 hcp' or '4 controls', etc except when addressing a specific aspect in light of the auction. Focusing on one metric to the large exclusion of others prevents one from learning that hand evaluation is an holistic endeavor, and the weight to be given to individual aspects varies with every round of the auction.

I'd suggest starting with the 4321 count, learn to add for length, learn to adjust for shortness only after the auction tells you that it may be valuable, and be aware that one needs to re-evaluate with every call at the table. Once one has that reasonably learned, start looking at how things like LTC can cause one to perhaps upgrade some hands and downgrade others.

Thanks so much for the reply. We do use 4321 and other evaluation methods for hands. But, we do tend to somewhat rely on LTC when making a final decision on level to bid. I think controls is something we need to learn more about when bidding. We use controls when considering slam, but not so much when considering game. Not sure how to communicate controls at a level low enough to assist with a game try. On the hands given, due to competitive bidding, it was hard for us to communicate the total number of trumps that we had. We definitely see LTC as only one tool. Interestingly, using Modified LTC actually got the "right answer" in each case. We are not advance players - although we are certainly trying to get more proficient - and we have been generally impressed with the results LTC has given us. I think we are very weak at looking at controls other than when there is slam interest.

[billyjef]

"I don't consciously use LTC anymore" It is, as it seems it was for you, of great value to examine hands afterwards, using various metrics, even metrics we will never use at the table, because that builds up our bridge judgment that becomes intuitive at the table. When I look at hands now, I often and immediately, see their intrinsic strength, in the moment and situation, and I am able to observe fluid like, how that changes as the situation and moment changes. This comes, from what OP is doing, in my opinion.
"It also afforded a very simple way of evaluating how high to bid, which had gained importance with contract." This skill is still very important, useful and shouldn't be, IMO, dismissed as marginal at best. I know you aren't dismissing LTC outright, and it is right not to glorify it as the only evaluation tool that one needs, and I didn't feel the OP was saying that, but it is still a very valuable tool for discussing the general intrinsic trick taking value of a hand objectively, even if it isn't absolute and money in the bank.

So yes, no good player speaks of any metric as the Rosetta stone of a hands true value, but all the ones I know, do look at and speak about a hands referencing losing/winning count.

Finally, stating LTC as outdated yet recommending other metrics that are equally dated, is too much of a dismissal for me to let ride. As clearly you know, much can be mined from that which others call outdated.

I really did think this was a great post if you removed the implications that LTC was any more or less inferior to any other single metric. The warning is appreciated, but that same warning goes to worshiping HCP, Honor Tricks, Quick Tricks, Binky points etc.