[jdeegan]

 kenberg, on 2013-August-27, 08:43, said:

The first response to the OP was from nigel_k who said that it depends on exactly what question is being asked. I agree wholeheartedly. The following hand perhaps supports this point. I am South. Mathchpoints (I forgot to put that into the daigram).

---

Assuming (I can always hope) that you do not think that I have done anything particularly moronic yet, what now? The five level belongs to the opponents? Does LOTT help? Just how many total trump are there? I post the full hand below, with a bit of a gap, so you can think before seeing it.








The full hand:

---

Huh! EW have 12 clubs, NS have 9 hearts, doesn't LOTT say that the number of tricks in a club contract plus the number of tricks in a heart contract is supposed to be 12+9=21? We are a little short. In hearts, the defense can be Ace of spades, spade to K, ruff, heart Ace holding hearts to 9 tricks. In spades it may be harder to find, but a diamond lead appears to hold a spade contract to 9 tricks via 2 spades, a heart and a third round diamond ruff (LOTT does assume best defense, right?) . In clubs, take the first 2 diamonds and sit and wait for a heart, holding clubs to 10 tricks.

OK, so this time "Obey the LAW" is the wrong adage, "The 5 level belongs to the opponents" is the right adage. BUT. Passing 5♣ scores a little under 40%, doubling 5♣ scores a little under 50%. I bid 5♥ and scored 93% since the opponents forgot to beat it: Club lead ruffed, heart lead ducked (now I will be down 1 at worst), heart continuation won by E, another club, run diamonds, score it up.


OK, E should beat it. After I ruff the first club, just how was E planning on beating this if I was holding the Ace of spade? So yes, he should place that card in his partner's hand and then the defense is easy. If it should turn out that declarer (that's me) should turn up with the spade Ace then that's just too bad, they were never beating it.

So I should have been down. Yes. But I wasn't. And was 5♥ a bad bid? Wrong on this hand, but at the time I made my decision I was far from convinced that we were beating 5♣.

It seems to me that this is the usual situation. We make close decisions, we are sometimes wrong, we are sometimes lucky. I guess tis hand exhibits my bad bidding since we can beat 5♣ and even bad bidding from the oppoents since they can beat 4♠. And it exhibits bad play, since they can beat 5♥. Otoh, everything seemed reasonable at the time.

Btw, my real objection to LOTT is not that the totals don't always work out right. The problem is that at the time of decision, the trump totals are usually, as here, not known. I knew we had 8 spades and probably not 9, I did not know we had 9 hearts, and I knew EW had "a lot" of clubs but translating "a lot" into 12 was beyond my powers. And then the total trumps were 2 more than the total tricks anyway. What's a guy to do?


May we all get better. I think I still bid 5♥ next time, but perhaps I am nuts. I have been called worse.

The real hand only proves that north can't bid a lick. Not doubling 5♣ with the worst playing hand imaginable is beyond God knows what. Pobrecito south. Imo, 5♥ is correct under the incorrect assumption that my partner is not a clueless idiot who bid like he/she wanted me to take the push.

[jdeegan]

 JLOGIC, on 2011-February-22, 14:37, said:

Winner!

Seriously, I have often been lucky enough to often be in a situation where I was playing some regional KO against a reasonable but not great flight A teams. I would say every single time the biggest factor in beating them was that they were just outgunned in cardplay. Bidding at imps is pretty easy, you try and bid close games and not do anything stupid. If there were a close slam hand or a freak hand, I would not be happy, since if it was just partscores and games we would probably always beat those teams, slams to me just made the match more random.

The fact is, if your team is significantly stronger in cardplay, you will be a big favorite in every match, and if you are significantly weaker, you will be a big dog.

It doesn't matter whether you open aggressively or not, or preempt aggressively or not, or bid 40 % games or not, none of those things offer much edge. Maybe you are gaining .05 imps a board with a certain bid. On the other hand if you read the cards well and guess a queen better than your opponent, you might be gaining 3 imps in equity on it. It is not even close how much more important card play is. The only really big edge to be had in bidding at imps is slam bidding, but that doesn't come up that often and presumably your opps won't be completely hopeless at it even if they're inferior, so it's just not a big enough edge.

If you are ever lucky enough to play a top 8 seed in the spingold, I think it will be painfully obvious to you that the reason you lose is because your opponents are not making mistakes in the cardplay, and it's winning them a lot of game (and maybe partscore) swings. That is the bread and butter of knockout matches. I find it laughable that anyone thinks they will come away from that experience thinking "Wow, I just got outbid!" or "80 % of the imps I lost were in the bidding "

And that is imps, MPs is even more about taking tricks.

There are 2 reasons people like to think bidding is 80 % of the game.

1) As has been said here, maybe when the blue team played the aces, or the diamond team plays the fleisher team, 80 % of the swings were bidding because their card play was near perfect. I can accept that, but that is only because they are at the very highest level and they all play the hands very well. This does not apply to you if you are not on one of the top 10 teams in the country.

2) Cardplay is harder to improve in, and it is more boring and concrete. Bidding on the other hand you can change your system all the time, make a few things more optimal, and really feel like you are making big improvements to your game. It is also more fun. It is just people lying to themselves. It is the same reason that almost all threads are about bidding and not play, no one can be proven wrong in a bidding thread, and people can debate it endlessly. On a cardplay thread we just get rainer posting the solution and everyone nodding...not that conducive to discussion, or being able to hold a view and keep thinking that you're right! It is the same reason that bridge teachers even at the lowest level teach bidding classes rather than play classes, and *cringe* teach conventions. People want to come away from a lesson feeling like they learned something, like they made some tangible gain. Going home and saying "look, I know kickback, it's going to save me so much room!" is a lot more rewarding than saying "well we went over some hands and I counted winners and losers and figured out what to do with my losers, blah, basically the same stuff I already knew..."

As roger said, there is no shortage of people who WANT to believe that bidding is 80 % of the game, unfortunately it is more like 10 %.

I would like to add a minor addendum to the, imo, definitive answer (shown in its entirety above) to the original question. You can actually learn a lot about dummy play from books and computer lessons. Both defense and dummy play require the skill of counting which in turn requires practice and mental conditioning. Expert dummy play also relies on many different recurrent themes. Everything from strip endplays to the Morton's Fork coup. A little book learning applied here can go a long way. On defense, you need a partner, and that makes it quite a bit more complex to learn.

[jogs]

 jdeegan, on 2013-August-31, 21:02, said:

On defense, you need a partner, and that makes it quite a bit more complex to learn.

For flight A players most points are lost on defense. Defense requires
the partners to play as a team. Often the partners are defending at
cross purposes.
In your example the East player defended without thinking. He did not
attempt to count his side's tricks or declarer's tricks.
Your 5♥ bid would have been a loser against competent defenders.
The correct contested bid depends on the skill of our opponents.
Bidding is a larger part of the game on contested boards than uncontested
boards. Bidding 35-40% of the game on contested boards. At most 20%
on uncontested boards.

[Wwchang]

I think it depends. Bidding system is relatively unimportant, but I think bidding judgment is quite important: making 200 when the other side is making 420 is no good, and if you make a poorly judged sacrifice, then particularly at MPs there may be no difference between -300 and -500. I would be shocked if Spingold top seeds do not have significantly better bidding judgment, on the whole, than Spingold bottom seeds. (Whether that superior bidding judgment works out in a particular sample of 64 boards is luck.)

One partner of mine (an intermediate), whose stats I've tracked through 800 cross-IMP boards, loses on average 0.57/board on bidding (0.37 of this is competitive, 0.12 is non-slam constructive, and 0.09 is slam bidding), 0.30/board on dummy play, and 0.12/board on defense, compared to "sound" play (which is limited to my skill in the post mortem, aided by an engine); I look at each play and try to determine whether it is reasonable, even if it does not lead to the double dummy best result. I don't track what could have been done with a different system, so these are all related to misjudgment. I am sure that there are additional errors that I don't pick up in my analysis, probably in the range of 0.2-0.3 imps/board in total. This also does not include any errors that end up not costing actual IMPs on the particular board.)

So at least for this individual, working on improving bidding judgment would appear not to be time wasted.

Errors are errors; they can all cost. At least in my set of stats, the average bidding error cost 6.9 imps, the average dummy play error cost 4.7 imps, and the average defensive error cost 2.7 imps.

[PhilKing]

 Wwchang, on 2013-September-01, 12:37, said:

Errors are errors; they can all cost. At least in my set of stats, the average bidding error cost 6.9 imps, the average dummy play error cost 4.7 imps, and the average defensive error cost 2.7 imps.

Most card play errors are not fatal, so the average cost is much lower. For instance, partner misplays a game, but the defence don't seize the chance, or they play a 70% line and make when a 100% chance was available. Card play errors tend to get punished more at a higher level, and the defenders put you under way more pressure. I'd be willing to bet that a typical intermediate player makes an average of at least three card play errors per hand.

And in the bidding you get the same thing. For every three errors your partner makes in the bidding perhaps only one leads to the wrong contract, so take care with your stats.

[jogs]

 kenberg, on 2013-August-27, 08:43, said:

Btw, my real objection to LOTT is not that the totals don't always work out right. The problem is that at the time of decision, the trump totals are usually, as here, not known. I knew we had 8 spades and probably not 9, I did not know we had 9 hearts, and I knew EW had "a lot" of clubs but translating "a lot" into 12 was beyond my powers. And then the total trumps were 2 more than the total tricks anyway. What's a guy to do?


May we all get better. I think I still bid 5♥ next time, but perhaps I am nuts. I have been called worse.

LoTT is not an absolute. It is just a rough guideline. 30/40/30.
30% of the time there are more tricks than trumps.
40% of the time tricks equal trumps.
30% of the time there are less tricks than trumps.

Only that 40% is actually less than 40%. Whenever there are more
than 18 total trumps, expect tricks to be less than trumps.
Whenever patterns are flat, expect tricks to be less than trumps.
Skewed patterns produce more tricks than trumps. Skewed patterns
are ones which contain many singletons and voids.

[nige1]

 Wwchang, on 2013-September-01, 12:37, said:

I think it depends. Bidding system is relatively unimportant, but I think bidding judgment is quite important: [SNIP]
Errors are errors; they can all cost. At least in my set of stats, the average bidding error cost 6.9 imps, the average dummy play error cost 4.7 imps, and the average defensive error cost 2.7 imps.

Agree with Wwchang.

 PhilKing, on 2013-September-01, 12:52, said:

Most card play errors are not fatal, so the average cost is much lower. For instance, partner misplays a game, but the defence don't seize the chance, or they play a 70% line and make when a 100% chance was available. Card play errors tend to get punished more at a higher level, and the defenders put you under way more pressure. I'd be willing to bet that a typical intermediate player makes an average of at least three card play errors per hand.

And in the bidding you get the same thing. For every three errors your partner makes in the bidding perhaps only one leads to the wrong contract, so take care with your stats.

Agree again with PhilKing Sometimes, we review matches as a team. Whether we win or lose, we find that we make about 2 obvious errors per player per deal. Few errors cost but bidding errors seem to cost more often than play errors.
Obvious points:

• Bridge is a game of mistakes: minimising your own mistakes; helping partner to avoid his mistakes; recognizing your opponents' mistakes; co-operating with partner to exploit them.
• Good bidding requires the ability to play the hand in your imagination. So the best bidders are expert card-players.
• In the auction, we make only a few bids; but as a declarer or defender we make lots of plays. The later the action, the more immediate the effect on the result.
• Nevertheless, people skills differ. Some are good at bidding; others good at defence. Some excel at pairs; others at teams. In each area, only the difference between your skill-level and that of your opponents matters.

[kenberg]

 jogs, on 2013-September-01, 18:44, said:

LoTT is not an absolute. It is just a rough guideline. 30/40/30.
30% of the time there are more tricks than trumps.
40% of the time tricks equal trumps.
30% of the time there are less tricks than trumps.

Only that 40% is actually less than 40%. Whenever there are more
than 18 total trumps, expect tricks to be less than trumps.
Whenever patterns are flat, expect tricks to be less than trumps.
Skewed patterns produce more tricks than trumps. Skewed patterns
are ones which contain many singletons and voids.

I don't quarrel with these (presumably estimated) statistics, and I know there are a lot of adjustments. But I want to mention again something I seldom see discussed. Quite often, when the decision must be made, I don't know the total number of trumps. That certainly was the case in the hand I posted.

I am not on any campaign against The Law, I'll leave that to Mike Lawrence. But it does, in some quarters, assume the status of Holy Writ. Too much.

I try to keep in mind the question "What have I shown? Do I have more?". This applies to points but perhaps even more to distribution.


Anyway, I am not sure what, if anything, the posted hand proves or even suggests. As was suggested above, there is probably something to be said for N doubling to discourage a 5 level bid by S. And there is a lot to be said for EW beating 5♥.

Earlier on this thread I posted a hand from the Spingold. This later hand was not from the Spingold.

I guess I just want to stick with the idea that identifying the importance of plat versus bidding is not as simple as some believe.

[jogs]

Couldn't find the delete key for deleting a post.

Decided to just change the post.

LoTT: we actually do not know the total trumps as often
as we would like. Still in many auctions where opponents
have preempted, it is the only available tool for estimating
our tricks.

[jogs]

The game of bridge is complex. If it were easy the game would be solved
and uninteresting.
Cohen was too absolute in stating the relationship of total tricks equal to total
trumps. Lawrence is also wrong in his claim that there is no relationship
between tricks and trumps. It is somewhere in between.

Back to the board where North should probably double.

---

Exchange the location of the ♠6 and ♥6. Now 5♣ is unbeatable.

---

[GreenMan]

 jogs, on 2013-September-02, 09:10, said:

Cohen was too absolute in stating the relationship of total tricks equal to total
trumps. Lawrence is also wrong in his claim that there is no relationship
between tricks and trumps. It is somewhere in between.

Lawrence hasn't made that claim as far as I know.

Cohen focuses on the mean and ignores the standard deviation. Lawrence highlights the SD as a corrective.

The question IMHO is whether the high variance in the equation means that the Law causes as many problems as it solves. I believe it does, once you have enough experience to apply more sophisticated evaluation techniques. Cohen's work helped me get through an important formative period, and I do think it's useful for beginners and low intermediates, just like second hand low and so forth.

[jogs]

 GreenMan, on 2013-September-02, 09:37, said:

Lawrence hasn't made that claim as far as I know.

http://www.newbridgelaw.com/

This is Lawrence's website: Discussing the Law of Total Tricks

Quote

If you are a questioning person you may wonder why trumps and tricks should go hand in hand. Lots have been written about the Law, but practically no explanation has been given. Larry Cohen skips the issue, just like the other main proponent of the theory, Cohen's former partner Marty Bergen. And there is a very simple reason for their silence: There is no connection. If we have a deal where the total tricks are 16 and the total trumps also are 16, the tricks are not a function of the trumps. They are related to each other only tenuously and indirectly.

Trumps is just one of many variables for estimating tricks.

[kenberg]

 jogs, on 2013-September-02, 09:10, said:

The game of bridge is complex. If it were easy the game would be solved
and uninteresting.
Cohen was too absolute in stating the relationship of total tricks equal to total
trumps. Lawrence is also wrong in his claim that there is no relationship
between tricks and trumps. It is somewhere in between.

Back to the board where North should probably double.

---

Exchange the location of the ♠6 and ♥6. Now 5♣ is unbeatable.

---

Proving another point: Bridge is a very interesting game.

[GreenMan]

"no connection" and "related tenuously and indirectly" are contradictory. Lawrence appears to be saying "no connection" as hyperbole, then qualifying it. (Obviously if trumps were not related to tricks at all, then a 1-1 fit would produce the same # of tricks on average as a 6-6.)

[blackshoe]

Vernes found a correlation between the number of trumps and the number of tricks. Lawrence appears to be saying there's no causal link. That's a horse of a different color.

[GreenMan]

I doubt anyone would say that there's no causal link between how many trumps you have and the number of tricks you can take. Unless you can make a case that causality runs in the other direction, or that some third variable has causal effects on both.

To me, Lawrence is simply saying that the correlation is not strong enough to rely on.

[jogs]

Lawrence is claiming his short suit totals of the side suits
play a greater role than total trumps.
In selected scenarios this is true. When there is a double
fit and each partner has only four cards in the short suits,
the distribution of the short suits determine our winners.

5=4=?=? facing 4=5=?=?

For simplicity all our points will be in the majors.

1) AKxxx, QJxx, xxxx, - facing QJxx, AKxxx, - , xxxx
2) AKxxx, QJxx, xxx, x facing QJxx, AKxxx, x , xxx
3) AKxxx, QJxx, xx, xx facing QJxx, AKxxx, xx , xx

There are no obvious losers in hand 1.
Two losers in hand 2.
Four losers in hand 3.

[akwoo]

I think it is definitely useful to view the Law of Total Tricks as a first approximation to the Law of Total Losers (the total number of tricks lost by the two sides in their contracts is the sum of the shortest suit lengths in each of the four suits).

[pdmunro]

 Fluffy, on 2011-February-21, 13:02, said:

you can't live without one or the other.

Fluffy, I agree. Maybe questioning whether bidding or play is more important in bridge is like asking whether sex or money is more important in marriage.

[Free]

 pdmunro, on 2013-September-10, 08:57, said:

Fluffy, I agree. Maybe questioning whether bidding or play is more important in bridge is like asking whether sex or money is more important in marriage.

That's a poor comparison. By having one you can get the other, so neither is important depending on how monogamous you are ofcourse