BBO Discussion Forums: How do good players really play - BBO Discussion Forums

Jump to content

  • 2 Pages +
  • 1
  • 2
  • You cannot start a new topic
  • You cannot reply to this topic

How do good players really play

#1 User is offline   thepossum 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,362
  • Joined: 2018-July-04
  • Gender:Male
  • Location:Australia

Posted 2019-February-04, 22:29

Dear all

I have been very interested in this question since I started playing on here 8 months ago. Ive been playing for many years but until I started reading about computer bridge, double dummy solvers, GiB etc, I never really thought about top players and the statistical precision that many discuss now. Obviously many fairly simple probabilities are known by intermediate to good players - probabilities of splits, finesses, adjustment for leads and bidding, 50% targets for MPs and whatever the probabilities are for IMPs. I also know there are very complex methods of hand evaluation, other than expertise, which attempt to give a number to a good hand. However most people would not be capabale of calculating these on the fly. Obviously there are people in the world who could easily calculate KR on the fly. But most of us, and from what I read, most good players rely on their experience combined with counts and heuristics and memory during the pay, and of course excellent deductive and inferential abilities.

After reading about GiB and double dummy approaches it seems that the philosophy towards double dummy was flawed. I do not believe that an expert player uses any form of simulation to approach a play. They use their memory, knowledge, expertise, simple probabilities (orat least reasonably calculable by a human brain in a second or so), judgement etc ....... to judge what to do. Most of our human guidelines and rules and heuristics have been based upon 100s ?? of years of play and outcomes, not really simulation (although an almost uncountable number of hands from a very large number of people is a simulation of sorts)

When I read forums here and response to queries I often see people discussing fairly precise probabilities, and having done some quick probability calculations in order to advise people on a bid or a play

So my question as a (an ordinary but experience) player, a mathematician and statistician, psychologist (I forgot to mention psychology above), ex AI researcher, and generally curious person is .... what is the reality behind good players and how that has translated into computational theory. I think many questions eed to be asked about the approach to computational bridge. Ive read a bit over the last 6-8 months and it is acknowledged as a very complex informational problem. Is there a lot of space left for research and is the simulation approach always going to result in less than expert players. My feeling is, that, along with many big data approaches, that they tend to exclude a great deal of heuristic, behavioural and other knoweldge that a top bridge player can bring to the table

I would be grateful for any discussion from experts in the field of computational bridge

regards Possum
0

#2 User is offline   sfi 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,576
  • Joined: 2009-May-18
  • Location:Oz

Posted 2019-February-05, 00:09

View Postthepossum, on 2019-February-04, 22:29, said:

But most of us, and from what I read, most good players rely on their experience combined with counts and heuristics and memory during the pay, and of course excellent deductive and inferential abilities.


When it comes to hand evaluation, every good player I've discussed this with (and it's been many over the years) uses little more than 4-3-2-1 count, losing tricks, number of trumps, key cards and judgement. The judgement of a top class player is so much superior to any numerical computation that it's not even a close call.

In fact, any book above a true beginner level that focuses on a more accurate calculation (rather than evaluation) should be thrown straight in the bin, IMO. It directs the student's attention to entirely the wrong place.

Calculating odds in the play is somewhat easier, due to three fairly straightforward things that can be applied at the table and lead to close approximations:
  • Know the odds of a break for when there are 2-7 cards in the opponents' hands. Any good player should have these cold - post them on your fridge if you have to. The phrases "52-48", "78-22", "40-50-10", "68-28-4", "36-48-14-2" and "62-30-7-1" should be ingrained in the memory. (8 is "47-33-17-3-0" if you feel a bit more adventurous.)
  • Understand the theory of vacant spaces and how to apply it. This is the crux of how people calculate numbers at the table.
  • Understand how restricted choice is to be applied, both in simple situations and in vacant space calculations.


All of these are easily calculable at the table, so you can often work out things like, based on the information you have so far, the relative odds are precisely 4:7 that East has the trump queen. It gets easier with practice, and it's worth practicing. Working out the absolute percentages is often tougher and less valuable because you frequently have to take into account things you can dismiss using the techniques mentioned above.

After all this, in close decisions good players tend to trust their instincts over a 1-2% difference. Table feel does come into play and there are very few players that don't give away any information with their demeanour. Great players will be able to explain exactly why they didn't go with the odds. Don't play them for money.
1

#3 User is offline   sfi 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,576
  • Joined: 2009-May-18
  • Location:Oz

Posted 2019-February-05, 00:17

View Postsfi, on 2019-February-05, 00:09, said:

When it comes to hand evaluation, every good player I've discussed this with (and it's been many over the years) uses little more than 4-3-2-1 count, losing tricks, number of trumps, key cards and judgement. The judgement of a top class player is so much superior to any numerical computation that it's not even a close call.


For example, this hand came up 3 years ago. How does your evaluation of this hand change between the first and second rounds of the auction:



The initial pass is certainly supportable, but you should now be closer to jumping to 6 than passing again (5D is the sensible bid). Is it really worth trying to spend lots of time trying to model that mathematically?
0

#4 User is offline   mikeh 

  • PipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 12,835
  • Joined: 2005-June-15
  • Gender:Male
  • Location:Canada
  • Interests:Bridge, golf, wine (red), cooking, reading eclectically but insatiably, travelling, making bad posts.

Posted 2019-February-05, 01:33

Assuming that you are asking sincerely, the answer is probably both simpler than you imagine and more complex than you’d believe.

Hand evaluation is relatively straightforward initially. Count hcp, look at suit lengths and internal texture. I don’t know if any good player who computes a number, beyond the 4321 count. Having established that it is, say, 11 hcp, the expert adjusts unconsciously for controls, combined honours, suit length, suit texture and decides whether he or she likes the hand.

The process is ongoing with every action During the auction. Adjustments are made, not in terms of numbers but in terms of liking or disliking the hand. How does our hand fit partner? What impact does the opponents’ bidding, if any, have on our hand?

For competitive bidding, degree of fit and shape are the main factors. For game, it’s more overall strength, which includes hcp, shape and degree of fit. For slam it’s all of that plus controls.

In the play, sure, every expert knows roughly the a priori odds, but far more important are such things as drawing inferences not only from what the opps played but also from what they did not play. Negative inferences are as valuable as positive ones. Visualization is key....some squeezes play themselves, but seeing a trump squeeze or a crisis cross squeeze requires, for most, some real concentration. Counting needs to become almost automatic. Concentration is absolute. Back when I played frequently, I could tell you, at the end of the session, the auction on each hand and precisely who played which card. Every expert I know can rattle off those details when discussing a hand later.

Reese once write, and I paraphrase, that if the average player could read the mind of an expert on a difficult declarer play problem, he would not believe it.

Many plays are very difficult to work out at the table. Most of the advanced technical plays were discovered only by looking over hands where the play was missed. This means that reading books is invaluable. I defy anyone to plan and execute a stepping stone squeeze if one has not read about it.

You will n,ever become an expert playing robots. You will never become an expert without playing with and against experts.and asking questions....always asking questions. How,ever don’t be one of those bores who ask an expert a question and then argue heatedly that the expert is wrong. I and friends of mine have endured countless questions about ‘what would you do’ only to find out that the questioner wasn’t trying to learn but to get approval for his silly bid or play. Those people can’t learn. Unfortunately you have exhibited those characteristics in this forum which is why I have largely ignored your posts. Hopefully I have misjudged you but my answer is intended for anyone who wants some insight into how experts think.

Few players ever become experts, other than the several million who claim expert status on BBO. Those who do will work hard at learning the game and will pay their dues playing with non-experts, and working their way up gradually until they get to play with a local expert, whether that player is a true expert depends on where one plays.
'one of the great markers of the advance of human kindness is the howls you will hear from the Men of God' Johann Hari
0

#5 User is offline   sfi 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,576
  • Joined: 2009-May-18
  • Location:Oz

Posted 2019-February-05, 03:28

View Postmikeh, on 2019-February-05, 01:33, said:

Back when I played frequently, I could tell you, at the end of the session, the auction on each hand and precisely who played which card. Every expert I know can rattle off those details when discussing a hand later.


Interestingly, I've tried long and hard to train myself not to retain this information after the end of the hand. So much so that I have often been surprised when looking at my scoresheet later in the match (I remember being genuinely shocked to see a -1700 recorded four boards beforehand some years ago). Even in the discussion afterwards I tend to have to recreate the thought process that was going on at the time to answer questions about the hand.

Is there a reason that transferring previous hands to long-term memory is valuable (apart from not looking vague at post mortems)?
0

#6 User is offline   thepossum 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,362
  • Joined: 2018-July-04
  • Gender:Male
  • Location:Australia

Posted 2019-February-05, 07:04

Dear Sfi and Mike

Thankyou for the responses. I am asking sincerely and professionally. It is something I am very interested in and I am challenging/questionning some of the current computational bridge theory. But I need to understand it better. I genuinely have the broad experience (in other fields) to ask these questions if not the expert bridge knowledge, despite having played for a long time and having read a great deal.

regards Possum
0

#7 User is online   pescetom 

  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 7,203
  • Joined: 2014-February-18
  • Gender:Male
  • Location:Italy

Posted 2019-February-05, 07:05

View Postsfi, on 2019-February-05, 00:09, said:

[*]Know the odds of a break for when there are 2-7 cards in the opponents' hands. Any good player should have these cold - post them on your fridge

Are divorces expensive in Oz ?
0

#8 User is offline   mikeh 

  • PipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 12,835
  • Joined: 2005-June-15
  • Gender:Male
  • Location:Canada
  • Interests:Bridge, golf, wine (red), cooking, reading eclectically but insatiably, travelling, making bad posts.

Posted 2019-February-05, 07:34

View Postsfi, on 2019-February-05, 03:28, said:

Interestingly, I've tried long and hard to train myself not to retain this information after the end of the hand. So much so that I have often been surprised when looking at my scoresheet later in the match (I remember being genuinely shocked to see a -1700 recorded four boards beforehand some years ago). Even in the discussion afterwards I tend to have to recreate the thought process that was going on at the time to answer questions about the hand.

Is there a reason that transferring previous hands to long-term memory is valuable (apart from not looking vague at post mortems)?

It’s a side effect of the level of focus that good players bring to the game. You will never be an expert unless you are aware, throughout the hand, of who bid what and who played which card in which context. The recall aspect is merely a useful side effect. It is useful for at least two reasons. Being self-critical helps. One will often, at or following an event, wake up at 2 or 3 am and realize a mistake one made, being able to visualize the hand perfectly allows for a detailed mental review. Another is that a very important means of keeping sharp, and improving, is to discuss the hands with other good players after the session. Of course, I learned most of what I know before hand records were common in club games. I used to play at a strong club, and we’d all go to a pub afterwards and discuss the hand. Nowadays, with universally available hand records, perhaps one doesn’t need this as much. However, as I say, this is a side-effect, not the primary goal.

I had to laugh at your story about the 1700. I won’t say that experts never go for 1700, but I will say that I can’t imagine forgetting it quickly. For one thing, one should pay attention to how it happened. -1700 is the sort of result that one should have at most about once every 2,000 hands, at most, unless one is saving against a vulnerable grand. I remember doing that exactly once, lol. (A 1700 save, that is).
'one of the great markers of the advance of human kindness is the howls you will hear from the Men of God' Johann Hari
0

#9 User is online   Cyberyeti 

  • PipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 13,897
  • Joined: 2009-July-13
  • Location:England

Posted 2019-February-05, 08:15

View Postmikeh, on 2019-February-05, 07:34, said:

It’s a side effect of the level of focus that good players bring to the game. You will never be an expert unless you are aware, throughout the hand, of who bid what and who played which card in which context. The recall aspect is merely a useful side effect. It is useful for at least two reasons. Being self-critical helps. One will often, at or following an event, wake up at 2 or 3 am and realize a mistake one made, being able to visualize the hand perfectly allows for a detailed mental review. Another is that a very important means of keeping sharp, and improving, is to discuss the hands with other good players after the session. Of course, I learned most of what I know before hand records were common in club games. I used to play at a strong club, and we’d all go to a pub afterwards and discuss the hand. Nowadays, with universally available hand records, perhaps one doesn’t need this as much. However, as I say, this is a side-effect, not the primary goal.

I had to laugh at your story about the 1700. I won’t say that experts never go for 1700, but I will say that I can’t imagine forgetting it quickly. For one thing, one should pay attention to how it happened. -1700 is the sort of result that one should have at most about once every 2,000 hands, at most, unless one is saving against a vulnerable grand. I remember doing that exactly once, lol. (A 1700 save, that is).


There are some 1700s that are unavoidable, using a weak no trump, occasionally you run into them, I've collected 2000 before from 1N. Those you don't particularly wake up in a cold sweat about, occupational hazard, the others fortunately are very rare. Being able to temporarily blank bad results from your mind for some players is a useful skill if they would worry about them if they didn't.

Sometimes I remember the exact cards, more often I will remember that west led the 4th highest of 5 where it doesn't matter whether it was the 3 or 4 that he led then played the 2 when I could see the other one.

I absolutely agree that you learn most in post mortems with better players.
0

#10 User is offline   barmar 

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Group: Admin
  • Posts: 21,398
  • Joined: 2004-August-21
  • Gender:Male

Posted 2019-February-05, 09:47

View Postmikeh, on 2019-February-05, 01:33, said:

Hand evaluation is relatively straightforward initially. Count hcp, look at suit lengths and internal texture. I don’t know if any good player who computes a number, beyond the 4321 count. Having established that it is, say, 11 hcp, the expert adjusts unconsciously for controls, combined honours, suit length, suit texture and decides whether he or she likes the hand.

Maybe I'm not the "expert" you're talking about, but it's not always unconscious. I often say to myself things like "Wow, this is a really nice 11 count" or "This may have 20 HCP, but it's still pretty lousy because of the 4333 shape and all the spots are 2s and 3s -- but I guess I have to open 2NT anyway".

It's nice to see good players admitting that they still use 4321 count with adjustments. I've encountered many players who claim that this is just a crutch for beginners, and they just look at a hand and evaluate it holistically, but this has always seemed disingenuous to me. I think it's often an excuse for complaining about convention and alert regulations that refer to HCP: "But I don't use HCP when evaluating my hand, how can I be expected to use it in explanations?"

#11 User is offline   mikeh 

  • PipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 12,835
  • Joined: 2005-June-15
  • Gender:Male
  • Location:Canada
  • Interests:Bridge, golf, wine (red), cooking, reading eclectically but insatiably, travelling, making bad posts.

Posted 2019-February-05, 10:30

View Postbarmar, on 2019-February-05, 09:47, said:

Maybe I'm not the "expert" you're talking about, but it's not always unconscious. I often say to myself things like "Wow, this is a really nice 11 count" or "This may have 20 HCP, but it's still pretty lousy because of the 4333 shape and all the spots are 2s and 3s -- but I guess I have to open 2NT anyway".

It's nice to see good players admitting that they still use 4321 count with adjustments. I've encountered many players who claim that this is just a crutch for beginners, and they just look at a hand and evaluate it holistically, but this has always seemed disingenuous to me. I think it's often an excuse for complaining about convention and alert regulations that refer to HCP: "But I don't use HCP when evaluating my hand, how can I be expected to use it in explanations?"

I did not mean that one is unaware of how one sees a hand: quite the opposite. However, speaking for myself, I do not consciously add up my 10's and 9's and 'add a point' or use any sort of conscious arithmetic. At most I will occasionally, on borderline hands with shape and light on hcp, use the LTC as a sort of final indicator.

AKxxx Axxxx xx x is obviously a nice 11 count, and while less experienced players might use, say, the rule of 20, or might add points for length or shortness (hopefully not both), I just look at it as a clear opening bid. Qxxxx Qxxxx Ax K: same shape, same hcp, but an obvious pass even tho the LTC is sufficient to open. I don't arrive at any 'number' to represent the value of that hand. I'd certainly 'count my hcp', but that is reflexive, and serves as an introduction to the largely sub-conscious valuation process.
'one of the great markers of the advance of human kindness is the howls you will hear from the Men of God' Johann Hari
0

#12 User is offline   hrothgar 

  • PipPipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 15,372
  • Joined: 2003-February-13
  • Gender:Male
  • Location:Natick, MA
  • Interests:Travel
    Cooking
    Brewing
    Hiking

Posted 2019-February-05, 10:54

Few quick comments:

1. I'm not sure whether Gladwell's "10,000 hour" rule is sound of or, however, I do think that there is something to be said for the basic idea that practice makes perfect. If you want to get really good at playing bridge hands against strong competition, the best way is to play a lot of bridge hands against strong competition.

2. Historically, there have been a bunch of written resources that players can use to try to improve their play. In particular, books like

Bridge Odds for Practical Players
The Dictionary of Card Combinations
Watson's Play of the Hand
Love's book on Squeezes

can all be useful

3. I think that there is significant value in computerized double dummy analysis and hand simulations; however, I don't think that the results of the sims are particularly useful. Rather, what is helpful is jumping through all the hoops to get the simulations set up in the first place and then debugging your code.

4. I think that the state of research wrt to applying AI to bridge is pretty weak, in part because the areas where there are the biggest opportunities for advancement (design of bidding systems) seem to be off limits because of regulatory issues. With this said and done, I think that the bridge is going to get solved accidentally when someone gets bored and points a Deep Learning cluster that is busily solving Starcraft or Go or some other "important" game at bridge and the whole thing will get wrapped up in a weekend.
Alderaan delenda est
0

#13 User is offline   sfi 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,576
  • Joined: 2009-May-18
  • Location:Oz

Posted 2019-February-05, 13:00

View Postmikeh, on 2019-February-05, 07:34, said:

I had to laugh at your story about the 1700. I won’t say that experts never go for 1700, but I will say that I can’t imagine forgetting it quickly. For one thing, one should pay attention to how it happened. -1700 is the sort of result that one should have at most about once every 2,000 hands, at most, unless one is saving against a vulnerable grand. I remember doing that exactly once, lol. (A 1700 save, that is).

It's not that I don't learn from it later. This particular result was an unmitigated disaster that I started with a bad choice and then made it worse by not trusting partner. I haven't made either error again in the decade since.

But, putting it aside meant we won the match comfortably and went on to win the event. I can't imagine doing that while having a board like that running through my thoughts.
0

#14 User is offline   JanisW 

  • PipPipPipPip
  • Group: Full Members
  • Posts: 129
  • Joined: 2017-September-12
  • Gender:Male
  • Location:Germany

Posted 2019-February-05, 15:19

I'm certainly no expert in bridge, but I can give some insight into chess. My playing strength there is around 2200 so not an International master, but it's still a reasonably high number and I've talked to enough grand masters about this topic, to have an informed opinion. The reason I think, that it might be helpful in bridge is, that I found mikehs statement about remembering the play of a whole session quite interesting. In chess, it is actually quite the same. Even my level is sufficiently high that I can easily replay the games of a tournament after it has finished from memory without a scoresheet for a reasonable long time. Furthermore, there are games that just stick to memory, which I could recite from memory although 10 years have passed since playing them. This is a feat weaker players often lack. If you translate that into bridge, it becomes obvious that mikehs statement is true. I truly believe, that in the progress to become a stronger player, you have to be able to at least remember a few boards for a reasonable time.

Here I can only talk about chess again, but I guess it's true for bridge as well, that being able to remember the deals, helps in visualizing hands at the table. You can obviously play chess without a board and it is kind of what you do when you calculate longish variations. Every better chess player I know is able to play a whole game of chess without even looking at a board at a comparable high playing strength. Remembering where the pieces are is just second nature and you actually have no trouble "seeing" the board in your head. An that is kind of what happens in bridge to good players. A very good player will simply know, and here I mean without using much (or maybe even any) effort who played what card. If somebody showed out in a suit, automatically the suit split will appear in their head and with that possible breaks in the other suits pop up. Correct me if I'm wrong there mikeh. If these things become second nature, your brain is freed up to calculate several possible winning lines. If you struggle to remember the cards played and have to count to 13 to calculate how a suit split, you waste too much energy and have less brain power left to actually plan the play.
This is obviously a trainable feat, but to some extent, it is also innate and not everyone can reach the same level.

A second issue I want to talk about is pattern recognition. In chess, there is a lot of research done into that matter and there is a quote stating, that a grandmaster knows about as many patterns as a linguist knows words in a tongue. This obviously comes with practice, practice and practice again and is easily translatable into bridge. A good bridge player "recognizes" some layout of cards and instantly some possible solutions come to mind. I used quotation marks because most of this happens subconsciously. The more patterns (=layouts/solutions) you have available the deeper you get in your thinking. You can sort of think about it as having a head start in the thinking process. If you have to work out some possible solutions first you obviously cannot get as deep in your thinking like somebody who already has those available and starts his thinking at a deeper stage, Something like: "I recall that solution was working there because the A was offside, but that cannot be the case here. If the A was onside it was actually better to play like that. But if I play this suit that way, the problem is a possible 1-4 split with East which I can only pick up if, and so on and so forth.

To sum it up the two most important features in the thinking of a good bridge player (which should also be sound training advice)
Visualizing the hand (suit splits, played cards,...) because it is necessary for number 2 and saves energy at the table
Pattern recognition, because the solution process starts several steps ahead.

regards
JW
4

#15 User is offline   mikeh 

  • PipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 12,835
  • Joined: 2005-June-15
  • Gender:Male
  • Location:Canada
  • Interests:Bridge, golf, wine (red), cooking, reading eclectically but insatiably, travelling, making bad posts.

Posted 2019-February-05, 16:15

View PostJanisW, on 2019-February-05, 15:19, said:

I'm certainly no expert in bridge, but I can give some insight into chess. My playing strength there is around 2200 so not an International master, but it's still a reasonably high number and I've talked to enough grand masters about this topic, to have an informed opinion. The reason I think, that it might be helpful in bridge is, that I found mikehs statement about remembering the play of a whole session quite interesting. In chess, it is actually quite the same. Even my level is sufficiently high that I can easily replay the games of a tournament after it has finished from memory without a scoresheet for a reasonable long time. Furthermore, there are games that just stick to memory, which I could recite from memory although 10 years have passed since playing them. This is a feat weaker players often lack. If you translate that into bridge, it becomes obvious that mikehs statement is true. I truly believe, that in the progress to become a stronger player, you have to be able to at least remember a few boards for a reasonable time.

Here I can only talk about chess again, but I guess it's true for bridge as well, that being able to remember the deals, helps in visualizing hands at the table. You can obviously play chess without a board and it is kind of what you do when you calculate longish variations. Every better chess player I know is able to play a whole game of chess without even looking at a board at a comparable high playing strength. Remembering where the pieces are is just second nature and you actually have no trouble "seeing" the board in your head. An that is kind of what happens in bridge to good players. A very good player will simply know, and here I mean without using much (or maybe even any) effort who played what card. If somebody showed out in a suit, automatically the suit split will appear in their head and with that possible breaks in the other suits pop up. Correct me if I'm wrong there mikeh. If these things become second nature, your brain is freed up to calculate several possible winning lines. If you struggle to remember the cards played and have to count to 13 to calculate how a suit split, you waste too much energy and have less brain power left to actually plan the play.
This is obviously a trainable feat, but to some extent, it is also innate and not everyone can reach the same level.

A second issue I want to talk about is pattern recognition. In chess, there is a lot of research done into that matter and there is a quote stating, that a grandmaster knows about as many patterns as a linguist knows words in a tongue. This obviously comes with practice, practice and practice again and is easily translatable into bridge. A good bridge player "recognizes" some layout of cards and instantly some possible solutions come to mind. I used quotation marks because most of this happens subconsciously. The more patterns (=layouts/solutions) you have available the deeper you get in your thinking. You can sort of think about it as having a head start in the thinking process. If you have to work out some possible solutions first you obviously cannot get as deep in your thinking like somebody who already has those available and starts his thinking at a deeper stage, Something like: "I recall that solution was working there because the A was offside, but that cannot be the case here. If the A was onside it was actually better to play like that. But if I play this suit that way, the problem is a possible 1-4 split with East which I can only pick up if, and so on and so forth.

To sum it up the two most important features in the thinking of a good bridge player (which should also be sound training advice)
Visualizing the hand (suit splits, played cards,...) because it is necessary for number 2 and saves energy at the table
Pattern recognition, because the solution process starts several steps ahead.

regards
JW


Hi Janis

I agree with what you have written.

Consider dummy with KQ10x and you are sitting in front of dummy with J9xx, declarer marked with Axxx. The usual line followed by a competent declarer is to cash the King, then cross to the Ace, and hook your jack. What can you do? Well, if he starts with the King, and partner plays a meaningless low spot, play the 9.

A declarer who has not seen this before will realize that he can now pick off your partner's Jxxx, so long as declarer has the requisite spots...say A8xx opposite KQ10x...cash the KQ and lead through partner's J7 into declarer's A8.

I once defeated a 6D contract with this false-card, but obviously it has to be played in tempo. Hesitate at all, to work it out, and you give the show away. I was able to play in tempo because I was familiar with this and other deceptive plays from reading, long before I ever saw this layout at the table.

A similar situation arises when one holds an Ace in front of dummy's KJ(x)(x) holding and declarer leads small very early in the play. Whether one ducks or flies depends on whether one thinks that one has to take the trick, either because declarer has a stiff or because we need to do something active, or one thinks that we can afford to duck and hope for a misguess. The problem is that the slightest twitch will give the show away. Mike Lawrence once read my micro-twitch in this situation in a World Pairs event: my partner, a fine player, asked Lawrence why he guessed right, since my BIT was miniscule. Grant Baze, on whose teams my partner and, a couple of times, I had played in regional events, so was a friend, said it was because I had twitched. Lawrence, a superb player, had no problem picking up on that tiny break in tempo: again, recognition, coupled with being truly 'at the table'.

One has to train oneself to recognize the KJx problem as soon as dummy comes down. Most good declarers will take a little time before playing to trick one, and if they don't a good defender in third chair should always pause for a few seconds. One has to be 'at the table' and put that time to use, and to decide before declarer wins a trick what one is going to play if declarer immediately plays low towards dummy. Conversely, as declarer, if you know that you will need to make a guess in this suit, lead the suit as early as possible.

An expert will either already know he has to win or will duck on general principles but will do so in perfect tempo: which is impossible unless one is familiar with the situation. Obviously I failed in that against Lawrence: I don't think I've ever broken tempo in that situation since (which means that I have on occasion ducked when I should have risen, since my rule is 'when in doubt duck').

By the way, this sort of thing is another reason, amongst others, why playing against robots will never lead to becoming a good player (except against robots!). Robots are never 'at the table' any more than they are capable of losing focus. I don't play against robots but my understanding is that one can take as long as one wants before playing to a trick and the robots won't draw any inferences. Nor will they hesitate when put to a guess, not at any speed that a human can detect.

Adverting to Richard's reference to Gladwell's 10,000 hour rule, the point isn't simply how much time one spends: it is how one spends that time. Play bad bridge for 10,000 hours, with mediocre players, and one will become a very mediocre player. Read challenging books, and re-read them, and play in as tough a field as one can, with as good a partner as one can find, especially a partner who wants to get better, and one will build this pattern or situational recognition ability (or find out that one is never going to be an expert. The smartest person I ever met in university was a post-doc in nuclear engineering, who loved bridge but was hopeless at the game: we weren't that good, although our game featured several players who went on to become amongst the best in the country, but he was utterly hopeless).
'one of the great markers of the advance of human kindness is the howls you will hear from the Men of God' Johann Hari
0

#16 User is offline   ggwhiz 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 3,952
  • Joined: 2008-June-23
  • Gender:Male

Posted 2019-February-05, 17:44

The biggest difference is in card play.

Meticulous attention to detail such as declarer facing a lead of low to the Queen winning the Ace from Ace King and always hiding as many cards as possible for as long as possible from the defence.

Having the vision on defence to win trick one and switch to a club from Jxx into dummies AQx catching partner with the King and enough time to capture that trick later but not too late.

Inspired plays and falsecards are written up far more often than bids imo.
When a deaf person goes to court is it still called a hearing?
What is baby oil made of?
0

#17 User is offline   HardVector 

  • PipPipPipPip
  • Group: Full Members
  • Posts: 479
  • Joined: 2018-May-28

Posted 2019-February-14, 17:37

Read Rodwell's book. That will get you the best glimpse into the mind of an expert. Advanced players will incorporate lesser versions of the same.
0

#18 User is offline   michel444 

  • PipPipPipPip
  • Group: Full Members
  • Posts: 241
  • Joined: 2022-September-10

Posted 2022-September-15, 07:23

View Postbarmar, on 2019-February-05, 09:47, said:

Maybe I'm not the "expert" you're talking about, but it's not always unconscious. I often say to myself things like "Wow, this is a really nice 11 count" or "This may have 20 HCP, but it's still pretty lousy because of the 4333 shape and all the spots are 2s and 3s -- but I guess I have to open 2NT anyway".

It's nice to see good players admitting that they still use 4321 count with adjustments. I've encountered many players who claim that this is just a crutch for beginners, and they just look at a hand and evaluate it holistically, but this has always seemed disingenuous to me. I think it's often an excuse for complaining about convention and alert regulations that refer to HCP: "But I don't use HCP when evaluating my hand, how can I be expected to use it in explanations?"

the 4321 is a good starter for hand evaluation
consider this hand QJX QJX QJX QJXX YOU GET AN ARE THE DEALER
do I open ?
I will open a weak 1 NT 11-14 . 12-14 any day
but with a Strong 1 NT opening
I may open 1 Club but could decide to pass
0

#19 User is offline   nullve 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,228
  • Joined: 2014-April-08
  • Gender:Male
  • Location:Norway
  • Interests:partscores

Posted 2022-September-15, 09:34

Assume expert partner and standard 2/1.


0

#20 User is offline   mikeh 

  • PipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 12,835
  • Joined: 2005-June-15
  • Gender:Male
  • Location:Canada
  • Interests:Bridge, golf, wine (red), cooking, reading eclectically but insatiably, travelling, making bad posts.

Posted 2022-September-15, 12:09

View Postnullve, on 2022-September-15, 09:34, said:

Assume expert partner and standard 2/1.



Was this supposed to be a new topic?

Not sure how it fits with this thread

Seems pretty clear to pass. Obviously i intend to back into the auction later. Equally obviously, opening (1C) could well work out better than passing. But, playing whatever is meant by ‘standard 2/1’ (there being, to my knowledge as someone who has played methods based on 2/1 for more than 40 years with many partners, no such thing), ‘standard’ implies that one doesn’t open on aceless 8 counts, unless one is preempting. Partners understandably tend to think that their decent misfitting 13 or 14 count makes game or that their decent 8 or 9 count, with red tricks, makes for a lucrative penalty double at some point. I like my partners to be able to play me for the hand I show in the auction.

Let’s say I open 1C and later bid spades twice, showing 5=6. Maybe I have AJxxx x x AJxxxx….a hand with which I’d happily be bidding. That’s a far different hand, minimum in hcp as it is, than this hand.
'one of the great markers of the advance of human kindness is the howls you will hear from the Men of God' Johann Hari
2

  • 2 Pages +
  • 1
  • 2
  • You cannot start a new topic
  • You cannot reply to this topic

1 User(s) are reading this topic
0 members, 1 guests, 0 anonymous users