Determining Distribution

[phikappaph]

While I have many weaknesses in my game, what frustrates me the most is my inability to "Picture" distribution. If one opponent shows out, I use that knowledge to manage that suit, but I don't take advantage of the concept of using that data to get a complete picture of distribution as a whole. As such, end plays and squeezes are far more challenging.

 

Can anyone suggest a web site or technique for practicing on how better to grasp the unseen hands as the play develops? I've learned much here on bidding, systems, conventions and leads. But I've not come across practicing the dynamics or tricks for determining distribution. I did just learn from the interesting article here on the Law of Total tricks how the total number of trumps can help determine distribution in the oppositions hands.

 

Any help would be appreciated.

 

Please, no suggestions about taking a memory course. My wife reminds me all the time.

[Hanoi5]

Try to practice on every hand you play. Read about these kind of hands. Make exercises/problems with this theme.

 

There's not an easy recipe/formula. You have to practice hard.

[SteveMoe]

Counting shape can be important. Here's a link to the BBO counting game developed by Fred Gitelman - It trains us to immediately recognize the 4 number pattern totaling 13 for suits in a hand or a suit across 4 hands. http://bridgewinners.com/pages/counting-game/

 

Eddie Kantar's books on defense are very good at showing how to count.

 

When it comes to counting shapes of hands, first ask what the bidding shows. You can know what shapes declarer is likely to have from the bidding in many cases. Best is to note bidding and likely shapes as the bidding evolves. This helps whether you are on lead or declaring. By the time it comes to make a play you will have started building your image of the missing hand shapes.

 

When declaring, the opening lead offers hints at the suit split around the table (4th from longest and strongest).

 

During the play, opponents count signals might help (though many do not signal all the time).

 

Show - outs (when a defender or declarer discards) paint a clear picture of the distribution. The important thing is to do the mental work to note what the show-out says about the suit around the table, and the other suits in that hand.

 

What's important is to take a few seconds and focus on the information as you receive it. Each bit of additional information (bidding, passing, lead, play) should help you refine your estimates.

 

Importantly. sometimes it's right to count shape, sometimes HCP, sometimes tricks. You need the develop the discipline and the presence of mind to know which matter more and focus on that information.

 

Suppose you declare 3N and get the ♦5 lead. With ♦98 opposite ♦A32, you know immediately that RHO has 3 cards higher and the ♦ are likely splitting 5-3 or 4-4. If RHO has the ♦4 you know they are 4-4. If LHO has 5♦s, then 5332 or 5431 are their more likely shapes. You can then begin forming inferences based on the rest of dummy and your hand.

 

Here's something you might enjoy: http://www.cincybridge.com/Lessons/20100506_The_1st_90_Seconds.pdf

[nigel_k]

Paradoxically, I think people can really improve in this area by spending more time on analysis at the end of play, with all four hands exposed. You get a lot better at identifying the layouts where your play will matter and mentally discarding the ones where it doesn't, which greatly simplifies the task of deciding which layout to play for. Often just trying to write down the alternative layout you were playing for will be enough to show that it was inconsistent with the bidding and previous play.

[phikappaph]

Paradoxically, I think people can really improve in this area by spending more time on analysis at the end of play, with all four hands exposed. You get a lot better at identifying the layouts where your play will matter and mentally discarding the ones where it doesn't, which greatly simplifies the task of deciding which layout to play for. Often just trying to write down the alternative layout you were playing for will be enough to show that it was inconsistent with the bidding and previous play.

 

 

I appreciate the comments and hopefully they will assist others. I had not considered prioritizing tricks, HCP's, etc., mentioned above, rather I was intent on learning a rote skill. I have found that kibitzing games whether I have all four hands shown or just one tends to hurt my efforts. The pace of the expert is just too fast. I learn a lot about bidding but can only marvel at the understanding declarers often have on setting up end plays and effectively using squeezes to offer a choice of losing options to defenders.

[ggwhiz]

It takes work and a shortcut to me when I started was our postmortems.

 

We learned to "talk" a bridge hand as in I had a 4-3-5-1 12 count in the proper order, SHDC and remember the relevant high cards from hands played hours ago and before the advent of hand records.

 

The notion of the 4 numbers adding up to 13 crept into identifying important patterns as declarer. Having hand records may actually hold you back from developing this skill.

 

One of our local players still writes every hand down with x's for spot cards in the margins of her score card and is now a Grand LM.

[gwnn]

Pretty stupid advice I know but it helped me a great deal to learn a relay system! Then you really learn the most common distributions by heart and you also get to think a lot about how potential contract would play out during the auction.

 

edit: This reminds me a bit of what a language teacher told me. She was talking to a French guy and made a joke about their strange number system. He didn't understand. So she said 'you say four twenties and fifteen instead of just saying ninety-five.' He said 'no we don't!' and he had to think for a while and then he realised that the language teacher was right. For him 'quatrevingtquienze' was just one word and he never broke it up into parts. So, too, at some point must we also think of 4351 as a hand instead of some vague indications of spades and diamonds and even more than just 4+5+3+1, just as 4531. Relay systems definitely help with that.

[phikappaph]

Pretty stupid advice I know but it helped me a great deal to learn a relay system! Then you really learn the most common distributions by heart and you also get to think a lot about how potential contract would play out during the auction.

 

edit: This reminds me a bit of what a language teacher told me. She was talking to a French guy and made a joke about their strange number system. He didn't understand. So she said 'you say four twenties and fifteen instead of just saying ninety-five.' He said 'no we don't!' and he had to think for a while and then he realised that the language teacher was right. For him 'quatrevingtquienze' was just one word and he never broke it up into parts. So, too, at some point must we also think of 4351 as a hand instead of some vague indications of spades and diamonds and even more than just 4+5+3+1, just as 4531. Relay systems definitely help with that.

 

 

Have been trying very hard to think about the totality of hands, i.e., the "4531" hand, etc. Very new to me. I did start doing something that has proved to be an enormous help, perhaps something every advanced player but me does as a set routine. After the opening lead, I count the outstanding unseen cards by suit and now keep track. So, when dummy comes down, I get a suit count that initially adds up to 26 such as "5939". I then reduce the number by two for each trick so if diamonds are led and followed, 5939 becomes 5919. A round of clubs becomes 5917, etc. I use the 4 hand robot option with BBO and do not even care about the hands, just keeping track of the count slow enough to be accurate but fast enough to make play bearable when actually playing. I am getting to the point that the bidding can tell me minimum card holdings and one player showing out is starting to give me better distribution visualization. The link showing the pause, plan and execute gave me the thought. It is making a difference already. My goal is read the board and keep up with an expert that I will kibitz, something that I am not close to yet. However, my game has really improved already. Once again, thanks

[ggwhiz]

What you are doing is very workable but for me I'm dealing with 4 sets of numbers ie. I have 9 trumps and will soon know if they split 1-3 or 3-1 etc.

 

The 4 sets of numbers may (or may not) be a touch faster if you have to pause and extract from your one large set as you most often only need to decide how to tackle one key suit later in the play. Just a guess cause this is the only method I've ever tried and whatever works for you is good.

[phikappaph]

What you are doing is very workable but for me I'm dealing with 4 sets of numbers ie. I have 9 trumps and will soon know if they split 1-3 or 3-1 etc.

 

The 4 sets of numbers may (or may not) be a touch faster if you have to pause and extract from your one large set as you most often only need to decide how to tackle one key suit later in the play. Just a guess cause this is the only method I've ever tried and whatever works for you is good.

 

Good point. Initially, it's easy to lose track of the actual cards played while keeping count. But it is getting easier. I also find that my count is often wrong at the end of the hand, that is I should know the suits of the last 2-3 tricks. But, As I keep going, I am getting better at developing winners. I also find that knowing the total count makes it a lot easier to decide which suit to develop as play progresses. I now able to effectively track discards, something that is making a big difference.