[HeartA]

Jlall, on Jan 7 2005, 06:34 PM, said:

in my opinion the fun and beauty of these combinations are in the analysis. Thats one of the great things about bridge. No matter how hands you play, you see so many interesting things. No matter what level you are at you learn something new. Suit combinations would not be fun to me if i had to punch them into some program to work them out. Even if I saw some combination was better than another, I wouldnt truly understand why unless i analysed it. I would miss out on the beauty of the game. I would also become a weaker player at the table.

.......
My 2 cents.

Agree with Justin.

[mpefritz]

What most analyses of suit combinations lack are:

1) The real likelihood of the combination. As I have mentioned several times and Stephen has spelled out, each split has a slightly different likelihood. So just listing the possibilites is a great first step, but incomplete. Using a program to calculate these exact numbers helps with the feel for what is REALLY correct as opposed to some guesstimate. Since I can easily create the odds by hand (and calculator) or on a spreadsheet myself, I have no qualms about using a program because I know the underlying principles. I also agree with Fred (and Stephen) that the odds calculators are probably a detriment to those who are still learning how to figure out suit combinations. (well..I think everyone is still learning..)

2) The odds that a player will play a certain card from a certain combination. for example, versus a newbie AJT98 7654

Lead 7,3,J,K

Do you finesse the 2nd round by RC? What if the Q showed up?

The point is that most newbies play smallest available, so the K rules out the Q and the Q gives no information.

This was relevent on Fred's KQT987 32 hand. The strategy of what to do when EAST wins the A on the first round depends on EASTs likelihood of winning the A from Ax. In fact, if EAST wins the A from Ax more than 1/3 of the time, then NOT playing EAST for AJ when he wins the A on the first round is the winning play. (Source data and assumptions available upon request)

SUITPLAY cannot calculate these numbers.

Jlalls QT4 A98765 is another example. If we can estimate that WEST will duck from Kx all of the time, then his line is best. If not, low to T and then hook is better (as he stated in his solution).

So SUITPLAY can help with the calculations, but you still need to apply table feel and player odds to those numbers.

fritz

[Flame]

Some ppl has great memory and for them it might be best to do it your way memorizing combo's by books and comps. For me i dont even sure I remember the few combos we have seen here this week, if i have them at the table ill have to think again but since i remember something and i practiced i might get it right. Memorizing doesnt work for me.

[hrothgar]

One quick comment here:

Memory is a very tricky thing...

The brain's ability to commit information into long term memory depends on its ability to associate information with different "keys". In some cases, those keys are related to specific physical cues (for example a smell or a sound). In other cases, the key could be some type of logical framework that you use to store the information.

"Physically" solving problems forces the brain to go through many more processing steps than reading the information from a computer program. According, I'm willing to bet that this "enforced" busywork will help most players learn the fundamentals involved.

I readily admit that there are players who don't need this type of reinforcement. However, I am guessing that that these same players are able to recognize these problems as one specific example within a larger class.

Richard

Whose skill at these sorts of problems is truly sad...

[fred]

I agree with everything that mpfritz says, but (as I have said before) I think that when a new problem is presented here there should be a presumption of "best defense". These problems are complicated enough without have to think about the possibilities of misdefense. Of course if some people want to include information about "less than perfect defense" considerations, that's great, but unless we are all trying the same problems to begin with, there is going to be a lot of confusion.

I personally like solutions that do not involve any real math (and the problems that are solvable this way). Both these problems and their solutions are more elegant in my view. If you can clearly illustrate a solution simply by listing caes why not do it?

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com