Questions and Answers


Question 23

Hi Anders & Mike,

I am an engineer, with a statistic background. I have always been a law-follower, but with the feeling, that something was wrong. Thank you for the statistic work, you have shared with us in your book. The law is about right, the average trick count equals the trump count, but your work shows, that the variance is too big. I am looking forward to the coming bridge season where I am going to analyse a lot of deals with SST/WP.
   Have you read the book 'Saknade point' (Missing Points) by Bertil Johnson, from 'Jannerstens BridgeBibliotek'? I think your method, with SST/WP and his with the missing points/cover poins are much the same. Agree? Your method are far more simple; easy to learn and use.
   You seem to avoid an important point in your analysis. 3-points does not mean a trick, it is the average change to take a trick. Points are used to give an estimate of tricks, it's an evalution method. Instead of counting tricks, we use the more refined method of counting 1/3 'chance to get a trick' (and we hope that average/variance in long term are right)
   It feels wrong, when you count points after you know the position of other cards than your own. For one you can only count 3 point at a time. Of course you can see where the tricks come from, given a good guideline for points to trick evaluation, but they are tricks, not points. If you use the bidding to place cards, the definition of 1/3 point looses its statistic meaning. I don't know what to use instead of. As a good friend of mine says: 'how many points to count for that ten of trump' when a slam depended of him having it.
   I hope you will collect a lot of input and gather them in a new book, where your methods will be refined. I will hope for a tighter definition and a systematic scheme/review for the method.
   If you want, say 20 players to evaluate 100 randoms hands with the SST/WP method, to get a statistic material, I will be glad to help.

Steen Bøhm, Denmark


Answer

Yes, points and tricks are not the same. The idea of '3 points equal 1 trick' is derived from the fact that if you have a suit of, say, A-K-x opposite Q-x-x, you take three tricks with your honors: one with 4 HCP, one with 3 HCP and one with 2 HCP. The average trick taken is with 3 WP. So, we could instead say that 3 WP is what we use on average in order to take a trick with an honor.
   Yes, we have read Bertil Johnson's book, which is based on a method developed by Swedish internationalist Alvar Stenberg roughly 50 years ago. Stenberg introduced the method of Missing Points (mp) for strong unbalanced hands (Acol two-bids or even stronger), but it is possible to use it in other areas as well. Since that method also considers the two prime factors: distribution and working honors, I wouldn't be surprised if it comes to similar results as WP + SST.


More questions and answers:

[ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 ]


Copyright © 2016, Mike Lawrence & Anders Wirgren