Bridge Software to Analize Situations PART 1
On 9 May, 2013 At 13:50
Responses : Comments are off for this post
Source: 4th EUROPEAN OPEN BRIDGE CHAMPIONSHIPS Sanremo, Italia
When I learned bridge some (ahem) 40 years ago, it was widely taught that you needed 26 points to bid and make 3NT and 4/4 and 29 points for 5/5. But over the years a few things have happened to lower these benchmarks. The most important is that the quality of declarer play has improved dramatically, so why therefore hasn’t defence equally improved?
I would love to be able to answer these questions and I certainly invite written submissions on the subject. However, in the meantime and to test the theory, I used a piece of software called Bridge Browser, which was written by Stephen Pickett of Canada. What this allows one to do is to call up all of the results of millions upon millions of hands played on OK-bridge when the software was first developed and more recently on BBO and to statistically analyse them. Further you can analyse by excluding players who do not have a particular rating. (http://www.microtopia.net/bridge).
So I called up ten thousand hands played over a period of time in 3NT at IMP scoring (this took the computer some twenty hours) including doubled contracts, regardless of vulnerability, where the declaring side had exactly 24 HCPs and found that the average number of tricks made when holding a combined 24 HCPs was 8.65.
I can report that the standard deviation of the number of tricks was a mere 0.01 indicating that the variances from this 8.65 tricks was exceptionally low. Some may say that bidding 3NT with a combined 24 count is therefore questionable. HOWEVER the average gain by doing this at IMPs was 1 IMP per board, making it a very solid action indeed and if you don’t bid it then you will likely be a long-term loser.
This might seem like a small difference from the 26 points which we were taught. However, your side is now holding 60% of the points rather than 65% or 8% less. As a further check I also ran three thousand hands with a combined 23 HCPs to see if there was a significant difference and there was. The average number of tricks was 8.25, but interestingly,
even doing this gained 0.44 IMPs per board on average.
I don’t, however, recommend this as a long term strategy unless you are an excellent declarer player. So just in case you haven’t been told today — bid’em up!!!!
Esta entrada también está disponible en: Spanish