Spingold Survey

When Boye wrote that he had looked at the hands at the other table from his quarter-final match and determined that something was wrong, I decided to look closely at them myself and see if I noticed a pattern. One or two hands by themselves prove nothing. Everybody takes non-mainstream actions, and sometimes these actions succeed. But if there is a pattern, that is something else.

I did find what I believed to be such a pattern. I have been in private contact with Boye, and he has told me that what I noticed was what he had noticed also.

How does one demonstrate that such a pattern really exists, as opposed to being a chance occurrence? One must first form a hypothesis about exactly what information is being signaled. Then, one must examine every deal, not just the deals which looks suspicious.

To illustrate, consider my analysis of the German doctors last year. I formed the hypothesis that they were signaling for opening leads. I then looked at every hand they were on opening lead.

Most of the hands are not relevant. The partner of the opening leader might not have had anything clear to signal for. The opening leader might have had a likely lead of the signaled suit of the suit anyway. In that case, when the suit is led it is not an indication of guilt. It is simply good bridge. Or the opening leader might have a likely lead of some other suit and would override the signal. In that case, if the opening leader overrode the signal that would not be an indication of innocence. It would simply be good bridge.

For a hand to be relevant, the following conditions would have to apply. The partner of the opening leader would signal for suit A. The normal lead would be suit B. However, with the help of the signal, suit A would be the likely lead.

If the above conditions exist, then if suit A is led that is an indication of guilt. If suit B (or some other suit) is led, that is an indication of innocence. By examining all such relevant hands, one can see how often the "guilty" lead is made and one can mathematically determine how likely this result is to occur by chance if the pair is honest. From this information, one can then make a judgment as to whether or not signals were given.

In order to determine what a likely action is, one person's judgment is not good enough. That person may have a bias. However, the judgment of a lot of players will almost certainly be accurate. I posed the problems to Bridgewinners in a survey. From the responses, I was able in an objective manner to determine which hands were relevant.

I am doing the same thing with Fisher-Schwartz. I have formed a hypothesis, and constructed a survey of all the hands they played in the Spingold which were on vu-graph. The results of this survey will determine which hands are relevant. Once this is done, we can see the actions actually taken on the relevant hands and form a conclusion.

In this case, I am not doing a public survey for several reasons. It is important that those polled are top experts, since expert opinion is needed to determine what the likely actions really are. Also, I do not want to risk the poll results being tainted by factors such as a voter responding more than once. What I did with the doctors was simply an exercise in confirming what had already been established. With Fisher-Schwartz as far as I know there is no established evidence about their signaling, so only the deals can speak.

It should be noted that the experts taking the survey will not be making any decisions on guilt or innocence. In fact, they will have no idea what actions Fisher-Schwartz took on the hands. They will simply be answering bridge questions as to what they would do. The only purpose of the survey is to determine which hands are really the relevant hands.

At this point I do not want to reveal what my hypothesis is. I am not trying to hide things. It is simply that I do not want to risk the survey being tainted. Many of the experts who will be polled are Bridgewinners readers, and it is best if they do not see this in advance. I promise you that once the survey is completed I will publish the full results, which will consist of every hand, the results of the survey which determines which hands are relevant, and the actions Fisher-Schwartz took on the relevant hands. This will be the information needed for anybody to form their own conclusions.

I have a request of Bridgewinners readers. Some of you will examine the deals, and may have speculations about what the pattern is which we have noticed. I ask you to NOT publish your speculations, since doing so may taint the results of the survey. This is very important. We want to do things in the right way.