"A patient goes to see a doctor. The doctor performs a test with 99 percent reliability--that is, 99 percent of people who are sick test positive and 99 percent of the healthy people test negative. The doctor knows that only 1 percent of the people in the country are sick. Now the question is: if the patient tests positive, what are the chances the patient is sick?"
The intuitive answer is 99 percent, but the correct answer is 50 percent...."
The base rate fallacy requires two things:
a) a very low incidence of the issue in the broader population:
AND
b) some sort of binary test which generates (even a small number of) false positives.
I can't think of any situations as a poker player where this would be relevant in the game, but it is certainly an important issue to be aware of when designing anti-fraud security systems around online poker sites. For example, given the relatively tiny numbers of cheaters in the poker world, false positives are a huge problem to any security system.
The base rate fallacy requires two things:
a) a very low incidence of the issue in the broader population:
AND
b) some sort of binary test which generates (even a small number of) false positives.
I can't think of any situations as a poker player where this would be relevant in the game, but it is certainly an important issue to be aware of when designing anti-fraud security systems around online poker sites. For example, given the relatively tiny numbers of cheaters in the poker world, false positives are a huge problem to any security system.
I think not being aware Bayes theorem leads to the perpetuation of automated cognitive errors. In particular when you think a line can look so strange that it has to be a bluff. In reality, when we compare the chances of a player having air vs simply having a strong hand, even though both are rare, the strong hand comes out as favourite (Hope I'm explaining that correctly). Obviously visa versa can be true given a certain line.
Basically, we have a bias to be suspicious of strange/rare things. Without Bayes we lean more towards categorising these rare/strange things as a threat or in my example perceiving a rare line as a bluff.
