BGonline.org Forums

Flaw in PR / Is this wrong? - Xavier wants to recall his previous message

Posted By: eXtreme Gammon
Date: Thursday, 19 December 2013, at 10:04 p.m.

In Response To: Flaw in PR / Is this wrong? (Bob Koca)

Bob "No. They should be counted again. Note that the PR hit for the first error is based on playing correctly from then on. If it is a repeating position you lost additional equity with the second error."

Ah!!! bob, you are absolutely right. I take back what I said before, all errors need to be counted even if the position repeats...

Here is a simple case to clarify what Bob explained to me (and make sure I got it right):
Imagine that the probability of repeating the position is 50%, that End = 0.4 (no double) and Ed=0.5 (double/take)

End can be broken down in End = 0.5*Ed (dance and then D/T) + 0.5*E1 (no dance).

Where E1 is the average equity if you do not double and the position does not repeat. That makes E1=0.3

If you make the commitment to not double, your expected error should be 0.2 (Ed-E1).

XG charges you only 0.1 the first time and any subsequent time it happens.
In average you will be charged 0.1*(1+0.5+0.25+0.125...) which converge to 0.2

So Rod, you should count these 2 errors and be glad you did not dance longer :)

Post Response

Subject:
Message: