Quote:
Originally Posted by hedgie43
That's not the value of OT. Sports betting isn't about mean outcomes.
Apples and oranges. The discussion was around this:
Quote:
But generally speaking in Basketball, more points are scored in the second half.
I was referring to expected points, not distribution. The formula I noted above is correct in determining the value of OT in estimating the expected points in 2H+OT. That is,
expected value.
To address what you are introducing, we need to discuss the affect on scoring distributions.
A first order approximation of course would be to derive the value of OT noted above. Then, determine if the frequency of OT is independent IRT 2H scoring. If it is, then including OT should basically shift the pdf right about a point.
If the frequency of OT is NOT independent on 2H scoring, for example let's say OT was actually "extra 5 minutes only if 2H scoring was <70 points", then the distribution above, say 90 or higher points, would be less affected than <90 points. The left tail of the pdf would shift right to account for OT, while the remainder would remain about the same.
IRT to recent NBA scoring, you got me there. For years 2H+OT expected scoring was > 1H. Although I see it is not much of a difference. I'm guessing that might be caused by more foul calls in 1H, more uniform distribution on winning margins, and changes in clock management late in games. I'm sure there are factors I'm overlooking as well.