Quote:
|
Originally Posted by Razz
Just remember that the lower the total (or expected score) the more valuable the points.
|
There's no question that this is indeed a factor. Its theoretical justification is that due to the greater sparsity of likely outcomes for "low" totals as opposed to "high", there's also a greater liikelihood of realizing any given MOV under a "low" total scneario. And indeed, this certainly has a large imapct in baseball. In basketball, however, it gets a little more complicated.
If you look at NCAA Basketball spreads over the past 8 or 9 years and partition the data set into a "high" total subset and a "low" total subset what you find is that while for most spreads push probabilities when you move from the "high" total subset to the "low" do increase (as predicted by theory), there are a few spread clusters (most notable of which may be the 3-4-5 spread cluster) where figure probabilities actually
decrease. Assuming that this isn't a data anomaly (and the texture of the probability differences surrounding the clusters suggest that indeed it isn't), this suggests that there exists some other factor, which for certain spread groups and as a function of expected total score, exerts an impact on push probabilities opposite to that of sparsity. (And of course it stands to reason that for other spread groups the factor's effect is in the same direction as that of sparsity.)
The macro explanation for this theoretical factor is certainly up for debate. I do have my own hypothesis about it but I'd be interested in seeing if anyone else has any thoughts before I poison the well with my own bombastic flavor of expression.
Buy or Sale points
Linear Mode
