Virginia Oddsmaker's Math

By: chippenham
Published On: 8/26/2007 1:40:34 PM

It seems like everyone has been checking the Virginia Oddsmaker blog lately.  I've got two problems with his site though.

First, he's obviously the hometown bookie of a Republican party nominating convention, these odds are so weighted to Republicans.  Even Faisal Gill is an even shot with Paul Nichols, despite all of the bad news coming out about that race and the poll which leaked on NLS a few weeks ago which shows Nichols already up nearly 20 points.

Second, this guy doesn't know anything about math.  In practically every two man race the odds add up to less than 100% that one of the two people will win.  Either he is counting on an incredible showing for write in candidates or he just doesn't understand how this stuff works.

Let's look at the aforementioned 51st House District, where both candidates are given even odds of 3-1 that they will win.  Hmm, let's see in real life that means if I bet $1 on each horse I'd be guaranteed $3 in winning no matter who won.  So, a $2 bet automatically wins me $1.  Sorry, that's just not how it works.

A simple explanation for the Virginia Oddsmaker and all aspiring handicappers.  3-1 odds means that an event will happen 1 (the 1) times and not happen 3 (the 3) times.  So, in 4 occurences, it will happen 1 time, or 1/4 of the time = 25% of the time.  So according to the Oddsmaker, Faisal Gill will win 25% of the time and so will Paul Nichols.

Well, who would win the other 50% of the time?  I agree that Gill probably has odds of 25% or less, but that means that the odds of Gill NOT winning are 75% or higher.  If Gill doesn't win, Nichols does.  These things are kind of zero sum.  So, if Gill's odds are 3-1, we need to express Nichols as winning 3 times for every 1 loss, or 1-3.

Oh my god!  It's as simple as that.  In the majority of Virginia races where there are only two candidates, the odds for one candidate are the inverse of the odds for the other.  Who could get something as simple as this wrong?  Well I guess the same person who thinks Tricia Stall, Nick Rerras, Richard Stuart, Jeannemarrie Devolites-Davis and Ken Cuccinelli (among others) are all huge favorites to win their districts...not to mention that in most races his odds would predict that there is at least a 50% chance that neither of the two declared candidates will win.


Comments