In the March 21 issue, Entertainment Weekly handicaps this year’s Oscar races, offering odds in the major races.
Caveats before I ask my question:
a) I am not a math whiz
b) I realize EW is doing this for fun, and not making an effort to calculate actual probabilities
Now:
I seem to remember that, when laying odds on an event, the total of the probabilities has to equal 100%. Therefore, in a given Oscar race, if everyone had an equal chance of winning, the odds on each person or film would be 5-1. (Because a 20% chance of winning = 1/5 = 5-1 odds. This could be a major misunderstanding on my part.)
If that’s the case, then EW’s odds really don’t make sense, statistically. (They may make sense from a bookmaker’s standpoint, what with having to cover the vig and all, but I really don’t know. Have I mentioned my confusion about all of this?)
For example, they handicap the Best Director race as follows:
Pedro Almodovar: 5-1, or 20%
Stephen Daldry: 15-1, or 6.7%
Rob Marshall: 3-2, or 67%
Roman Polanski: 10-1, or 10%
Martin Scorsese: 2-1, or 50%
Given those numbers, there’s a 153.7% chance that someone will win this race, when a 100% chance would be the right answer.
Am I correct in thinking EW’s bookies are mathematically challenged? If so, do I actually understand where they’ve gone wrong, or am I misremembering what probability theory I did know?