|
BGonline.org Forums
OT: Estimating real winning probablilities from decimal odds
Posted By: leobueno
Date: Friday, 7 September 2012, at 9:16 p.m.
I am trying to figure out whether my calculation of actual winning chances from published odds makes sense.
Below are the decimal odds of 4 events, A, B, C, D.
The second colum shows the calculated implied odds from the decimal odds (i.e., 1/odds), which if you add, exceed 100% (hey, the book has to have an edge).
In the third colum, I try to normalize the implied odds by dividing them by their sum (109.1%), so that the total winning probabilities will be 100%.
The question is: are the derived "Actual" winning probabilities the real ones? In other words, is this a good way to remove the book's vigorish and approximate the "true" odds?
Outcome|Odds|Implied|Actual
A|1.40|71.4%|65.4%
B|2.84|35.2%|32.3%
C|71.00|1.4%|1.3%
D|91.00|1.1%|1.0%
Sum|----|109.1%|100.0%
By the way, the events are the winning party in the 2012 USA elections (Democratic, Republican, Any Other, Independent).
The odds are the median odds published by OddsChecker.com.
|
BGonline.org Forums is maintained by Stick with WebBBS 5.12.