View Single Post
  #57  
Old 09-11-2007, 05:19 PM
Limpfold Limpfold is offline
Senior Member
 
Join Date: May 2004
Posts: 267
Default Re: Daliman\'s NFL Wong teasers, Season Deux: Week 1

I used the charts available from smartcapper linked in the FAQ to estimate the probability of Bengals winning with 3-9 pts then calculated Bengals win probability from Pinny's moneyline.
I'm sure more mathknowledgable people will find flaws in this method but I suspect the number you get aren't that far from a more exact analysis.
Then I compared EV using the Cin -3 +108, Cin -7.5 +215 and let it ride options and found the Cin -3 +108 to be the best (~7% better than letting it ride)
Basically I was just out to minimize my loss since betting Wong's roundrobin is huge variance.
Losing 1 leg is very acceptable while losing 2 legs or more cause a huge BR dent.

I spent some time today calculating possible outcomes when roundrobining Wong teasers so for those of you who are curious here's the result doing full roundrobins of 3,4 and 5 teamers.

I used 75% as the chance of a 6pt teased wongable game to win.

3 teamer: 0-1 games won = 15.62%
2 games won = 42.19%
3 games won = 42.19%


4 teamer 0-1 games won = 5.08%
2 games won = 21.09%
3 games won = 42.19%
4 games won = 31.64%

5 teamer 0-1 games won = 1.56%
2 games won = 8.79%
3 games won = 26.37%
4 games won = 39.55%
5 games won = 23.73%

Now what I'm really interested in is how to decide/adjust the unit size to keep the chance of going busto being to high.
I also question using 75% as winning probability overall, I think it should be more accurate to divide it into home dogs, home favs, away dogs and away favs.
Using 75% the bookies would be, um, poorer.
Reply With Quote