arbing EV situation help

[beetman]

The difference between -110 and +112 is a litle bit more than 5%, not 6%. (-110 is 52.38%, +112 is 47.16%, for a difference of 5.22%.) One thing I forgot to mention is that the 3 has historically been worth a bit more with away dogs than home dogs, because home dog score distributions have been higher variance.

You can bet much more than 3% of your bankroll on either side because for every 3% you bet on either side, you're actually only risking 0.15%. If you bet 525 to win 500 on away team +3 -105, and 525 to win 500 on home team -2.5 -105, you're only risking 25 to win 500 that home team wins by exactly 3, which will happen about 10% of the time. Essentially you're getting +2000 on a +1000 proposition.

If you're betting 1/4th Kelly, which is pretty conservative, you can bet % of bankroll on both sides, which means you're only risking 1.12% Assuming a $10,000 bankroll, the optimal Kelly size would be risking $2280 to win $2068 on both sides, in which case you're only risking $112. And there's nothing wrong with betting 1/2 Kelly or even full Kelly there. Opponents of the Kelly criterion as applied to sports betting argue that you never really know if you have an edge, which might be true when dealing with issues of handicapping, but "how often will the home team win by 3" is simple mathematics, not as nebulous as most handicapping issues.

I don't have a particularly large bankroll compared to most serious sports bettors but whenever I find a middle that profitable, I invariably run into problems getting enough money down before I run into problems with overbetting my bankroll.

[Thremp]

An example like this came up the other day for me with NFL halftimes.

-105 u17.5
+103 o17

Needless to say I think I'm comfortable putting my whole bankroll on this every day of the week.