[BobJoeJim]

[ QUOTE ]
[ QUOTE ]
arbitrage is a waste of time. Too much effort for too little money.

[/ QUOTE ]

Not really. Arbitrage (scalping) is the way that many successful sports bettors make a LOT of money. You have to have a decent-sized bankroll to do it, but it is clearly not a waste of time.

[/ QUOTE ]
I've been thinking about this a bit, and doing some math on it. The data I'm missing is this: How often can you find scalps if you take the time to look, how big do they tend to be, and how many books do you need to have in order to make those frequencies the case?

My reasoning is essentially as follows: If your scalp is on a spread (rather than a moneyline) then you have about a fifty percent chance of winning, and you have to risk $2.10 to win, so an 8 cent scalp has an EV of about 2 cents per dollar in your bankroll, if you put everything on it in the right ratio. An average 2% return, if you could find just one a week, could cause your bankroll to grow very quickly indeed, but is that plausible?

Also, since scalps often disappear quickly, it seems like you wouldn't tend to have time to put more money into your account on the appropriate book before betting, you'd just have to ride with however much you have available, which means the lesser of your two accounts on the two books. So if you're only betting scalps you'd want to keep your bankroll evenly distributed among your books. Therefore if you have five books, each with 20% of your roll on them, then that 8 cent scalp earns its average of 2% of only the 40% of your roll on the two involved books, meaning less than 1% return on your bankroll, which becomes much less exponential.

So average scalp size, how often you can find them, and how many books you use are the determining variables in how profitable betting scalps is (as opposed to, say, putting your money in a mutual fund or something, and generating some minimal revenue with no work at all).

Anyone have any numbers for me?