Middling Totals in Football

[J_Treez]

This is a question about sports betting, but it's the probability aspect I'm interested in. So, I hope nobody's shorts get in a knot over it being posted here.

I'm reading Amarillo Slim in a World Full of Fat People the other day, and in the Doyle,Sailor and Slim Chapt. he goes into a discussion about middling totals in college football. He calls two bookies, and one has the total on an Alabama-Tennessee game at 45.5, whereas another has the same total at 50. So, he puts $800 on the under at 50 and another $800 on the over at 45.5. So he's risking the $80 vig for 4 numbers (46,47,48,49) to win $1600, and another number (50) to win $800. So, he's getting 20-1 on that $80 vig, and 10-1 if one bets a push at 50.

Now, HERE'S what I don't understand how to calculate. He says the TRUE ODDS of the total landing on one of those five numbers is 6-1. How the hell do you figure that?

[AngusThermopyle]

[ QUOTE ]

Now, HERE'S what I don't understand how to calculate. He says the TRUE ODDS of the total landing on one of those five numbers is 6-1. How the hell do you figure that?

[/ QUOTE ]

Empirical evidence.
If the game were baseball or basketball, his odds would clearly be wrong.
He is basing his numbers on his belief that the total scores of such a contest ( relative strengths of offense v defense ) will be distributed around 48 points, and that historically, 1 in 7 similar contests will have a total of between 50-46 inclusive.

"Handicapping" rather than "probability"

[J_Treez]

No wonder I couldn't wrap my head around it. I thought something like that may be the case, but the way everything was phrased confused me. Thanks.

[tylerdurden]

I'm wondering how often he finds two bookies offering o/u numbers that different. I've never seen differences bigger than 1.5 or so.

[J_Treez]

That particular instance was pretty unusal, according to Slim. He goes on to detail a hustle he pulled off a couple of years ago where he got bookies to let him move the totals by 1.5 points, provided he makes a string of five bets a week. Then he'd try to find discrepecies in the numbers and then move the totals.

Say we got Dallas vs. Philly totals from seperate books at 45 and 46 . He'd then buy the 45 down to 43.5 and bet the over, while buying the 46 up to 47.5 and bet under. Slim says this was "stronger than Nellies breath". I don't know Nellie, but I'm guessing she doesn't floss.

[WarDekar]

You can look at historical evidence and find games based on the spread and total what the chances of it hitting certain #s are and then find the % chance you should hit your middle.