Exposure

One game, graded both ways

Say a book has taken these bets on a Sunday game at -110: $33,000 to win $30,000 on the favorite, and $11,000 to win $10,000 on the underdog. Now grade it both ways.

Result	The book keeps	The book pays out	The book's result
Favorite covers	$11,000	$30,000	Loses $19,000
Underdog covers	$33,000	$10,000	Wins $23,000

The exposure on this game is $19,000 on the favorite. That is the figure on the bookmaker's screen, and it updates with every bet. I used to watch that number more than I watched the games.

The same $2,000, two ways

Take the same $44,000 and split it evenly, $22,000 on each side. Whoever covers, the book keeps $22,000, pays $20,000 and makes $2,000. No sweat at all.

Now look at the lopsided book again. If the game is a true coin flip, it wins $23,000 half the time and loses $19,000 the other half, which on average is a profit of $2,000 a game. The same $2,000. The vig is worth exactly as much to the lopsided book as to the balanced one, and what changes is the swing: one book makes $2,000 every time, the other makes $2,000 on average, in lumps of plus $23,000 and minus $19,000.

So the question a bookmaker asks is not whether his book is balanced. It is whether he can afford to lose $19,000 on this game. A big shop can, and will leave the position alone. A local bookie with $30,000 behind him cannot, and has to do something about it.

What one game doesn't show you

Straight bets are the easy part. Three other things make the real figure harder to read, and the first is the one I'd worry about most.

A parlay carries its liability forward. A $50 five-teamer at -110 a leg wins $1,218, and after four legs have won, that whole payout is riding on the last game, on top of the straight bets. A book can look balanced on the Monday night game and still be buried on it because of live parlays. Futures are slower but no kinder: the book carries a liability on each team for months, and a longshot that a lot of small players took in August can become the biggest exposure in the building by January. Teasers are the third. Players like to tease the same favorites, and when they all come in, they all come in together.

This is why books cap payouts on long parlays. The page on how payout limits work covers it from the player's side.

Credit is exposure too

A book that takes post-up money has the player's stake in hand before the game starts. A bookie working on credit does not, so his exposure has a second part: the player who loses and can't pay.

Every credit player has a limit on how much he can owe before he has to settle. A careful bookie sets that figure by what the man can pay, not by what he wants to bet, and I wouldn't give much for the chances of one who does it the other way round. A winning week that can't be collected is a losing week.

Four answers to a big number

Most of the time, at a big book, the answer is to do nothing: if the number is right and the bankroll can take it, sit. Otherwise the book has three moves. It can move the line to make the heavy side less attractive (see odds movement). It can cut the limit on that side or that market. Or it can lay off, betting the heavy side at another book and giving up part of the profit to cut the risk; the sums are under risk management.

When a book takes your $50 on a side and refuses $500 on a prop, it isn't personal. It is telling you where it is willing to carry risk and where it isn't.

Your own exposure

The same sum applies to your bankroll. A book that risks $19,000 of a $30,000 roll on one game is gambling, whatever its edge, and a player who puts half his bankroll on one side is doing the same thing with the vig against him. I'd keep any one bet to a small part of the bankroll, 5 percent at the very most. Then you can be wrong for a while without going out of business.

Having the edge isn't enough. You have to still be there when it shows up.