Hoping some of you veterans can help with an answer. Like medium sized operation with about 45 players $13,000-15,000 cash reserve to insure solvency, 12-15,000 a day in action during football, how does one figure out a maximum amount of action to take on a select game like this weeks jets game, so with a $1,000 per game per player limit, when does a sportsbook decide to layoff action. Please any example will help, just very curious about this, im sure the answer is simple
-
-
Should be simple, every bookie tries to make a guaranteed profit regardless of the outcome of the event.
Near the start of each event, look at your set of bets and calculate how much you would win\lose for each given outcome of the event. These numbers will usually be different at the start since the line is never perfect. Then, for the side of the bet you would lose money on, begin to place bets so that you will win (or decrease your loss) if that event happens, until they even out.
Example. I'm a book and I hold 10 bets on the Redskins +3 and 8 bets on Seattle at -3 (same game, all $11-to-win-$10). Those bets netted me $198 (18 bets of $11).
If the Skins win my outflow is $210 (10 bets paid $21 each, so in that case I profit $198-$210 = -$12. Bad!
If Seattle wins my outflow is $168 (8 bets paid $21 each, so in that case I profit $198-$168 = +30. Good!
I'm a bookie, not a bettor, so I need to even those out. I don't want to lose money if the Skins win. So I go to another book and put $20 on the Skins +3. Why? My $20 will pay $38.18 if the Skins win, but I lose $20 either way. So now:
If the Skins win my outflow is still $210, but now $198-$20-$210+$38.18 = +$6.18.
If Seattle wins my outflow is still $168, and profit is $198-$20-$168 = +10.00.
You could fine tune that $20 further to equal them out, but the idea is that now you make less on Seattle than before, but either way you are +$$$. Of course it gets more complicated if you take bets at a range of spreads, or you can't get good spreads for your bet, that's where having a programmer working for you comes in handy.
Actually, that sounds like an interesting program to write. Input a range of bets at different spreads and different sides and input current available odds\spreads and figure out exactly how much to bet to equalize the payouts....
Hope that makes sense,
podonneComment -
the most common misconception is that books want equal action on both sides of a play. i have writing code for local books for many years and the easiest way to answer the question is that they want balanced plays in terms of amount - it doesn't matter who it is on. in other words a local guy wants all $100 plays or all $1000 plays or whatever. if they get totally out of whack on a particular game, sometimes they layoff - but there are many that let everything ride. think of a parlay card. say it has 100 plays on it (sort of like the number of plays on a typical saturday). what the guy running the cards wants, as an example, is all the cards to be 4 team plays for $10. If he gets 1000 cards like that - he knows what his take will be and he he doesn't give a damn whether or not the action is balanced between teams on the same play, but if some guy drops $1K on one play it throws the dyanmic out of whack. say a book has 100 players and they each make 6-$100 plays on any given day. there might be more action on certain plays - like today a lot of people are taking PIT and points angainst HOU. but if the book is balanced like the example it doesn't matter. as a rule with 600 plays (of each the same amount) 55% should lose. so if each player goes 3/3 they they collect 270 and the book gets 330. but the distribution should continue 6/4, 4/6, 5/1, 1/5, an 6,0 and 0,6. it is independent of the amount bet on any particular side of a particular play.Comment -
An old timer I know will layoff if he has like one or two games top heavy , more than that he lets it ride.Comment -
pokernut what does he consider top heavy really though because i really find that as interesting point. Not that the others aren't just very technical, and assuming the old school guys where way less technical in calculating the amount. Not less intelligent please don't get me wrong, just less technical in formula. Still knowing what he considers top heavy would be interesting. Thank you all for the answers so far.Comment -
Really? They should like even money on all sides. Leaving the book unbalanced would be the same as the book having an opinion on the game. I thought that was the whole point of a being a book, you make money either way.
I get (and have seen research to the effect) that sometimes very large books\sites\vegas will take positions on games, purposefully allowing the book to become unbalanced so they can make more money, but no way your average bookie\book can play it like that. The games aren't that easy to beat.
I can't say books don't do it, or whether they "like" having thier books unbalanced, but damn they sure shouldn't like being unbalanced on games as a long-term strategy.Comment -
Comment -
Feel free to continue on the conversation guys but the OP won't be returning. It was a shill hiding links in his posts.Comment -
Nevertheless, my point was that they are leaving money on the table.
This was actually an interesting problem, so I whipped up a program to simulate the effect of laying off imbalances or not. Generates a series of 100,000 drastically simplified events with random bets on each side and resolves randomly with or without laying off.
Suprisingly, laying off didn't make that much of a difference. I mean, the difference between not laying off and laying off all the time clearly increases profit, but only by between $1 and $2 per game. My guess is that the time required to make the layoff bet, fund the transaction, transaction fees, its not worth it for a local bookie. If it were me, I'd just do it for BIG imbalances, maybe where my negative liability is greater than 10% of my bankroll.Code:Book profit with no layoff: $461158.9, $4.611589/game Book profit with layoff when larger is 2x smaller: $504696.5, $5.046965/game Book profit with always layoff any imbalance: $584215.8, $5.842158/game
Bigger books, especially computerized ones, take note though. Money is left on the table if you leave a book unbalanced.
Code attached if interested. The calculation of exactly how much to lay off to equalize the profit is suprisingly simple (assuming you can get 10/11 at the same spread of all your taken bets).
podonneCode:[SIZE=2][FONT=Consolas][SIZE=2][FONT=Consolas]layoff_dollars = (1 - larger / smaller) * smaller[/FONT][/SIZE][/FONT][/SIZE]
Comment -
You're doing something wrong. Laying off costs you EV and your simulation has nothing to do with EG.Comment -
Put down the crack pipe, realize that making -ev bets doesn't increase your ev, verify that with your numbers in post 2, and then debug your own program.Comment -
i always wonder if locals take a position at an online book opposite the bet of the 90 percent player to offset some of the losses if it hits. it wouldn't have to be 1 to 1 either and with a reduced juice book you would still not be laying the -110 offshore.Comment -
Getting +110 on a 50/50 bet suggests a kelly stake of ~4.5%Comment -
my local guido once told me they never lay-offComment -
This is a great post you guys started. I know locals as well, and some do go broke. What I find interesting is the books in Las Vegas are not permitted by law to lay off any money to other places. Therefore, they are always rooting for one side over the other. A balanced book is something that rarely happens. I may not understand the higher math, but I was always under the impression that despite being top heavy on many games, the book will win 50% and the players will win 50% of those contests. Naturally, the losers will pay 11-10 for their losers and the house will profit. And here lies the reason why the books are always changing their numbers. It is their magic 'weapon' for enticing balanced action, because even though it rarely happens, it is always to be strived for. Every place cooks their hamburgers differently, and likewise with books they can decide to employ a strategy that they prefer. I was always told that books that are well financed rarely lay off, while smaller offices might have a lower tolerance for swings. I've heard there are offices that are both betting and booking offices. A 'sharp' might come in with a play they respect, and they might (1) move their number, (2) bet that game with other offices at advantageous numbers, and (3) decide to 'gamble' on that game. It is my understanding this is what might create what I see referred to as 'reverse line movement.' However, I was very fascinated with your interesting posts and would hope that you might reply to my note here and lend your experience and knowledge and math abilities to the discussion. I, too, would like to know what the sharp books and locals do, but what I wrote above is what I was always told.Comment -
Books may not be getting +110 overall though.
If they open at -4 -110 and have to move it to -4.5 -110 then the early money is the equivalent of -4.5 +105 or whatever half the 4 is worth.
I suspect most bookies manage by the puke number. When your potential liabilities hit a number that would make you throw up in your mouth a little then you lay off the excess.
Can Vegas books take "insurance" on a sporting event if they can't lay off at another book?Comment -
depends how much you are shading the action
use kellyComment -
Correct- use Kelly. Unbalanced action doesn't matter as long as too much of your bankroll is not at risk. If Kelly says that your optimal stake is 4.5% of your bankroll, you would layoff the unbalanced side when it is greater than 4.5% of your bankroll. And by the way, a $13-15k "reserve" if not nearly enough for 45 players. You would half to lay off bets when the potential loss is less than $700 on one side. With a bankroll this small, you are doomed for failure, because there will be deadbeats that will not pay or will slow pay. And when people win, you will have to pay out.
With a bankroll this small, you would have to limit people to 100 bucks a game. Remember with your 4.5% theoretical hold, you would need to book 1 million dollars in bets to win a measely 45k in profit. When you add in stiffs and slow plays, you are dancing way to close to the edge of the cliff with a 15k bankroll. You also have to pay your online shop a % that needs to get paid whether you get stiffed or not. I am assuming that you can not answer the phone for 45 people all trying to call 5 minutes before the games start.
And remember it only takes one pissed-off loser, one jealous guy, one guy who has a crush on your girlfriend- to drop a dime on you and the cops will bust down your door. You will have a criminal record and never be able to get a professional job.
The only way to book is to find 10-15 middle aged, professional guys with families and a high net worth. And do not let anyone get in over their head with respect to their income. But guys like this will bust out your 15k bankroll. You will end up being a runner for your lay-off book and make very little money yourself.
Honestly, there are so many legal ways to make money; you'd be better off saving some money and opening a legit business.Comment -
editComment -
What bookies have to their advantage (beyond the juice) is players are almost always betting way too much relative to bankrolls.Comment -
I'm confused. Isn't podonne's example the only sure way for a book to make money no matter what?Comment -
It is expectation rather than hope.
The exposure/risk the book is willing to take on each side/total should be commensurate with EV.
In a totally lopsided market where the line has moved against you, it would more often than not make sense to hedge out but doing so when the numbers are in your favour is just giving money away in the long run.
The risk of ruin is factored in using Kelly/Fractional Kelly.Comment
