How do sportsbooks make money on this...

tacomax · 09-09-06, 03:08 AM

Originally posted by rainbowworld

I thought that the spread was supposed to even things out so that 50% of the money would be bet on the favorite and 50% would be bet on the underdog?

Please help me understand this.

Say you've got a book that works on a -110 spread handicap. If a book took even money on each side then they can expect a 4.5% return on the bets placed due to the juice. So, say the spread is -4 and there is a 50/50 chance of winning/losing then the sportsbook will make 4.55% on equal wagers on either side.

However, in reality, sportsbooks don't get even money on both sides which they can use to their advantage. Let's say that the majority of the money is always on the favourite and that the square players will be on the favourite whatever the line is (within reason). So, the sportsbook offers a -6 spread and still gets more money on the favourite. Sure, it's a lop-sided account sheet for the book but they are still offering a 50/50 odds (i.e. odds of -110) to bet on a favourite but the odds of hitting -6 might only be 40%. So they're offering 50/50 odds on a play that's only ever going to hit 40% of the time.

fearless · 09-09-06, 03:34 AM

Originally posted by tacomax

Say you've got a book that works on a -110 spread handicap. If a book took even money on each side then they can expect a 4.5% return on the bets placed due to the juice. So, say the spread is -4 and there is a 50/50 chance of winning/losing then the sportsbook will make 4.55% on equal wagers on either side.

However, in reality, sportsbooks don't get even money on both sides which they can use to their advantage. Let's say that the majority of the money is always on the favourite and that the square players will be on the favourite whatever the line is (within reason). So, the sportsbook offers a -6 spread and still gets more money on the favourite. Sure, it's a lop-sided account sheet for the book but they are still offering a 50/50 odds (i.e. odds of -110) to bet on a favourite but the odds of hitting -6 might only be 40%. So they're offering 50/50 odds on a play that's only ever going to hit 40% of the time.

Thanks but you didn't touch the money line which is what I was asking about. I mean, if over 90% of the public thinks that seattle is going to win this game, you'd expect some number close to that to bet seattle on the moneyline, right? Because, in reality, Seattle should definitely win this game, all the sportsbooks are gonna lose on the moneyline, right (regardless of whether or not they cover the spread)? So, how do sportsbooks make money in a situation like this?

rm18 · 09-09-06, 03:40 AM

You don't really understand what thinks means, vs the line. I think Seattle wins, but I am still betting Detoit ML at +275 or so.

tacomax · 09-09-06, 03:44 AM

Originally posted by rainbowworld

Thanks but you didn't touch the money line which is what I was asking about. I mean, if over 90% of the public thinks that seattle is going to win this game, you'd expect some number close to that to bet seattle on the moneyline, right? Because, in reality, Seattle should definitely win this game, all the sportsbooks are gonna lose on the moneyline, right (regardless of whether or not they cover the spread)? So, how do sportsbooks make money in a situation like this?

Same difference, really. Most of the money is on the spread rather than the ML so it's less of an issue. But anyway, say that Seattle should statistically be at odds of -220 to win the game, if the sportsbook (knowing that their customers are going to back the favourite whatever the price) offers -280/+260 on the game then they know that they're going to get a lot of money on a price which is very unfavourable to their customers.

fearless · 09-09-06, 03:53 AM

Originally posted by tacomax

Same difference, really. Most of the money is on the spread rather than the ML so it's less of an issue. But anyway, say that Seattle should statistically be at odds of -220 to win the game, if the sportsbook (knowing that their customers are going to back the favourite whatever the price) offers -280/+260 on the game then they know that they're going to get a lot of money on a price which is very unfavourable to their customers.

Thanks, I understand it now. I guess it works out somehow. I mean the difference between -220 and -280 is only 5% (-280 = 74% chance of winning and -220 = 69% chance of winning), I don't consider that to be significant when the real odds should be like 90%. But, I'm sure a lot of people consider 5% to be very significant and it is when you're talking about huge bets. But, to me, a win's a win.

Ganchrow · 09-09-06, 04:18 AM

Originally posted by rainbowworld

I mean, if over 90% of the public thinks that seattle is going to win this game, you'd expect some number close to that to bet seattle on the moneyline, right?

Just as rm points out, the percentage of the general public (a notion not inconsequentially different from the betting public) which might believe a Seattle win most likely is fundamentally quite different from the actual likelihood that Seattle will win.

When a value bettor (as all professional handicappers are) places a bet on Seattle to win at -280, he's not necessarily saying that he definitely expects Seattle to win (such a claim would be foolish), but rather that he believes Seattle has a probability of greater than 73.68%. Similarly, when rm bets Detroit at +275, he's not doing so because he believes that Detroit is more likely to win the game than Seattle, but rather that he believes that Detroit has a greater than 26.67% probability of victory.

I suspect that the overwhelming majority of the 10% believing a Detroit victory most likely are not sportsbettors -- and the even more overwhelming majority of those who are almost certainly not profitable ones. There's a big difference between binary (yes/no) and continuous (0%-100%) propositions. If you roll a 6-sided one time do I expect to see a 6? No. I don't. And I'd put a lot of money at even odds on that. But offer to pay me a 100-1 on that 6 showing up? Well, of course you'd find me betting the other way ...

Ganchrow · 09-09-06, 04:20 AM

Originally posted by rainbowworld

Thanks, I understand it now. I guess it works out somehow. I mean the difference between -220 and -280 is only 5% (-280 = 74% chance of winning and -220 = 69% chance of winning), I don't consider that to be significant when the real odds should be like 90%.

Then I do hope you're loading up the truck with Seattle -280. If Seattle's chances truly are 90% than -280 is an absolute steal ...

diamond · 09-09-06, 04:41 AM

Originally posted by rm18

You don't really understand what thinks means, vs the line. I think Seattle wins, but I am still betting Detoit ML at +275 or so.

Spot on. Its about betting value. Even though I think a team would win, I could still end up betting the other side because of the value.

fearless · 09-09-06, 05:13 AM

Originally posted by ganchrow

Then I do hope you're loading up the truck with Seattle -280. If Seattle's chances truly are 90% than -280 is an absolute steal ...

Well, if the LVSC is saying that Seattle is a -6 at Detroit, that would mean that the spread would normally be about Seattle -13 if the game were played in Seattle and at -13, Seattle would have a greater than 90% chance to win the game, according to the oddsmakers. Personally, I think detroit's home field advantage is very low against a great team like seattle, possibly even non-existant. If you go by the LVSC, about -12 = 90% probability of winning and if I were going to make a line on this game, that's probably what I would say. But, it's only the first game, there are too many x factors and things we don't know.

About this particular game though, I would like to point out that the spread has gone from Seattle -3.5 to Seattle -6.5 at some books which is a huge move.

moses millsap · 09-09-06, 07:36 AM

All very well explained by the posts above, when working with MLs, it's all about value.

For instance, say, I rate Federer to win 90 times out of 100 against Roddick. Theoretically, I should bet on Federer if I can find a line better than 1.11 and the amount that I should wager on how much value I think the play presents. This is up to you to decide. Simple elementary formula is with this example:

1.20 odds x 90% chance of winning = 108

The true odds in your mind will always equal 100, so the discrepancy from that number to the number you get from the book's number should dictate your confidence rating for that specific play.

fearless · 09-10-06, 06:57 PM

Originally posted by ganchrow

Then I do hope you're loading up the truck with Seattle -280. If Seattle's chances truly are 90% than -280 is an absolute steal ...

I would like to point out that Seattle won the game 9-6. Beautiful, they won but everyone who bet on them from -3.5 to -7 (where the spread eventually ended up) lost their wagers. That's why I only bet the money line.

onlooker · 09-10-06, 07:06 PM

Even the Moneyline bettors had to be sweating this one out. They were getting out played and penalties is what killed the Lions in this game. Be glad you squeaked out this Seahawk win.

JohnnyBoy928 · 09-10-06, 07:10 PM

Right!!!!!! Sometimes I just rather be lucky then smart... Fa Sho

Just too bad I am neither

Originally posted by onlòóker

Even the Moneyline bettors had to be sweating this one out. They were getting out played and penalties is what killed the Lions in this game. Be glad you squeaked out this Seahawk win.