-
You are viewing Orangepower as a Guest. To start new threads, reply to posts, or participate in polls or contests - you must register. Registration is free and easy. Click Here to register.
I don't gamble, and so I only think about this stuff occasionally, but reading what you wrote makes me think Vegas has to think something like -10 is the true line.
It seems to me that there's only two distinct ways for Vegas to go about it: get all bets to be split 50%-50% about the line (because they have more confidence in the public than themselves) and pocket the rake, or even better (when doable), set the line right in between the two Guassian (presumably) distributions of: #1 what the public thinks, and #2 what Vegas knows reality to be (because Vegas fully trusts themselves more than the public). #3 would be some combo of the two/somewhere in between (to hedge). Lastly, Vegas would never want to set the line such that both distributions were entirely on the same side of the line because that would mean the public would always be correct (and Vegas would always lose).
We know they aren't doing #1 fully because you say 75% of the bets are on OSU** rather than 50-50, so that means they are either doing #2 fully, or partially (which would be #3). So going through scenarios...if they place the line right down the middle of the "what the public thinks" distribution, then half the public loses all the time and half win all the time, and it's a wash other than the rake. Alternatively, if Vegas places the line down the middle of the "reality" distribution, then half the time all the public loses, and half the time all the public wins, and its a wash once again. But if they place the line directly in between the "public perception" distribution and the "reality" distribution, then the entire public loses all the time*, which is the best possible scenario for Vegas.
So in our case, if "public perception" is -4 ISU and Vegas is at -7, then Vegas must think reality is -10 ISU if they're doing #2. Vegas has no motivation to put the line anywhere that makes them less money, so, again, that means the line gets put right in between the two distributions.
For anyone that thinks Vegas is probably doing #3, hedging a bit towards #1 (public perception) because Vegas only partially trusts their own sports experts, that would actually mean Vegas thinks "reality" is even further beyond -10 ISU (for example, -13 ISU).
None of this means Vegas is correct, of course.
*For simplicity, all this assumes the two distributions don't overlap at all, but even if they do, it still holds, but holds progressively not as strongly the more the distributions overlap.
**Actually, you said bets, not money. I'm a little unsure how that complicates things. So my above post is probably not quite right just based on this, if not for some other reason too.
It seems to me that there's only two distinct ways for Vegas to go about it: get all bets to be split 50%-50% about the line (because they have more confidence in the public than themselves) and pocket the rake, or even better (when doable), set the line right in between the two Guassian (presumably) distributions of: #1 what the public thinks, and #2 what Vegas knows reality to be (because Vegas fully trusts themselves more than the public). #3 would be some combo of the two/somewhere in between (to hedge). Lastly, Vegas would never want to set the line such that both distributions were entirely on the same side of the line because that would mean the public would always be correct (and Vegas would always lose).
We know they aren't doing #1 fully because you say 75% of the bets are on OSU** rather than 50-50, so that means they are either doing #2 fully, or partially (which would be #3). So going through scenarios...if they place the line right down the middle of the "what the public thinks" distribution, then half the public loses all the time and half win all the time, and it's a wash other than the rake. Alternatively, if Vegas places the line down the middle of the "reality" distribution, then half the time all the public loses, and half the time all the public wins, and its a wash once again. But if they place the line directly in between the "public perception" distribution and the "reality" distribution, then the entire public loses all the time*, which is the best possible scenario for Vegas.
So in our case, if "public perception" is -4 ISU and Vegas is at -7, then Vegas must think reality is -10 ISU if they're doing #2. Vegas has no motivation to put the line anywhere that makes them less money, so, again, that means the line gets put right in between the two distributions.
For anyone that thinks Vegas is probably doing #3, hedging a bit towards #1 (public perception) because Vegas only partially trusts their own sports experts, that would actually mean Vegas thinks "reality" is even further beyond -10 ISU (for example, -13 ISU).
None of this means Vegas is correct, of course.
*For simplicity, all this assumes the two distributions don't overlap at all, but even if they do, it still holds, but holds progressively not as strongly the more the distributions overlap.
**Actually, you said bets, not money. I'm a little unsure how that complicates things. So my above post is probably not quite right just based on this, if not for some other reason too.
No one is going to share their formula for setting lines. It takes years to fine tune the idea of weather into an equation that equals points. Or in injuries. A starting qb out can cost as much as 7-10 points while a starting WR doesn’t move the line at all.
That said, Iowa State is reporting zero injuries.
Currently, even with a 7 point spread, 72% of the betting is on us.
That said, Iowa State is reporting zero injuries.
Currently, even with a 7 point spread, 72% of the betting is on us.
Sure. Take total points for OSU offense. Divide by number games played. 159/6 = 26.5
Total points given up by OSU. 108/6 = 18
Giving us a positive score of 8.5 total points over the average victory. 26.5-18= 8.5
Now do the same for opponent.
ISU total offense points. 202/6 = 33.66
ISU total defense points. 98/6 = 16.33
Giving ISU a positive score of 17.33
Compare the 2 final positive scores ( not always positive ) 17.33 - 8.5 = 8.83
So the statistical base line spread involving these 2 teams is 8.83.
Like I said earlier, I didn’t have the time to do the full analysis. That takes 30 or so minutes. Even with free information at your finger tips. What you do with that baseline thread from here on is propriety protected. Variables as numerous as 13 to 15 out. Weather. Who the refs are and are they liberal in their calls? Injuries. Qb girlfriend troubles. It goes on and on. Just eyeballing it I said earlier 7. Not too bad for old eyeballs but if you look back you’ll see say where I could argue that it should be 10.
8.83. Hilarious.
And that’s why I wondered earlier why hasn’t Vegas factored in some home points for ISU?
That’s their secret. But I’ll tell you. Two games on the ISU schedule are of whack.
107-10. KU and UNLV. Giving them a two game positive score of 48.75. So they weighted those probably by around 20%.
And it’s not enough. Spread down to 6.5 now.
Total points given up by OSU. 108/6 = 18
Giving us a positive score of 8.5 total points over the average victory. 26.5-18= 8.5
Now do the same for opponent.
ISU total offense points. 202/6 = 33.66
ISU total defense points. 98/6 = 16.33
Giving ISU a positive score of 17.33
Compare the 2 final positive scores ( not always positive ) 17.33 - 8.5 = 8.83
So the statistical base line spread involving these 2 teams is 8.83.
Like I said earlier, I didn’t have the time to do the full analysis. That takes 30 or so minutes. Even with free information at your finger tips. What you do with that baseline thread from here on is propriety protected. Variables as numerous as 13 to 15 out. Weather. Who the refs are and are they liberal in their calls? Injuries. Qb girlfriend troubles. It goes on and on. Just eyeballing it I said earlier 7. Not too bad for old eyeballs but if you look back you’ll see say where I could argue that it should be 10.
8.83. Hilarious.
And that’s why I wondered earlier why hasn’t Vegas factored in some home points for ISU?
That’s their secret. But I’ll tell you. Two games on the ISU schedule are of whack.
107-10. KU and UNLV. Giving them a two game positive score of 48.75. So they weighted those probably by around 20%.
And it’s not enough. Spread down to 6.5 now.
-
3
- Show all
Also this is why I quit. You have to do this for every matchup on your own. Then you compare to their lines. Look for anomalies. Find out why they are. Act accordingly. We were doing it preinternet days. Very valuable information back then. Still is. But a lot of it is misinformation just like everything else on the net. Working that amount time makes wayyyyy more money in the stock market.
There’s more. There’s an entire industry devoted to just statistics involving the spreads only. They track them back to the very first bet in some cases. Lots of believers in this system as well.
For instance:
Under is 5-0 in Cowboys last 5 games as an underdog.
Under is 4-0 in Cowboys last 4 games on grass.
Cowboys are 4-0 ATS in their last 4 games overall.
Cowboys are 4-0 ATS in their last 4 games following a straight up win.
Cowboys are 4-0 ATS in their last 4 conference games.
Just based on that info alone. Many will make a parlay ( multiple picks must happen ) and take the pokes and the under.
After all this time, I still feel that flipping a coin or picking a winner is always a 50% proposal. However, although it’s 50% every time, I can’t wrap my head around someone picking right/wrong 25 times in a row. I’ve calculated it as you can imagine, 2 to the 25th power is extraordinarily. Yet, I’ve seen it happen year in and year out.
For instance:
Under is 5-0 in Cowboys last 5 games as an underdog.
Under is 4-0 in Cowboys last 4 games on grass.
Cowboys are 4-0 ATS in their last 4 games overall.
Cowboys are 4-0 ATS in their last 4 games following a straight up win.
Cowboys are 4-0 ATS in their last 4 conference games.
Just based on that info alone. Many will make a parlay ( multiple picks must happen ) and take the pokes and the under.
After all this time, I still feel that flipping a coin or picking a winner is always a 50% proposal. However, although it’s 50% every time, I can’t wrap my head around someone picking right/wrong 25 times in a row. I’ve calculated it as you can imagine, 2 to the 25th power is extraordinarily. Yet, I’ve seen it happen year in and year out.
The formula isn't difficult. They use something similar to Sagarin and other similar statistical models. It serves as a starting point. The model probably includes something on betting history. In other words do fans of certain teams tend to bet one way or another? i.e. is there a betting trend for that team.
After that it really isn't about who they think will win and by how many points. The point spread is adjusted based on the money that's being bet for one team versus the other. Ideally, they have the same dollars bet for both teams, that way Vegas always wins.
How to Calculate the House Edge
The main reason the house always wins in sports betting is because they are able to charge an addition tax on all their bets, known as the juice. Standard juice is -110, meaning you have to pay the house an extra 10 cents on every dollar you bet. However, this juice could be even higher, up to 20 cents or more (-120).
To put this in perspective, let's say you place two NFL spread bets. One is on the Miami Dolphins + 3 and the second is on the Dallas Cowboys -6. In both instances you are paying -110 juice on both bets. Let's say you split the games and go 1-1 with your two bets, with the Dolphins covering but Dallas failing to cover or losing straight up. If you go 1-1, that means you broke even, right?
Wrong. Because you had to pay -110 juice on both bets, this means you had to risk $110 in order to win $100. So in total you risked $220 overall. You won $100 on your Dolphins bet, plus you got the $110 you risked back. But you lost the $110 you risked on your Cowboys bet. So overall, you risked $220 and ended up with $210. This means you actually lost $10 dollars despite splitting your plays.
https://www.vsin.com/sports-betting-101-how-to-calculate-the-house-edge/
After that it really isn't about who they think will win and by how many points. The point spread is adjusted based on the money that's being bet for one team versus the other. Ideally, they have the same dollars bet for both teams, that way Vegas always wins.
How to Calculate the House Edge
The main reason the house always wins in sports betting is because they are able to charge an addition tax on all their bets, known as the juice. Standard juice is -110, meaning you have to pay the house an extra 10 cents on every dollar you bet. However, this juice could be even higher, up to 20 cents or more (-120).
To put this in perspective, let's say you place two NFL spread bets. One is on the Miami Dolphins + 3 and the second is on the Dallas Cowboys -6. In both instances you are paying -110 juice on both bets. Let's say you split the games and go 1-1 with your two bets, with the Dolphins covering but Dallas failing to cover or losing straight up. If you go 1-1, that means you broke even, right?
Wrong. Because you had to pay -110 juice on both bets, this means you had to risk $110 in order to win $100. So in total you risked $220 overall. You won $100 on your Dolphins bet, plus you got the $110 you risked back. But you lost the $110 you risked on your Cowboys bet. So overall, you risked $220 and ended up with $210. This means you actually lost $10 dollars despite splitting your plays.
https://www.vsin.com/sports-betting-101-how-to-calculate-the-house-edge/
-
1
-
1
- Show all
The formulas we are discussing are two different ones. I know nothing of the formula that uses saragin and whatnot to decide the ESPN edge. I was showing how oddsmakers set the baseline. That part is easy. How they tweak it to get 50% action both ways is a house secret. That’s why camera phones were blocked in casinos at first. They use to use spies anyway and with the internet, house odds are no longer a secret to the competition.