I started developing my lines many years ago. I use game stats and typically start betting week 3. I compare my line to the Vegas line and look for 'mispricing.' I had great results for many years but last year my model took a shit. I've been making some adjustments however this season I'm winning because I'm fading the model. My adjustments just aren't working. My model uses 11 game statistics which try to quantify turnover, time of possession, and field position advantages. Each team has their own set of stats and I essentially set the teams against each other to determine the line. My thought is that due to the changes on defense....hand check rules, def holding, hitting defenseless receivers, etc....there are now major advantages to offensives. 30% of my lines use to be within a point of Vegas now I get only 1 or 2 a week. I think I'm right on the offensive explosions....last year the NFL broke scoring records for 7 weeks in a row. I think Vegas has made the proper adjustments in their spread lines but I am still lagging. Any thoughts or advice would be great. thanks all.
Models work pretty good. You have to make good predictions actually at the start of the season and stick w them. Ie predict points for, points allowed etc and use those throughout the season. If a team has injuries just subtract some pts for the injury for that week.
If you use actual pts throughout the year avg week to week it can get skewed up and down due to the schedule, injuries etc.
Just my .02
If you use actual pts throughout the year avg week to week it can get skewed up and down due to the schedule, injuries etc.
Just my .02
NFL use to be all about defense, & noe has switched to mostly an offensive show........NFL wants fans to stay glued to their TV, they don't want people turning the TV off cause of a defensive battle.
Its all about making money, hence the rule changes to protect the qbs & to have the offenses moving up & down the girls.......in today's NFL, I definitely would not want to play on the defensive side.
As for your prediction models, its kind d of hard to predict turnovers......turnovers kind of get streaky with defenses.
You still have to have a decent defense in the NFL to win the bowl.......
Its all about making money, hence the rule changes to protect the qbs & to have the offenses moving up & down the girls.......in today's NFL, I definitely would not want to play on the defensive side.
As for your prediction models, its kind d of hard to predict turnovers......turnovers kind of get streaky with defenses.
You still have to have a decent defense in the NFL to win the bowl.......
I think your right about the skew factor. I've never considered this type of system. How do you account for teams that play above/below expectation. If your setting your value at the beginning of the season. I found that identifying teams that are playing above/below expectation to be very profitable in the past and by using a week to week system;I think the model can see the changes before the public does?
I think it's about the money as well 'computer group'. I've been betting the overs in London because I think higher scores will entice London to strike a deal with the NFL. I know they want that. As for turnovers my thought is that teams that consistently turn the ball over will likely continue to do so in the future. I guess turnover may not be the right turn as much as giveaways. Do you think that's true.
What I've noticed over the years, is that good defenses usually get the lucky bounces to go their way, which converts to turnovers vs bad qbs like Cutler that are prone to giving the defense an easy way out.
Another way to look at turnovers is by giving a value to qbs & how many times they turn the ball over.
Imo, very few qbs can correct the problems they have with turning the ball over......I don't know if its their football IQ, if its their mechanics, if its the offensive line.......I believe to have a successful qb, you have to have a good offensive line, BUT, you need a smart qb with a high football IQ that can read defenses,etc.......
Some qbs like Geno Smith are playing for a terrible organization & whether he has a future in the NFL as a qb, remains to be seen.
As I stated before, look at which qbs are prone to turn the ball over, especially when playing vs a top 5 ranked defense.......thing is, who rates the defenses & how?
I recommend you come up with your own system for defenses cause a lot of stuff you see on ESPN, etc is worthless........
Another way to look at turnovers is by giving a value to qbs & how many times they turn the ball over.
Imo, very few qbs can correct the problems they have with turning the ball over......I don't know if its their football IQ, if its their mechanics, if its the offensive line.......I believe to have a successful qb, you have to have a good offensive line, BUT, you need a smart qb with a high football IQ that can read defenses,etc.......
Some qbs like Geno Smith are playing for a terrible organization & whether he has a future in the NFL as a qb, remains to be seen.
As I stated before, look at which qbs are prone to turn the ball over, especially when playing vs a top 5 ranked defense.......thing is, who rates the defenses & how?
I recommend you come up with your own system for defenses cause a lot of stuff you see on ESPN, etc is worthless........
You can check your predictions vs actual results and make adj accordingly. Sometimes you will over/underestimate teams. TB for ex is playing awful for their talent and a team like CLE is playing better than expected. So you can adj your expectations too once you see some consistency in improvement or decline. This is Dan Gordon's method more or less from his book "Beat the Sportsbooks"
My offensive/defensive ratings are all based on my own system. I think you are right about thinking about who is a top 5 ranked def. Currently, I compare offensive from team A to offensive to team B and defensive from team A to defensive from team B. Perhaps I should incorporate offense to defense......If for example a top ten offensive is playing against a top 5 defense how much of a line movement would you factor in? Also, do you account for top 10 for rush/pass/total when comparing to def top 5 rush/pass/total
I agree with both of those teams. Haven't had the guts to bet either yet though. I'm a browns fan and hate betting on teams that I 'root' for. I will look into that book. My model is based upon articles from the following: "Testing Rationality in the POint Spread Betting Market by John Gandar" Journal of FInance. "Testing Market Efficiency: Evidence From the NFL Sports Betting Market" by Philip Gray. Journal of Finance. "Evidence of Betting Market Intraseason Efficiency and INterseason OVerreaction to Unexpected NFL Team Performance" by Steven Sapra Journal of Sports Economics. "The Effects of Risk Aversion on Wagering: Point Spread versus Odds" by Bill Woodland. Journal of Political Economy. I actually have read several articles by Bill and Linda Woodland on sports betting. They have alot of interesting ideas.
I work in finance and have advanced degrees in Finance and have studied statistical modeling. Many of my methods are actually based upon predicting stock prices. I thought I could look to apply some of those concepts into the sports betting world. I did not consider looking into a book like what you suggested all of my research has been in academic journals. Have you reviewed any of the articles that I mentioned?
I work in finance and have advanced degrees in Finance and have studied statistical modeling. Many of my methods are actually based upon predicting stock prices. I thought I could look to apply some of those concepts into the sports betting world. I did not consider looking into a book like what you suggested all of my research has been in academic journals. Have you reviewed any of the articles that I mentioned?
I agree with both of those teams. Haven't had the guts to bet either yet though. I'm a browns fan and hate betting on teams that I 'root' for. I will look into that book. My model is based upon articles from the following: "Testing Rationality in the POint Spread Betting Market by John Gandar" Journal of FInance. "Testing Market Efficiency: Evidence From the NFL Sports Betting Market" by Philip Gray. Journal of Finance. "Evidence of Betting Market Intraseason Efficiency and INterseason OVerreaction to Unexpected NFL Team Performance" by Steven Sapra Journal of Sports Economics. "The Effects of Risk Aversion on Wagering: Point Spread versus Odds" by Bill Woodland. Journal of Political Economy. I actually have read several articles by Bill and Linda Woodland on sports betting. They have alot of interesting ideas.
I work in finance and have advanced degrees in Finance and have studied statistical modeling. Many of my methods are actually based upon predicting stock prices. I thought I could look to apply some of those concepts into the sports betting world. I did not consider looking into a book like what you suggested all of my research has been in academic journals. Have you reviewed any of the articles that I mentioned?
I work in finance and have advanced degrees in Finance and have studied statistical modeling. Many of my methods are actually based upon predicting stock prices. I thought I could look to apply some of those concepts into the sports betting world. I did not consider looking into a book like what you suggested all of my research has been in academic journals. Have you reviewed any of the articles that I mentioned?
Thanks. Will review those articles. I have a math/finance background too btw. Find this stuff fun for a hobby. For the posting as well, if you want to respond to a post, just click on the quote button in the bottom corner of the post and then type your response ( as I did here). It helps to sort your responses when you have many people posting in a thread.
I post the leaders picks for the NFL Las Vegas Westgate contest in a thread also FYI
