MIT Math Wiz Has Created A Program To Pick Winners for College and Pro Ball

[gavin]

Ok.
So I l grew up with a few kids that ended up going to and graduating from MIT. I got them hooked on betting probably 2 years into college for all of us. They created some program using statistical data from like every game ever played to create simulations, breakdowns, etc. To be honest I dont know how they got it all but they spent about 2 weeks creating the program. They are already pulling together money from friends etc to try and get this thing going if it ends up working. They let me have access to the program which I had to download on to my PC. They are going to tell me who to bet on which games look better than others etc. Will it work? I dont know. I am skeptical but these guys really know what their talking about in regards to math, statistics, probability etc....It only works for football for some reason. I will keep a record tracker and post the picks they tell me to. I guess its worth a shot!
Back during fball season.

 

[BigdaddyQH]

Good Luck with the new system, but to get an accurate read, you will have to give it 5 years or more. Anyone can have a great, or terrible season, and one season does not make a system good or bad.

 

[russ1945]

Good Luck with the new system, but to get an accurate read, you will have to give it 5 years or more. Anyone can have a great, or terrible season, and one season does not make a system good or bad.

Not only that but they have to take into consideration coaching changes, and a host of other factors that are not mathematical. I have seen a modified slide rule for handicapping horses. I can tell you right now they are better off focusing on handicapping horseracing over college football.
You can factor in linage, types of tracks, jockey and owner info all of which there is a huge accessible data base on. I have accessed the NCAA statistics and they do keep an archive but I think in the end coach's 11 yr old grandson may out pick these guys.
If they persist any way, tell them to get some access as to how lines are set in general and a valid data base for that. I know this sounds exciting and who knows it could work. Be really interesting to keep track of if it doesn't cost you any money doing it. I am not trying to discourage them but school of hard knocks is a post graduate degree in and of itself.

 

[UoweDucks]

Why not bet via simulated games on NCAA 2010 xbox360. Takes all the fun out of handicapping...

 

[Conan]

That's allright. I'll just catch it when it comes out on HBO.
Do you think Kevin Spacey will be the MIT professor in this one too?

 

[russ1945]

Why not bet via simulated games on NCAA 2010 xbox360. Takes all the fun out of handicapping...

Ducks: They are probably running their program against prior seasons. Does MIT give a degree in sports betting? Probably not and there is a reason, it is not a science. You would probably have a better chance if Rainman was into college football. Truth of the matter is there is way to much math involved and the predictability would be something that would have to be adjusted and readjusted constantly. They are probably thinking, well there are just 60 or so games that should be easy to hone in on. In terms of the kind of math they are usually dealing with it actually is a small target. It is that they are going to have to deal with variables and they will not even know they need them until they need them. Again, the old dart board comparison, how accurate do you have to be when the odds are 50/50. Well conversely, 50/50 also means all or nothing. Right now they are laughing their asses off at us for even questioning their intentions or their abilities. After all, we did not go to MIT. I say let's book 'em ourselves.

 

[gavin]

I asked them about how things like weather are applied into the program and they said that part should be ready soon. I really dont understand the stuff they are talking about...I didnt go to school there and arent like these guys but apparently there is some variable they will plug in on saturday mornings with approximate weather for the remaining day throughout the country...who knows?!? I just figured anything is possible. They did say they went back only 25 years for all of the statistical data they used etc...f&*& its better than flipping a coin for me! I'll keep everyone posted on what they tell me to bet come fball season and well see how well it works. They are seriously into this and looking at every angle possible. They even hired some "degenerate" to tell them about every in and out of the game etc...whatever

 

[TheComputerGroup]

This sounds very interesting! I will be paying attention to see what games they come up with. Posting all there plays & keeping a precise record ATS will show us if they have the right variables in this program. If these guys are from MIT, thats the start you want for creating a sports program, because the math involved is insanely diffuclt to the ordinary person. If this program does work & its hitting at 60% or higher, there problem will be looking for the best line & finding bookies willing to take there bets. There always moving to Vegas First things first, lets see how well it works & go from there.

 

[gavin]

yea should be...I will keep bumping thr thread

 

[sixers76]

there was a group like that years ago called the computer boys And kosher boys they worked for billy walters.

 

[SoonerBS]

Why not bet via simulated games on NCAA 2010 xbox360. Takes all the fun out of handicapping...

That one has already been done by MistaFlava . . . . . .

 

[gavin]

yea I mean the strangest thing is is that they dont even care about making money really...they are more interested in beating the odds or system or whatever but they are going to pick the ten strongest games each week...i think its for their thesis or something but these guys are some serious nerds.

 

[TheComputerGroup]

Have they back tested this program to see how well it has done? Im impressed they have this program running in only 2 weeks. Before the spreads come out for week 1, you should ask them to run the program to see what Percentage or lines they come up with for those games, or will they wait until several weeks to go by before implementing there program.

 

[gavin]

I asked them about the backdating and they said they already ran like a million scenarios....they must have used old lines from last season and they said there was like a 63% success rate. I really dont know how they are getting it but they plan on giving me the ten strongest games each week. We will see how it goes. They are continuing to work on the program through the summer..i guess they got some grant so they dont have to work...just work on this...tough job huh? I'm glad I grew up with the kid....maybe it will pay off!

 

[russ1945]

I asked them about the backdating and they said they already ran like a million scenarios....they must have used old lines from last season and they said there was like a 63% success rate. I really dont know how they are getting it but they plan on giving me the ten strongest games each week. We will see how it goes. They are continuing to work on the program through the summer..i guess they got some grant so they dont have to work...just work on this...tough job huh? I'm glad I grew up with the kid....maybe it will pay off!

If you follow Steele very much you should know this. Before the season starts, I guess before he even gets the Magazine out, he has a system that runs all of the seasons matchups and that is how he comes up with his projected point differential. I think he uses more than one format to simulate all of the games. So there is a base of data out there and it is just a matter of what variables you inject into the projection (new HC, new QB, home field advantage, etc.). To me the interesting thing is that you would have to be able to project a line to make it work ATS. You could project the final scores and then wait until the actual lines come out. The trouble with those projections is that they can and probably are inherently wrong because of factors like injuries, TO margins, bad weather, development of new QB's, etc. They have probably gone back to prior years and taken real scores against real lines but their relevance could be out the window the first few weeks of a new season but might get more relevant as a season progresses and they make the right corrections. It may hurt some of our feelings that these guys might succeed but if greed is not their motivation they are probably more likely to succeed. I have a friend who opted for the AF Academy instead of MIT who is better at math than I am. He knows nothing about football and has no interest. He is in the process of looking over my spreadsheet and we are interested in who he picks based on the numbers alone.
The beauty of math is that it has infinite possibilities. It has helped define the world, why not football.

 

[TheComputerGroup]

Good stuff gavlam, keep us posted!

 

[TheComputerGroup]

If you follow Steele very much you should know this. Before the season starts, I guess before he even gets the Magazine out, he has a system that runs all of the seasons matchups and that is how he comes up with his projected point differential. I think he uses more than one format to simulate all of the games. So there is a base of data out there and it is just a matter of what variables you inject into the projection (new HC, new QB, home field advantage, etc.). To me the interesting thing is that you would have to be able to project a line to make it work ATS. You could project the final scores and then wait until the actual lines come out. The trouble with those projections is that they can and probably are inherently wrong because of factors like injuries, TO margins, bad weather, development of new QB's, etc. They have probably gone back to prior years and taken real scores against real lines but their relevance could be out the window the first few weeks of a new season but might get more relevant as a season progresses and they make the right corrections. It may hurt some of our feelings that these guys might succeed but if greed is not their motivation they are probably more likely to succeed. I have a friend who opted for the AF Academy instead of MIT who is better at math than I am. He knows nothing about football and has no interest. He is in the process of looking over my spreadsheet and we are interested in who he picks based on the numbers alone.
The beauty of math is that it has infinite possibilities. It has helped define the world, why not football.

Russ, will you place the games your spreadsheet comes up with on this forum?

 

[russ1945]

All you have to do is ask - of course

Russ, will you place the games your spreadsheet comes up with on this forum?

First let me point out I have some experimental ratings I am trying out. I am not absolutely sure if I can predict with them this year. They should be of help next year after I have a year to work with them. What I will post on here is if I see a team who on this rating shows potential to beat the line by a certain number higher than the established line. I will also post after the games if something pops up that could have been beneficial but was unpredicatable until I got the post game results. I know that sounds funny but that is the way it is. For instance, we were all throwing out some numbers on the Memphis/Ole Miss game. I think I threw out a line something like Ole Miss -17 and someone else thought it might be closer to 7 or so. The rating I am talking about shows Ole Miss 25.5 pts
better than Memphis. Do we go with that right off the bat? Do we let it go find out if it is right or wrong and see what we find in week 2. If we wait the numbers will tell me if I was right on one or both teams. Maybe the number for Ole Miss is good as is and Memphis has to be adjusted, or vice versa, possibly both. But I have other factors in the mix too. I will post five games a week (games I will bet) but I may not put every game I bet on here. One thing you can bank on, if I put a game on here I am betting it. I am also interested to see if someone that has one of my spreadsheets can find something that can help us all. That is why I put the game breakdowns on here. My hope is that these MIT kids will be so good I will never have to do any research ever again.

 

[TheComputerGroup]

First let me point out I have some experimental ratings I am trying out. I am not absolutely sure if I can predict with them this year. They should be of help next year after I have a year to work with them. What I will post on here is if I see a team who on this rating shows potential to beat the line by a certain number higher than the established line. I will also post after the games if something pops up that could have been beneficial but was unpredicatable until I got the post game results. I know that sounds funny but that is the way it is. For instance, we were all throwing out some numbers on the Memphis/Ole Miss game. I think I threw out a line something like Ole Miss -17 and someone else thought it might be closer to 7 or so. The rating I am talking about shows Ole Miss 25.5 pts
better than Memphis. Do we go with that right off the bat? Do we let it go find out if it is right or wrong and see what we find in week 2. If we wait the numbers will tell me if I was right on one or both teams. Maybe the number for Ole Miss is good as is and Memphis has to be adjusted, or vice versa, possibly both. But I have other factors in the mix too. I will post five games a week (games I will bet) but I may not put every game I bet on here. One thing you can bank on, if I put a game on here I am betting it. I am also interested to see if someone that has one of my spreadsheets can find something that can help us all. That is why I put the game breakdowns on here. My hope is that these MIT kids will be so good I will never have to do any research ever again.

You say your want to see what the results are from week 1 to see how close you were to the lines, havent you back tested your formula to say last years lines?

 

[russ1945]

You say your want to see what the results are from week 1 to see how close you were to the lines, havent you back tested your formula to say last years lines?

My base numbers are brand new sets of numbers based on points scored last year by players who are returning this year, last years defensive avg ppg, number of returning starters, and a number I plug in for special teams. These numbers cannot be back tested because they are a new base, arbitrary, and may not apply to all teams. However, they may be very relevant, they should be adjustable so as to bring them current by making weekly adjustments based on current statistics as compared to the base numbers (ly's performances). I am afraid that I may have problems with teams where the QB leaves but all the key receivers return and maybe vice versa. If a new QB takes off right away (look how Texas Tech reloads at QB) the numbers may flow right along. I am pretty sure that defensively I account for losing numbers of players but not necessarily the quality of the players although that should be reflected in the avg points allowed per game. After week one I will see what would have happened if I had accounted for All Americans who returned and/or did not return and take a look at the QB/receiver scenarios. In the case say of Memphis/Ole Miss I could see a 25 pt blowout because there are 18 factors involved. Also, the first couple of weeks there are many mismatches (strong vs. weak teams) and that data may only apply to those types of matchups. I am not rushing anything or forcing anything. I am just going to assess the numbers and see if they indicate any kind of predictable performance. This will not be a system. At best it will just make something pop out that might not otherwise have been detected by not comparing the base numbers. But to answer your question you would have to have old rosters, and go over all the statistics for every team for x number of years and account for players who returned and who did not for every one of those years. I do not think that much info is available so I will have to do the trial and error thing. I am not creating a formula I am merely observing numbers to see if they are accurate and can assist in predicting the outcome of certain matchups.