Statistics tell the whole story. Independently verified?

[Obsessed]

In conclusion, based on the formula... UT should win ___ games. Anyone?

9

I ain't scared. Ill say it.

 

[daj2576]

I'd like to see the numbers run on returning experience at qb. This is just as big of a predictor as coaching is IMO. You could conclude that they are correlated but not necessarily causative

Ask and ye shall receive. Dave Bartoo (cfbmatrix.com) sent me this link yesterday after he apparently read your question in this thread. I think the numbers will surprise you.

College Football Returning Starters Kickers and QB 2013

 

[daj2576]

This isn't "theoretical physics" this is football. Iv'e been watching football for almost 50 years now and have seen countless games where the team that was supposed to win lost and the team that was supposed to lose won. Football is played on the field not in calculations. There probably is something to these theories which is all they are but there are too many variables in the human equation to place stock or even bet on it. Here is my take. Watch a few games being sure to have some Jack Daniels!! Enjoy the games and save physics for the classroom. I kinda like the chaos theories myself. :hi:

I do not doubt that in 50 years you have seen plenty of upsets. There are a few key questions:

1) what is an upset? Is it when some talking head tells you that one team is better without any objective data to conclude that? Is it your idea of who would be a better team, is it Vegas's idea of who should win? Those would be key questions in determining if what you witnessed was actually an upset or a perceived upset. If you haven't noticed, this (and similar systems) seem to shake up the idea of perception vs. reality.

2) In your fifty years, I am certain that you have also seen many games where the predicted winner, won. I am also certain that those games don't make much of an impact on our memory. Probably because they are far less exciting and less memorable. In fact, the vast majority of college football games are played between two teams who are both subjectively and objectively in different tiers, thus predicting the outcome before the game is even played is very simple. That is, in fact, why Vegas uses spreads to gamble on a game, because simply predicting who wins or loses is far easier than most give credit for.

 

[daj2576]

9

I ain't scared. Ill say it.

I believe the odds go something like this:
10% 6-6 or below; 60% 7-5; 20% 8-4; 10% 9 wins or more (not including bowl game).

 

[McDad]

I believe the odds go something like this:
10% 6-6 or below; 60% 7-5; 20% 8-4; 10% 9 wins or more (not including bowl game).

Your bell curve looks a bit lopsided.

 

[daj2576]

Your bell curve looks a bit lopsided.

lol Clever.

A bell curve would place the probable outcomes as impossibilities. You can't win 7.4 games. so I tried to force the rough probability into real numbers. This isn't scientific, just a gut feeling based on having reviewed this data for so long.

 

[McDad]

lol Clever.

A bell curve would place the probable outcomes as impossibilities. You can't win 7.4 games. so I tried to force the rough probability into real numbers. This isn't scientific, just a gut feeling based on having reviewed this data for so long.

Daj, how would your preseason ranking look when factoring in the recruiting rankings from your matrix and the +/- of the coach's effect on talent?

 

[daj2576]

Daj, how would your preseason ranking look when factoring in the recruiting rankings from your matrix and the +/- of the coach's effect on talent?

That is an interesting question, actually. I would like to look at that. However, I doubt I will have time. I am currently covered up with running a comprehensive set of evaluations for my employer. I doubt I will have any time to advance my system much before the start of the season. By then, I will undoubtedly be ready to just watch football as I have been buried in numbers all summer.

By the way, you can call me Derek. I think sometimes we (generally, not specifically) tend to lose some civility because of the anonymity on message boards.

 

[McDad]

That is an interesting question, actually. I would like to look at that. However, I doubt I will have time. I am currently covered up with running a comprehensive set of evaluations for my employer. I doubt I will have any time to advance my system much before the start of the season. By then, I will undoubtedly be ready to just watch football as I have been buried in numbers all summer.

By the way, you can call me Derek. I think sometimes we (generally, not specifically) tend to lose some civility because of the anonymity on message boards.

You're right. It is unfair of me to ask that considering this is just a hobby. Maybe the websites you have previously mentioned will put together a preseason ranking.

Thanks, Derek. You can me, His Most Magnificent and Magnanimous McDad.

 

[Obsessed]

I believe the odds go something like this:
10% 6-6 or below; 60% 7-5; 20% 8-4; 10% 9 wins or more (not including bowl game).

If I remember correctly 7-5 was on talent alone correct? Then with the coaching effect CBJ should be worth 2 more wins correct?

( I understand this is just numbers)

Ill just split the differences and go with 8-4

 

[drodvol]

I do not doubt that in 50 years you have seen plenty of upsets. There are a few key questions:

1) what is an upset? Is it when some talking head tells you that one team is better without any objective data to conclude that? Is it your idea of who would be a better team, is it Vegas's idea of who should win? Those would be key questions in determining if what you witnessed was actually an upset or a perceived upset. If you haven't noticed, this (and similar systems) seem to shake up the idea of perception vs. reality.

2) In your fifty years, I am certain that you have also seen many games where the predicted winner, won. I am also certain that those games don't make much of an impact on our memory. Probably because they are far less exciting and less memorable. In fact, the vast majority of college football games are played between two teams who are both subjectively and objectively in different tiers, thus predicting the outcome before the game is even played is very simple. That is, in fact, why Vegas uses spreads to gamble on a game, because simply predicting who wins or loses is far easier than most give credit for.

You remind me a'lot of Sheldon Cooper from the big bang theory. Everything doesn't have to be about non objectivity theory. Can't we just sit down throw some beers back, shoot some jack and enjoy the games without theorizing everything to death? your obviously a smart guy but you need to relax a little bit.eace2:

 

[daj2576]

You remind me a'lot of Sheldon Cooper from the big bang theory. Everything doesn't have to be about non objectivity theory. Can't we just sit down throw some beers back, shoot some jack and enjoy the games without theorizing everything to death? your obviously a smart guy but you need to relax a little bit.eace2:

Fair enough.
I admit that football is more fun when you don't know the outcome before hand.

And yes, I will take a Jack and Coke, or a white russian (because the dude abides).

 

[daj2576]

If I remember correctly 7-5 was on talent alone correct? Then with the coaching effect CBJ should be worth 2 more wins correct?

( I understand this is just numbers)

Ill just split the differences and go with 8-4

The reason I am hesitant to apply Jones' coach effect in year one is that he was +0 his first year at Cincy. Maybe that indicates that the first year is all about installation?

 

[daj2576]

You're right. It is unfair of me to ask that considering this is just a hobby. Maybe the websites you have previously mentioned will put together a preseason ranking.

Thanks, Derek. You can me, His Most Magnificent and Magnanimous McDad.

I am into brevity, so I will stick with McDad.

And, I wasn't suggesting that you were asking too much. I was simply explaining that as much as that is a good idea and I would love to see that, I just don't have the time.

 

[Obsessed]

The reason I am hesitant to apply Jones' coach effect in year one is that he was +0 his first year at Cincy. Maybe that indicates that the first year is all about installation?

Hmmmm. Good point.

Thanks again Derek

 

[Obsessed]

The reason I am hesitant to apply Jones' coach effect in year one is that he was +0 his first year at Cincy. Maybe that indicates that the first year is all about installation?

Double post.

 

[McDad]

I am into brevity, so I will stick with McDad.

And, I wasn't suggesting that you were asking too much. I was simply explaining that as much as that is a good idea and I would love to see that, I just don't have the time.

Didn't take it that way. I would like to see it, too. Preseason rankings the way they are currently done is a joke. With the matrix at 70% and other's technique at 80%, no doubt that a valid preseason rank could be put together.

 

[Rifleman]

The reason I am hesitant to apply Jones' coach effect in year one is that he was +0 his first year at Cincy. Maybe that indicates that the first year is all about installation?

That is probably wise.

I think the +0 is because Kelly left Butch with players unsuited for Butch's particular flavor of the spread.

Kelly liked to send 4 receivers vertical on seemingly every play at Cincy and let them adjust their own routes.

Butch demands precise route running and is more a 4 curls, controlled passing guy.

Our receivers don't know anything, so they don't have to unlearn anything. Not sure if that is good or bad. I think Pig Howard is going to be in hog heaven in this offense, though!

 

[sjt18]

I was simply trying to make two points to the man for whom "the bell has wrung".

1) A statistical anylysis can easily be scewed.

If the data is skewed then it ceases to be "statistical" in the strictest sense, right? I completely agree that someone good with numbers can dazzle and amaze the unwary. That does not seem to be the objective of the OP or those involved in this kind of modeling.

2) Just because you present numbers behind your prediction does not make them inheriently more accurate then the person who lists reasons.

That's true but those things are not mutually exclusive. Those "reasons" either are or could be quantified almost all the time.

That is really all. As for your points in my post they are all correct. I hope this cleared it up a bit.

Yep.

I guess I have a bit of a kneejerk reaction to people who go off of emotion and whim... not necessarily in football but in business. I do not need every jot and tittle to be known... but I don't want to shoot in the dark either.

 

[sjt18]

Fair enough.
I admit that football is more fun when you don't know the outcome before hand.

And yes, I will take a Jack and Coke, or a white russian (because the dude abides).

You never know the outcome before hand with 100% accuracy. Because many factors are knowable, we can get close. Because some factors are remote and rare, we cannot know for sure.

 

[daj2576]

You never know the outcome before hand with 100% accuracy. Because many factors are knowable, we can get close. Because some factors are remote and rare, we cannot know for sure.

What I was trying to communicate, and you beautifully point out, is that on any given Saturday I can watch several football games and generally know the majority of the results before the games are even played.

I am not now, nor have I ever, tried to communicate that I ALWAYS know how these things turn out. This certainly makes me wish I had never found these evil numbers. I mean that I like not knowing, more than I like knowing.

The curse of knowledge is real. lol

 

[daj2576]

Didn't take it that way. I would like to see it, too. Preseason rankings the way they are currently done is a joke. With the matrix at 70% and other's technique at 80%, no doubt that a valid preseason rank could be put together.

Sometimes when I am busy with work and check in here for a minute or two, I find that I write with too much force. I never intended to communicate anything negative, in fact the opposite. I am simply trying to convey that it is "all good" that I am flattered that you trust me enough to not only provide ideas on how to sharpen this "system" but to ask me to provide you with more data. I wish I had the time, seriously.

 

[Rifleman]

Just because an event only has two possible outcomes doesn't mean that the likelihood of each outcome is 1/1.

Just because likelihood is variable doesn't mean the number of observable factors leading to a binary outcome is infinite.

AND, this is key: even if the number of variables leading to a binary outcome approaches infinity--which one could imagine in field like astrophysics but certainly not in a prescribed game like football--there will always be a smaller subset of factors involved which have strong correlation to one outcome or the other.

This is why an astrophysicist can know that a large planet, which that cannot be seen via telescope, must orbit a star light years away because of a tell-tale wobble in the rotation of said star. Gravitational wobbles correlate highly to gravitationally significant objects being nearby.

Obviously, the number of factors involved in putting that planetary mass into orbit around that specific star is beyond incalculable. Still, only the key factors which indicate an orbiting planet need to be evaluated in order to predict with a great deal of certainty that an unseen planet is there. If math can do that, why is it shocking to some that it can be useful in sports both in prediction of outcomes and game planning.

All pretty elementary stuff I learned in the Tennessee Public School System. Suck it, Bammers!

 

[daj2576]

Just because likelihood is variable doesn't mean the number of observable factors leading to a binary outcome is infinite.

AND, this is key: even if the number of variables leading to a binary outcome approaches infinity--which one could imagine in field like astrophysics but certainly not in a prescribed game like football--there will always be a smaller subset of factors involved which have strong correlation to one outcome or the other.

This is why an astrophysicist can know that a large planet, which that cannot be seen via telescope, must orbit a star light years away because of a tell-tale wobble in the rotation of said star. Gravitational wobbles correlate highly to gravitationally significant objects being nearby.

Obviously, the number of factors involved in putting that planetary mass into orbit around that specific star is beyond incalculable. Still, only the key factors which indicate an orbiting planet need to be evaluated in order to predict with a great deal of certainty that an unseen planet is there. If math can do that, why is it shocking to some that it can be useful in sports both in prediction of outcomes and game planning.

All pretty elementary stuff I learned in the Tennessee Public School System. Suck it, Bammers!

And they say 11Bravos are dumb.

 

[Rifleman]

And they say 11Bravos are dumb.

HA! My grandpa took a German sniper round a week after dropping into Normandy, came back with a steel plate in his head, and was still sharper than any Bama fan I've ever met.

Also, he really helped TV reception in those days before cable!!!