Prep Time 2012

Fine. I don't care about that study. You posted it with its claims. I repeated it.

Now, fwiw, I took your data for extra time and non-extra time GT rushing.

I then added an extra column for that team's yds/rush allowed against BCS AQ (from cfbstats). I then added a column for the Ratio of GT's actual yds/rush over Opponents yds/rush average against BCS AQ.

I then calculated the average over the 4 years.

For Teams without extra prep
Average of GTAct/OppDef Ave = 1.33
StDev = .31

For Teams with extra prep
Average of GTAct/OppDef Ave = 1.33
StDev = .31


No difference.

Hopefully this works:

https://docs.google.com/spreadsheet/ccc?key=0AgBtvZUVCL--dEFSbU9TWkwyWHBWRGxYVUZqZ3FnalE#gid=0

Oh well, I'll worry about making the spread sheet public some other time.
You just have to hit 'file' and then 'publish to the web' which will give you a link that you can post. If you want to pm me a copy paste of the numbers, I can just put those in my spreadsheet as well.

Pretty much the same ratio was in my spreadsheet as well. I had 1.38 vs. 1.37 using defenses season average including everything not just BCS. Yours is better but result obviously about the same.

The killing happens when you look at yards per play not just yards per rush. In other words, it's seems the passing is hurting us. It's definitely very interesting.
 
good start sam, could you share list of games and the W-L records? Google docs is free.

btw, those numbers look very low. it seems like pretty good bet to bet against those teams.

Yes it does seem very low, I was surprised, and might start gambling! I'll see if I can get a cleaned up excel sheet. Getting this was bitch because I had to look at all the ACC games for the last 4 years. I have a site visit today so it may not be ready until tomorrow.
 
good start sam, could you share list of games and the W-L records? Google docs is free.

btw, those numbers look very low. it seems like pretty good bet to bet against those teams.

I think this also shows most teams, against the spread anyways, have a harder time when other teams have time to prepare. Hell, we do pretty good compared to the rest of the ACC.
 
I think this also shows most teams, against the spread anyways, have a harder time when other teams have time to prepare. Hell, we do pretty good compared to the rest of the ACC.
Where did you get the vegas spread data? I am curious if there is an easier way than I used to get it. I use SBR and it has some missing information. (http://www.sbrforum.com/college-football/odds-scores/20111119/ I just change the date to the date I want to get spread.)

When I looked at GT's number, I didn't get .44 but I got .375 with one push. I suppose you got .44 because that one game (NC State) wasn't a push for you.

Maybe looking at the game differentials with respect to the spread is a better idea when the sample size is small. That way you don't worry about the pushes and capture information on how well or bad we did with respect to the spread.
 
You just have to hit 'file' and then 'publish to the web' which will give you a link that you can post. If you want to pm me a copy paste of the numbers, I can just put those in my spreadsheet as well.

Pretty much the same ratio was in my spreadsheet as well. I had 1.38 vs. 1.37 using defenses season average including everything not just BCS. Yours is better but result obviously about the same.

The killing happens when you look at yards per play not just yards per rush. In other words, it's seems the passing is hurting us. It's definitely very interesting.

Wow. I don't know why I didn't see those numbers at the end of your chart; I guess I didn't scroll over enough.

Regardless, the difference between the passing yards/play ratio from extra-prep to prep is .33 while the standard deviation for each number is about .6.

I think that still shows that extra-prep isn't a major factor.

By looking at simply yards/play, I think you are errantly adding the rush sample-size to the yardage disparity from the passing game.
 
By looking at simply yards/play, I think you are errantly adding the rush sample-size to the yardage disparity from the passing game.
It uses total yards and total number of plays to get yards per play. Are you saying yards per play is not a good indicator of an offense?

I get using the ratio approach to normalize may not be the best approach since offensive success may not grow linearly with YPP, but it sounds like you have a problem with YPP itself.
 
It uses total yards and total number of plays to get yards per play. Are you saying yards per play is not a good indicator of an offense?

I get using the ratio approach to normalize may not be the best approach since offensive success may not grow linearly with YPP, but it sounds like you have a problem with YPP itself.

What? I don't understand how you draw some universal conclusion that I'm saying ypp is not a good indicator of an offense when I'm saying that it doesn't serve the purpose to which you are applying it.

Let me be clear. W-L in an indicator of relative team strength. However, I don't think one should reasonably use it as an indicator of defensive preparedness for the offense since the team's defense also falls in the mix.

YPP is an indicator of relative team explosiveness in comparison to other teams.

However, when you are trying to measure how an offense does against a defense, I think it is better to recognize that both offenses and defenses may have differing strengths with respect to the run and the pass.

Since, I used the ratio of GT-actual/D-ave myself, I have no problem with that metric. However, neither with respect to the run or the pass does the data show that extra-prep gives an advantage outside the standard deviation.
 
What? I don't understand how you draw some universal conclusion that I'm saying ypp is not a good indicator of an offense when I'm saying that it doesn't serve the purpose to which you are applying it.

Let me be clear. W-L in an indicator of relative team strength. However, I don't think one should reasonably use it as an indicator of defensive preparedness for the offense since the team's defense also falls in the mix.

YPP is an indicator of relative team explosiveness in comparison to other teams.

However, when you are trying to measure how an offense does against a defense, I think it is better to recognize that both offenses and defenses may have differing strengths with respect to the run and the pass.

Since, I used the ratio of GT-actual/D-ave myself, I have no problem with that metric. However, neither with respect to the run or the pass does the data show that extra-prep gives an advantage outside the standard deviation.
I think YPP indicates more than explosiveness. I think it indicates the success of the offense. As far as explosiveness, I would probably use standard deviation of the play yardages as a good indicator.

68% of observations from a normal distribution fall within 1 standard deviation of the mean, so I disagree on measuring an effect based on whether they are beyond a standard deviation or not.
 
So according to you, every firing of every head coach is unjust.

Lol, are you really this dumb?

No, I just don't believe that W/L record is the "most . . . significant statistic in all of football," which is what you said.

If it was, undefeated teams would automatically win the national championship every year, regardless of who they played. The Heisman would go to a player on that undefeated team, regardless of whether someone else has better stats. You could predict who would win a game simply by looking at what team had the better record.

Winning and losing may be important to the bottom line, but it doesn't come close to being the most significant statistic. If it you were right, LSU would be the champion, not Bama (since they had won the first game).

The point I was making, which you apparently couldn't comprehend, is that all football statistics are relatively meaningless because there are too many variables. A passer rating reflects not only how good the QB is, but also playcalling, offensive scheme, offensive line, WRs, quality of opponent, opponent's scheme, weather, score and game situation (does the team need to air it out to come back, is it ahead by a lot and throwing short, safe passes?), etc. With 16 games a season, something as simple as a game in heavy wind or in the snow can have a big effect on a season's stats.

Compare it to sports with meaningful statistics - in basketball a shooting percentage is largely determined by the shooter and the person guarding him. A batter's average is largely determined by the batter's and pitcher's skill. Yes, there are variables there too like fielding and weather, etc. But those balance out over 162 games and 600 ABs. That's TEN TIMES the number of football games. That's why there are advanced statistics in baseball, like BABIP, xFIP, wOBA, wRC+, WAR, but not in football.
 
FWIW the total offense stats were already in there, so I went ahead and got ratios for those as well.

Calculated a ratio for each game: (yards GT gained on a defense)/(yards that defense gave per game).

Extra-time mean ratio is 1.08 with standard deviation of 0.26.
No Extra-time mean ratio is 1.21 with standard deviation of 0.27.

It reflects what we saw in yards per play.
 
Lol, are you really this dumb?

No, I just don't believe that W/L record is the "most . . . significant statistic in all of football," which is what you said.

It's what gets coaches hired and fired, and it's how the national championship is decided. Nobody bats an eyelash at using four years worth of win/loss record to determine how to spend tens of millions of dollars. Yet here you are saying it's statistically meaningless.
 
It's what gets coaches hired and fired, and it's how the national championship is decided. Nobody bats an eyelash at using four years worth of win/loss record to determine how to spend tens of millions of dollars. Yet here you are saying it's statistically meaningless.

It's certainly not how the national championship is decided. If it were, they would have given the ---- thing to LSU after the regular season rather than have them replay a team they already beat.

Again, I'm not saying its meaningless for certain considerations and you are a whore-mothered mentally-challenged ****tard for even trying to suggest that's what I'm saying. Wait until you have the reading comprehension of a 2nd grader, re-read my posts, figure out what the hell I'm talking about, and then get back to me. Before then, spare us all the pain and suffering of reading through your dribbling posts and refrain from posting.
 
and you are a whore-mothered mentally-challenged ****tard for even trying to suggest that's what I'm saying. Wait until you have the reading comprehension of a 2nd grader, re-read my posts, figure out what the hell I'm talking about, and then get back to me. Before then, spare us all the pain and suffering of reading through your dribbling posts and refrain from posting.

gtphd, you have been outdone.

/sig
 
FWIW the total offense stats were already in there, so I went ahead and got ratios for those as well.

Calculated a ratio for each game: (yards GT gained on a defense)/(yards that defense gave per game).

Extra-time mean ratio is 1.08 with standard deviation of 0.26.
No Extra-time mean ratio is 1.21 with standard deviation of 0.27.

It reflects what we saw in yards per play.

That's a pretty big difference, imo. Wonder what its like for the defense?
 
It's certainly not how the national championship is decided. If it were, they would have given the ---- thing to LSU after the regular season rather than have them replay a team they already beat.

Again, I'm not saying its meaningless for certain considerations and you are a whore-mothered mentally-challenged ****tard for even trying to suggest that's what I'm saying. Wait until you have the reading comprehension of a 2nd grader, re-read my posts, figure out what the hell I'm talking about, and then get back to me. Before then, spare us all the pain and suffering of reading through your dribbling posts and refrain from posting.

I know exactly what you're saying.

You're saying that the sort of statistical standards people use on FDA drug tests to make sure folks don't have heart attacks from the Purple Pill don't apply to college football win loss records. And I agree. But I don't give a shit, because by that rationale Chan Gailey could have been as good a coach as Nick Saban, mathematically, and nobody can prove other wise.

And that's dumb.
 
FWIW the total offense stats were already in there, so I went ahead and got ratios for those as well.

Calculated a ratio for each game: (yards GT gained on a defense)/(yards that defense gave per game).

Extra-time mean ratio is 1.08 with standard deviation of 0.26.
No Extra-time mean ratio is 1.21 with standard deviation of 0.27.

It reflects what we saw in yards per play.

I would be interested to know you reaction to your ypc rush, ypa pass, and YPP ratios if you exclude 2009.
 
I would be interested to know you reaction to your ypc rush, ypa pass, and YPP ratios if you exclude 2009.
Our YPP ratio average for 2010 and 2011 is about the same as to the YPP ratio average when teams have extra time for us.

So when teams have extra time, we can expect the offense to be as crappy as they were in 2010 and 2011 against BCS teams on average.

Thanks for making my point in another way...
 
Back
Top