Recruiting Breakdown - the objective viewpoint

#51
#51
I don't guess OP realizes that people leave early for the draft or people get kicked out of school..I like your optimism but your theory is flawed..According to your theory bray,hunter,patterson will be in orange.
 
#52
#52
Gruuden said:
I don't guess OP realizes that people leave early for the draft or people get kicked out of school..I like your optimism but your theory is flawed..According to your theory bray,hunter,patterson will be in orange.
Click to expand...


While you are theoretically correct, the theory also assumes that LSU has its 9 juniors no returning included in the analysis, the Georgia UBS who are leaving are included in the analysis, the Bama guys leaving are included, etc.

In other words, everyone has players leaving and therefore all are included in that analysis. It's not as though only Bray, Hunter and Patterson are leaving.

Short of a time consuming team by team analysis, it's about as fair as one can get.
 
  • Like
Reactions: 2 people
#53
#53
Gruuden said:
I don't guess OP realizes that people leave early for the draft or people get kicked out of school..I like your optimism but your theory is flawed..According to your theory bray,hunter,patterson will be in orange.
Click to expand...

What I have said repeatedly is that doesn't matter for this evaluation. I know it is hard to imagine that without accounting for specific attrition that a model can be so successful at prediction. It is. What you aren't realizing is that when reviewing 120plus teams, attrition of the best talent is pretty similar among the teams who have talent capable of leaving. So, class rankings, in general, follow the same arc over time. Sure it is flawed. Anything that is not correct 100% of the time has flaws.
 
#54
#54
WOW Daj Good info . after almost 2 years , I see some of what I came here to find . Good analyses and Information . Fine work , sent you a friend request based off this OP . Thanks . LOL and No Homo .
 
#55
#55
Minus Bonner,Taylor,Phillips,Sentimore,and Bourque, last years class is likely in the 30's or worse. Should be taken into account.IMO
 
#58
#58
For an already thin roster, I tend to agree.

Seems an unstable program wiil suffer more than usual attrition, skewing the results.

Still enjoyed the analysis.
 
Last edited:
#59
#59
mtowntim said:
For an already thin roster, I tend to agree.

Seems an unstable program wiil suffer more than usual attrition, skewing the results.

Still enjoyed the analysis.
Click to expand...

The thing with averaging 4 years of classes together, and then comparing those averages to other schools is that it takes huge swings in numbers for a single year for it to really effect the predicted outcome by more than 1 or 2 games. I actually spent some time last year trying to determine how bad the attrition from the 2009th class could have hurt UT, and even if it dropped to 99th, it would have only accounted for enough loss of talent (numerically) to explain losing to South Carolina but kept us well above Miss. State. Now, there is an argument that unusual attrition actually has more latent effects because the lack of depth means that the 1s scrimmaging the 2s will be such a large disparity in talent as to actually hurt the 1s. With any statistical comparison, there are always anomalies. Perhaps UT is one. I have been looking at this since September, and have compared Dooley's performance at LaTech to that at UT and it seems the common denominator is that Dooley under performs with talent, not that talent was Dooley's problem.
 
  • Like
Reactions: 2 people
#60
#60
Love the info Daj, gives me hope that Butch can turn this program around quicker than expected. It's not a perfect predictor but if that info is correct 70% of the time it's a lot more reliable than anything else out there.
 
#61
#61
jortfreeTN said:
Love the info Daj, gives me hope that Butch can turn this program around quicker than expected. It's not a perfect predictor but if that info is correct 70% of the time it's a lot more reliable than anything else out there.
Click to expand...

It is correct roughly 70% of the time, within a deviation of 1, over a long-term time period.

For a single season, in the SEC, the numbers fluctuated a bit more than expected last year, but were still strong indicators. Here is 2012.


Team Predicted Actual
Alabama 8-0 7-1
Florida 8-0 7-1
LSU 6-2 6-2
Georgia 7-1 7-1
Auburn 5-3 0-8
Tennessee 5-3 1-7
South Car. 4-4 6-2
Texas A&M 4-4 6-2
Ole Miss 2-6 3-5
Arkansas 2-6 2-6
MSU 1-7 4-4
Missouri 2-6 2-6
Kentucky 1-7 0-8
Vanderbilt 0-8 5-3

8 of 14 (almost 60%) were within a deviation of 1 from their predicted outcome, with 4 of those teams performing exactly as predicted.

If you do this over long periods of time, you begin to see the fluctuations flatten out and that talent becomes an incredible predictor.
 
Last edited:
#64
#64
Good stuff OP, I still think we are going to surprise some teams and coaches this year! I believe we will see a huge difference between a CBJ's team and a CDD's team period..
 
#65
#65
Great thread.

I would stop trying to explain averages to people though. Not meaning any disrespect to them, but a lot of people really cannot grasp the "true" concept of averages over a period of time.
 
  • Like
Reactions: 1 person
#67
#67
OrebankVol said:
Good stuff OP, I still think we are going to surprise some teams and coaches this year! I believe we will see a huge difference between a CBJ's team and a CDD's team period..
Click to expand...

I am trying to be pragmatic. His first year at Cincy, he performed exactly as his talent would predict. If that translates the same to UT, we should expect 7-4 with the tossup being Oregon. If however, now that he has a much more talented and deep team than he has ever had, he can over perform by his +2 average, we could be looking at a 8-10 win season.

Bottom line: even if he under performs, I will refuse to judge his first year. After his first full year to recruit, and a first full year to perform we don't see marked improvement, the jury should be charged and sent to deliberate.
 
#68
#68
BaldBiker said:
Great thread.

I would stop trying to explain averages to people though. Not meaning any disrespect to them, but a lot of people really cannot grasp the "true" concept of averages over a period of time.
Click to expand...

I have noticed.

It is really like people don't want to see that there is a level of predictability to a system such as college football. They want to believe that the media's self created "intangibles" and other such nonsense are what matter. Sure, "intangibles" have an effect but that effect appears to be so small as to be considered the exception and not the rule.
 
#69
#69
daj2576 said:
I have noticed.

It is really like people don't want to see that there is a level of predictability to a system such as college football. They want to believe that the media's self created "intangibles" and other such nonsense are what matter. Sure, "intangibles" have an effect but that effect appears to be so small as to be considered the exception and not the rule.
Click to expand...


Yes, and your OP and explanation were very good to anyone who really understands numbers and the law of averages.

The same theory or concept can be applied to many facets of life. Is it full proof? No. But a 70% + or - 1 avg over a period of time is pretty damn good for a fairly simple formula. A lot of people have made lots of money using similar formulas in business.
 
  • Like
Reactions: 1 person
#71
#71
Excellent post - my "heartfelt" thanks for the analysis and additional insights...
 
#72
#72
Daj you may have mentioned this in one of the other threads, but have you applied this data to other coaches that have transitioned to the SEC besides the Doofus?
 
#75
#75
tm3 said:
Daj you may have mentioned this in one of the other threads, but have you applied this data to other coaches that have transitioned to the SEC besides the Doofus?
Click to expand...

I would love to do this with Saban/Miles and others, the problem is using Rivals (my arbitrarily preferred information provider) I can only find data back to 2002. Beyond Petrino and now Sumlin, I don't think that one could do a similar analysis.

I would love to find Saban and Mile's recruiting data before they both went to LSU. It is more complicated than that, obviously, because I need the recruiting data for everyone they played too.
 
New posts
You must log in or register to reply here.

VN Store



Back
Top