How Number of Returning Starters Affects W/L

Submitted by RakeFight on

I ran across this interesting analysis of how the number of returning starters affects a college football team's number of wins the following season. 

http://collegefootballmatrix.wordpress.com/college-football-rosters-depth-charts-2011/4966-2/think-you-have-returning-starters-impact-solved-dont-bet-on-it/

Followed by a little more specific/on-topic breakdown of the Big Ten

http://collegefootballmatrix.wordpress.com/college-football-rosters-depth-charts-2011/4966-2/think-you-have-returning-starters-impact-solved-dont-bet-on-it/big-ten-returning-starters-in-the-matrix-2012/

I was surprised to see that there's not much difference in the middle numbers... ie: it doesn't seem to make much difference if you return anywhere from 4-9 starters on either side of the ball.    Discuss.

Croatian_Blue

March 27th, 2012 at 10:51 AM ^

I don't understand how they only have 5 returning starters on offense for us.

Probable starters this year that started last year are Denard, Fitz, Lewan, Omameh, Barnum, Schofield, Gallon, and Roundtree.  That's 8, maybe 7 if you don't count Barnum since he was injured most of last year, but still more than 5.

coastal blue

March 27th, 2012 at 11:24 AM ^

Also on defense they have us at 6, but we only - obviously more significant than that - lost Martin, Van Bergen, Heininger and Woolfolk, unless I'm missing something. 

The big thing for Michigan is that we don't have a unit (i.e. the secondary in 2010) that is trotting out 5-6 freshmen/RS freshmen in the two deep with the only real experience coming from a couple of sophomores and a senior who was playing wide receiver the season before. The guys who are stepping in as starters this year have at least played and/or been in the system a few years. 

AFWolverine

March 27th, 2012 at 10:58 AM ^

I was using the handy-dandy MGoSearch for that first link you posted. I guess I've been a member of this site long enough to be considered an upperMGoclassman now? The search function works in my head too!

Edit: I guess not. Only a sophomore. *sigh...

justingoblue

March 27th, 2012 at 11:03 AM ^

By the way, this isn't off topic.

 

ONTOPIC

  • Anything Michigan sports related
  • Anything related to other Big Ten teams or upcoming opponents
  • Stuff about the blog itself
  • University of Michigan topics that don't relate to sports

GRAY AREA

  • College sports in general
  • Ann Arbor

UNWELCOME, KTHX

  • Politics

OFFTOPIC

  • Pro sports of any variety
  • Everything else

mark5750

March 27th, 2012 at 12:42 PM ^

Anything that has to do with using numbers of returning starters or particular position groups like kickers or QBs is using such incomplete information that it seems insignificant at best.  At face value it seems that how long a team has been in a particular system would make a bigger difference than returning starters simply due to the logic of reloading with players that know the system would have less impact than returning a starter that has to learn a new system.  To stretch reality for the sake of a point it seems like replacing a senior walk-on like Will Heininger with a sophomore 5* like Ondre Pipkins (I know they play different interior line positions and OP is an incoming freshman) would be far more desirable than having a returning starter that is essentially playing as a stop gap due to thin numbers.  Also, things like strength of schedule and number of wins the previous year will have a large impact.  Michigan for example would either have to have no less than one loss to beat the previous year while we could go 11-3 with a B1G championship and a Rose Bowl win and no more wins and one more loss.  Just thoughts on the process.

RakeFight

March 27th, 2012 at 1:56 PM ^

I completely agree.  There are any number of variables in this analysis that are not controlled for and that may influence a season's outcome more than returning starters.  Coaching changes?  System changes?  How long did the returning starter actually start?  Strength of schedule?  Etc.

But, I would argue that it's still a valid analysis simply because we as fans often do not consider any of those other variables when we make statements like, "Team X is going to be great this year because they have returning Y starters." 

mark5750

March 27th, 2012 at 2:57 PM ^

I agree with the simple analysis when it is applied more specifically.  For example, I dont remember the exact amount but historically speaking when you have an offensive line returning a high number of total starts (ie. the five starters have a combined 75 starts) that will be a strong indicator for offensive success.  Yet another example that I was thinking of and didnt include because my rant was getting excessive was the thought that Penn State fans would probably rather have a true freshman blue chipper this year like Gunner Kiel than the incumbant Matt McGloin, so quality of returners has to be considered even in a simple analysis of returners.

maizenblue87

March 27th, 2012 at 7:48 PM ^

I put much more faith in the competency of the coaching staff, rather than number of starters tetirning. As was proven last year, great coaching can practically make a "silk purse out of a sow's ear."