[gojetsgomoxies]

given that i'd say most lists predicting most improved and deteroirating teams this year are based on returning starters, how would you weight it.

this may be painfully obvious, but i think i'd do Offense plus Defense with extra weight (in declining order) on QB, Oline and then further down punter. not sure about defensive line.

two more things i'd look at returning starters broken down by junior/senior vs. below and returning tackles.

love any responses/suggestions...

[WorkHorse]

The first thing I want to know about returning starters is are they any good.

Doesn't mean a hill of beans if you have 11 starters back on a defense that was 112th in total defense the year before. Even with some improvement...you still got problems.

I love to see a pretty good past defense where two or three starters have been beaten out.

[gojetsgomoxies]

i agree... i always wonder about the 120th ranked defense having 10 starters back. i shake my head and wonder if that's good (it might be).

i did cursory analysis of ats results of teams starting alot of sophomores and freshmen and it wasn't surprising good (small sample size alert). maybe it means much, much better talent. i said i don't worry about a new starter being a senior, but maybe it means he has little talent.

[suicidekings]

The quality of the returning players is key, but the quality/experience of the coaching staff is almost as important. I would rather see a coaching staff that's all been in place for 3+ years and 5 returning starters than 2-3 new coaches and 8 returning starters. The number of returning players can also be undermined by system changes where the players are asked to play different roles if not an entirely different scheme for a different coach.

Among players, I emphasize offensive linemen and the defensive front 7 quite a bit more than any skill positions. If a team can dominate opponents at the line of scrimmage, they can accomplish more with less talent at other positions. However, if they're weak at the line, the team will be inconsistent and vulnerable regardless of what else they're putting on the field.

[gojetsgomoxies]

I meant to say i checked ATS results of heavy sophomore/freshman starting teams (many more sophomores almost always) and it WAS SURPRISINGLY GOOD..... I had a NOT in my original post... small sample size alert.

suicidekings, thanks. i appreciate your comments

[WorkHorse]

This is a topic Gojets has brought up that needs more discussion. So much emphasis is placed on returning starters at the beginning of the year that sometimes I think its misleading. Experience is important in CFB for sure...but assuming returning experience is always positive can be dangerous for sporting men.

Love to see some more input on this topic.....worthy topic for this forum.

[gojetsgomoxies]

yeah, it's hard to really pin it down.

i never know what to think when an absolutely horrendous D has 9 returning starters or 3 returning starters.....

i'm going to put some work into this. steele, who has the best data, has returners broken down by major position groups.

i do think returning starters are pretty much most analysts' first cut at most improved vs. bigger freefall.

[gojetsgomoxies]

gonna try to post a list of total returning starters on OL and DL for all the teams. or OL, QB and returning tackles (who actually make the tackle, not DT or OT)

[SamDiamond]

So, and take this for what it's worth.

But I met a guy, who considers himself a quant, that is a stats guy who creates power numbers for each program. Here is how he explained what he tried to do (He did not give me his recipe), but he spoke in overviews, and he touched on base what you guys have already posted.

He takes returning starters and multiplies it by last year's performance-- adding in for consideration, the strength of schedule for each program. For example, Louisianna Tech's offensive numbers may be top 5 in the NCAA, but if you put them up against SEC competition, they wouldn't be the same.

So in a nutshell, take returning starters, multiply/add a combination of factors (yards per play, 3rd down % defense, pass efficiency defense, sack % and attempt to quantify the strength of schedule into the equation, and you have a rating for each side of the ball.

[chunk]

Relative team strength assessment is huge, but we all know (or should know) that this is not an exact science. We make an educated guess and go from there. That's why these Kelly criterion backers that claim to have nailed their edge down to the nat's ass crack me up. When it comes to actually handicapping a game and taking a stand, it's impossible.... Total BS.

[gojetsgomoxies]

i found it interesting that one analyst thought returning kickers is very important. i can definitely see a good returning punter as important. fg kicker not as confident it's important. it's important if you have a poor one. FSU vs. miami a few times.... i'll try to google it again.

i am going to see if i can chase down detailed returning starters (steele of course) from past seasons and try to compare to great and terrible ATS records that year.

[gojetsgomoxies]

here's the guy thinks kickers are very important......

http://cfbmatrix.com/returning-start...-kickers-2013/

[gojetsgomoxies]

and here are the very detailed returning starters for 2013. i will sort by conference........ what is a good cheap hosting site so i could post a link to my work?

http://www.philsteele.com/Blogs/2013/JAN13/DBJan21.html

[gojetsgomoxies]

looks like this may be first year phil steele has published returning starters by position groups..... but i should be able to work something up.

[Huckleberry Pig]

gojets, thanks for the hard work!

[gojetsgomoxies]

i did a basic microsoft excel perusal of last year's list.

top 19 returning starter teams 2012 (O + D + K +P simple sum). number of teams really just determined by break in rankings

106-121 ATS...... if i exclude kstate and vandy (complete overachieving teams), ats is 88-114. if i exclude UTSA (new teams seem differnent), it is 81-111.... dangerous once you start excluding but i might look at 2011 ATS to filter further (hopefully that justifies kstate and vandy exclusion).... only a few ats winners. more ats losers and breakeven...... 4 teams without both kickers back were ATS losers. and now that i look at rankings

2012 least returniing starters..... very interesting. alot more extreme ATS records. winners and losers.

28 teams or 13 teams depending on break chosen.

73-86 ATS on smaller break..... 172-163 on larger break (larger meaning more teams).... so last group is 99-77 but not sure there's significance to that.

some huge winners: miami_fl, nw (12-1??), cincy, sjsu (11-2), ark state .... will check 2011 ats records as filter.

some huge losers: iowa, so miss, hawaii, idaho, virginia, cal, nms, air force .... again will check 2011 ats records as filter.

and will also check returning QB and OL. probably as far i will go.

wonder if you could get a list of teams with low number of returning starters and then use first game ATS 2013 (assuming against FBS team) as filter for the rest of season.

[gojetsgomoxies]

yup, kstate and vandy were both big 2011 ats winners. so it perhaps validates their exclusion..... utsa was first year, so i'd exclude and if i had to go one way or the other blindly i would always go LONG a new fbs team from a good football state.

on the 2012 low returning starters who had excellent or terrible ats records......... excellent 2012 ats records were 33-27 in 2011. poor 2012 ats records were 50-52.... zero returning kickers ran towards quite positive ATS......

as mentioned i will try to differentiate by QB, OL and returning tackles...... probably not too soon though.