Finding good football statistics
I need some help finding stats that are better than simple raw yardage totals as we all know how flawed those are. What are some good suggestions. I want tempo free type stats, efficiency and all that. I know of Kenpom for basketball. There has to be something awesome for football that I don't know about yet.
April 22nd, 2011 at 12:04 AM ^
at www.footballoutsiders.com is probably what you're looking for
April 22nd, 2011 at 12:08 AM ^
for some reason the site won't work right now, but that's the one i was thinking of. thanks
April 22nd, 2011 at 12:10 AM ^
right now either. But search the site here if your looking for UM stuff. Brian posts enough about it you should be able to find some stuff to keep you up late tonight
April 22nd, 2011 at 12:20 AM ^
reading FanGraphs pretty hard for baseball lately. I'm still amazed that fans look at stats like batting average, RBIs, slugging percentage and crappy stats like those.
April 22nd, 2011 at 10:07 AM ^
I've been meaning to email Brian about this but, does anyone know how FEI is truly calculated. The most information that I could find was on this site: http://bcftoys.blogspot.com/2006/10/efficiency-in-college-football.html
But, that still leaves me with some questions.
For FEI Game efficiency=((Points for-Points against)/7)/(Total competitive possessions/2)
But, we don't know how "Total competitive possessions" is calculated. For example, if a team is down 14 points with 30 seconds left in the first half, that's still a competitive possession (or least if they're at the 50 or closer to the end-zone it is). But, if a team is down 14 with seconds left from their own 5 that probably isn't a competitive possession. So, at the very least time remaining, field position, and score must be used to calculate total competitive possessions. But I can't find an explanation for how they do it.
Additionally, the site says that game efficiency is adjusted for each opponent. But I cannot find any explanation of how this adjustment is made. Is it down purely on other teams' FEI. So are pre-season polls used to "seed" the data and then their effect is slowly removed until at the end of the season pre-season polls aren't used in the calculation at all? Or is it down some completely different way.
I would really appreciate it if someone could help me find these answers. FEI says that our offense was amazing last year, but since I don't know how the stat is calculated I don't know how reliable that ranking actually is.
April 22nd, 2011 at 12:10 AM ^
April 22nd, 2011 at 12:22 AM ^
yeah, football is the hardest to figure by far
April 22nd, 2011 at 12:38 AM ^
so it's hard to strip it down to evaluate individual peformance since the player are dependent on other player's performance.
Baseball is easy because it's an individual battle within a team sport. Basketball is somewhat the same way but you can make conclusion based on individual performance.
http://web1.ncaa.org/mfb/natlRank.jsp?year=2010&div=IA&site=org
^^^ Last year's stats they will have a new page for this coming season
I personally like http://cfbdatawarehouse.com/
But w/l is not what he looking for.
I love that site for quick references for smack talking. My line of work I run into all fan bases and that site is gold (gold Jerry, gold).
I hear “S-E-C, S-E-C”. I say “Sorry UM is 20-6-1 vs ESSS-EEE-SEEE.”
LETS ROCK!
April 22nd, 2011 at 10:27 AM ^
They can be very deceiving especially based on your competition level. Seeing as how you obviously took a shot at last year's team in your opening sentence. Watch the games and you'll find out all you need to know. We played seven of the top 30 Ds in the country last year. If you're looking into insight on the new staff they played 6 of the 20 worst Ds and a 4-7 FCS school.
Everyone loves to point out SDSU game with TCU last year a 40-35 loss. SDSU actually jumped out to a 14-0 on a flea flicker and a D TD on a fumble recovery in the end zone. TCU scored TDs on five of their next six possessions while SDSU didn't get a single first down. SDSU scored 3 TDs and racked up 199 of their 300 total yards in the final 16 minutes. SDSU had 21 carries for 38 yards and threw he ball 26 times completing 11 for 262 yards. The four TDs scored by the O came on or after long passes- 33 and 35 yard TDs and one yard runs after 49 and 50 yard completions. Not one sustained drive.
http://espn.go.com/ncf/recap?gameId=303172628
Against Utah, SDSU threw the ball 54 times while rushing 21 times despite having a 17 point lead in the second quarter. Against Missouri threw 44 times ran it 33 times. Hillman had 23 carries for 228 with 168 on two runs. Against BYU 12 rushes and 36 passes. Don't see much smashmouth football in these games. Of course Borges' preferred O is the West Coast Offense which is more finesse than "manball".
In SDSU's two wins against teams with winning records, Air Force and Navy (one dimensional offensive teams) they were split almost 50/50 with a slight advantage to running the ball. The other seven games I won't even look at because the combined records of the opponents - 46 games under 500.
Borges has said Denard's rushing totals will drop this year to around 1,000 yards or roughly 80+ per game. He only had three such games last year (MSU, Ill, and Purdue) and he threw 7 of his 10 interceptions in those games.
April 22nd, 2011 at 12:32 PM ^