Sagarin

I just used the wayback machine to look at the Sagarin predictor ratings from early January (games through 1/3). Notice anything?

1. Duke
2. Louisville
3. Florida
4. Indiana
5. Kansas
6. Syracuse
7. Michigan
8. VCU
9. Ohio State
10. Minnesota

Wichita State was 19, Michigan State was 23, Marquette was 54, FGCU was 127.

Marquette was a bad miss, but on the whole these were a better predictor of tournament results than the ratings in March. You'd have done pretty well, filling out a bracket based on these.

Sometime later I'll check to see if this was true in prior years, but in the meantime I'll make a suggestion: is it possible that performance in non-conference games, against the full variety of styles of play that you'll eventually see in the tournament and against teams that you aren't familiar with (nor they with you), is a better predictor than conference play? It's been a while since we've seen M light up a zone like we just saw for a few minutes in the first half against Florida, because except for Northwestern nobody in the B1G plays it.

According to the Sagarin ratings, which predict actual game outcomes:*

IN BB, UM is #1 in the B1G.  In the nation, UM is #3,  IND #4, and Ohio is not in the top 10.*

IN FB, on a neutral field UM would still be favored over Ohio (by 0.15 pts).   Why? UM’s nonconference opponents included the two teams now favored to be in the national title game.  By contrast, Ohio’s “marquee” nonconference game this year supposedly was Cal, which is now not even in the top 70 nationally.  Moreover, Ohio’s  other wins were often very narrow.  In fact, even for the UM game, they were only +2 pts after subtracting 3pts for home field (not even considering the injury to UM’s starting QB). 

What then should we make of Ohio’s claim that they could win the AP national title?  That claim should be laughable to any educated voter.  In fact, Ohio is not even ranked in the top 20 nationally by Sagarin (they are #24, whereas UM is #22).  Also, Ohio will not be tested in a competitive bowl or conference championship.  Indeed, in the latter, on a neutral field, they would be  favored by less than half a point vs WISC, 2.5 pts vs NW and they would be underdogs to both NEB and UM.

Interestingly, ND's delusions of grandeur also should be tempered.  Although human pollsters will no doubt put them #1, would ND actually be favored to beat all the other teams according unbiased Sagarin PREDICTOR ratings?  No, not at this point. Why?  ND had many narrow wins, even over marginal teams and teams expected to be huge challenges--like USC--turned out not to be so great.  Thus, by Sagarin's ratings, while ALA is #1 and Oregon #2, ND is only #3  Both ALA and OR both would be favored over ND by large margins on a neutral field.  In addition, unlike ALA but like Ohio, ND will not be tested in a conference championship game.

*I report only those ratings that predict actual game outcomes.  For BB I take the average of ELO and PREDICTOR ratings.   For FB, I report only PREDICTOR ratings (not the ELO-CHESS, which is used by the BCS but does not consider margin of victory or predict actual game outcomes).

http://usatoday30.usatoday.com/sports/sagarin.htm

I like that computers are part of the BCS rankings.  They're objective.  That doesn't mean they're smart, though, since any computer ranking is only as smart as the people writing the ranking algorithm.

I'll pick on Jeff Sagarin, since his rankings are the best known and he's the most extreme with the Big 12.  Here are the Big 12 teams in his BCS rankings:

2. Oklahoma State (10-0; AP#2)
6. Oklahoma (8-1; AP#5)
7. Kansas State (8-2; AP#16)
9. Baylor (6-3; AP#25)
12. Texas A&M (5-5; no AP votes)
13. Texas (6-3; AP#31)
17. Missouri (5-5; no AP votes)
26. Texas Tech (5-5; no AP votes)
28. Iowa State (5-4; no AP votes)
66. Kansas (2-8; no AP votes)

For reference, he has Michigan at #27 and his top ranked Big Ten team, Michigan State, is #22.  In other words, he believes that 70% of the Big 12 has had a more impressive season than every single Big Ten team.  Every Big 12 team but Kansas has been better than Wisconsin.

It seems like these rankings are far too kind to teams struggling in conferences that perform well in a few nonconference games.  The Big 12 performed relatively well, granted, but here are all of their games against AQ conference teams:

Wins (6): Arizona (OSU), Florida State (OU), Miami (K-State), UCLA (Texas), Iowa (ISU), UConn (ISU)  [note: they also had a few solid wins against TCU, Tulsa, etc.]
Losses (3): Arkansas (TAM), Arizona State (Missouri), Georgia Tech (Kansas)

This is a trivial thing to get worked up about, but these rankings make a big difference in determining who plays in BCS games and the championship game.  Problems here can have major consequences.

Note: I've split the statistics to provide subtotals by OOC and Big Ten games. Also, I added a couple of line charts to show these summaries. This is also the week FEI adds rankings for Offense Efficiency, Defense Efficiency, Field Position Advantage, and Field Goal Efficiency.

Synopsis: After 7 games, Michigan is currently ranked #17 in scoring offense and #82 in scoring defense. Based on these rankings, M maintains a 43% chance for a +5 WLM (9-4 or better) season and an 83% chance for a winning season. The offense was plagued for the second week in a row by TOs (and by penalties), which turned 522 yards of total offense into just 28 points (yuck!). The defense actually played relatively well and allowed just 383 yards (the lowest yardage total since UConn – excluding the baby seal BGSU). But, the offensive (pun intended) TOs eliminated scoring chances for M, created a short field for the defense, and put the D back on the field earlier and more often. Iowa's second and third TDs both started at about mid-field after an interception and a blocked FG.

I use scoring stats because yardage stats are inherently flawed. (If you don't believe that, I guess M won the game with Iowa – 522 yards to 383 yards.) Being #82 in scoring defense is not good but U-Ms defense is not as bad as the #105 (an improvement of 7 places from last week) in total defense indicates. According to the FEI rankings at Football Outsiders, Michigan is ranked #83 in total defense.

Based on the FEI (Fremeau Efficiency Index), Michigan is predicted to win between 7.8 and 7.9 games (excluding bowl game but adjusted with +1 for U-M's one FCS opponent). Based on the FEI, M would have been expected to win 3.9 FBS games to date (we have won 4.0 FBS games to date).

Using a rough calculation based on the FEI, Michigan will be favored by 16 points over Penn State (I'll add the real FEI prediction when it comes out after our bye week– usually Thursday after the bye). Using the Sagarin Predictor, M will be favored by 2.8 points. (Vegas Odds Opened with M favored by ??). All these will be updated after the bye week but will probably not change very much.

Just as I was confused the game with Iowa was predicted so close last week, I am confused this week why Sagarin has the PSU game relatively close. Unless M implodes with TOs, this should be a win.

This line chart differentiates between OOC and Big Ten points per possession. It shows what has happened since the start of conference play. In the Big Ten, U-M is averaging only 2.5 points per possession (PPP) and 42 YPP. The defense is giving up 3.1 PPP and 42 YPP. With an average of 12 possessions per game for each team, this translates into a 7.2 point disadvantage for Michigan. (In OOC games, this was a 20 point advantage.)

For those who want yardage stats, here they are – split by OOC and Big10 games. Offensively, total yards per game are moderately lower in conference play (13% less), rushing yards are significantly lower in conference play (35%), while passing yards are moderately higher (18%). Defensively, total yards allowed in conference play have increased significantly (24%), rushing yards allowed are up just moderately (16%), while passing yards allowed are up significantly (28%).

 DETAILS: Here are the FEI numbers ( FEI Forecasts and Football Outsiders FEI ). FEI is a weighted and opponent-adjusted season efficiency, and is expressed as a percentage as compared with an average FBS team. The average team will have an index of approximately 0.00. Teams below average have negative index values.

Note that FEI completely excludes all non-FBS data (the W-L record is only for FBS games, etc.). Therefore, you need to add 1 to FBS-MW to get the final predicted wins for M this year. Or, if you use FBS-RMW, you need to add 1 to the current W-L record to get the final predicted wins for M this year. BTW, the difference between FBS-MW and FBS-RMW is the number of FBS games each team would have been expected to win to date.

The FEI is a drive based analysis considering each of the nearly 20,000 drives each year in college football. The data is filtered to eliminate garbage time (at the half or end of game) and is adjusted for opponent. A team is rewarded for playing well against good teams (win or lose) and is punished more severely for playing poorly against bad teams than it is rewarded for playing well against bad teams. I've included the GE basic data so you can see the impact of adjusting for opponent.  (See: Football Outsiders Our Basic College Stats )

Here are the Sagarin Ratings.

Sagarin uses two basic ratings: PREDICTOR (in which the score MARGIN is the only thing that matters) and ELO-CHESS (in which winning and losing only matters, the score margin is of no consequence). The overall rating is a synthesis of the two diametrical opposites, ELO-CHESS and PREDICTOR.

Per Sagarin: ELO-CHESS is “very politically correct. However, it is less accurate in its predictions for upcoming games than is PREDICTOR”.

Here is the U-M vs. Penn State National Statistical Rankings with the advantage for each category indicated (all categories within 10% are considered a "push").

Here are the week by week National Statistical Rankings for Michigan (cumulative thru the week indicated):

I have included the major rankings for offense and defense but scoring rankings show the best correlation to winning and losing. Scoring rankings are based on PPG. Rushing, Passing, and Total rankings are based on YPG.

Here is the basic data for Michigan (each individual week followed by totals and then average per game). I've included Total Possessions for Offense & Defense along with the calculated data per possession. Number of possessions do not include running out the clock at the half or end of game. Offense Plays and Defense Plays are better indicators than Time of Possession.

Using Scoring Offense and Scoring Defense National Rankings for the past 5 years (FBS AQ teams only), this table shows the percentage of teams that finish the season with a +WLM and a +5 WLM. For example, teams that finished in the Top 40 in both offense and defense had a 100% chance to be +WLM and an 82% chance to be +5 WLM (9-4 or better).

Each year, of the 66 FBS AQ teams, 65% (43 teams) end up with a + WLM and 36% (24 teams) end up with a +5 WLM.

  Synopsis: After 6 games, Michigan is currently ranked #14 in scoring offense and #75 in scoring defense. Based on these rankings, M has a 43% chance for a +5 WLM (9-4 or better) season and an 83% chance for a winning season. The offense definitely had its worse game of the season scoring only 17 points (primarily due to TOs). Rushing yardage was about 50% of the average for the year whereas passing yards were fairly close to the average. The defense continued to struggle allowing 9 more points than average and about twice the average rushing yardage. Defense passing yardage was slightly below the average. Michigan's current PPG is 26.8 so the defense must hold teams below this number to improve their performance.

I always use scoring stats because yardage stats are inherently flawed. Being #75 in scoring defense is not good but U-Ms defense is not as bad as the #112 in total defense indicates. According to the S&P+ rankings at Football Outsiders, Michigan is ranked #62 in total defense. 

Based on the FEI (Fremeau Efficiency Index), Michigan is predicted to win between 7.5 and 8.1 8.1 and 8.5 (ED: corrected) games (excluding bowl game but adjusted with +1 for U-M's one FCS opponent).

Based on the FEI, Iowa is favored by just a single point with just a 51% PWE (projected win expectation). Using the Sagarin Predictor, Iowa is favored by 3.4 points (note the Sagarin Elo-Chess actually has M as the favorite by 6.7 points and the Sagarin overall rating has M favored by 1.4 points). (Vegas Odds Opened with Iowa favored by 3.0).

I am surprised at how close this game is predicted. The SoS adjustments are especially interesting because the FEI SoS Algorithm (explained here) shows Iowa with a more difficult schedule than Michigan but Sagarin has just the opposite. Iowa has also had an extra week to prepare due to their bye last week. IMO, Michigan will have to play its best game of the year and end up with a positive turnover margin to win.

  Overall this year, U-M is averaging 3.2 points per possession (PPP) and 46 YPP. The defense is giving up 2.2 PPP and 38 YPP. With an average of 12 possessions per game for each team, this translates into a 12 point advantage for Michigan.

DETAILS: Here are the FEI numbers ( FEI Forecasts and Football Outsiders FEI ). FEI is a weighted and opponent adjusted season efficiency and is expressed as a percentage as compared with an average FBS team.

Note that FEI completely excludes all non-FBS data (the W-L record is only for FBS games, etc.). Therefore, you need to add 1 to the projected numbers for FBS-MW to get the final predicted wins for U-M this year. Or, if using the FBS-RMW, add 1 to the current win-loss record to get the final predicted wins for the year.

The FEI is a drive based analysis considering each of the nearly 20,000 drives each year in college football. The data is filtered to eliminate garbage time (at the half or end of game) and is adjusted for opponent. A team is rewarded for playing well against good teams (win or lose) and is punished more severely for playing poorly against bad teams than it is rewarded for playing well against bad teams. I've included the GE basic data so you can see the impact of adjusting for opponent.  (See: Football Outsiders Our Basic College Stats )

Here are the Sagarin Ratings.

Sagarin uses two basic ratings: PREDICTOR (in which the score MARGIN is the only thing that matters) and ELO-CHESS (in which winning and losing only matters, the score margin is of no consequence). The overall rating is a synthesis of the two diametrical opposites, ELO-CHESS and PREDICTOR.

Per Sagarin: ELO-CHESS is “very politically correct. However, it is less accurate in its predictions for upcoming games than is PREDICTOR”.

Here is the U-M vs. Iowa National Statistical Rankings with the advantage for each category indicated (all categories within 10% are considered a "push").

Here are the week by week National Statistical Rankings for Michigan (cumulative thru the week indicated):

I have included the major rankings for offense and defense but scoring rankings show the best correlation to winning and losing. Scoring rankings are based on PPG. Rushing, Passing, and Total rankings are based on YPG.

Here is the basic data for Michigan (each individual week followed by totals and then average per game). I've included Total Possessions for Offense & Defense along with the calculated data per possession. Number of possessions do not include running out the clock at the half or end of game. Offense Plays and Defense Plays are better indicators than Time of Possession.

Using Scoring Offense and Scoring Defense National Rankings for the past 5 years (FBS AQ teams only), this table shows the percentage of teams that finish the season with a +WLM and a +5 WLM. For example, teams that finished in the Top 40 in both offense and defense had a 100% chance to be +WLM and an 82% chance to be +5 WLM (9-4 or better).

Each year, of the 66 FBS AQ teams, 65% (43 teams) end up with a + WLM and 36% (24 teams) end up with a +5 WLM.