Mid-Week Metrics Retreats to a Cabin in the Woods

Submitted by The Mathlete on November 3rd, 2011 at 3:12 PM

Kick it off like we always do…


The early touchdown saw the unadjusted numbers drop below 30% but the spread adjustment kept the expectation at about 70% or above for the whole game.

Top 3 Plays:

Play #45, +12%, Robinson to Roundtree for 49 yards on 3rd and 20.

Play #73, +11%, Robinson to Gallon for 42 yards and 1st and Goal.

Play #67, +10%, Avery picks off the Marve deflection

Bottom 3 Plays:

Play #6, –18%, TerBush to Bush for 48 yards a Purdue lead.

Play #36, –9%, Robinson picked on third down.

Play #46, –9% Gardner picked for the first time this year.

Game Recap

The story of Saturday was mostly Fitz Toussaint, and rightfully so. His +7 on the day was the best mark for a Michigan running back on the season. In fact, other than Vincent Smith’s +6 against Eastern Michigan no Michigan back had even crossed +3 on the season.

The defense continues to do enough to allow the offense to take hold of the game. After four B1G games this season the defense has had a best Win Percent Added (WPA) of +6% from Saturday to a worst of –7% against Northwestern. That is a incredibly tight window to operate in and means that defense has essentially held serve in every B1G game this season. The offense is still doing the heavy lifting, but at least the defense isn’t adding to load this season.


Grades are in PAN (pts/game) and opponent adjusted.

Rush Offense: +3

Fitz: +7

Vincent Smith: +1

Denard: -7

Pass Offense: +5

Denard: +7

Devin: -1

Rush Defense: +2

Pass Defense: +0

Special Teams: +1 (best of the year)

Heisman Tracking

A little frustrated with the coverage on the Heisman this year. Apparently the NFL GM’s are now deciding who the best performer is in college football. This isn’t a knock on Andrew Luck, but just because the NFL says he is a sure thing, doesn’t automatically mean he is going to have the most deserving year. He has piled up good stats against bad teams so far and the conversation could still change.

Here are my ratings of the top contenders looking at both PAN and WPA.

Player, School: PAN (Rank), WPA (Rank)

Andrew Luck, Stanford: +6 (25th), +1.7 (8th)

Trent Richardson, Alabama:  +4 (2nd RB), +.4 (11th)

Kellen Moore, Boise St: +11 (4th), +2.1 (4th)

Russell Wilson, Wisconsin: +13 (1st), +2.4 (1st)

Case Keenum , Houston: +13 (2nd), +2.2 (3rd)

Denard Robinson, Michigan: +6 (22nd), +2.2 (2nd)

Ron Zook Dumb Punt of the Week

Pickings were a little slimmer this week but this week’s award goes to the Will Muschamp and the Florida Gators. Trailing by 4 in the World’s Largest Non-Alcoholic but Actually a Total Drunkfest Party with a little over 8 minutes to go, the Gators faced 4th and 2 at the Georgia 37. This one works out for the Gators, even though they go on to lose, anyway. After taking a Delay of Game to give the punter more room (the first sign of a dumb punt), Florida manages to down the ball at the 4 and get a three and out. Florida gets the ball back at the 36, trading 2+ minutes on the clock to get a yard and a fresh set of downs, the Gators throw three straight incompletions and then go for the 4th and 10 but fail, never getting the ball back.

Projections and Ranking

If you missed it Monday I posted detailed odds for the B1G championship game. Michigan stands at 9.5% overall. Going 4-0 down the stretch bumps it up to nearly 40%. A loss to division foes Iowa or Nebraska effectively kills the chances where going 3-1 with a loss to either Illinois or Ohio still leaves the odds around 20%.

After an expected win last week against Purdue, the overall win projection for Michigan is relatively unchanged in between 9 and 10 wins.

Game Odds

Opp (Change vs last post)

@Iowa: 67% (-1%)

@Illinois: 55% (+6%)

Nebraska: 56% (-8%)

Ohio: 70% (-7%)

My Top 5

1. Oklahoma St

2. Boise St

3. LSU

4. Alabama

5. Stanford

B1G Ten

9. Nebraska

12. Michigan

13. Wisconsin

15. Michigan St

19. Penn St

38. Purdue

42. Ohio

47. Illinois

73. Northwestern

88. Iowa

104. Minnesota

113. Indiana

Iowa Preview

PAN, National Rank (leader), B1G Rank (leader)

Rush Offense

Michigan: +5, 3rd (Georgia Tech), 1st

Iowa: +1, 53rd, 7th

Pass Offense

Michigan: +2, 32nd (Boise), 4th (Wisconsin)

Iowa: –6, 118th, 11th

Rush Defense

Michigan: +2, 25th (LSU), 5th (Michigan St)

Iowa: +3, 18th, 3rd

Pass Defense

Michigan: +1, 43rd (Oklahoma St), 7th (Michigan St)

Iowa: +3, 23rd, 2nd

Special Teams

Michigan: 0, 88th (Florida St), 10th (Purdue)

Iowa: +1, 49th, 7th

The next three games are all slight Michigan favors before matching up with an improving Ohio team. A home game versus Iowa would make me more comfortable but I still think it goes our way, 37-30 Michigan.

One final note is that based on a little twitter prompting from @cdbarker I have begun work on a game-theory manifesto and it's going to be long, probably to be posted in December. Planned topics include: how to use timeouts, suprise on-side kicks, a better 2-point conversion chart and possibly a revisit of 4th down stategy. Hit me in the comments or @The_Mathlete with other things you would like to see.



November 3rd, 2011 at 11:28 AM ^

I think Coach Beckman at Toledo deserves special mention in your Zook Punt Of The Week for his crappy timeout management Tuesday night

Looking forward to that manifesto, Mathlete

One Inch Woody…

November 3rd, 2011 at 11:44 AM ^

You do opponent adjusted stats, but do you do play-calling adjusted stats? Or garbage time adjusted? For example, Eastern Michigan was able to rack up nearly 200 yards on the ground and came into the game ranked 6th in the nation in rushing yards per game. They only passed a grand total of 3 times, though. Michigan State also didn't pass as much because of the trash tornado, which is a factor in their high rushing total in that game.

Similarly, once Purdue fell behind by a lot, they were forced to almost exclusively pass and by the 4th quarter they had only accumulated 170 or something yards total so we put in our 2nd string defense and they were finally able to move the ball. The vast majority of Purdue's passing yards came in garbage time.

One Inch Woody…

November 3rd, 2011 at 12:39 PM ^

How is Iowa 3rd and 2nd in rush and pass defense in the Big 10? I have no idea how you can arrive at those numbers.. (even after looking at your methodology). The only medicore competition they have beaten(unless you count Pitt) is Northwestern, and have lost to some terrible competition (Minnesota, Iowa State). Not to mention Indiana, Northwestern, Minnesota, Iowa State, and Pitt racking up huge amounts of yardage against them.

In addition, Iowa is ranked 9th in scoring defense, 9th in total defense, 12th in pass defense, 9th in pass efficiency defense, and 9th in sacks. In other words, they have marginally better defense than Minnesota, Indiana, and Northwestern. This is all while playing their cupcake schedule.

Maybe you're counting the Penn State game as bumping-up their stats? But you have to realize then that Penn State is 12th in the big 10 (below Minnesota) in pass offense efficiency , 9th in pass offense, and 10th in total offense. I don't see how they can be one of the top 3 defenses in the Big 10 this way.

steve sharik

November 3rd, 2011 at 1:09 PM ^

Florida's probability of scoring on the drive at the point of facing 4th and 2 from the 37 compared to their probability of scoring on the following drive with 1st and 10 from the 36 post punt.

The Mathlete

November 3rd, 2011 at 2:07 PM ^

Going has a 38% chance of scoring a TD and the 1st and 10 has a 62% chance of scoring a TD. However, to get back to the same spot, Florida had to get a couple breaks. Most punts from the 42 don't get downed at the 4, the average is the 12. Second break was getting a stop. Third was a weak punt by Georgia that gave the Gators back the ball in good shape. Usually trading punts in this territory is a net loss for the driving team. 


November 3rd, 2011 at 2:27 PM ^

What is contributing the most to our bad special teams ranking? Is  there an area that is good? FG kicking is improved, but we haven't had to kick much. How bad are we doing with kick returns? I thought our punting was supposed to be a strength.

The Mathlete

November 3rd, 2011 at 2:56 PM ^

Even though the rank is low the effect is minimal. Special teams really only matter if you are really good or really bad at them which Michigan is neither. Kickoff is a strength, FG kicking is slighly above average. Punting has been a disappointment and both return games are the biggest source of the ranking.


November 3rd, 2011 at 3:32 PM ^

I'm still wondering about the Illinois decision to go for 2 when up by 7. I tried doing the math myself, but there were too many variables to make my napkin math reliable in the least bit. It seems to me, though, that if you are near certain that the opponent's offense will move down the field to score (1) you should be near certain that they will make a 2pt conversion, (2) they probably have a better chance to win in overtime than you, and (3) you should take every opportunity you can to seal the game. I view it in the same way that I view our final TD against ND. We could have gone the safe route, kicked the FG, and gone to overtime, but if you have a chance to seal the game, you take it, assuming that the gamble is small.

Anyway, I think it would be interesting to see which scenarios, if any, warrant going for 2 and which variables (opp off, your def, time, hfa, etc.) would be the most influential in the decision.


November 3rd, 2011 at 3:32 PM ^

How are the "Game Odds" calculated?

After Illinois layed an egg against Purdue and their recent losses to PSU and OSU, it would seem that we should be heavily favored, even if @ Illinois.

The reverse for Nebraska, which took MSU to task this past week.

The Mathlete

November 3rd, 2011 at 3:39 PM ^

Each week game scores are re-adjusted for all prior games based on each teams work for the season in total. In week 7 Illinois was a +11. If they played this week they would be a +5. I don't forecast a trend, but their recent results have brought them down by nearly a touchdown in the last month.

Philip A. Duey

November 3rd, 2011 at 3:42 PM ^

OT theory: e.g. Playing for OT or a win at the end of regulation? Percentages on taking the ball 1st in OT vs. taking the "bottom of the 9th" advantage?  Score a TD to be down 1 before 3rd OT: when do you go for 2?  4th and goal inside the 2 with the first possession: what probability of scoring a TD suffices to overcome the necessity of taking the 3 points?

Hoke vs. Dantonio theory: Hoke has shown himself to be aggressive on 4th down this year, especially with short yardage in the "No Man's Land" where we often get the Dumb Punt of the Week.  But he's so far just played straight up going for it.  Dantonio, as we know, has a proclivity for the fake.  A subsection of fourth down theory might look at going for it straight up vs. faking a FG/punt.