By The Numbers - Purdue Gut Punch

Submitted by The Mathlete on November 9th, 2009 at 2:18 PM
All numbers are points per game vs an average team. They are adjusted for strength of opponent.  No 1AA games or stats are included.  For more detailed questions on how the numbers come about, click here.

Expected Points

The pace and field position of this game set up perfectly for Michigan.  Despite giving away a possession to Purdue on the on-side kick, Michigan actually had the advantage in expected points in this game, 28 to 27.  A pretty even game all around on the field position and the number of drives.  Purdue had an extra possession but Michigan had better field position, helped by a rare +1 in the turnover margin for the game.  Both teams had about the same opportunity and did the same things with them, 5 TD's and a field goal each.  Obviously the missed PAT provided the difference.

Rushing Offense

Predicted: +3
Actual: +13

The way these two teams have been playing, this could be anything from a huge advantage for M or Purdue, or it could be a total draw.  There doesn't appear to be much consistency to either of these units.
It turned out to be a huge advantage for Michigan.  Only two teams in the country did better than a +13 on the ground last Saturday and none of them were from the Big 6 conferences.  Minor posted an opponent adjusted +7, the 10th best performance in the country.  Even after adjusting for quality of opposition, this performance was just behind Iowa and the 2nd best rushing performance in Big 10 play this year.  A lot of positives for the ground game this weekend.

Passing Offense

Predicted: -2
Actual: +4

It does not appear that there is much in the past several games that indicates Purdue being a team Michigan can significantly exploit through the air.
Although not a huge volume through the air or anything particularly spectacular, no team this year has created more points/play in the passing game against Purdue than Michigan did on Saturday.  Northwestern and Toledo both had solid games in the air against Purdue, but both did on 47 and 67 plays, respectively.  Michigan produced 5 points of value on only 28 attempts, its best performance through the air since opening
week against Western.

Rushing Defense

Predicted: -3
Actual: +1

This appears to be as good of a chance as Michigan is going to have to shut down a running game the rest of the season.
I don't know if shut down is the right word to describe what Michigan's run defense did on Saturday, but after the hemorrhaging that was our run defense against Illinois, this was a solid but not spectacular bounce back.

</good news>
<bad news>

Pass Defense

Predicted: +2
Actual: -10

How Purdue's passing attack fares could swing the game as there has been 4 touchdowns worth of variance in their performance this year in Big 10 play alone.
After showing glimpses of disaster all Big 10 season, the Michigan pass defense finally put together a full show this weekend.  It's a good thing Michigan got the two picks or this number would have really been ugly. 

Special Teams

Make a 51 yard field goal, miss an extra point.  In a game in which Michigan did a lot of things well it was obviously the negatives that ended up winning out.  After a much better than expected start to the season, Michigan kicking game finally was an issue.  The missed PAT was obviously a difference in the game and all in all, Michigan posted its first negative kicking game of the season, one of only 6 teams in the country to make it this far without a negative performance from their place kicker.

Overall, however ,the special teams were outstanding. The final result was a +5 for the game after outstanding results from all of the other four units.  In totality, this was Michigan's best special teams performance of the season.

Conclusions?

Somehow I feel much better about the team after writing this than I did at noon on Saturday (post-game out here on the west coast).  Great performance from the running game, solid game through the air, rush defense that wasn't a sieve and quality results in the return game. 

There are always two types of outcomes to look back at a football game for.  The W or L coutcome.  Obviously in that regard Saturday was a failure.  The other outcome is how did you perform?  So many facets of the team performed as well as they have all season on Saturday.  But we also saw how one unit can ruin it for the rest of the class and man, we lost to Purdue at home for the first time since the days of Sterling Cooper and that really sucks.  So in other words, I don't know what to say. 

Comments

bigmc6000

November 9th, 2009 at 2:29 PM ^

I don't know how far back you've got these numbers but is there any way you could find out our historical performance vs Wisconsin? I.E. Expected vs results. I'm just curious, if you don't have the data then by all means don't go wasting your time on it but if you do I'd be interested to see if we consistently perform better or worse against them than we should.

As always - great job w/ the metrics, always a great read!

Huntington Wolverine

November 9th, 2009 at 2:43 PM ^

As frustrated as I was with the loss, I thought our guys played pretty well and lost to an underrated Purdue team that gave close calls to quite a few other teams and upset OSU quite handily a few weeks ago.

Though we can't make a direct 6 Degrees To Kevin Bacon connection, we played Purdue closer than OSU did. I'm not giving up that the team plays well against Wiscy or OSU and pulls off a win. You never know when it starts clicking for a young team, when they turn that mental corner.

Remember the momentum shift at Illinois when we curled up and died. What was the reaction after the quick second half turnaround by Purdue- we kept fighting. That's progress and maturity that wasn't there a week ago.

umberkeley

November 9th, 2009 at 3:16 PM ^

Taking net actual to predicted values:

+10 rushing offense
+6 passing offense
+4 rushing defense
-12 passing defense
+4(?) special teams (inferred from a +0.7 Michigan ST in your Purdue preview)

Plus

+1 in turnover margin

Did Purdue manage better scoring efficiency to get the win?

The Mathlete

November 9th, 2009 at 3:39 PM ^

A couple things not noted in the write-up should help fill in the gaps.

Michigan was -4 in penalties on the game and interceptions are included in the passing numbers but fumbles are excluded from all offensive numbers, and Michigan's fumble was another -4.