Illinois: Beyond the Box Score

Confused?  Catch up with the BTBS Primer.  It's a Tuesday during football season, so that can only mean one thing: BTBS GAME REVIEWS!!!!  For the newer readers, the BTBS posts (which stands for Beyond the Box Score...an homage to the great SBN baseball blog of the same name, since my love of baseball stats led me to pursuing college football stats) dive into the Mizzou games using advanced stats, mostly those that I created at Football Outsiders and have discussed here a few times.  I do highly recommend reading the primer above.  You don't have to be a complete and total numbers geek to get something worthwhile out of these posts, I hope ... though it certainly doesn't hurt.  These are my favorite posts to write, so I hope you don't just let your eyes go blurry at the sight of numbers and skip it.  Even if you ignore the numbers, there is word-based analysis here as well.

To help get your bearings in the "BTBS Box Score" below, I am including a column that shows you the 2009 national average for a given statistic.  That should give you a reasonable idea for where the teams performed well or poorly.  With that said, let's dive in!

	Illinois (13)	Mizzou (23)	2009 Nat'l Avg
Close %	100.0%
Field Position %	34.4%	58.8%	50.0%
Leverage %	65.6%	63.6%	68.1%
TOTAL
EqPts	10.9	21.3
Close Success Rate	35.9%	45.5%	41.9%
Close PPP	0.17	0.28	0.32
Close S&P	0.529	0.732	0.741
RUSHING
EqPts	8.6	5.0
Close Success Rate	41.0%	46.4%	42.7%
Close PPP	0.22	0.18	0.29
Close S&P	0.632	0.642	0.717
Line Yards/carry	3.06	2.73	2.90
PASSING
EqPts	2.2	16.3
Close Success Rate	28.0%	44.9%	41.1%
Close PPP	0.09	0.33	0.35
Close S&P	0.370	0.783	0.766
Std Downs/ Pass. Downs Sack Rate	8.3% / 7.7%	0.0% / 4.2%	4.8% / 7.4%
STANDARD DOWNS
Success Rate	35.7%	51.0%	47.7%
PPP	0.10	0.26	0.34
S&P	0.458	0.773	.812
PASSING DOWNS
Success Rate	36.4%	35.7%	29.9%
PPP	0.30	0.30	0.29
S&P	0.666	0.660	0.594
TURNOVERS
Number	4	1
Turnover Pts	15.0	4.3
Turnover Pts Margin	-10.7	+10.7
Q1 S&P	0.719	0.626	0.747
Q2 S&P	0.808	0.641	0.733
Q3 S&P	0.188	0.904	0.750
Q4 S&P	0.301	0.691	0.716
1st Down S&P	0.257	0.755	0.766
2nd Down S&P	0.888	0.590	0.733
3rd Down S&P	0.468	0.903	0.701
Projected Pt. Margin	-21.1	+21.1
Actual Pt. Margin	-10	+10

What can we glean from this?

• Mizzou should have won by much more.  The projected scoring margin was approximately three touchdowns, not just ten points.  This is what happens when a) you miss a field goal, b) the other team makes an improbable field goal, c) big penalties go your opponent's way (namely, two iffy roughing penalties and two iffy pass interference penalties), and d) you struggle nearer to the opponent's end zone.  You leave points on the board.
  
  Depending on whether you are an optimist or a pessimist, you can look at this in different ways.
  
  Optimistic: They truly dominated this game, and if they played ten times, Mizzou would win 8-9.  It took luck, timing and the element of surprise for Illinois to stay close, and Mizzou really proved they're going to be a good team this year.
  
  Pessimistic: Mizzou lacks a killer instinct and has no idea what plays to call in the red zone.  This could lead to a season full of worse-than-the-stats-suggest results.
  
  
• Mizzou dominated the field position battle.  Almost three of every five plays Mizzou ran were in Illinois territory.  It took a lot to hold the Tigers to just 23.
  

Explosiveness could be a problem.  Mizzou's efficiency numbers were decent, but they still found themselves in far too many passing downs (Leverage % = Total Plays Run on Standard Downs / Total Plays).  Their 0.28 PPP (EqPts Per Play) was below last year's national average, and while it's great that we've solved one problem -- we now have two wonderful bailout options on 3rd-and-6 in T.J. Moe and Michael Egnew -- this team is going to really need Wes Kemp, Jerrell Jackson and others to get downfield and eat up chunks of yardage.

A one-game sample is not enough to make any concrete conclusions, but here's something close to one: if Mizzou is struggling to make big plays, especially on standard downs, they will struggle to score more than 23 points in a game.  We've been really spoiled in recent years, with Jeremy Maclin, Danario Alexander, etc., turning a 4-yard gain into a 40-yarder.  If we can't display more of a downfield threat in the coming games, Big 12 defenses will start cluttering up the play near the line of scrimmage, and the offense will struggle.

Mizzou's defense dominated once the element of surprise wore off.  As mentioned on Sunday, there is no way Mizzou could have determined that Illinois would spend much of its day in what I'm calling the Cocked Pistol (Pistol with an off-set fullback).  Thanks to that and a couple of the aforementioned iffy penalties, Illinois' S&P was solid in the first half, and they put 13 points on the board.  But in the second half, things were determined more by talent and athleticism, and the Mizzou defense had much more of it than the Illinois offense.  Considering Will Ebner was out, Luke Lambert ended up out, and Jasper Simmons was gimpy, this is certainly an encouraging development.  A healthy, fully-manned Mizzou defense might be pretty good.  Might.  The Illinois offense also might be pretty bad, especially since nobody else will be caught offguard by the Cocked Pistol.

 

Rushes and Directions

This doesn't happen every game, but this week's official play-by-play included directions and descriptions of runs and passes.  I love this, and I desperately wish this would always be a feature (sometimes it just says "Rush" or "Pass" with no further description).  I realize there are limitations -- we are completely relying on how one person in the press box interpreted each play -- but we can still probably glean interesting information from this.  When this information exists, I plan on using it this season.

Below is a table looking at the direction of each Mizzou rushing attempt.  Three rushes were not given any further description -- they are listed in the "Undefined" column on the righthand side.  We will focus on the "defined" carries.

Player	Over Left End	Over Left Tackle	Over Left Guard	Up The Middle	Over Right Guard	Over Right Tackle	Over Right End	Undef.
De'Vion Moore	4 for 35 (8.8)		2 for 13 (6.5)	3 for 13 (4.3)	1 for 0 (0.0)	1 for 8 (8.0)	2 for 6 (3.0)	3 for 3 (1.0)
Kendial Lawrence		1 for 0 (0.0)		3 for 7 (2.3)	1 for 7 (7.0)	1 for 8 (8.0)	1 for (-2) (-2.0)
Henry Josey							1 for 10 (10.0)
Blaine Gabbert			1 for 3 (3.0)	2 for (-3) (-1.5)		1 for 7 (7.0)
TOTAL	4 for 35 (8.8)	1 for 0 (0.0)	3 for 16 (5.3)	8 for 17 (2.1)	2 for 7 (3.5)	3 for 23 (7.7)	4 for 14 (3.5)
TOTAL (RBs Only)	5 for 35 (7.0)		10 for 40 (4.0)			6 for 30 (5.0)

Of the 21 "defined" rushes by the running backs, 10 were basically up the middle (or close to it), while 11 went outside.  Mizzou averaged almost six yards per carry on the outside runs, four yards per carry between the tackles.  Hopefully this data exists for future games too, as I'm curious about both the inside-outside distribution of the carries and the success level.

One has to wonder if Derrick Washington would have made a positive impact, particularly on those inside carries.  I can confidently say that the Moore-Lawrence-Josey trio did as well our better than Washington would have in running to the outside.

One other note about Mizzou rushes, and this point gets directly credited to The Beef, who pointed this out during the game: on the plays where the ball is snapped while the running back is still in motion (which is to say, most rushing plays), the ball is being snapped too early.  These plays are relatively slow to develop as a whole, but if the blocking is there, I'm okay with that.  What happened a lot on Saturday, however, was really really slow to develop.  It had nothing to do with the play itself, but with when Blaine Gabbert called for the ball to be snapped.  If he waits another beat to snap the ball, then one has to think it would help matters considerably.  Mizzou was fine running the ball on Saturday -- not great, not terrible -- but that extra beat probably cost them at least a few extra yards.  At this point, we are what we are; we know how Mizzou is going to attempt to run the ball this year, and there really isn't any point in pining for something different.  But if the timing improves, these runs will improve.

One more thing: I echo Dave Matter's sentiments.  More Josey, please.  Very intrigued by this guy.

Targets and Catches

Now to the passing game.  For those who have bought the 2010 Missouri Football Preview (still available at the Mizzou Bookstore and soon the Tiger Team Store as well!), you've probably noticed the existence of "target" data on the Receivers preview.  This refers to a two-part series from December (Part One, Part Two) in which we looked not only at who caught passes, but also the "intended for" data that typically comes with play-by-play data.

Here's a quick refresher:

• Targets: The number of times a pass was intended for a receiver, whether the ball was caught or not.
• Catches: The number of times the player caught the ball.
• Catch%: The percentage of targets that a player caught.
• Target%: The percentage of total passes intended for one specific player.
• Yards Per Target: Self-explanatory.  You take a player's receiving yards and divide it by the total number of targets instead of catches.

Typically, you would expect to see a TE's or RB's Catch% in the 80+% range, as they are likely catching more high-percentage passes.  Meanwhile, a WR will usually end up anywhere between 60% and 80%, depending on how far downfield most of his attempts are.  This is where Yards Per Target becomes an interesting stat to watch. If you are more of a downfield receiver (like Wes Kemp), you will have a lower Catch%, but your Yards Per Target might be similar to guys with higher Catch%'s.

Here's the Target/Catch data for the Illinois game:

Player	Targets	Catches	Catch%	Target%	Rec. Yds.	Yds. Per Target
T.J. Moe	14	13	92.9%	29.2%	101	7.2
Michael Egnew	12	10	83.3%	25.0%	60	5.0
Wes Kemp	8	3	37.5%	16.7%	46	5.8
Jerrell Jackson	6	4	66.7%	12.5%	50	8.3
L'Damian Washington	2	1	50.0%	4.2%	11	5.5
Kendial Lawrence	2	2	100.0%	4.2%	1	1.0
De'Vion Moore	2	0	0.0%	4.2%	0	0.0
Brandon Gerau	1	1	100.0%	2.1%	12	12.0
Marcus Lucas	1	0	0.0	2.1%	0	0.0
TOTAL	48	34	70.8%	100.0%	281	5.9
TOTAL (WR)	32	22	68.8%	66.7%	220	6.9
TOTAL (RB)	4	2	50.0%	8.3%	1	0.3
TOTAL (TE)	12	10	83.3%	25.0%	60	5.0

Five immediate observations:

1. Wow, T.J. Moe.  Not even Tommy Saunders caught 93% of the passes thrown at him.  In fact, he never topped 78%.  Of course, in two of three season, Saunders also averaged at least 8.0 Yards Per Target.  Moe's role is almost more like that of a tight end than a wide receiver.
   
   
2. The WR-RB-TE distribution is almost exactly that of 2008.  The difference, of course, is that if Moe is more like a tight end than a wide receiver, the yards per target is not going to be the same.
   
   
3. Wes Kemp had a lot of missed opportunities.  For the season, I expect Wes Kemp's Catch% to be somewhere between 50-60%.  With Moe and Egnew taking most of the possession duties, Kemp's role will be as primary blocker on the sideline passes (he was infinitely better than any of the tight ends in this role on Saturday) and a deep threat.  Obviously his numbers would look much different if he hadn't been called for pass interference on a particular long ball (that would have bumped his Yds. Per Target to somewhere between 9.0 and 10.0), but he's going to need to take better advantage of his opportunities from here on out.
   
   
4. Jerrell Jackson is probably the key.  Jackson is the best combination of short-long threat on the team.  He can play the possession guy if he needs to, but he appears to be a very good intermediate route runner as well.  If he starts seeing more Targets as he gets healthier, Mizzou's passing game could open up considerably.
   
   
5. Derrick Washington is missed in the passing game.  Even more than running between the tackles, the area in which we might miss Washington the most is in the passing game.  Four passes to the running backs resulted in one single yard on Saturday.  Kendial Lawrence did nothing with his two receptions, while De'Vion Moore had a bad drop (he was running with the ball before he caught it) and wasn't able to reel in a more high-difficulty screen pass.  Washington quite possibly would have made both catches.  Obviously Washington is gone, and we have to move on just like the team has; but I wanted to point out where I think Mizzou's offense might suffer the most without him.

One other thing we can look at when the play-by-play provides it: type of pass.  This is even more subjective than the rushing data (where's the line between right middle pass and deep out pass?  8 yards? 12?  And what's the difference between a sideline pass and a right middle pass?  4 yards?), but since the data exists, I thought I'd throw it out there.

I'm going to roughly define these passes into three ranges:

• "Shorter": screen passes, sideline passes and slant passes.  Slants might technically go in the "medium" category as well.
• "Medium": right middle, middle, left middle, and post passes.
• "Longer": deep and deep out passes.

Six passes were undefined.  Here's the data.

Type	Att.	Comp.	Comp. %	Yards	Yds/ Att.
Screen Pass	6	5	83.3%	21	3.5
Sideline Pass	17	14	82.4%	126	7.4
Slant Pass	7	6	85.7%	62	8.9
Right Middle Pass	1	1	100.0%	7	7.0
Middle Pass	4	3	75.0%	30	7.5
Left Middle Pass	1	0	0.0%	0	0.0
Post Pass	3	1	33.3%	12	4.0
Deep Out Pass	2	2	100.0%	17	8.5
Deep Pass	1	0	0.0%	0	0.0
Undefined.	6	2	33.3%	6	1.0
TOTAL	48	34	70.8%	281	5.9
TOTAL (Shorter)	30	25	83.3%	209	8.4
TOTAL (Medium)	9	5	55.6%	49	5.4
TOTAL (Longer)	3	2	66.7%	17	5.7

Thirty of the 42 defined passes were of the "shorter" variety (23 if you don't count the slants).  I don't have a ton of problem with this.  With the open-field blocking we are likely to get with the Kemps of the world, then a lot of these shorter passes might result in huge gains.  My biggest concern moving forward is just that the deep balls are occasionally there ... and that the shorter passes actually do result in huge gains occasionally.  Illinois did a good job of tackling and preventing the screens and sideline passes from breaking open.  Mizzou will have to do a better job of making that happen in the future.

Three Positives

1. Always. Be. Closing.  I can honestly say I wasn't particularly worried about the 13-3 halftime deficit.  I vowed to hold off on being worried until or unless Illinois scored on its opening drive of the second half.  They did not.  This is, by my count, the fifth time in the Gabbert Era that Mizzou has won after trailing at halftime.  We were used to the clinical destruction of the Daniel Era, but Gabbert takes a little while longer to get rolling.  In the end, Mizzou closed, and Illinois has to put that coffee down.
   
   
2. T.J. Moe's routes are phenomenal.  When Mizzou had 3rd-and-4, Moe ran exactly four yards and caught the pass.  When it was 3rd-and-8, he ran precisely eight yards.  I have expressed my overall concern with our downfield passing -- maybe it comes around by conference play, and maybe not -- but it is so nice having that third-down bailout option we didn't have last year.  Every great offense needs that.
   
   
3. 1-0, baby.  Mizzou has started 1-0 on the season every year since Gary Pinkel's first in Columbia.  It doesn't take much statistical analysis to come to the conclusion that that is much, much better than starting 0-1.
   

Three Negatives

1. Early LB Play.  When it came to Illinois' funky offense, no unit seemed more perplexed and on-its-heels in the first half more than the linebackers.  On Illinois' one touchdown drives, LBs lost receivers underneath on two different occasions, including on the touchdown itself.  Plus, they just were not tracking Scheelhaase very well.
   
   The defense made its adjustments, and the LBs (Zaviar Gooden and Andrew Gachkar in particular) came up huge down the stretch. But it was worrisome there for a while, especially considering the LBs are supposed to be the strength of this D, with or without Will Ebner.  They may have still ceded the "strength of the D" mantle to the D-line, however.  Aldon Smith was every bit as monstrous as we hoped he would be, even against the run.  Meanwhile, Brad Madison made a gigantic hit and recovered a fumble, Michael Sam was in Nathan Scheelhaase's face on multiple occasions, Dominique Hamilton surged in the second half, and Jacquies Smith was ... well ... he was on the field a lot.  J. Smith and Marcus Malbrough will need to step up their game if they don't want to be overtaken by Madison and Sam.
   
   
2. Mizzou fans.  Seriously, smile.  Mizzou won, and they won by almost exactly the amount that Vegas expected.  Sure, we all wanted a 52-0 blowout ... but we always want that.  It doesn't usually happen.  Mizzou encountered adversity -- of both the self-inflicted and surprise varieties -- and they won by 10.  If they do not improve one lick over the next month, they are still likely to be 5-0 heading to College Station.  Our goals are obviously quite high, but there's still something to be said for doing what is expected, and Mizzou did precisely that.  Ole Miss and Kansas lost to FCS teams.  Oklahoma barely beat Utah State.  Mizzou handled their business and moved on.
   
   
3. Defer.  Honestly, I don't have much of a problem with being a second-half team, so to speak.  We were very spoiled when Chase Daniel was running the show; quarterbacks who aren't human computers sometimes take a little while to get going.  As long as Gabbert comes through in the second half, then I'm mostly okay with that.
   
   But here's a request: we need to acknowledge that maybe we are slow starters now, and if we win the toss, we need to start deferring to the second half.  Mizzou once again won the coin toss and elected to receive to start the game.  I understand the mentality behind this, and I even like it: we want you to be behind the first time you touch the ball.  The only problem is, Mizzou isn't scoring to start the game anymore.
   
   In the 13 games of the Gabbert Era, Mizzou's first drive has resulted in three touchdowns, three field goals and seven punts.  As a whole, six scoring drives out of 13 isn't terrible.  But electing to receive was a much more appealing option when Daniel, Maclin, and Coffman roamed the field.  Things are a bit more of a grind for us right now, and deferring is, to me, the correct call.
   

Summary

Seriously, smile.  Saturday was a very good day.  Not a perfect day, but a very good one.  Now, we turn the page to McNeese State.