Derrick Green not traveling to Citrus Bowl

Derrick Green not traveling to Citrus Bowl

Submitted by MGoStrength on December 27th, 2015 at 5:18 PM

I know we've heard speculation that Derrick Green may wind up transferring after getting burried on the depth chart and even more so now after the verbal committment of Karreem Walker.  It's stil unclear if that will happen, but this is just another peice of information regarding the once highly touted running back.  Here's the link:…

2014 RB Performance

2014 RB Performance

Submitted by fergusg on August 24th, 2015 at 10:18 PM

OK - so I know there have been a number of posts on RB performance and who the best RB is (mods - I think this is a new take, but feel free to remove if its redundant).  

I did a little more digging into the RB performance last year to see if I could quantify who performed better.  I came up with the following charts, which I think definitively shows Drake was our best running back by far.  (click to make bigger).  

Chart 1 - Big Ten Opponents only

This is based on approximately 32 attempts by Green, 56 by Johnson and 82 by Smith (Note: see data disclaimer below).

How do I read this chart?

Basically, the chart tells you what % of each RB's rushing attempts went for more than X yards.  The x axis is "X yards", the y-axis is the % of attempts.  Being higher on the chart is better.  Data is limited to Green, Smith and Johnson.

E.g.  25% of Green's rushining attempts went 7 yards or more; 34% of Johnson's did the same and only 20% of Smiths.  Likewise, 18% of Johnson's carries went 10 yards or more. 

Note: Read "15+" as "16 or more yards" (its a little nuance I oversaw when creating the charts)

How did you create this chart?

I basically copied and pasted the play by plays from ESPN for every game into a spreadsheet, then ran some conditional formulas and pivot tables  to identify:

1. Was a UM running back in the play?

2. If so, was it a rushing attempt?

3. If so, what was the result in yards.

4. Filter, summarize, etc

Data disclaimer: The underlying data may not be 100% correct, there may be minor discrepancies, but based on the digging and testing I've done, the impact to the results is limited.  I'd put the confident interval at >95% on the results.

What should I to take away from the chart?

Basically, Drake Johnson was the most efficient back by some margin.  71% of his carries went for 3 or more yards, compared to 44% for Green and 51% for Smith.  The talk of him not being a Power 5 RB seems like nonsense to me, based on the data.

I made also made up a metric I called the explosion/implosion index, calculated as follows: % of carries 10 or more yards divided by % of carries for 0 or less yards.  Johnson kills the other two here...

Explosion/implosion index results

  • Johnson: 2.0
  • Green: 0.57
  • Smith: 0.43

What if you include all games?  What if you exclude Indiana?

The answer is it gets closer, but doesn't change the story substantially.  

If all games are included, Drake still sits higher at all point on the curve if all games are included.

If all games except Indiana are included, Drake sits higher at all but 5 points (9-12 yards), where all the running backs are within 1-2 percent of each other.

Chart 2  - all games excl.  Indiana

If all games except App State, Miami (OH), and Indiana are included (Chart 3), Drake is higher at all points, except 15+ yard (5% of Drakes carries when 15+ yards, 7% of Green's did). 

Chart 3  - all games excl. App State, Miami (OH) and Indiana


Grantland: All of the Non-Harbaugh Reasons to Get Excited About Michigan Football in 2015

Grantland: All of the Non-Harbaugh Reasons to Get Excited About Michigan Football in 2015

Submitted by JClay on August 6th, 2015 at 2:33 PM

" A nine-win season should be within reach. Of course, Michigan could always limp to another disappointing finish, but if a few things break right and the Wolverines wind up playing on New Year’s Day, it won’t only be because the new head coach is the Messiah. It will be because this team has underachieved for a while, and it has the pieces it needs to do well right away."…

Time is a Flat Football: Running Backs

Time is a Flat Football: Running Backs

Submitted by MilkSteak on July 23rd, 2015 at 1:08 AM



"Time is a Flat Football" is a series of posts which will explore players from Michigan football history members of the 2015 team resembles the most. Tackled in these posts will be the offensive "skill" position groups: Quarterbacks, Running Backs, and Receivers/Tight Ends. My apologies go out to the offensive line, but it's very difficult to get o-line statistics, and more difficult to compare the groups. I used Python and Pandas almost exclusively for this quick trip to the past. Any "predictions" can be described as unscientific, but kind of fun.


Disclaimer: Obviously caveats do apply here. These are namely the effects of other position groups, coaching, and style of offense on the players being analyzed. Also, the past probably has no bearing on what current players will do, unless you believe Rust Cohle. I plan to deal with these issues by completely ignoring them. It's the off season, people.


Michigan has an interesting mix of running backs this year. Junior backs Derrick Green and De'Veon Smith were highly rated coming out of high school, but neither have locked up the feature back role at this point. Drake Johnson is heading into his senior year after a promising junior season which unfortunately ended after a knee injury. Newly eligible Ty Isaac will be a RS sophomore after taking a year off following his transfer from USC. It's a crowded but talented backfield, and at this point, not much separates them. Let's take a look at their stats, gathered from Here are their stats throughout the years they have been active.  


Yr Cls Player Rush Att Rush Yds Rush Avg Rush Td Rec Rec Yds Rec Avg Rec Td Plays Tot Yds Tot Avg Tot TD Tot
2013 1 Derrick Green 83 270 3.3 2 0 0 0 0 83 270 3.3 2
2013 1 De'Veon Smith 26 117 4.5 0 0 0 0 0 26 117 4.5 0
2013 1 Ty Isaac 40 236 5.9 2 4 57 14.3 0 44 293 6.7 2
2014 2 De'Veon Smith 108 519 4.8 6 3 26 8.7 0 111 545 4.9 6
2014 2 Derrick Green 82 471 5.7 3 2 26 13.0 0 84 497 5.9 3
2014 3 Drake Johnson 60 361 6.0 4 1 11 11.0 0 61 372 6.1 4


We can also look at the totals for each player:  


Yr Cls Player Rush Att Rush Yds Rush Avg Rush Td Rec Rec Yds Rec Avg Rec Td Plays Tot Yds Tot Avg Tot TD Tot
2014 2 Derrick Green 165 741 4.5 5 0 0 0 0 167 767 4.6 5
2014 2 De'Veon Smith 134 636 4.65 6 0 0 0 0 137 662 4.7 6
2014 3 Drake Johnson 60 361 6 4 1 11 11 0 61 372 6.1 4
2013 1 Ty Isaac 40 236 5.9 2 4 57 14.3 0 44 293 6.7 2


  Green and Johnson were each having promising seasons last year before going down with injuries. De'Veon Smith put up a relatively good season, especially considering he spent most of the year splitting carries with Green and Johnson. Isaac's freshman year at USC was good for a freshman back who was not the featured guy. Let's find some comparisons to past Michigan running backs from past years.  


Derrick Green

Derrick Green

Yr Cls Player Rush Att Rush Yds Rush Avg Rush Td Rec Rec Yds Rec Avg Rec Td Plays Tot Yds Tot Avg Tot TD Tot
1984 2 Bob Perryman 76 393 5.2 5 2 15 7.5 1 78 408 5.2 6
1991 1 Tyrone Wheatley 86 548 6.4 9 10 90 9.0 0 96 638 6.6 9
2000 1 Chris Perry 77 417 5.4 5 0 0 0 0 77 417 5.4 5
2007 2 Carlos Brown 75 382 5.1 4 0 0 0 0 75 382 5.1 4
2014 2 Derrick Green 82 471 5.7 3 2 26 13.0 0 84 497 5.9 3


Derrick Green's sophomore campaign ended after just 6 games, so his stats ended up looking like players coming off the bench. As you can see in the chart above, Green is in good company. Freshmen Tyrone Wheatley and Chris Perry are very similar to sophomore Green. No one on the list is a prolific pass catcher, which makes comparisons easier. Arguably the best metric to judge running backs by is Rush Avg, AKA yards/carry. Let's see who's similar here, and throw in TDs just to compare.



Green's 5.7 Yds/Carry looks very similar to freshman Chris Perry's 5.4 average. Freshman Tyrone Wheatley's 6.4 Yds/Carry represents the top of the comparisons, and he was much more of a TD vulture than Green has been. Carlos Brown's sophomore campaign looks somewhat similar as well. Let's see how these running backs fared in their next year.  


Yr Cls Player Rush Att Rush Yds Rush Avg Rush Td Rec Rec Yds Rec Avg Rec Td Plays Tot Yds Tot Avg Tot TD Tot
1985.0 3.0 Bob Perryman -11.00 -154.0 -1.5 -5 5 61 3.4 -1 -6.00 -93.00 -0.80 -6.0
1992.0 2.0 Tyrone Wheatley 99.00 809.0 0.9 4 3 55 2.2 3 102.00 864.00 1.00 7.0
2001.0 2.0 Chris Perry 35.00 39.0 -1.3 -3 0 0 0 0 41.00 85.00 -1.10 -3.0
2008.0 3.0 Carlos Brown -46.00 -260.0 -0.9 -4 0 0 0 0 -40.00 -237.00 -1.00 -4.0
  2.5 Mean 19.25 108.5 -0.7 -2 4 58 2.8 1 24.25 154.75 -0.48 -1.5


Every running back outside of Tyrone Wheatley saw a decrease in their Yds/Carry. The average running back saw an increased workload of about 20 carries, good for an extra 100 yds. Coach Wheatley is really skewing the numbers here. He took the leap from "damn, that guy's good for a freshman" to "damn, that guy's good". This is the type of jump we are hoping for with Green.  

If I had to make one prediction based on this data, I'd say that Green's Yds/Carry will go down this year. Should he win the feature back role, I have no doubt that he'll be relatively consistent. However, the progression for the comparable backs above shows that if you're not Tyrone Wheatley (and most people aren't) you'll come back to earth.  

Bottom Line: Derrick Green should have a season similar to Sophomore Chris Perry (2001).  

  Yr Rk Player Cls Rush Att Rush Yds Rush Avg Rush Td Rec Rec Yds Rec Avg Rec Td
  2001 2 Chris Perry 2 112 456 4.1 2 6 46 7.7 0


The 2001 team won 8 games and lost 4. Chris Perry split carries that year with B.J. Askew, who had a good season as well. At this point it's tough to see much separation between all four candidates for the feature back role. Barring a surprise breakout, this should translate to a running back by committee simply for the sake of fresh legs.  

De'Veon Smith

De'Veon Smith, thanks to


De'Veon Smith is another applicant for the feature back position. In the wake of losing Green to a broken clavicle, Smith saw the most carries on the 2014 team. However, Drake Johnson started stealing carries towards the end of the year before he too went down with an injury. Let's see to whom Smith is most comparable.  


Yr Cls Player Rush Att Rush Yds Rush Avg Rush Td Rec Rec Yds Rec Avg Rec Td Plays Tot Yds Tot Avg Tot TD Tot
1984 1 Jamie Morris 118 573 4.9 2 14 131 9.4 0 132 704 5.3 2
1993 2 Ed Davis 93 441 4.7 2 11 89 8.1 0 104 530 5.1 2
1997 1 Anthony Thomas 130 529 4.1 5 21 205 9.8 0 151 734 4.9 5
2001 2 Chris Perry 112 456 4.1 2 6 46 7.7 0 118 502 4.3 2
2008 1 Sam McGuffie 118 486 4.1 3 19 175 9.2 1 137 661 4.8 4
2014 2 De'Veon Smith 108 519 4.8 6 3 26 8.7 0 111 545 4.9 6


The comparison which leaps out is Smith's sophomore season to Jamie Morris' freshman season. Attempts and Rush Avg are very similar. Morris was more involved in the passing game than Smith, but their Rec Avgs are similar (Smith's sample size is miniscule, though). Let's explore the similarities graphically because we can! DSmithComps

In addition to Jamie Morris, De'Veon Smith looks a lot like sophomore Ed Davis with more TDs. De'Veon Smith, much like Green, is in good company with the other players included in the comparisons. Chris Perry and Anthony Thomas were eventually drafted in the 1st and 2nd rounds, and Sam McGuffie jumped over multiple guys (a habit he has yet to break). McGuffie left after one year, but let's see how the rest of the guys progressed.  


Yr Cls Player Rush Att Rush Yds Rush Avg Rush Td Rec Rec Yds Rec Avg Rec Td Plays Tot Yds Tot Avg Tot TD Tot
1985 2.0 Jamie Morris 79.0 457.0 0.30 1.0 19.00 85.00 -2.90 1.00 98.0 542.0 0.10 2.0
1994 3.0 Ed Davis -13.0 -102.0 -0.50 1.0 -2.00 -53.00 -4.10 0.00 -15.0 -155.0 -0.90 1.0
1998 2.0 Anthony Thomas 16.0 232.0 1.10 7.0 -6.00 -65.00 -0.50 0.00 10.0 167.0 0.70 7.0
2002 3.0 Chris Perry 155.0 654.0 0.10 12.0 8.00 110.00 3.40 0.00 163.0 764.0 0.20 12.0
  2.4 Mean 59.25 310.25 0.25 5.75 4.75 19.25 -1.03 0.25 64.00 329.50 0.03 5.5


Most of these players took a major leap forwards in multiple stats. It seems as though once a player has reached a De'Veon Smith level of contribution, the next year they are expected to take on a more significant role in the offense. The average running back got about 60 more carries, 310 more yards, and gained 0.25 more yards/carry. The players most similar to Smith (Morris and Davis) represent opposite trajectories. Morris became the feature back and would keep that role until he graduated. Ed Davis continued to split carries until he graduated.


Bottom Line: If he wins the feature back role, De'Veon Smith could have a season similar to Sophomore Jamie Morris (1985).


Yr Cls Player Rush Att Rush Yds Rush Avg Rush Td Rec Rec Yds Rec Avg Rec Td Plays Tot Yds Tot Avg Tot TD Tot
1985 2 Jamie Morris 197 1030 5.2 3 33 216 6.5 1 230 1246 5.4 4

  Morris had a great season by all accounts. Most impressive was his 5.2 yards/carry and the first of three straight 1,000 yd seasons. The 1985 team went 10-1-1 and finished ranked #2 in the country, with Morris being a large part of the offense. Past Wolverines show Smith to be the running back in the most prime position to break out.

Drake Johnson

Drake Johnson, courtesy of CBS

Drake Johnson came on strong last season before going down with a knee injury. By all accounts he's been putting in the work to be ready for this season, and with his vision he is definitely in the hunt for the feature back role. Let's see who Johnson's 2014 season was reminiscent of.  


Yr Cls Player Rush Att Rush Yds Rush Avg Rush Td Rec Rec Yds Rec Avg Rec Td Plays Tot Yds Tot Avg Tot TD Tot
1977 2 Roosevelt Smith 57 308 5.4 4 5 46 9.2 0 62 354 5.7 4
1989 3 Allen Jefferson 65 380 5.8 3 3 27 9.0 1 68 407 6.0 4
1992 1 Ed Davis 61 374 6.1 3 3 11 3.7 0 64 385 6.0 3
2007 2 Carlos Brown 75 382 5.1 4 0 0 0 0 75 382 5.1 4
2009 1 Vincent Smith 48 276 5.8 1 10 82 8.2 2 58 358 6.2 3
2010 3 Michael Shaw 75 402 5.4 9 10 75 7.5 0 85 477 5.6 9
2011 3 Vincent Smith 50 298 6.0 2 11 149 13.5 2 61 447 7.3 4
2014 3 Drake Johnson 60 361 6.0 4 1 11 11.0 0 61 372 6.1 4


Drake Johnson's profile is similar to just about every change of pace back Michigan has had in recent years, minus the screens. Ed Davis pops up again, and Vincent Smith appears twice. Perennial change of pace backs Carlos Brown and Michael Shaw also appear. The most apt comparison seems to be Ed Davis' freshman campaign, followed by Vincent Smith's junior year. You know the drill: let's look at the progress they made from the comparison season to the next.  


Yr Cls Player Rush Att Rush Yds Rush Avg Rush Td Rec Rec Yds Rec Avg Rec Td Plays Tot Yds Tot Avg Tot TD Tot
1978.00 3.00 Roosevelt Smith 41.00 102.00 -1.20 -1.00 4 69.00 3.60 3.00 45.00 171.00 -0.80 2.00
1990.00 4.00 Allen Jefferson -13.00 -111.00 -0.60 3.00 -1 -5.00 2.00 -1.00 -14.00 -116.00 -0.60 2.00
1993.00 2.00 Ed Davis 32.00 67.00 -1.40 -1.00 8 78.00 4.40 0.00 40.00 145.00 -0.90 -1.00
2008.00 3.00 Carlos Brown -46.00 -260.00 -0.90 -4.00 0 0 0 0 -40.00 -237.00 -1.00 -4.00
2010.00 2.00 Vincent Smith 88.00 325.00 -1.40 4.00 5 48.00 0.50 0.00 93.00 373.00 -1.40 4.00
2011.00 4.00 Michael Shaw -44.00 -203.00 1.00 -6.00 -9 -63.00 4.50 0.00 -53.00 -266.00 1.00 -6.00
2012.00 4.00 Vincent Smith -12.00 -204.00 -3.50 0.00 -1 -75.00 -6.10 -1.00 -13.00 -279.00 -3.80 -1.00
  3.14 Mean 6.57 -40.57 -1.14 -0.71 1 8.67 1.48 0.17 8.29 -29.86 -1.07 -0.57


The basic gist of this table is that many of these backs have reached their ceiling, and might actually take a step back year-to-year in Rush Avg. While the mean Rush Att and Rush Yd changes are +6.57 and -40.57, respectively, the running backs themselves seem to fall into two categories. The first category includes freshman to sophomore Vincent Smith, freshman to sophomore Ed Davis, and sophomore to junior Roosevelt Smith. Each of these backs saw a significant increase in their carries and a decent uptick in yards. The second group saw just as significant a decrease in usage between years.  

The question of which of these groups Drake Johnson will fall into is difficult to answer. Perhaps the most relevant distinction between these two categories is Class. Most upperclassmen with Drake Johnson level production the previous year saw a decrease in touches the following year. All seniors experienced this decrease, as did junior Carlos Brown, although his decrease was entirely injury related. Only junior Roosevelt Smith did not see lower numbers. The two underclassmen, Ed Davis and young Vincent Smith saw increased usage.  

Bottom line Given that he's going into his ***RS Junior year, I'm most inclined to say Drake Johnson's final year will mirror that of Senior Michael Shaw (2011).  ***Thanks for the correction.

Yr Cls Player Rush Att Rush Yds Rush Avg Rush Td Rec Rec Yds Rec Avg Rec Td Plays Tot Yds Tot Avg Tot TD Tot
2011 4 Michael Shaw 31 199 6.4 3 1 12 12 0 32 211 6.6 3


The 2011 went 11-2 and won the Sugar Bowl. Shaw saw a small share of the rushing load with Denard, Fitz Toussaint, and junior Vincent Smith getting more carries. Carries were difficult to come by for Shaw behind these backs, a problem which will also be faced by Johnson.  

Ty Isaac

Ty Isaac, expertly photo shopped by someone here at MGoBlog


Ty Isaac is the outside challenger this year. After a decent freshman season with USC, Isaac transferred to Michigan, finally completing the Justin Fargas trade. Isaac's 2013 freshman season was similar to a few familiar names.  


Yr Cls Player Rush Att Rush Yds Rush Avg Rush Td Rec Rec Yds Rec Avg Rec Td Plays Tot Yds Tot Avg Tot TD Tot
2006 1 Brandon Minor 42 238 5.7 2 1 9 9.0 0 43 247 5.7 2
2008 1 Michael Shaw 42 215 5.1 0 6 32 5.3 1 48 247 5.1 1
2009 1 Vincent Smith 48 276 5.8 1 10 82 8.2 2 58 358 6.2 3
2013 1 Ty Isaac 40 236 5.9 2 4 57 14.3 0 44 293 6.7 2


If you've made it this far, you're acquainted with two of these names already. Freshmen Michael Shaw and Vincent Smith were similar to Isaac in most stats. Brandon Minor's first year was similar to Isaac's in many ways as well. Let's see how the sophomores stack up.  


Yr Cls Player Rush Att Rush Yds Rush Avg Rush Td Rec Rec Yds Rec Avg Rec Td Plays Tot Yds Tot Avg Tot TD Tot
2007.00 2 Brandon Minor 48.00 147.00 -1.40 -1.00 2 -8.00 -8.70 0.00 50.00 139.00 -1.50 -1.00
2009.00 2 Michael Shaw 0.00 -30.00 -0.70 2.00 -4 -27.00 -2.80 -1.00 -4.00 -57.00 -0.80 1.00
2010.00 2 Vincent Smith 88.00 325.00 -1.40 4.00 5 48.00 0.50 0.00 93.00 373.00 -1.40 4.00
  2 Mean 45.33 147.33 -1.17 1.67 1 4.33 -3.67 -0.33 46.33 151.67 -1.23 1.33


Minor and Smith saw upticks in usage and yards, but all three regressed a little in Yds/Carry. I really do not believe that Isaac will be like one of these guys, simply because I'm pretty convinced he'll get at least a third of the carries.

Bottom Line: No idea. Ty Isaac is a bit of an unknown at this point. Just over 40 touches on a different team is not much to go on. Isaac's his build suggests he'll be more Minor than Smith, but his pass catching ability seems closer to the latter than the former.  

What Does It All Mean?

  I'm not sure that Michigan has a running back on the roster who is ready to be "the guy". Derrick Green and De'Veon Smith have each shown flashes of being able to handle the entire load, and Johnson looked good in limited action last year. Isaac had a promising freshman season at USC, but it's difficult to know just what he is. A look at the Yds/Carry shows reveals strong numbers for each back. RbCurAvg

These numbers are even more impressive when plotted against every Michigan running back with greater than 20 carries since 1975. The x-axis is the Class (1-freshman, 2-sophomore, etc) and the y is Yds/Carry.


The dotted line represents the absolute average Yds/Carry, and all four backs are well above the line. Those with fewer carries are well above the line, but even De'Veon Smith's >100 carry season last year put him above average. It looks like Michigan has at least four solid backs, meaning that at the very least we will have a strong committee. At best we might have the next Chris Perry or Jamie Morris. Just who that might be is impossible to answer at this point.  

Rushing statistics and the offensive line

Rushing statistics and the offensive line

Submitted by ST3 on October 7th, 2014 at 3:00 PM

In my most recent diary, I claimed that the offfensive line was not terrible. I'm still seeing people claim that they are, so I thought I'd check some statistics. I prefer quantitative analysis over qualitative conjecture. The first thing I checked was individual rushing yards per carry.…

There are 124 teams in I-A. De'Veon Smith is 50th in I-A with 6.0 yards per attempt. Derrick Green checks in at #71 with 5.7 yards per attempt. So that "terrible" offensive line has allowed not one, but two running backs to be in the top 75. That's not great, but it's not terrible either. I'd say it's about average.

The other criticism you see is something like this, "of course we gained yards, we were playing Rutgers." So let's look at our opponents.…

Utah is 14th in yards per attempt, and ND is 17th. Those are two pretty good defenses. Minnesota is 45th and Rutgers is 63. Those are two pretty average defenses. And last but least, Miami (OH) is 96 and Appy State is 119. Those are two pretty bad defenses.

So in conclusion, the running game has been average against average defenses.

We are 2-4 because of turnover margin and some head-scratching coaching decisions. I think 5-7 or 6-6 are still real possibilities. Which is not what I expected coming into the season, but the offensive line is not the reason why we are underperforming expectations.

Dilithium Quantified

Dilithium Quantified

Submitted by MCalibur on August 19th, 2014 at 4:39 PM

[Author note: This thing is long and pretty technical. That said, I think there will be  sufficient payoff and value for you the reader. Still, be ye warned.]

madscientist Have you ever wished there were a convenient way to rate rushers the same way we rate passers? Sure, passer rating has its weaknesses—all mathematical formulas do—but despite it's issues, I've come to appreciate passer rating as a very useful framework to evaluate a player/team when it comes to passing the ball. In the same way that finding a corner piece to a jigsaw puzzle helps you figure out it's entire quadrant, once you have an idea of what to expect from the passing game you can leap to other touchstones to determine what to expect from the running game. A rusher rating would be just the sort of touchstone needed to really start messing around for those of us who are so inclined. This diary lays out what I think should work for these purposes.

To recap some of my previous work: passer rating combines four important factors—completion percentage, yards per attempt, interception rate,  and touchdown rate—and blends them into one number. For rushing stats, important information for coming up with an analogous metric has been hard to come by until came along. Tons of fascinating and useful data, for free. God bless the internet.

To come up with the rating, I looked only at positions that would be considered normal rushers (QB, RB, TB, FB, HB, SB, WR) that have an average YPC greater than zero. If you can’t meet those criterion, then you cant represent a normal rusher, thus sayeth the me. Other positions register rushing attempts but allowing the odd rush by a punter to color your view of what normal looks like would be dumb. See the chart below for more information. Also, if a guy averages negative YPC, uh, find something else to do, kthx. Other than that, no other filter was applied but some math wonk tricks were and I’ll talk about those as we go.

data discrimination

Parameter Mapping

Completion Percentage → Gain Percentage : Parsed play by play is necessary to generate a replacement for completion percentage. I opted to go for Gain Percentage: the percentage of attempts that resulted in more than zero yards. I figured the basic goal of a pass is to complete it (brilliant insight, I know) and the basic goal on a rush attempt is to gain positive yardage so…any gain of more than zero yards is mission accomplished. This parameter is as much about team skill as it is about player skill but the same can be said for Completion Percentage.

MikeHartYards Per Attempt: Direct analogue.

Interception Rate → Fumble Rate: The direct analogue would of course be fumbles lost per attempt but that’s not the right way to do it IMO.  The luck factor that influences whether or not the team actually loses possession has nothing to do with the fact that bringing possession into question is a terrible idea. So, all fumbles whether lost or not  are counted in the calculation.

There is also a bit of mathematical wonkiness deployed as well. Mike Hart is famous—at least around here—for his deftness at protecting the rock. It was awesome: 991 carries, 5 loose balls, 3 losses of possession. That was an aiight career, but these guys were kinda, sorta, maybe, better (!) at protecting the rock:

Player Team Att FMB
Jacquizz Rodgers Oregon State 789 1
Javon Ringer MSU 843 3
Montee Ball Wisconsin 924 4

OK, so the wonkiness…a lot of people who register meaningful rushing attempts do so at a pretty low level of opportunity. Even stud RBs often split carries with other backs: Eddie Lacy siphoned off carries from Mark Ingram before becoming the man, and T.J Yeldon  did the same to Eddie Lacie. So in order for fumbles to make sense for players that get meaningful carries in low doses,  we need to consider the question: at which point does a low fumble rate cross the threshold from wait-and-see to holy-crap-check-that-dude-for-stickem?



What we have here is a chart comparing the observed percentage (red dots)  and the mathematical probability (blue line) that a player will have at least 1 fumble versus the number of carries he has registered. The red dots are binned in increments of 1 so the sample sizes out past 150 are pretty thin but if bigger bins are used, you’d see a scatter of points that more closely follow the mathematical fit, because… math. The blue line was derived using logistic regression.

The weirdness at zero for the mathematical expectation might be concerning as it suggests that there’s a 20% chance you’ve fumbled despite not having a single carry to your credit. However, that is just an artifact of the data. It is possible to fumble on your one and only carry as actual observations show. What the math does, though, is it considers the sample size of the observations and then finds the best fit possible to the overall dataset. There are ways of dealing with that issue, but…I rather talk about football. Also, KISS. This is good enough for my intended purpose.

Anyway, the point of doing all that is it allows me to apply what I’ll call the Phantom Protocol. Basically, I take that curve, subtract it from 1, and add the resulting value to the player’s fumble total. As the number of carries increases, the effect of the phantom fumble recedes thus leveling the playing field and letting us evaluate players with low sample size as best we can. The result of this bit of data manipulation is that a guy with no fumbles in 16 carries is assigned an average fumble rate and by the time 100 carries are registered, the penalty is not perceivable. Below 16 carries, the assigned penalty is pretty stiff but this trick levels the playing field to let us look at guys with few carries and not just dismiss them with the low sample size red card. Sure, 16 carries is still a low sample but at least the rating self corrects for the fact that fumbles take time to manifest.

Most importantly though, the protocol adequately acknowledges players with low fumble rates even though they have a lot of carries. It’s easier to have a 1% fumble rate after 100 carries than it is to have the same rate after 789 carries.  That said, after a while the fumble rates should be allowed to speak for themselves. Quizz Rodgers and Mike Hart need their proper allocation of DAP; nothing more, nothing less. I think the ghost protocol concept accomplishes exactly that.

Touchdown Rate: This one is also directly analogous but here again I’ve deployed the ghost protocol to credit guys with low sample the expectation of an eventual TD. TDs come about much more freely than fumbles do with goal line attempts and the like so this credit vanishes very quickly. But fair is fair: the protocol giveth and it taketh away.

Those are the components directly analogous to the ones used in passer rating and these would be enough to go about the business at hand. However, whereas a passer’s job is to get the ball into the hands of a play maker, players that are given the ball whether by pass of handoff are called upon to be the playmaker. Certainly the scheme, play call RPS, and execution of the supporting cast all have major influence on the results of a play but the ball carrier can do things that elevate the call from good to great. I wanted to be all formal-like and call this the Impact Run Rate but this [stuff] is s’posed to be fun, man. Hence—

Another Dimension: the Dilithium Quotient


The 20 yard threshold is usually referenced as registering a play as a big play. That would certainly qualify as a big play by any standard but that threshold seems to have been established somewhat arbitrarily in my opinion. On average, a generic runner on a generic team in a generic game gains about 4 yards per attempt with a standard deviation of about 7.5. Its called the standard deviation for a reason as a huge swath of observations (about 2/3rds) occur within 1 SD of the mean, or between –3 and +11 (remember: discrete data). The other 1/3 of observations get split evenly with 1/6 below -3 yards and 1/6 above 12. I’ve used objective criterion, you know, math, to define Impact Runs as those that register 12 yards or more. To register one of these the player’s entire team has to execute the play correctly, then the carrier he has to do something special (i.e. juke a dude, break a tackle, be fast). This is the real life manifestation of the Madden Circle Button and its informative. It’s the difference between Barry Sanders and Emmitt Smith.

Denard Robinson was great at this but it might be surprising to hear that he wasn’t the best. Percy Harvin in the spread option was ridiculous in this category. Percy had touched the ball a lot when he was a Gator and 27% of the time, he darted for an impact run. By Contrast, Denard’s DQ% was ‘only’ about 15%. Could you imagine Denard breaking loose almost twice as often? Of course, the scheme, the team’s execution of the scheme, and the player’s deployment within the scheme has a lot to do with this number. Florida circa Percy Harvin was galaxies away from Michigan circa Denard Robinson. Percy Harvin was the 3rd rushing option in Florida’s spread and shred, Denard Robinson was options 1-10. Also, being the QB in the spread-option means you are concern #1 for defenses: the cornerstone. That was triply the case when facing Michigan with Denard in the captain’s chair. Harvin was usually one-on-one with a guy 10 times slower than he was who was also probably pooping his pants.

Denard’s DQ% was pretty stable around 15% (scheme be damned) but his utility rate (723 career carries) was second to none save minor conference QBs. His closest proxy Pat White (684 career carries) broke loose at a 19% clip in RichRod’s Scheme.  However, the Big EEEast sans Miami and Virginia Tech wasn’t quite the Big TEEEN. Denard went up against stout defenses way more often than Pat White did and did so without the benefit of Steve Slaton or Noel Devine and the benefit of a revolutionary offensive scheme. When Pat White lost RichRod is DQ% dropped to under 12%, Denard didn’t bat an eye. Everyone *knew* they had to stop Denard and only him on *every play* and they still had their hands full trying to actually do it. The fact that Michigan could never position itself for him to win the Heisman trophy will always be one of my sports fan laments. For ever and ever and ever.  He better get a Legends Jersey or I’m qui’in’. I don't care if that’s silly. You’re silly. Where’s my bourbon?

Blending It All Together

RBRatCoef Passer Rating was developed such that an average QB would end up with a rating of 100 according to the data set that was used to develop it, which was gathered two maybe three football eras ago when linemen couldn’t really block and scholarship limits weren’t so much. I’m not sure how they went about the process of pinning the rating to average==100 and I don’t have the data to try an replicate the results…so, I kinda, sorta, you know, pulled something outta my [hat]. That is to say: I did what I think is correct or at least valid. I normalized each parameter by it’s par value, summed them together, then forced resulting rating to equal 100. Ultimately the 100 thing is completely arbitrary, but negative numbers are weird, I guess. All said, a rating of 100 means the player was a solid runner but not special, below that you wonder if he should be running at all.

Where in the World is Carmen San Diego Mario Mendoza

mendoza Now that we have a calibrated formula its time to get down to business, application. I calibrated the rating so that 100 was a normal guy, but to figuring out what par should be is a little more complicated. I mentioned earlier that if you cant get to a rating of 100 I don't think you should be a primary running option and I also think we should only look at primary running options to establish our benchmark. But being a primary running option means different things depending on where you’re lining up.

When trying to crack a nut like this I often find that the data itself will help you figure out where to chop it. In the chart below I have plotted Average Rating vs. Amount of Carries. Obviously, the better runner you are, the more carries you should see but runners that are REALLY good are few and far between…this chart shows that dichotomy very nicely. I like to look for population gaps and/or inflection points in a performance curve. Those usually a good places to drop an anchor as far as I’m concerned. When they are near each other it’s a dead giveaway. Based on the data itself I’m using 115 for RBs, 70 for QBs, and 120 for WR as performance benchmarks.




Laugh Test

So, this is all well and good but the real test is whether or not things make sense. Here the values for the B1G in 2013:

Team Name Player Name RB Rat Attempt Yds/ATT TD% FMB% Gain% Dillitium%
OSU C. Hyde 188.35 208 7.31 0.072 0.005 0.942 0.135
IND T. Coleman 182.49 131 7.31 0.092 0.015 0.832 0.137
WISC M. Gordon 172.71 206 7.81 0.058 0.015 0.888 0.150
WISC J. White 169.86 221 6.53 0.059 0.000 0.810 0.122
IND S.Houston 157.12 112 6.72 0.045 0.009 0.786 0.152
NW T. Green 146.27 138 5.33 0.058 0.000 0.841 0.087
ILL J.Ferguson 141.77 141 5.52 0.050 0.007 0.816 0.113
MSU J. Langford 129.16 292 4.87 0.062 0.007 0.849 0.065
NEB A. Abdullah 116.18 281 6.01 0.032 0.018 0.875 0.100
MICH F. Toussaint 114.60 185 3.50 0.070 0.011 0.676 0.070
MINN D. Cobb 112.55 237 5.07 0.030 0.008 0.827 0.084
PSU Z. Zwinak 109.68 210 4.71 0.057 0.014 0.867 0.052
IOWA M. Weisman 106.12 226 4.31 0.035 0.004 0.832 0.058
PSU B. Belton 99.05 157 5.11 0.032 0.019 0.854 0.083

This generally looks pretty reasonable to me in terms of an overall ranking as well as a relative ranking. The players/team you’d expect to be at the top and bottom of the list are where they are supposed to be. If anything I’d criticize the Mendoza line at 115 given how we all feel about Michigan’s running game last year. Maybe 115 is just the threshold of suicide and 130 or better is what we fans really want from our teams. But, even this jibes with what I think.

As with passer rating, this rating depends on player skill, surrounding support, and offensive scheme. Toussaint’s YPC and Gain%—components heavily influenced by surrounding support (i.e. the O-Line)—are way under par. So is his Dilitium % which is a skill/talent/speed thing but the dude had a bum knee and he’s not that far off of par there. Makes sense. So, he hit the Mendoza line even though he had bad support in front of him, sorta like Gardner. These numbers make sense to me.

Re: Smith Vs. Green

I mentioned in my last diary that it was interesting to hear grumblings about De'Veon Smith being ahead/competitive with Derrick Green because I think the numbers bear this out. Check this out:

Player Name Att TD Fum Gain % Yds/ATT TD% Fum% DIL% RB Rat
F. Toussaint 185 13 2 0.676 3.50 0.070 0.011 0.070 114.60
D. Green 83 2 0 0.723 3.25 0.025 0.004 0.048 83.42
D. Smith 26 0 0 0.769 4.50 0.015 0.023 0.077 73.05

These guys played with the same support and in the same system so the differentiators on display here are essentially Skill and Opportunity. Neither Green nor Smith actually registered a fumble but the Ghost Protocol affect Smith’s rating more because he has far fewer carries. Indeed, Smith’s rating is also bolstered by a phantom touchdown, but this effect dissipates faster because TDs occur more frequently. So the math is screwing Smith over here a bit. Meanwhile, Smith’s Gain % and YPC (hitting the right hole at the right time) and DIL% (juking, speed, whatever) were the highest on the team last season. Yep, Small samples yadda yadda. Just sayin’.

Anyway, that's a lot of words and I hope this was worth the read. Of course, I will be referring to this information in future diaries. Thanks for reading and let please provide and criticisms or comments you might have in, uh, the comments section.

11 Days.

DeVeon Smith Slightly Ahead of Green Coming Out of Spring

DeVeon Smith Slightly Ahead of Green Coming Out of Spring

Submitted by Space Coyote on July 29th, 2014 at 11:32 AM

Hoke said today that De"Veon Smith had a bit of an edge over Derrick Green coming out of spring. Reminds that Drake Johnson back

— angelique (@chengelis) July 29, 2014


Also, here is says it is mostly because of pass pro.


Per GoBlueinMN below (so give him a +1 for this info below; put in OP to allow discussion) and Nick Baumgardner on twitter, who also explains that others beside Norfleet will get a look at returns because the team must get better, and says Kenny Allen will now do kickoffs:

Hoke says Glasgow is going to play center. Sounded like his shuffling to other spots is going to slow down.

— Nick Baumgardner (@nickbaumgardner) July 29, 2014