Michigan - still a far outlier in opp offensive holding penalties

Submitted by CarrIsMyHomeboy on October 1st, 2018 at 9:58 AM

https://mobile.twitter.com/jakeb4813/status/1046491152692785153

Summary from the last 4 years: At 0.76 holds called per 100 plays and 0.58 accepted per 100 plays, Michigan is a far outlier. And in treating it as a mere tally, Michigan's 11 holds across the last 4 makes Michigan's sum the only one that is unbunched from the group.

Unoriginal but true reasoning: Rampant DLs theoretically induce more such opponent holding not a record amount fewer.

Acknowledgement that a ref conspiracy is not the ONLY explanation: (See before)

A way forward:

I think it's OK for Michigan to politely demand an answer. And, since that answer will almost assuredly not be "oh we're just systematically biased against you," it's also OK/smart to discuss a way to test or confirm the answer they do get. 

For example, if the Big Ten says it's just outrageously bad luck/coincidence, then Michigan should expect reversion to the mean - maybe even quickly. If the Big Ten asks its refs and says it may be more due to something structural about Michigan's defense, then Michigan can go back and count holding penalties when they were/weren't doing that thing (to confirm). 

Either way, nothing bad can come from going to the Big Ten and saying "hey, here's a weird/concerning observation; let's talk."

umich1

October 1st, 2018 at 2:25 PM ^

Thanks for sharing the initial link to when I did the analysis.  For those who want to read more, I also posted a follow up here https://mgoblog.com/diaries/offensive-holding-b1g-and-beyond-part-deux.

I did share the analysis with John Bacon, who I'd call a friend, as well as another player on the football team. I assume at this point Harbaugh is aware.

I ended up doing a lot more analysis that never got posted.  For example:

  • Michigan's trend of getting shafted on holding goes back to 2010, with 2011 being the only year where Michigan was in the middle of the pack.
  • I tried to do an analysis of penalties by referee, including holding. The sample sizes just got too small, with varying opponents, to get any signal among the noise.  O'Neill actually graded out in the middle of the pack of B1G referees.

This will always be the image I use to illustrate how screwed Michigan has been:

3 B1G Holding.png

SAMgO

October 1st, 2018 at 10:04 AM ^

Nothing more to add beyond emphasizing how actually insane these numbers are. I'm not a ref conspiracy guy but I DO believe there is something to the O'Neil crew having an internalized bias against Michigan and that this needs to be addressed by our AD with the conference. Clear statistical outlier when, if anything, they should be an outlier the other way given the strength of our DL.

mlax27

October 1st, 2018 at 10:08 AM ^

I feel like this should be updated each week, posted and copy the big 10 each week, maybe the BTN crew.  Make it publicly known.  Michigan may not be able to realistically ask for an explanation, so it may take some work to get it noticed by someone who matters.

TrueBlue2003

October 1st, 2018 at 11:47 AM ^

The interesting thing about this chart is that the two highest teams here are two of the worst teams in the conference over this time period.

Maybe refs feel bad for bad teams?

The per game numbers mean nothing, btw, since better teams get off the field and defend fewer plays.  Certainly surprising that we're still the lowest on a per play basis but it's not as bad as the per game numbers look.

Still doesn't explain why we're so much lower than other good teams, but the chart is interesting.

MichiganTeacher

October 1st, 2018 at 12:21 PM ^

Hm. The part where you say "better teams get off the field and defend fewer plays" makes me think "Oh, yeah, we do tha-" until I remember our offense over the last few years and I think, "Wait, maybe our defense is actually on the field for more plays than a lot of defenses."

Tempo, schedule, etc. all play roles too, obviously.

TrueBlue2003

October 1st, 2018 at 1:31 PM ^

Well, we also play a snail's pace on offense so M plays fewer drives per game than most college teams.

And I agree that last year, our offense was bad and in some games, we probably did play more defensive snaps than the opposition (PSU and Wisconsin almost certainly). Still, for most of our games, even last year, we ground out long drives with the running game, got three and outs frequently (I think M leads the country in three and outs over the past few years) and didn't play many defensive snaps in most games.

In 2015 and 2016, our offenses weren't elite but they were fine, so we almost certainly didn't play many defensive snaps those years.

So compare the charts.  The per game data has FIVE really bad teams at the top when you just consider the raw numbers.  That's because they have to defend a lot of plays.  Then Wisconsin, MSU and OSU shoot up the list on a per play basis, and Michigan isn't as egregiously an outlier (although still bad), because all those teams defend fewer plays than the average team.

Rutgers on the other hand, is top 5 in per game but plummets in per play because they defend so many plays.  Hence the per game data is meaningless, really.  Per play is much more instructive.

Michigan leads the country in three and outs over the past couple years or something like that.  The M defense doesn't play many plays.  Doesn't take them long to get off the field.  That's usually the hallmark of an aggressive scheme.  Opposite of bend don't break defenses.

Brhino

October 1st, 2018 at 10:07 AM ^

On the subject of penalties, but not worthy of its own thread:

I'm looking for video of the play against Nebraska where the Michigan runner suffered what appeared to be a blatant face mask.  No flag was thrown initially, but the officials huddled and then threw a flag.  Does anyone have that on video?

mgokev

October 1st, 2018 at 11:01 AM ^

I was at this game and the entire stadium saw the facemask and boo'd the refs incessantly. Only after it was put on the big screen and clear and blatant facemask (and the refs looked up and watched it) did they throw the flag. 

I'm convinced that if the stadium screen operator hadn't put that on the video board, it wouldn't have been called. It truly felt like a "oh shit, well now we have to call it" scenario. Not an "ah we must've missed it" scenario. 

Brhino

October 1st, 2018 at 11:37 AM ^

Thanks.  I forgot it was on a punt return - probably why I couldn't find it.

Some people have pointed out that the defender did not grab the face mask, but rather the top of the opening of the face hole, near the forehead.  Your still photo above would look to confirm this.  Some casual observers claim this means it wasn't a penalty, but they are wrong - the "facemask" penalty covers any opening of the helmet, not just the mask.

So I see two possible explanations for the penalty:

1. None of the refs saw the penalty, in spite of the blatantness.  That's bad.  They only threw the flag upon seeing the video scoreboard, a sort of unofficial video review. While we're glad they got the call right eventually, that's not something they're legally supposed to do, so that's bad too.

2. The refs observed the head yank but noted that the actual face mask was not grabbed.  All refs in position to throw the flag were not aware that the rule covers all helmet openings and thus did not throw it.  In the ref huddle someone pointed out the exact rule, and based on that standard, the flag was thrown.  That's bad too, because it suggests most of the refs on the field don't know all the rules.

 

lhglrkwg

October 1st, 2018 at 10:09 AM ^

Saw that on reddit this morning. I think it’s fairly outrageous that we - probably having had the best D-line in the conference over the last few years - are also getting the least number of holding calls. It’s like the refs have a subconscious inclination to ‘help’ some teams not others or in our case, it has to be a really egregious hold to warrant a flag. We should probably be drawing the most holding flags. Certainly not the least. 

pescadero

October 1st, 2018 at 11:43 AM ^

" probably having had the best D-line in the conference over the last few years - are also getting the least number of holding calls. "

Based on a look I did a year or two ago - that actually seems to be the trend across college football. Better DL tend to get LESS holding calls.

If I remember right, over the couple year period I looked at, Alabama was even getting significantly LESS holding calls than Michigan.

 

LeCheezus

October 1st, 2018 at 10:49 AM ^

I don’t blame you, he is quite memorable for all of the wrong reasons.  This is not just a Michigan thing.  O’Neal’s crew has a knack for missing blatant stuff and asserting themselves into the game at terrible times with “I guess you COULD call that but why now” calls.  They are consistently  inconsistent.

CTSgoblue

October 1st, 2018 at 1:16 PM ^

The O'Neill crew was 2015 MSU game.  Ugh.

And the 2016 Iowa game where they called us for a dubious targeting call, roughing the punter, and grazing a face mask that set up the game-winning FG.

And the 2015 Oregon State game for roughing the punter (that was running outside of the tackle box), which elicited the thrown clipboard from Harbaugh (the B1G later apologized for the wrong call). 

Are You Not En…

October 1st, 2018 at 10:12 AM ^

I am a huge hockey fan. https://scoutingtherefs.com is maybe my favorite site ever. I wish they would go slightly more in depth but they break each ref down in terms of statistics. Very cool stuff. So you know the refs taht call a higher percentage of minors, majors etc per game, what they are calling, etc.

Wish this was a thing for CFB.

VintageBlue

October 1st, 2018 at 10:13 AM ^

That leads me to wonder if Michigan's DL is explicitly coached not to exaggerate the impact of holds, ie, not flop.  If that's the case, the logic must be that if you flop and don't get the call you might take yourself further out of the play. 

Anyone else got a non-conspiratorial explanation (and I'm not saying it's not bias)?

LSAClassOf2000

October 1st, 2018 at 10:18 AM ^

Unoriginal but true reasoning: Rampant DLs theoretically induce more such opponent holding not a record amount fewer.

It's interesting to see the numbers because the phenomenon is truly strange when it comes to Michigan. You would think our DL would create far more offensive holds than it does, and indeed, especially this season, I am sure quite a few yards have been left out there as a result of these uncalled holds.

drjaws

October 1st, 2018 at 10:22 AM ^

I just wish other teams DBs got called for holding/PI.  Sure, Michigan grabs quite a bit more but blatant non-calls when the same is done to our WRs is infuriating as well.

Ghost of Fritz…

October 1st, 2018 at 11:45 AM ^

Most of the PI calls against M this year have been within what the rules define as PI.

The problem with PI calls in general is that there is great variance from week to week on how closely ref crews will enforce PI rules.

Some crews will call PI on some very minor stuff (that is arguably technically within the rule definition of PI).

Other crews will call PI only on egregious PIs and allow a lot of technical PI violations to slide. 

Gotta be consistent within games and also week-to-week.

And then there are the non-PI calls late in the 2016 M-OSU game...  Don't get me started.

drjaws

October 1st, 2018 at 12:04 PM ^

Yea most of the PI and holding calls against Michigan's secondary were legit. 

My argument is that when the other teams DBs are doing the same thing as Michigan's DBs, those flags aren't thrown nearly as much.  I see Michigan's receivers getting mugged and rarely is there a PI or holding call on the other teams DBs

OwenGoBlue

October 1st, 2018 at 10:28 AM ^

The idea of an anti-Michigan referee cartel is, of course, insane. The penalty/opponent penalty numbers are also insane.

This years-long gaslighting is not fun!