Monday, December 28, 2020

Anatomy of a Running Back Substitution, Part 2


Zach Charbonnet (image via Freep)

The other day I posted about Michigan's running back rotations (LINK). Today I will be addressing why running backs might get rotated.

Use the Amazon links here to support TTB (LINK):

There are a lot of reasons to rotate running backs from play to play. Ideally, you have a perfect, three-down running back who never gets hurt or makes mental mistakes . . . but those are rare.

Hit the jump for a discussion of eight reasons to sub in new running backs.

1. Fatigue: This one is pretty self-explanatory, but fatigue can set in for even the best trained athletes. Sometimes a guy needs a breather and some water, especially if he's being fed the ball over and over again. Or if he just broke off a 60-yard run.

2. Injury: Injury.

3. Discipline: I included this one not because it's a common issue, but there is a chance that any player might need to be reined in for unsportsmanlike behavior.

4. Coaching tips: This is rare that a running back would need to be removed from the game in the middle of a drive for a coaching tip, but occasionally, defenses do something to confuse a back in pass protection or on his run keys. But if you run an offense predicated on inside zone, for example, and a defense does a particular thing (charging the mesh, for example), the running backs coach might want to pull him out for a play or two to talk to or remind him of some things. Maybe the running back messed up on play #2 on the drive, and there's a good chance that inside zone will be run again two or three more times on that same drive and it needs to be addressed immediately.

5. Pass protection: This is a big one, and we have seen it play out many times for Michigan over the years. A few players who were excellent pass protectors: Mike Hart, Vincent Smith, De'Veon Smith, Zach Charbonnet. A few players who struggled in pass protection: Carlos Brown, Derrick Green, Ty Isaac. Michigan's coaches have made a lot of adjustments over the years, such as Brady Hoke bringing in Vincent Smith on third downs to help protect the quarterback.

6. Pass receiving skills: Speaking of Vincent Smith, it also helped that he was great at running and catching screens. Pass receiving backs need to have reliable hands, and they also need sufficient quickness to beat linebackers on routes and make somebody miss. A good receiving back often catches the ball in the flat, where he has to make at least one guy miss to gain any yardage.

7. Trickery and subterfuge: One reason to use a particular back for a play is his ability to throw the ball, hand off the ball, etc. Michigan does not typically run a lot of trick plays, but guys like Vincent Smith and Hassan Haskins have attempted passes in the past. Some guys also just have a better feel for pitching the ball (on a reverse or flea flicker) or handing off the ball (on a reverse or double reverse). Also, on 4th-and-2, you may want to use your best short yardage back to sell a hard fake on a play action pass.
Scenario: Karan Higdon goes for 9 yards on a 3rd-and-11 draw, setting up for De'Veon Smith to enter on 4th-and-2 for a play action pass.

8. Run scheme efficiency: I saved this for last, because I think it requires the most nuance when it comes to coaching and the analysis of substitutions. Different running backs are better at running certain schemes than others. Some examples from Michigan history:

  • Power: Anthony Thomas. Thomas had a good combination of patience, burst, and size/strength to run this play.
  • Counter: Anthony Thomas, Tim Biakabutuka. I couldn't choose between these two as being the best example, because I loved how they both set up counter. They got good depth and timed it up pretty perfectly, while also displaying that aforementioned patience, burst, and power.
  • Inside zone: Hassan Haskins. Haskins has the patience, vision, and toughness to run inside zone, and he also makes some really good cuts for being a bigger back (6'1", 220 lbs.).
  • Outside zone: Mike Hart. Hart displayed perhaps the best vision of any Michigan running back. His lack of speed could be frustrating at times, but he made so many people miss, found so many cutback lanes, and broke so many tackles that he could be very efficient.
  • Toss sweep: Ty Isaac. In recent memory, Isaac has been the best at running the toss. He seemed most comfortable at running outside the tackles, and had the vision/patience in space to make things happen with linemen pulling in front of him.
  • Pin and pull: Karan Higdon. Higdon was a slasher with pretty good speed. He made pretty good reads and, like the guys who ran power/counter well, he showed the patience and burst to gain chunks of yards.
  • Down G: Tim Biakabutuka. Biakabutuka was a soccer player growing up, and he wasn't the best at running between the tackles. Where he excelled was running off tackle, because while he had some power, he also had the quickness to bounce it all the way outside if necessary.
  • Iso: Tyrone Wheatley. I probably could have chosen Wheatley for a few of these categories, but he played almost 30 years ago, so memories might be vague. Listed at 6'1" and 225 lbs., he was probably the most physically talented back in Michigan history, taking into account his sprinter's speed.
  • Power read: Fitzgerald Toussaint. Toussaint was a short-ish back (5'10") whose combination of toughness and speed made him a threat getting to the edge. You want someone who can both threaten to make a big play down the sideline and also run through tackle attempts of players he might find out there (outside linebackers, safeties, and corners).

Of course, there are all kinds of run concepts I didn't include here (wham, trap, etc.), but those are some of the most commonly used ones from the last 30 years.

I understand those runners are cherry-picked from the last three decades of Michigan football, so let me apply reason #8 to Michigan's 2020 roster:

Zach Charbonnet: Charbonnet is mostly an inside runner as a bigger back with some power. He had a 70-yard touchdown on power against Minnesota and has had some big plays on inside zone. I also think he has the quickness to be used on power read, pin and pull, etc., things that go off tackle, but that's not how Michigan used him. Charbonnet is also Michigan's best pass protector.

Blake Corum: Corum is the smallest and quickest back on the team, which ideally puts him on the list for outside zone, toss sweeps, power read, etc. While he has admirable toughness and vision to run inside, it's just tough to do at that size. He also has the quickness to be an asset in the passing game, on wheel routes, etc.

Chris Evans: Evans has good hands and was recruited by some to be a receiver in college. You can create mismatches with him by splitting him out wide or into the slot. He has pretty good quickness, so he likes to bounce things outside whenever possible. But if there's a large enough hole inside, he can hit it with drive through some tackles.

Hassan Haskins: Haskins is the best inside zone runner on the team, and he can also run pin and pull, one back power, power, etc. He lacks the speed to do much outside the tackles, so you want him to run plays where he can keep - or at least get - his shoulders square to the line of scrimmage.

Christian Turner: Turner never really got enough playing time for me to identify where he excelled. My operating theory is that he has the least upside of any of Michigan's backs and does not necessarily excel in any aspect, though he did have a nice run against Florida a couple years ago.


With Michigan having an offense that employs multiple schemes - power, inside zone, pin and pull, etc. - the need for a variety of running skills is apparent. I think the most complete back is Charbonnet, but obviously Michigan's staff has not used him with a great deal of regularity.

Some teams really only focus on one or two schemes, such as the North Carolina Tarheels. It's probably not a coincidence that they had two backs run for 1,100+ yards, and they combined for 28 rushing touchdowns in 2020.

There was some talk that Michigan was putting too much on its players when Jim Harbaugh first arrived, and perhaps that is still the case with Josh Gattis as the offensive coordinator. Not every offense has to be as simple as North Carolina's, and Michigan was bound to struggle in 2020 with all the injuries and NFL departures on the offensive line. But perhaps Michigan would benefit from cutting down on the playbook, which would also cut down on the number of backs who would get reps.

The last part in the series will cover how substitutions work in game preparation and on game day. 


  1. Very nice write-up. Being the football amateur that I am, I do wonder: if a team has running backs that seem suited for only certain scenarios, doesn't bringing them in signal what play will be run? Or eliminate what plays will *not* be run?

    1. It could tip off what play is going to be run, yes. I think the key is to have enough counter/constraint plays, such as Hassan Haskins being brought in for a play action pass off inside zone. Or Blake Corum being brought in for a power read play action pass. But with so many schemes AND the constraints off of them, the defense still has to think, "Okay, who's in the game? What's he good at? And what constraints can they run off of that?" That's a lot of thinking from play to play.

    2. That's a good point Thunder but if you're talking about teams like OSU and MSU who mark Michigan on the calendar and scout everything, they are going to catch that stuff - and if not their assistants are going to call it out from the sideline.

      That's why you can't just have Ty Isaac or (going back further) Michael Shaw out there for sweep plays - he has to be able to pass block and run up the middle too.

      As for Corum, I disagree he is too small to run inside effectively. He's strong and stocky enough to hold up like Barry Sanders. I'd worry more about a skinnier kid like Evans but he seemed to do alright on inside runs too.


    3. That's fine but they still have to stop it. It doesn't matter. With UNC you know inside zone is coming, but teams can't stop it. With Michigan and Mike Hart, you knew outside zone was coming, but teams couldn't stop it (consistently).

      You can disagree if you want on Corum, but he averaged fewer than 3 yards/carry this year and his inside running was nothing special. Not saying he can't get better, but a 5'8", 200 lb. freshman? It's pretty pointless running with him inside, and the numbers bear that out.

    4. Well yeah -- that's what having an identity is all about. Everyone knows what is coming but you run it anyway and they can't stop it.

      I haven't seen the situational rush stats - most of Corum's runs that I remember were outside.


  2. I tend to think simplicity is a virtue when it comes to executing scheme, particularly when you have a lot of young players at the college and high school levels. Obviously Harbaugh had success at a low level of college ball too, but I don't think you can expect the same stuff from college players as you do the NFL. Maybe Stanford was an anomaly, but even their stuff had a lot of consistency in terms of personnel.


  3. First thing, how do you have time to write these blogs?

    It seems Alabama doesn't do a lot of RB changing. Or does it just seem like it?

    1. "Ideally, you have a perfect, three-down running back who never gets hurt or makes mental mistakes . . . but those are rare."

      Najee Harris is rare. Alabama is able to select the best of the very best. The string of great Alabama running backs stretches years.

    2. Alabama has a higher level of talent but rotates just like everyone else -- depending on how good their guys are relative to each other.

      Most years at Alabama they split pretty evenly because they have multiple good options. BUT this year Najee got the bulk of carries because he is special, ie better than the other backs, ie a great option (like thunder said this is rare).

      The last time before him was Derrick Henry's final season in 2015. Also special it turns out given his NFL success. That was right after the year Henry split carries with TJ Yeldon. Yeldon also a very good back went off to the NFL and so Henry carried the load after he left.

      I don't think its a scheme or philosophy thing. You rotate the ball around unless you have a standout. Maybe some coaches have some kind of preference but most teams are going to rotate among the top 2, 3 or whatever their personnel calls for. In Michigan's case this year they had 4 very good backs. Hardly a problem.


    3. Alabama has better coaches than Michigan.

      I've been watching Alabama. Najee Harris wouldnt have ended up a Heisman candidate at Michigan.

    4. How do I have so much time? Every other morning, I randomly wake up between 3:30-4:30 a.m. and can't go back to sleep. So early in the morning is a great time to write.

      Alabama rotates backs when they have the guys to do it. They pitch the idea of rotating guys to keep their legs/bodies fresh for a long career, but if they only have one standout...they'll beat him up in order to win games. A few years ago, Najee Harris, Damien Harris, and Josh Jacobs each had 117-150 carries.

  4. I don't see a connection to scheme and rotation.

    Compare the distribution at a couple other schools.

    UNC may be simpler but allocates a similar percentage to their top back. 33% last year and 40% this year, roughly. The only difference is they gave their 2nd guy a bigger share of the load.

    Then you look at Wisconsin which didn't really change scheme all that much but went from 1 guy getting 50% of the carries last year to an equal split among 3 top backs this year. Very similar situation at Boston College when AJ Dillon left.

    Miss State runs a famously simple Mike Leach scheme and they split carries similar to Michigan.

    Nobody is running a primary back out there unless he is just way better than the other options. Michigan hasn't had that. That might be a bad thing if you expect a superstar back but it might be a good thing if you have lots of good backs like Alabama and OSU. I think it's more of the second category for Michigan.


    1. I think you'll find better evidence if you look back at Leach at Washington State. Last year in Pullman, his lead back had 127 carries, and the #2 guy had...16. This was a weird year at Mississippi State where he was breaking in a new team, the best player on the team (RB Kylin Hill) got injured and then opted out, etc.

      Leach's run game is very simple, so there's not a whole lot to digest, and when you're going up-tempo, you don't want to be making a ton of substitutions. That's one issue that I have with Michigan. They're supposed to be up-tempo, but with so many tight ends/backs/receivers coming in and out of the game, it's impossible to run a play before the defense is ready.

      I don't know the last time I saw Michigan snap the ball before the defense could get lined up.

    2. Kind of proves the point though - when Leach has a guy he trusts above the others he goes with him and when he doesn't he rotates. At Texas Tech he has had a full-on rotation, split between two guys pretty evenly, and had one primary back. It varied from year to year. It just depends who he has - Like other places.


    3. I haven't seen any evidence Michigan is going to supposed to be up tempo. Gattis explicitly said "we won’t go extremely fast. We’ll protect our defense" when he arrived. He talk about wanting to score and putting defenders in conflict -- that's all stuff Harbaugh would say too.

      Faster than traditional 90s ball? Sure, anytime you get rid of huddles you can go faster, but that just means looking like everyone else. They did talk about DICTATING tempo but that's not the same thing as trying to catch teams unprepared or forcing timeouts by hiking the ball quickly. That's more about gameflow and balance, being able to run and pass as needed.

      An OC talking about dictating tempo is like a DC talking about aggression. Even guys like Mattison (who was criticized for not blitzing enough) talk about being aggressive. They're all going to talk about it but it doesn't mean anything 80%.


    4. There was definitely a lot of coachspeak coming from Gattis. But, how do you DICTATE tempo without some variance?

      We might as well huddle, considering the amount of time we take to un@ss ourselves with personnel & playcall ...

    5. Agree with je93. The offense automatically dictates tempo in any offense, because the play doesn't begin until the offense snaps the ball. There are really two reasons to go no-huddle: 1) to run check-with-me 2) to catch the defense off guard. Michigan rarely checks out of plays, and they never catch the defense off guard. They might as well be huddling.

    6. You can play faster without surprising anyone. Just cut down the time it takes to huddle gives you the OPPORTUNITY to play faster and save clock. That they haven't successfully done that very often is more an execution issue than a conceptual one IMO.

      I mostly agree with "they might as well be huddling" criticism, because they flat out haven't executed in most time-constrained scenarios (PSU '19 being maybe a notable exception?)


    7. It's either concept or ineptitude

      JH teams have had issues with the play clock dating back to the 9ers (not sure about stanford). Most likely because of his dependence of playcall by committee - it's just too much going on to handle

    8. It's ineptitude. Been consistent through Fisch, Drevno, Hamilton, Warriner, Gattis that Michigan has struggled to play fast when they need to. So clearly not something that Harbaugh emphasizes enough.

      My point is that this issue of playing fast when you need to is distinct from any talk about "going tempo" in the modern sense. I don't believe Michigan is making a concerted effort to try to dictate matchups and catch the D off guard. We haven't seen that and they haven't said that, beyond general coach-speak.

      Why do I say one type of going fast is different from another? Because the same need to go no-huddle and call plays quickly was there in 1988, before anybody ever used the word "tempo" to apply to football.

      The other stuff that falls under more recent "tempo" developments of the last 20 years is not something we should expect to see under Gattis, IMO.
      But I absolutely agree that Michigan needs to be able to put together fast scoring drives to close the first half and when down late.


  5. Another excellent series Thunder

    In today's CFB, you need options & depth - we "appear" to have that
    - what most teams have though, is an identity - they know what & who they want to be. They end up with either a primary guy, and a compliment or two
    - in year 6, we still don't have an identity, and end up with a hole lot of everything. We spent the season shuffling guys in & out, only to realize what you wrote above
    - I prefer keeping it simple, ESPECIALLY in a shortened, compromised season when replacing EVERYBODY ... we obviously did not go that route, and it showed
    - I think it starts with committing to an identity. Rep our guys at it, and find the 2-3 RBs who do it best. Some years we'll have a fairly consistent split - especially early on; some years we'll have a special talent who commands more

    *much appreciate the stats and historical reference. Good for the "RBs do matter" column

    1. If anything the distribution stats show that RBs don't matter - because even when you have a guy you like better than the others he's still getting maybe 1 or 2 more carries out of 10 as compared to a pure rotation.

    2. Disagree. Most RBs are going to have something they don't quite excel at (see OP), meaning who you play & when matters - RBs matter

    3. I agree that Michigan needs to find its identity. I think they've had an offensive identity at times under Harbaugh, but there was none in 2020. Michigan wasn't good at anything on offense, whether it was power, inside zone, RPOs, the dropback passing game, etc.

  6. Replies
    1. I don't like getting into speculation like that, but I'll say this: If there's another good situation out there, I wouldn't blame him for leaving.

    2. I think it's all but certain

      Sucks, as I thought he was the most well-balanced (and underused)

  7. I think Harbaugh has done a lot of great things to bring Michigan back to where it was. BUT. If I was to name the 3 biggest failings in the Harbaugh era they would be:

    1. Lack of consistency and identity on offense.

    2. High attrition, especially as it relates to management of injuries (Durkin mentality)

    3. Inability to play fast on offense.

    The other big thing (number 4) is letting the Mattison fiasco happen, especially now that Brown is out the door.

  8. I don't believe we need to play superfast, and certainly not all the time

    I do think we desperately need greater urgency at the end of a half or when down. I would also like to see VARIED tempo, catching a defense off guard. It's free, and it works. Not doing so has been costly, allows opponents to anticipate the call through personnel, alignment, and down & distance. Also gives big DL a chance to catch their breath, and pin their ears back on the next snap

    1. It's not free. Practice time is finite.

      Not saying it's a good idea or bad idea, just that Gattis hasn't promised it.