Is coaching overrated?

So asks Gregg Easterbrook, in an article titled “Coaching is Overrated”:

Changing the playcaller sure helped the Redskins!

In the cult of football, surely few things are more overrated than play calling. Much football commentary, from high school stands to the NFL in prime time, boils down to: “If they ran they should have passed, and if they passed they should have run.” Other commentary boils down to: “If it worked, it was a good call, if it failed, it was a bad call,” though the call is only one of many factors in a football play. Good calls are better than bad calls — this column exerts considerable effort documenting the difference. But it’s nonsensical to think that replacing a guy who calls a lot of runs to the left with a guy who calls a lot of runs to the right will transform a team.

One factor here is the Illusion of Coaching. We want to believe that coaches are super-ultra-masterminds in control of events, and coaches do not mind encouraging that belief. But coaching is a secondary force in sports; the athletes themselves are always more important. TMQ’s immutable Law of 10 Percent holds that good coaching can improve a team by 10 percent, bad coaching can subtract from performance by 10 percent — but the rest will always be on the players themselves, their athletic ability and level of devotion, plus luck. If the players are no good or out of sync, it won’t matter what plays are called; if the players are talented and dedicated, they will succeed no matter what the sideline signals in. Unless they have bad luck, which no one can control.

Yes and no. I wholeheartedly agree that playcalling is overrated, and he is right that much of the commentary after games involves a lot of second-guessing full of hindsight bias. Few ever pose the “should he have done X?” question in terms of the probabilities and tendencies at the time, or in the context of the 10 or so seconds available to make such calls. Indeed, I have even argued that there’s a case to be made that the best playcalling might be a controlled but randomized “mixed-strategy.”

The other coaching bogeyman is the aura surrounding “in-game adjustments” or “halftime adjustments,” both of which are supposed to be the “hallmarks of good coaching.” This is another thing where there’s a kernel of truth surrounding by a lot of speculation. Yes, a good coach will not do the same thing over and over again if it isn’t working, or if the other team has figured it out. And yes, coaching a game involves an ongoing process of what the other team is doing (this is one reason why I think, even if adjustments are part of the game, “halftime adjustments” are very much overrated). But if you want to see a bad coach then I’ll show you one who tries to “adjust” to everything the other team is doing with new schemes and ideas built-in midgame. Instead, teams with good coaching pretty much run only things within their plan — i.e. stuff they had practiced during the week. Indeed, much of what fans or commentators will pick out as an “adjustment” was something in the original gameplan that just didn’t get called until the second half because of the flow of the game. Yet how can good coaches both “adjust” throughout a game and also not deviate from what they have practiced?

This brings me to where I depart from Easterbrook, that coaching is minor. (I don’t really know how to judge “overrated” — in relation to what? overrated by whom?) While playcalling is definitely overhyped (hey, the talking heads get paid to talk about something), preparation is extremely important, and much of a gameplan involves contingency planning. It also means that the “base stuff” should have the counters built in, the constraint plays are already there, and the defensive adjustments are easy to make because they are a part of the system. A good offense “implies the counter,” meaning that if a defense adjusts in some way, then playcalling is simple because there’s an obvious counter play to be called. On defense you take away the other team’s best stuff, and focus on other things as it comes, though by dictating to the offense through aggressiveness and by trying to confuse it. Unlike Easterbrook I can’t hang a number on how many wins or losses “coaching” is responsible for (and if I could I’d imagine it varies by level), I can safely say that I think weekly preparation is underrated, because it is rarely talked about — other than platitudes like “we had a great week of practice” — has a long-tail in terms of continual refinement of technique and effort that can only improve incrementally, and that everything run in the games is stuff that has been practiced over and over and over.

Two final points on the Redskins situation.

I find it immensely weird that Sherm Lewis, a guy basically pulled off the street, can come in and call an NFL game and, for better or for worse, the offense looks basically the same. This further validates to me that the “base” NFL offense is almost totally generic across teams — how else would Sherman Lewis even be familiar with it? — and that the only variance comes in that weekly gameplanning and preparation. This is not to say it is bad or that such preparation is insubstantial and isn’t full of wrinkles (that’s what the whole post above tried to explain), but it is still built off a core set of offense that has changed little in recent years or from team to team. As a result Lewis could walk in and be on the same page with basically everyone else, and since he was present for the past couple of gameplanning sessions there wasn’t anything that would trip him up: It was either so old he was already familiar with it, or he had a hand in devising it for that opponent. Yet it remains strange to me that NFL teams are content to be so homogeneous.

Finally, Easterbrook accurately points out that the Redskins are about a bizarre an example as possible, as the whole Sherm Lewis switch seemed to be about much else apart from playcalling:

In the Redskins’ case, there is an added dimension — the rumor is Chainsaw Dan took away coach Jim Zorn’s play-calling authority hoping that Zorn would blow a gasket and quit. If Zorn quits, he gets nothing; if he’s fired, he receives the balance due on his contract. No doubt to Chainsaw Dan’s dismay, Zorn held his ground.

  • http://footballxos.wordpress.com Jon E.

    Coaching is underrated, playcalling is underrated, commentators and analysts are overrated.

  • omar

    Look how motivated the Redskins were on Monday night. They weren’t. That all falls on Zorn. He just stands on the sidelines like a guy that doesn’t know anyone at the party and is trying to stay out of the way.

    Playcalling: overrated
    Leadership: underrated

  • http://brophyfootball.blogspot.com brophy

    The best coaches are the ones who more often than not can steer teams away/past disasters. It (IMO) has more to do with probability and risk-assessment (liability management) than actual chalkboard BS. The more factors a coach can control (players, down & distance, preparation, execution, etc), the more ability he has to dictate each circumstance and set up the next. You could completely randomize play-calling, but still end up in a losing proposition because it ignores the fundamental concept of why plays are called and what they are attempting to accomplish versus all the given factors against their opponent.

    Winning is simple…it just isn’t easy (investment).

    In other matters (semi-related), just how much of the NFL is all smoke-and-mirrors, a facade of truth? Much of it is sports writer / fan-generated hype and snowballing preconceptions akin to WWE. How much of FOX,CBS, ESPN, NFL Network analysis plays into this cliched double-talk of hero(personality) worship? From the camera angles to the NFL Films dramatizations, to the “insider” stories, to the mystification of “coaching”, what is real and what is truly make-believe?

  • http://rc3.org/ Rafe

    I think Gregg Easterbrook is overrated. I honestly can’t believe he gets paid to do what he does.

    I think that coaching is underrated for the reasons you state. I was watching my beloved UH Cougars play Mississippi State a couple of weeks ago, and early in the game, they were throwing lots of screens. (OK, they always throw lots of screens.) MSU was overplaying the screen more and more, and all of a sudden, UH starts throwing this fake screen play at them, and they burned them on it repeatedly. There’s little doubt in my mind that this was something that was baked into the UH game plan from early on — when a team starts biting on the screen you fake it and get them another way. That’s coaching.

    By the same token, UH is terrible at defending the option. The defensive ends never seem to read the play and stay at home, and so quarterback keepers on option plays just tear them up. That’s coaching, too.

  • DM

    Sherm Lewis was able to come in off the street and not disrupt the offense too much in part because the offense that Zorn ran isn’t too far off from from the offenses that Lewis ran. Remember, they’re both ex-Holmgren coaches, so there’s quite a bit of familiarity. A good portion of his bread-and-butter plays are already there, he just needed to say them differently.

    It also speaks to, and you hit on this in your 2nd paragraph, how a team should and generally does stick to it’s gameplan throughout the whole game and that the notion of “halftime adjustments” truly is just empty talk by broadcasters. In Phil Simms’ book he says that he can recall only one time in his whole career that his team made a halftime adjustment, and even then it was simply the coach taking a play that they were running to only one side of the field and flipping it because they felt they could get a big play on the open space on the other side, which they did.

  • http://smartfootball.com Chris

    Brophy: I think I agree with your first point, but I’m a little cloudy on it. My point was that good coaching comes in what they do all week, preparing the players, fundamentals, techniques and reps geared to what they will actually be doing. The in game stuff is only sometimes relevant, and usually only at the margins. (Bad clock management might be the difference between winning and losing, but it didn’t change the game from a four touchdown game to a three point one — that was how the team as a whole performed.) And when I said the playcalling could be randomized I mean a structured randomization — i.e. an Airraid team would only have plays from their offense in the mix, or the same for flexbone teams, and even then on 3rd and 3 it’d only be from the universe of 3rd and 3s, with a tilt of say 60/40 run vs. pass or whatever was chosen. Just within that it’d be randomized. All the preparation would be built on what the coaches knew about the coaches knew and had done: They had worked on a certain play; their fullback can handle the other team’s defensive end; and the defensive tackle to that side is weak and they can get movement on him; etc.

    So if you’re saying you agree that preparation is king, and preparation is not even just limited to gameplanning, then I agree with you. That it’s about hard work. I look at gameplanning as a way to attack the other team yes, but also as a way to confine what areas your own team should focus on in its preparation. Sort of like a training manual in a way.

  • Steve

    I’m not sure what kind of halftime adjustments could possibly be made and communicated in the 12 minute halftime they have in the pros. It almost certainly has to be some wrinkle present in the original gameplan.

    On a side-note, I was once on a staff where the head coach tried to teach a different defense at halftime of a 30 point drubbing. It worked for one play in the second half. The other thirty or so they ran, not so much.

  • Brad

    I think what Brophy is hitting on that I think is a component of good and bad coaching is the ability to foresee a wide range of outcomes particularly disasterous ones and call the game in such a way to limit the downside.

    Here is an example that I see all the time. You are up a field goal just before halftime with a below average QB that has a history of throwing interceptions. You get the ball on your own 20 with about a minute left. You dedice to be aggressive and go two minute drill. First play you throw an out (or a throw to the flats)thinking you want to let the reciever run out of bounds. It is intercepted for a TD. That is bad play calling.

    Sure the QB screwed up, but he has a history of screwing up, he is not that good. In your play calling you should have anticipated that there was a reasonably high risk of the QB screwing up and given the field position that screw up might lead to a TD. And you chose a throw that is kind of hard to complete and that is often returned for yardage when there is a INT (no guys near by to tackle). Contrast that against the possiblility of driving 50-80 yds for a score in 1 minute with a mediocre QB.

    Many coaches want to “Be aggressive” and avoid “Playing not to lose” and default to stock playcalls regardless of talent and game situation.

  • http://brophyfootball.blogspot.com brophy

    the “coaching” (i.e. play-calling) that I was referring to was the rhyme and reason behind play calls. Long and short of it, think Venzini from “Princess Bride” (“I know you know that I know that you know…..”). D&D, field/boundary, personnel, time, etc factors predetermine much of what your opponent will do, to which end, you will play THOSE probabilities to set up (or paint you into a corner) a short-list of best matchups for the desired outcome (weighing the risk/reward of the foreseen situation). This isn’t randomized “sling-from-the-hip-see-what-sticks” approach to play-calling. Sure, you can out guess the guesser or throw tendencies on their head with something from left field, but it all is predicated on beating the front/coverage presented (vis versa for defense).

    [thats all I was attempting to say]

  • DM

    “I’m not sure what kind of halftime adjustments could possibly be made and communicated in the 12 minute halftime they have in the pros. It almost certainly has to be some wrinkle present in the original gameplan.”

    “Halftime adjustments” is an empty broadcaster phrase because football naturally has the entire offense or entire defense off the field at one time. If you’re going to make adjustments, you don’t wait until the half to do so, you have the ability to change things on the sideline. And that’s of course exactly what happens, you see it all the time when the QB is looking at photos of the possession the offense was just on, or players gathered around a coach on the sidelines. Plus you’ve got timeouts. If that counter play is not working on the first two possessions you don’t wait until the game is halfway over to make an adjustment, by which point it’s probably too late.

  • Tyler

    Easterbrook makes the mistake of equating coaching to play-calling. Coaching goes from the teaching of fundamentals, motivating players, managing practice time, to the preparation of a game plan and everything in between.

    Play-calling may be overrated but who is setting the standard? Is there some general opinion that has been set that I have missed? I believe some fans overrate play-calling and others underrate it. It isn’t as easy as picking random plays off a call-sheet and expecting your team to run them well. But I also can’t think of one “superior” play-caller that consistently outsmarts opponents. Everyone has been outsmarted at some point.

    So, I suppose, there are no conclusions that I can draw from Easterbrook’s article other than a bunch of situational “yes” and nos”.

  • JP – Chicago

    IMO, coaching is underrated and play calling is OVERRATED.The coaches leadership ability is of utmost importance. His ability to have the guys ready to play, to win, to compete. The mental aspects of the game are the most important thing a coach can address. Those are far more important than to call a play.

    I was taught in my 1st college game by my line coach, the game is won between the lines and his job(the coach) was to create confusion for the other team. Motions, sets etc, etc… We were to go out and take it to the other team and not to worry about anything else.

  • OldSouth

    Perhaps the easiest way to tell if someone is knowledgeable about coaching football is to ascertain his conception of what “coaching” is. At least in my experience, the following is almost always true:

    If he conceives of coaching as being largely about playcalling, adjustments, or driven by axioms, he probably understands little about coaching football. These folks give themselves away quickly by making statements that reveal that they’ve internalized these things. Easterbrook’s article conveys that to me.

  • Kevin

    I think coaching is overrated on GAME day, but underrated on the prep week. I think that’s what some of these other comments lean toward.

    A good coach makes himself and his team weaker when he thinks he can play that chess game on game day.
    Some of the best coaches give a lot of freedom to the QBs and players on game day from an adjustment/playcalling standpoint. They have the best view in the game but some coaches, I think especially high school coaches, are afraid to give up that power and control.

    You always hear the analogy of chess and football and it can be that mental, but I’ve also heard and agree that it should be checkers instead.

  • http://www.quantcoach.com QuantCoach

    Many Green Bay Packers players have reported that the great Vince Lombardi played almost NO ROLE AT ALL while the game was actually being played. Author Michael Lewis once wrote in the NY Times: “During a game, Lombardi played no strategic role; he didn’t do much but shout praise and criticism. ”The most useless guy on our sidelines” was how some of Lombardi’s players described him.” Bart Starr called the plays on offense and Phil Bengston ran the defense. But, man, could Lombardi prepare a team during the week. Like Belichick, Lombardi was a master at using film to prepare his team. Well done Smartfootball. Very well done.

  • MTK

    That coaches prevent their teams from losing is underrated. Without a coach who teaches sound fundamentals, football players would be prone to penalties, botched special teams, and turnovers. In a way, it’s easier to identify the behavior that will guarantee a loss than it is to put your finger on that which will guarantee a win.

    I do agree that play calling is overrated. Perfect execution is underrated. I’d guess 90% of all offensive plays in any given scheme could score against any defense, assuming a quarterback can check out of a clear mismatch.

  • Infinite5k

    To say that there are not in-game coaching adjustments is b.s. Of course there are adjustments made. Teams do not always do what we as coaches predict them to do against us. I remember game one of my first college game we thought they were going to run a 4-3 or a 4-4 defense. So for all of game-week we prepped for that. Then they ran a 3-3 stack. Well anyone who knows anything about offensive blocking assignments knows that it changes your blocking assignments and angles quite a bit. Also, there ARE halftime adjustments made – the thing is, most of them are not major adjustments. For instance, if a corner is playing wide in cover-3, the coach might have a wr stem his route differently. Adjustments are constantly made.

    Now, are the adjustments usually huge and drastic? No. The one example of not getting the defensive front we expected does happen – but not very often… maybe 2-3 times a year.

  • (Chess) Coach Mike

    “On a side-note, I was once on a staff where the head coach tried to teach a different defense at halftime of a 30 point drubbing. It worked for one play in the second half. The other thirty or so they ran, not so much.”

    Reminds me of a JV game I played in my freshman year of HS. The JV coach tried to install a new defense on the chalkboard twenty minutes before a game against our archrivals. Needless to say, it failed miserably (to the tune of a 34-0 halftime defecit). Fortunately, he was not back the year after.

  • http://www.bringitonsports.com.au Paul

    Offensive Coordoinators design plays with complimentry plays. Offensive coordinators are not mind readers. i think that Play Calling is overated but certain cheack and Audibles at the line of scrimmage are not….

  • Devin

    As a layman reading this, I agree that play calling is just something that fans harp on when they don’t know what else to blame. We don’t know what goes on during the week, so who knows why your offensive line can’t block well or your defense is missing tackles, but play calling is something that you can tangibly point to as a problem since you see the results immediately.

    One thing that I think is often overlooked is play calling and smart risk taking in high leverage situations. Look at Tennessee against Alabama, Kiffin chose to sit on the ball and try a 45 yard field goal instead of going no huddle and trying to get a play or two off. College kickers only make a little over half of kicks from 45 yards, moving that up by five or ten yards would have increased their chances of winning by a significant percentage, maybe 10 or 15 percent. Witness the different 4th down decisions by Pete Carroll and Jim Tressell in their game, USC scored their first touchdown on 4th and goal from the one while OSU kicked a field goal in the same situation and punted on a 4th and 1 from inside the USC 40. Paul Johnson this week went for it on 4th and 1 in overtime rather than kick a chip shot to go to a second overtime, I would bet that many if not most of the coaches in the country would kick the field goal rather than risk losing in the first overtime, even if the probabilities of scoring a touchdown from that position was much higher than winning in a second overtime. I think a lot of more conservative coaches could vastly improve their chances in high leverage situations by just sitting down and running the math in these kinds of situations rather than just blindly making the decision that carries the least immediate risk.

  • Carmon Maday

    Awesome Post, thanks for this fine Post. I will come back soon . Great tips also : stop biting my nails