What makes a good football coach?

Happy Monday, friends.  I hope everybody had a nice holiday weekend.  I spent the time in Los Angeles, visiting my brother and his family, and during my travels, I saw where Doug linked a Mark Kiszla article about Foxball possibly leading to John Fox getting fired after 2013.

That spurred an interesting discussion in the Lard comments, and it got me to thinking about the nature of football coaching.  It’s a 4.5-hour drive from LA to Yuma, Arizona, and that’s a lot of thinking time.

What I’d like to explore today are the various parts of the job description of an NFL head coach.  Once we identify them, I then want to try to get at the question of which parts are most important.  If we’re fairly judging a coach, having an idea of that in place is crucial. 

Finally, I want to talk about how we evaluate coaches in general.  I think that for the most part, those who effectively evaluate them are wholly unqualified to do so, and that that makes the job of head coach more difficult than it should be.

I would say that the following general set of responsibilities constitutes a reasonable model for a job description which we can call Generic NFL Head Coach:

  • Supervise the establishment of effective football schemes in all phases of the game.
  • Supervise the in-game play-calling in all phases of the game.
  • Ensure that all coaches and players continuously improve, and that the best performance possible is gotten from them.
  • Constantly evaluate the performance of those players and coaches, and provide regular feedback.
  • Establish, communicate, maintain, and evolve a coherent overarching on-field strategic framework.
  • In conjunction with the front office, participate in succession planning for soon-to-be-departed players, by developing effective replacements for them.
  • Maximize the effectiveness of practice and meeting time.
  • Make in-game tactical decisions, such as punt-or-go, and challenge-or-not.
  • Partner with the front office on college scouting and draft preparation.
  • Participate in recruitment of free agent players and coaches during the offseason.
  • Participate in team- and NFL-mandated media responsibilities.

Different organizations have slightly different structures, but in general, all head coaches are doing those sorts of things, right?  So, what are the most important things?

Briefly, before we go down that road, I would posit that no coach can give you all of those things at the highest level possible.  There aren’t enough hours in the day, and nobody is so talented as to be the best at everything.

For example, Bill Belichick is fantastic in-game, schematically, and with the organizational stuff, but I wouldn’t say that he’s anything special when it comes to driving player improvement, or maximizing the effectiveness of all available player resources.  Every coach has limitations; some are based on his own personal shortcomings, and others are based on mutually exclusive philosophical choices.  (Example – deciding whether players should fit the scheme, or whether the scheme should fit the players.)

Belichick is interesting, because he straddles the divide between the two types of coaches, maybe more than anybody else.  There are the technocratic scheme guys like Mike Shanahan and Josh McDaniels, who believe that the scheme is the star, and, effectively, that coaching wins games.  Then there are the fundamental football guys, who believe that players win the games, and that keeping things more toward the simple side helps them play fast and execute.  I view Belichick as a scheme guy who doesn’t totally buy that a great scheme is enough to win in the NFL.

So, back to the key question at hand – what responsibilities are most important for an NFL head coach?  I believe that in the modern NFL, with the salary cap and the increasing attitude that middle-class veterans need to be fungible, the most important trait is being able to maximize the use of player resources.

This comes down to a few things.  First, a coach needs to be good at developing players that the team drafts.  I’ll call this capitalization on talent, because a rookie comes in with physical talent, and needs to be taught how to play the NFL game.  Throughout the four years that a team has that guy, the coaches need to be helping that player improve from game to game and year to year.

The second factor is maintaining a flexible and learnable schematic framework that allows for players to move in and out every year.  To me, coaches who maintain schemes in which it’s nearly impossible for a rookie to play are doing self-defeating stuff.  A good example is the Steelers defense, which is very complex and specialized.  They do well at developing defensive linemen and linebackers there, but not before they waste a year or two sitting those guys on the bench to learn.

If you can do those two things as a coach, and as a coaching staff, your team is going to be competitive all the time.  Before a play is ever called, or a decision is ever made to punt or go for it on 4th and 1, the coach has brought significant value to the table by ensuring that he has schemes and players that can compete against anybody.

I think the next most important responsibilities of a head coach are the day-to-day tasks, like running effective practices, consistently evaluating player/coach performance, and delivering regular feedback on it.  I’m a believer that effective preparation is the best defense against randomness, and that more often than not, it will win out in football.

Third in importance, I think, is the effectiveness of play-calling and in-game tactical decision-making.  This is difficult to get your head around, because this is the part that most people see.  We’ll touch on that more a bit later.

Fourth is the organizational stuff, whether it be recruiting free agents or helping with draft preparation.  At the bottom, (but not to suggest that they’re unimportant), are the public-facing responsibilities of being a coach, like talking to the very tedious Jeff Legwolds of the world.

I’m one man, and that’s how I value the activities of coaches.  Others will undoubtedly value them differently.  How should we evaluate their performance, though?  Because, make no mistake, the media and fans are mostly the people doing the evaluations of them.

I think that somebody whose sports worldview is more heavily influenced by baseball (and resultingly, the use of statistics to make decisions) than mine is apt to dislike Fox, and to be turned off by his conservatism, and reliance upon traditional football thinking.

I’m not even necessarily talking about advanced statistics, per se; baseball managers have been using lefty/righty and specific pitching matchup stats to make decisions for decades.  I think the idea that a football coach should behave like a baseball manager, and make moves based on numbers is a natural idea for somebody who’s watched a lot of baseball over the years.

Baseball is a series of individual and measurable events, and for that reason, it’s the sport that can draw the greatest benefit from statistics.  Even before the smarter SABR guys came along, baseball was a game that was talked about (to paraphrase Bill James) in the language of statistics.  If you watch a game, the TV guys tend to couch the events in terms of choices that managers can make. 

Does he bunt the runner over here?  Maybe they hit-and-run.

I think that because baseball is presented that way, it’s easy to watch the game from the perspective of a manager, and to second-guess his decisions.  We can all relate more easily to an aging dude with a beer gut sitting on the bench than we can to Mike Trout, can’t we?

What can easily get lost if you get really hung up on a couple of dubious in-game tactical decisions is all of the behind-the-scenes value that a coach provided in even getting his team to that position.  I believe subjectively that John Fox, through his methods and processes, has coached his teams to as many or (usually) more victories than their talent says they should have had in every year that he’s been a head coach.  I don’t see a single season where his team underachieved its talent.  Remember, most of those seasons featured Jake Delhomme at quarterback.

What about 2012, right?  They underachieved in 2012!  The Broncos lost a nutty game that was filled with randomness and uncharacteristic mistakes at key moments.  Coaching didn’t make Rahim Moore lose his depth or make Peyton Manning throw an interception in double overtime, and it didn’t make Knowshon Moreno go out with an injury, and take the Broncos’ ability to effectively pass-protect with him.

Fox made a couple of questionable calls in the game, but those calls were not the reason the team lost.  In any event, we need be careful before we consider a whole season’s body of work to be tarnished by one game.  The same is also true of John Fox’s two-year reign as head coach; the larger sample size suggests that he does a good job, and one game (important or not) is just one game.

For the record, I don’t disagree in the least that Fox should optimally have a much better command of football math, and the associated probabilities of the decisions he makes.  I just view it as a shortcoming that’s less important than his obvious strengths, and frankly, as a shortcoming that he shares with about 28 other head coaches in the NFL. 

Not being great with the math represents a foregone opportunity to have an advantage over most other teams, but that’s not to say that Fox’s historically strong abilities in player and coach development, making the scheme fit the players, motivation, recruiting, and assisting in draft preparation, would be easily replaced by somebody who had better facility with math, and the gumption to go for it frequently on fourth down, despite the expected criticism from cretins in the mainstream NFL media.

You know, Josh McDaniels had a good feel for the football math, but some of his other shortcomings did him in during his short stay in Denver.  (Mostly it was his unwillingness to suck up to Woody Paige, if you ask me.)  Nobody has it all, despite what the average 21-year-old woman thinks.

When we evaluate coaches, I think that the most important thing to look for is organizational competency.  If it seems like the organization is running effectively, and that there’s improvement to be seen, that suggests that the head coach is likely doing a good job.  It's easy to look at some isolated decision in a game that we personally disagreed with.  It's even easier to just look at a won-loss record, and consider it without context.  Those are the most visible things to see, but when you focus on them, you can miss the forest for the trees.

I won’t judge Jeff Fisher on the win-loss record of the 2013 St. Louis Rams.  The organization has banked on a QB (Sam Bradford) who may or may not be good enough, and they play in what suddenly may be the toughest division in the NFL.  It’s beyond clear, though, that Fisher has put a functional organizational structure in place of the caliber that the Rams haven’t had since the Dick Vermeil/early Mike Martz days.  Fisher is doing a good job there, and the victories will eventually come, even if it’s not this season.

You know who kind of sucks at game management?  Andy Reid.  He’s really like an offense-focused John Fox, when it comes down to it, though.  He’s excellent at establishing and maintaining a consistently competitive organization, and I expect him to get the Chiefs in order pretty quickly.

I saw some comments in yesterday’s Lard about Marty Schottenheimer, and he’s a favorite target of Broncos fans, since John Elway beat him so much.  I think Marty was an outstanding football coach who consistently had his team in position to compete to win championships; it’s not his fault that he never had a championship-caliber QB to get him over the hump.

By the way, if John Fox goes 13-3 and loses in the playoffs again, and gets fired like Marty did at 14-2, I would consider that to be indefensible, just like Marty’s firing was.  That was all about his inability to get along with A.J. Smith, whether anybody remembers that at this point.  It worked out pretty poorly for the Chargers in the final analysis.

To me, you fire a coach when he isn’t getting any traction, and he can’t put his team in position to compete for championships.  You fire him when his messaging has gotten stale, or when he proves to be unable to adapt schematically to some growing league-wide trend.

You maybe fire him when the team is 24-24 over a three-year period, and he seems unwilling to change his approach.  You don’t fire him because his team got beat by another good team in a one-game sample.

Frankly, the idea that John Elway would or should do so is silly, in my opinion. 

1.  I’m not in the arguing business, I’m in the saying what I think business.
2.  I get my information from my eyes.

Follow me on Twitter  While you’re at it, Like our Facebook page

Ted's AnalysisYou Got Served

Recent Game Coverage

All Game Coverage