Book Review: Game Plan
Some reviews, like the one I recently did for the 2012 edition of the NFL Record & Fact Book, are relatively straightforward. Others are more complicated. This falls into the latter category. The book in question is Game Plan: A Radical Approach to Decision Making in the National Football League by Frank DuPont, aka FantasyDouche on Twitter, and I’m not sure quite what to say about it.
The animating thesis behind Game Plan is that NFL coaches are not good enough at decision-making, and they can improve. Now, for the past three years, I’ve co-written the Scramble for the Ball column at Football Outsiders. One of the weekly elements of that column is the awards. Those awards, which we have to hand out every week, currently include the Colbert Award, for bold decisions; the Mike Martz Award, for incomprehensible decisions; and the Keep Chopping Wood, for helping your team lose. I’ve had a couple people tell me the awards are the first and sometimes only part of the column they read, and I take handing them out seriously. The upshot of that is I spend a fair amount of time each week reading over gamelogs looking at the various tactical and strategic decisions coaches make. As a result, I’m quite ready to get behind that idea.
Some of DuPont’s cases are pretty easy. Take, for instance, Jason Garrett’s bumbling of the game management at the end of the Cowboys’ game against the Cardinals this year. It’s easy to believe that if Garrett had the same actual decision-making experience of a Madden player, he wouldn’t have flubbed the situation quite so badly. First-time NFL coaches, especially ones who’ve never been in charge before, need more experience making decisions. I noted this in my review of John Madden’s One Knee Equals Two Feet, where Madden, when hired as a very young head coach, decided to go to local high school games and essentially coach along with the actual high school coach as a means of getting practical experience. Simple, but smart, and perhaps Garrett would have chosen more wisely had he done the same.
At the same time, though, Raheem Morris, whom DuPont cites as having “majored in Madden,” purposefully and intentionally ordered his team to spike the ball every time after a first down in the two minute drill, regardless of the amount of time pressure the Bucs were under. As noted in The Essential Smart Football, that automatic spiking of the football is frequently a wasteful and very stupid move. An unnecessary spike played a key role in the Bucs’ loss to the Titans this year. I haven’t played much Madden online, but I’m guessing an automatic spike like that is not a common move among the better Madden players. Generalizing from n=1 is always a chancy business, but that an experienced Madden player could make a decision like that makes it clear that video game experience is no panacea.
The primary club DuPont uses to beat over the NFL is the recent revolution in poker, where the best players in the world went from relatively older, in their 30′s and 40′s, to in their 20′s. Younger people, DuPont claims, would make better decisions than these older coaches. The problem I have with that is, to steal a line Jurassic Park, “When Pirates of the Caribbean breaks down, the pirates don’t eat the tourists.” Playing cards are not independent decision makers; NFL players, and everybody who works for, under, and with an NFL coach is. That means there’s a lot more to coaching football than calling plays; a head coach is more the CEO in a 200-person organization, the head man in a number of different disciplines, more than he is a gameday coach; a common refrain of new coaches is the realizing of just how much of the job of being a head coach is Not Football. Optimizing gameday decisions is clearly good for an NFL coach, but the experience of, e.g., Andy Reid’s continued (a) employment and (b) success strongly suggests that it can be counterbalanced by being particularly good at other factors. I therefore found DuPont’s primary thesis under-proved and insufficiently convincing. It was never quite clear to me that DuPont understood or found the distinction important, and if he ever justified his argument that tactical decision-making is and should be the primary factor upon which NFL coaches should be evaluated, then I missed it.
Another good example of my frustration with Game Plan is his chapter on the draft resources NFL teams invest in running backs compared to offensive linemen. DuPont points out there are five offensive linemen on a field and often only one running back, but teams are drafting only twice as many linemen as running backs even though the relationship between draft position and production for a running back is relatively weak.
I’ll start with the problem from first principles, using rough estimates. I’ll use the Titans as my example, just because I know them the best, and keep in mind the league as a whole might fall a little differently. For most of 2011, they carried four running backs (three halfbacks and one fullback) and nine offensive linemen. The average running back career is shorter than the average offensive lineman career; particularly, you see a lot more offensive linemen playing 8-14 years than you do running backs. Without researching the typical career longevity of running backs and offensive linemen, just from that simple information, you might expect half of the running backs on a team and a third of the offensive linemen to be in the first 4 years of their career. This analysis is admittedly relatively crude, but from that positional distribution and career aging curves, we would expect two of the running backs to be in the first four years of their career and three of the offensive linemen to be the same. Assuming all of those players are draft, the team to be using fifty percent more draft picks on offensive lineman than running backs. That they are actually using twice the draft resources suggests, ceteris paribus, they value offensive linemen slightly more than running backs. Given that a higher percentage of offensive lineman start than do running backs, it makes sense that offensive linemen would be valued more, so teams drafting twice as many offensive lineman as running backs actually makes perfect sense to me.
This review almost certainly makes Game Plan seem worse than it was. In my typical style of reviews, I’ve highlighted a couple things from the book, and there’s more that I either agreed with or didn’t annoy me. If there hadn’t been those parts, or it had pissed me off in the same way as, say, Death to the BCS, you wouldn’t be reading this review; I’d have just given up on it. At the same time, though, it was as deeply frustrating a book as any I’ve read lately, all the more so because it was about something I’ve spent some time thinking about. You may well find it a more useful, more valuable, less frustrating book than I did, and some people I know certainly did. At only $0.99 for the electronic version, it’s worth even a speculative purchase.
For more about Game Plan, including the different ways to acquire a copy, see this page on Frank’s website.
FTC Disclaimer: I read a review copy of the electronic version, though I planned on buying it at some point before Frank emailed me with the offer.
UPDATE (8/7/12, 1253 CT): Frank’s response to my review is up on his site, and you should read that in addition to my review.