After the QB, judging player values is tough in NFL

If there is one key to building an NFL contender that no one disputes, it is that a star quarterback is a must.

And yet, the chaotic past week provides more evidence that paying for a star quarterback often kneecaps a franchise. The thing you must do is the thing you cannot do, unless you are really careful about it. The Detroit Lions lost Ndamukong Suh, the anchor of their defensive line, in no small part because of the $17-million US salary-cap hit carried by quarterback Matt Stafford. The Baltimore Ravens traded Haloti Ngata, the anchor of their defensive line, to save money against the cap, where quarterback Joe Flacco carries a $13.5-million number in 2015. The Dallas Cowboys let the NFL’s leading rusher, DeMarco Murray, walk in free agency, while quarterback Tony Romo has a whopping $27-million cap number this season. And, most unexpectedly, New Orleans moved Drew Brees’ favourite target, tight end Jimmy Graham, as they struggle to get under the cap. Brees, meanwhile, will suck up $26.4 million of the league-mandated $143-million per-team cap.

There are teams that pay their star quarterback a lot of money and manage to not cripple the rest of their roster — New England and Green Bay come to mind — but the only way to do it is to make sure that almost no one else is overpaid.

And this is where the NFL, in 2015, lags almost comically behind other team sports: there’s little consensus on what any position, after quarterback, is worth. The NFL needs its own version of baseball’s Wins Above Replacement, but it’s a long way from having one.

In baseball, though not everyone loves the metric, versions of WAR — a number that suggests how many wins a player would add to a team relative to an average replacement player — mean that front offices can assess and compare players regardless of their position on the field. But in football, individual player values don’t exist, at least not in a way that allows for easy comparison. A quarterback might have a passer rating of 99.8, which doesn’t mean much next to a running back that averages 4.2 yards per rush or a wideout who caught 75 passes. Things become foggier still on defence, where stats like sacks, interceptions and tackles are heavily influenced by who a player has playing next to him.

At the Sloan analytics conference in Boston a couple of weeks ago, the question of whether football can develop better player metrics was a popular one. As one front-office executive put it, the holy grail would see a player’s value boiled down to one number.

“If we could compare a guy who’s an eight to a four and a two, then it makes putting together a roster that much easier,” he said.

Getting there is a whole different challenge. There are so many moving parts in a single football play that grading individual contributions requires a great deal of work. But, a 60-minute game only includes about 11 minutes of live action, so a team that wanted to put the resources toward breaking down every play, player by player, could conceivably do it. Even so, teams would only be able to conduct this kind of research of their own teams, because it’s next to impossible to grade the action of a play unless the person doing the grading knows what play was called and the individual assignments on that play. You can’t fully grade a wide receiver unless you know what route he was supposed to run. These are the problems to which football doesn’t have an answer. At least not yet.