The NFL Draft: No One Knows Anything

It's not true that nobody knows anything about the NFL Draft. Scouts get paid lots of money, and GM's get paid even more, to learn as much as they can about evaluating prospects. They know more about this stuff than we do.

But they still don't know all that much.

The NFL Draft is not, nor will it ever be, a perfect distribution of talent. Teams are going to make bad selections and hell, even good selections can turn into bad players once you expose them to a bad organization with horrible coaching. There's no need to pretend that NFL teams know precisely what they are doing when it comes to drafting the talent.

This is well-educated guesswork. Everyone does their research, and they are better than just randomly throwing darts at a board, but it is still guesswork. No one really knows anything, they just hope to be less wrong than the other guy.

Take a look at a six-year sample of recent drafts and how often first round picks turn into Pro Bowlers:

Once we get to 2012, we see a sharp drop off just because players haven't been in the league long enough to establish themselves. We're not expecting guys to be Pro Bowlers right off the bat. So let's look at the five year period of 2007-2011.

A player taken in the top ten picks has nearly a 50% chance of being a Pro Bowler. That doesn't mean half are busts, but if you're drafting in the top ten, you want an elite player. The NFL is pretty good at identifying that talent.

Looking at the full first round, NFL teams draft a Pro Bowler 38.2% of the time. Picks outside the top ten become Pro Bowlers 34.5% of the time. That's a pretty good hit rate. Teams aren't anywhere near perfect, but they are doing a reasonably good job of properly rating incoming talent. Every year, the NFL drafts roughly 12 Pro Bowlers in the first round, 5 of them in the top ten picks.

Drafting has changed immeasurably over the years. Teams used to draft players by simply perusing a copy of Street & Smith's (ask your parents, kids), now we have the Combine and endless days of tests and evaluations. Has that made teams any better at drafting?

Actually, yes. Taking a sampling of how teams drafted through the decades, using years spaced a decade apart, we can see the improvement in NFL drafts:

In 1964, teams drafted 4 Hall of Famers in the first round and two more in the second. So, they knew something back then, but it does seem teams have gotten better at this. It seems top ten picks have been relatively constant, and 1974 is just the old NFL's answer to 2009 and not a trend, but teams have gotten more value out of the entire first round.

The Combine and the rise of the systematic measurable is one of the biggest changes in player evaluation. I'm a firm believer in just watch them play and you can tell who is good, but I have to concede that player evaluation has improved, and that very well could be the increased reliance on verifiable data gained from the Combine.

You don't need a weatherman to know which way the wind blows, and you don't need a 40 time to tell you who is a great player. Those players taken in the top ten are just great football players, taken as a group, and you don't need the extra help to tell who is an elite player.

That said, once you get out of the top ten picks, it gets harder to identify talent. Teams have a lot more data to work from than they did twenty years ago, and it shows in their performance.

It's easy to be dismissive of the Underwear Olympics. Guys with flashy 40 times and perfect body types fail in the NFL all of the time. However, teams are beginning to succeed more than they fail when it comes to identifying talent outside of the top ten picks of the draft.

When it comes to the late first round and beyond, those measurables matter. The reason they matter is because they seem to be good indicators of success. Production is great, and I'm a firm believer in production as a leading indicator of future production, but I can't pretend it's just random chance teams are getting better at drafting players now that they don't rely exclusively on college performance.

Which is a long way to say that NFL GM's are rightfully skeptical of Jarvis Landry. It doesn't mean they are right, as no one is even close to being right even half of the time, but that it is perfectly reasonable to be skittish after his Combine performance.