Over the last five seasons, four franchises have consistently been at the top of the NFL. The Packers, Seahawks, Patriots, and Broncos have had a strong run of success, with each team featuring a GM known for their draft savvy. The prevailing sense amongst Vikings fans is that our own GM, Rick Spielman, has similar savvy; the kind that can keep a team in contention for the long run.
Like many Vikings fans, I want to believe that our GM is one of the best in the business, so I decided to do an analytic-based comparison of Spielman's drafting ability to the GMs of the fantastic four franchises: Ted Thompson, John Schneider, Bill Belichick, and John Elway.
To test the drafting chops of these GMs, I want to focus on how much value each GM acquired relative to the amount of draft capital invested. In other words, I would naturally expect that the Vikings draft better players overall than the Patriots, simply because we are usually picking much higher in the draft. In order to compare the GMs, we first need a few analytic tools:
1) we need a way to assign value to a player's performance for their team
2) we need a way to properly account for the different outcomes expected based on where a player is drafted
3) we need a way to assign expectations based on where a player is in his career (i.e. what value do we expect a rookie to produce vs a second year player?)
Luckily, two sources provide us what we need for this study. Properly assigning value to a player's performance is an incredibly hard (impossible) thing to do, but Pro Football Reference's "Approximate Value" statistic is a decent proxy. You can read about how the statistic is calculated here. It is by no means a perfect measure, but it gives you a general sense for how much value a player has provided for each season they've played.
Once we have a way of assigning value to a player, we need to determine an expectation for the value produced given where the player was drafted. A player drafted in 2011 with a career AV of 10 would be considered a bust if they were taken in the first round, but a massive steal if they were taken in the seventh. Again, we aren't just looking at how many good players these GMs drafted, but how well they did relative to the draft capital spent. Chase Stuart at Football Perspective provides us with a handy resource to help answer this question. He created a draft value chart that assigned an expected total "marginal" AV for each pick in the draft for the first five years of a player's career. "Marginal" in this case simply means that Chase only counts AV for a given season if it was over 2, essentially giving a player no credit for just making the team (they need to contribute to add value).
So we now have a way to value players (AV), and a way to determine what total value should be expected given where they were drafted (Draft Value Chart). The last piece is determining expectations for value by year in the league. The 9th overall pick in the draft is expected to have 20.6 in total marginal AV for the first five years of their career. So if Anthony Barr's rookie year marginal AV was 5, is that better than expected or not? Chase Stuart again comes through for us by providing a breakdown of expected marginal AV by draft pick and by year in the league here.
With all three pieces, we can now look at which players were drafted by each team, determine how much value they provided relative to their expectation (by draft status and year in league), and see which of these five GMs come out on top. Before showing the results, I have two small caveats:
1) One downside of this approach is I won't be accounting for the value these GMs create by trading back in the draft, or by trading draft picks for proven players (something Rick has generally done well). This is simply a study of which GMs get the most value out of their draft selections.
2) I decided to base my comparison on drafts from 2011-2015. Spielman wasn't promoted to GM until 2012, so this comparison won't be perfect, but the new rookie wage scale started in 2011 and it seemed like a natural place to begin the study.
Okay, now for the final results:
|GM||Rick Spielman||Ted Thompson||John Schneider||Bill Belichick||John Elway|
|Actual Marginal AV||184||155||295||169||181|
|Expected Marginal AV||150.7||113.2||110.5||135.2||139.5|
As you can see, John Schneider has absolutely crushed every other GM in total value added, as well as value added relative to expectation. In fact, in the 2011 and 2012 drafts alone, Schneider added 267 points of marginal AV, a number that still blows every other GM's five year haul out of the water. This result wasn't entirely unexpected given the Seahawks managed to come away with Richard Sherman, K.J. Wright, Bruce Irvin, Bobby Wagner, and Russell Wilson in a two-year span. I shudder to think what his total marginal AV added would have been if he didn't trade first rounders for Percy Harvin and Jimmy Graham.
Beyond the Seahawks, the results are very interesting. Spielman came away with the second highest total marginal AV added over the five-year span, with Elway not far behind. When looking at returns relative to the draft capital invested, Spielman came in last, however his returns are right in line with those of Thompson, Belichick, and Elway. In fact, I would argue there isn't a statistically significant difference between their results. Our guy is just as good at drafting as the GMs for the top franchises in the NFL (other than Schneider's incredible two-year run), and in any given year, we can conservatively expect Rick to get 20% more value from his picks than the average GM based on this study.
To add even more optimism for Vikings fans, the early returns from Spielman's 2014 & 2015 draft classes (the two with Zimmer) have been 192% and 279%, respectively. The 2015 class doesn't even include much from Trae Waynes and Danielle Hunter, two players that combined for 1 marginal AV in 2015, but will likely be significant contributors down the road. Overall, Rick has done a fantastic job drafting throughout the last five years. The Vikings are poised to join the ranks of the NFL's elite teams with Rick Spielman (and coach Zimmer) leading the way.
If you'd like to see data from each draft pick in the last five years for each of the five teams studied, please see this google doc: draft data