## Analyzing Position Values in the NFL

Every draft pick has a value, as seen in my draft value chart.  When the first overall pick is used on a quarterback, that means the quarterback position gets credited with 34.6 picks. If you assign a value to every pick in each of the last ten drafts, you can get a sense of the amount of value spent on each position in the NFL in an average draft. The graph below shows the percentage of the draft value pie attributed to each position; for example, quarterbacks are selected with 7% of all draft capital:

The Pro-Football-Reference draft database separates defensive ends from defensive tackles, but groups linebackers together as well as cornerbacks and safeties together. In that light, it’s not surprising to see defensive backs taking the most prominent position on the chart. Now, let’s engage in some back-of-the-envelope math, and assume that each team plays 1 QB, 1.2 RBs, 1.3 TEs, 2.5 WRs, 2 Ts, 2 Gs, 1 C, 2 DEs, 1.5 DTs, 3 LBs, and 4.5 DBs. If we divide the percentages in the above table by these numbers, we can find a draft value per player on the field. That gives us a very different-looking table:

In the abstract, these values may not mean much. But what if we compare them to the 2014 NFL franchise tag values by position? There’s some obvious flaws with this sort of comparison — ten years of draft data versus one year of the franchise tag being a big one — but I thought it might serve as a useful benchmark. The graph below displays those franchise tag values, although note that the franchise tag groups all offensive linemen together; for defensive backs, I took the average of the tag value for cornerbacks and safeties.

So what conclusions can we draw? Quarterbacks seem slightly undervalued in the draft. How is that possible, in a a system that makes Jake Locker, Blaine Gabbert, and Christian Ponder high first round picks? I suspect this is an issue of (lack of) quantity over quality. On a per-starter basis (i.e., graph 2) quarterback was still the position that saw the most draft capital spent on it, even if it wasn’t as extreme as it looks on the franchise tag graph. But teams are not keen on using high picks on backup quarterbacks the way they are at other positions, which reduces the total draft capital spent on the position. In addition, I think the long life cycle of a quarterback — think of how the Patriots and Saints haven’t had much of a need to invest picks in quarterbacks — tends to enhance this effect, too.

Running back is a harder position to reconcile. Teams seem to love spending picks on running backs even though usually there is just one starting back on the field (although that person could be the focus of the offense). On the other hand, at least anecdotally, it feels like the market has begun to correct itself here. In the 2011, 2012, and 2013 drafts, Trent Richardson was the only player selected in the top 27, and it’s unlikely that any back will go that high in 2014, either. Still, it is kind of odd that, even assuming 1.2 running back starters per team, more draft capital was spent on running backs than any position other than quarterback. It will be very interesting to see how this looks over the next ten years, although there is still one reason for teams to draft running backs: it’s a young man’s position, and teams are wary to invest in older running backs.

One side effect of the emphasis on running backs is that wide receivers and tight ends seem undervalued in the draft: again, this may be changing as teams move towards more pass-happy offenses. The 2014 draft in particularly should be very wide receiver heavy.

On a relative basis, the tackle free agent tag seems pretty similar to the draft capital spent at the position (i.e., a hair behind wide receivers and a bit less than quarterbacks). We don’t have franchise tag data for guards and centers, but if we did, I imagine it would look pretty similar to the second graph chart, Alex Mack-deal notwithstanding.

The defensive end/defensive tackle comparison is pretty interesting. More draft capital is spent on ends on a gross basis, but I think it’s pretty safe to assume that teams tend to put more defensive ends on the field than defensive tackles (not only all 3-4 teams, but 4-3 teams in nickel and dime situations). The franchise tag for defensive ends is 3.5 million dollars greater than it is at defensive tackle, yet more draft capital is spent per “starter” on the interior players.

Any thoughts? I’ve got a couple of theories, although I don’t know if either is particularly sound. The planet theory of defensive tackles — i.e., there are only a few men on the planet who are extremely athletic and weigh 300 pounds — tends to push tackles up the draft board. On the other hand, wouldn’t this incentivize teams to pay money to the great defensive tackles, since they are so irreplaceable (which would drive up the franchise tag value?) Another is that relative to defensive ends, teams may view tackles as “safer” picks. Perhaps franchises think run stuffing monsters as easier to scout and less likely to bust, which leads to them being drafted with earlier picks?

My guess is the franchise tag for linebackers is driven up by 3-4 edge rushers, so I’m not sure if many conclusions can be drawn at that position. I think over the next ten years, we’ll continue to see a demphasis placed on all non-pass rushing linebackers. Linebackers who excel in coverage are extremely valuable, too, but they don’t seem to generate much draft buzz.

Defensive backs are considered slightly less valuable than defensive linemen by both the draft value chart and the franchise tag. That’s probably not going to change in the future, although the number of defensive backs on the field per play is only going to increase. In particular, it will be interesting to see if the success of players like Earl Thomas will drive up the value of the safety position.

• Sunrise089

Grouping defensive backs seems dicey since safety is so much less valued than CB.

I rarely want to suggest RBs are MORE valuable than might be suggested, but I think the starting 1.2rb’s figure distorts things and then leads to the draft value number being too high versus QB. RBs substitute way more than QB and I suspect more than #1 WR or TE. I’m actually not sure the RB should be punished for that when assessing their contribution.

• Chase Stuart

Unfortunately, that’s the way the data has been kept. Going forward, I think the plan will be to separate CB from S.

1.2 doesn’t refer to the number of RBs that play per game, but the number of RBs on the field per snap. To the extent running backs split time with each other, that should actually decrease the draft capital spent on running backs, since the curve is (reverse) logarithmic.

Should running backs per game be higher than 1.2? Seems lik it.

A backup quarterback has marginal value, while a backup running back sees action every game.

Maybe look at snap count per position and weigh for that?

• Chase Stuart

My assumption is that snap counts per position would be about 1.2 per snap count for running backs.

• Michael

Another theory on DT vs DE. You can always convert a LB into a DE so the pool of players at DE is bigger. There are a lot of athletic 250-280 lbs guys. But there aren’t as many athletic 300 lbs guys.

• Rob Harrison

That’s the planet theory, which he mentioned.

• wiesengrund

One aspect of the DE sv. DT disparity is maybe that the franchise tag numbers are “overvaluing” DEs due to their much more pronounced presence on stat sheets? I.E. the GMs and coaches who build a team know that interior lineplay is just as important as the edge (hence the relativly similar draft value), but agents of 10+ sack guys do not allow their clients to paid like 3-sack DTs and hence the overall skewage of contract (and thererfor: franchise tag) numbers?