What Happened to the First Round RB?

In the five-year period between 1970 through 1974, running backs made up 20% of all first round NFL draft picks. That's one out of every five. As recently as the 1985-1989 period, RBs made up 19% of first rounders. But by the most recent decade, from 2000 through 2010, RB selection was cut in half--down to about 10%. Last night, only 1 of the 32 players chosen (about 3%) was a RB, and he was chosen 28th, near the bottom of the round.

The graph below illustrates the trends in how teams favor each position over the past 41 years. Most positions are fairly stable. Click to expand.


I take this as a pretty clear indication of the relative values and scarcities of each type of player at the elite level. As passing becomes more and more prominent in the sport, and as teams realize that the success of the running game depends less on the singular talent of the ball carrier than on other factors, fewer RBs are selected.

So what positions are making up the difference? Defense backs primarily, having gone from 12% of first rounders in the '70-'74 period to 18% in the '05-'10 period. This squares with the ascent of the passing game.

One curiosity is the very high rate of selection of defensive linemen compared to offensive linemen. There are often four, and sometimes three, defensive linemen on the field at a time, and always five offensive linemen on the field. But defensive linemen are consistently selected more often than their offensive counterparts.

Notes: I included fullbacks as RBs in this analysis, but there were an insignificant number of them chosen in the first round: 1 in '80-'84, 1 in '85-'89, and 2 in '90-'94.

Data are from the Pro-Football-Reference.com Play Index Draft Finder.

  • Spread The Love
  • Digg This Post
  • Tweet This Post
  • Stumble This Post
  • Submit This Post To Delicious
  • Submit This Post To Reddit
  • Submit This Post To Mixx

21 Responses to “What Happened to the First Round RB?”

  1. David Myers says:

    Brian,

    I'd have to agree. The data set that drives the "Myth of 370" is compelling proof that the smart play are two pretty good running backs rather than a single stud for whom you have unrealistic expectations.

    David.

  2. Adam H says:

    My theory: just like offense is more important than defense, the quality of the D-line is more important than the quality of the O-line. The D-line attacks while the O-line prevents, especially in pass plays (which have become more prevalent). So, just like, in general, offense determines the success of the play more than the defense (interceptions, passer rating, run success rate, etc.), the attacking D-line is more deterministic than the defending O-line in the success of each play.

    I hope my thought process makes sense... To me, it all ties together. It is easier to "do" than it is to "stop someone from doing".

  3. Erik Jensen says:

    My guess is that D-line is preferred over O-line in the draft because of how the players develop over time. It strikes me that plenty of D-linemen (especially ends) have made a big impact in their first few years while most O-linemen take a while to develop. It might be interesting to take a look at average age of starters or Pro Bowl players at both positions over the years.

  4. Brian Burke says:

    Erik-That's my perception too. Immediate impact.

    One of my thesis advisors was named Erik Jensen.

  5. Vince says:

    I'd guess that the difference on the lines is on the inside rather than the outside. On the outside, both OTs and DEs need a special kind of body to be elite - giant and nimble at OT and explosive and pretty big at DE - so the ones that fit the mold are worth an early pick. On the interior of the line, elite DTs also need special bodies - either a bigger but still explosive body to be a penetrating one-gapper or a planet theory hulk to occupy 2 blockers - but the G/Cs don't need to fit such a rare mold so they aren't as worthy of first round picks.

    Do the numbers back me up? I checked PFR: in the first round from 1980-2010 (which is how PFR divides up the years) on the interior there were 74 DT/NTs vs. only 55 G/Cs and on the outside there were 121 DEs vs. only 98 OTs. So, not really - only about half the gap is on the interior (although the defense-offense ratio is more lopsided for the interior). Perhaps it's only LTs, and not RTs, who need a special body type?

  6. Joseph says:

    On the OL vs. DL difference--I think it also has to do with defensive lines rotating, vs. O-lines NOT rotating.
    Another thought--do some of the "DE's" get drafted to play 3-4 OLB's?

  7. Tarr says:

    I was going to say the same thing as Joseph - I would bet that more D-linemen see 10+ snaps in an average game than O-linemen.

    Even setting aside the pass/run efficiency gap, there's a number pf things that could drive the decline in RB drafting:

    1) Marginal value - if the production difference between an expected production of a top RB and a middling RB is smaller than the same gap for other positions, then RBs are less useful to draft. Of course, increased reliance on the pass would exacerbate this effect.

    2) Injury risk - if RBs get hurt more often than other positions, then the value of RB picks is lower.

    I'm not sure either of these are the case, but they could be factors that are becoming more clear as teams become more data-savvy.

  8. Alvin says:

    I agree, the risk of drafting a running back is way too high with the injury risk. Not to mention the fact that their careers are so short, NFL are trying to avoid the signing bonus money they have to pay when they pick a guy up in the first round.

    They're afraid they won't get a return on investment from the huge signing bonus.

  9. Anonymous says:

    i think it's just a matter that there are not many "can't miss" rbs are out there. Your Peterson's and Bushes will allways go high, but i don't see that quality any more. The spreads they run in college have placed a premium on smaller, more elusive backs, and the big studs don't get enough carries to show what they can do.

  10. Anonymous says:

    The OL vs. DL disparity is probably mainly the function of their marginal value. An elite DL makes a much bigger difference than the average DL compared to the difference between an elite OL (especially interior OL) and an average one.

    Teams shoot for elite production in the first round and expect average in the middle rounds.

  11. Anonymous says:

    If OL vs. DL draft position is a function of marginal value, shouldn't salaries correlate as well? DTs average $1.22 million, OLs average 1.27 mil, and DEs 1.58. Hopefully someone can find an OL salary breakdown by position.

  12. Bryan Morgan says:

    Is there a way to correlate this to "time missed for injuries". At least part of the reason RBs are increasingly viewed as interchangeable parts is that there's no such thing any more of a guy you can ride for 16 games per year at 25+ carries per game.

    Just wondering if there is a chance the RB position more than others has seen an increase in the "injuries per play" stat, compared to other positions, causing teams to be much more careful in investing large sums of money into this position.

  13. John Grenci says:

    Brian, This is probably hard to do, but is there a reasonably way (one would have to look at every play) to determine yards after contact for a running back? I think this would most strongly get to how valuable a running back is.

    One of the problems with player WPA's (or EPA's) is that it is hard to discern how much contribution is made by the rest of the offense.

    Certainly, Yards after contact is not the end all and be all for a RB, but I think it is one of the two biggest factors (finding the hole/hitting the hole being the other).

    If you were able to do this, and find that the running backs break tackles at similar rates (could use a coefficient of variation argument) and if the assumption that breaking tackles is the biggest factor for a running backs' ability, then it supports not taking running backs early in the draft.


    John

  14. Anonymous says:

    "If OL vs. DL draft position is a function of marginal value, shouldn't salaries correlate as well? DTs average $1.22 million, OLs average 1.27 mil, and DEs 1.58."

    Average is a bad way to look at marginal value. You should look at the spread between average salary and elite salary. Franchise tag for DE is 12.4 and OL (applicable almost only to LT, of course) is 10.7. The top five interior OL are paid much less.

  15. Anonymous says:

    smartfootball.com has a post on this also. Here's my comment from there:
    Part of the reason RB's are being drafted lower, as you've alluded to, is that the demand for them has greatly decreased. Part of it is the now somewhat standard use of 1-back sets, but also the fact that teams use less RB's in rotation, a by-product of more passing. Consider that only 1 RB prior to 1951 carried even 50% of the rushing load in any given season for his team, and that it wasn't until Jim Brown in 1960 (I believe, I may be off a year) that a player exceeded 60% of his team's carries. Teams used 3-5 backs as standard practice, and therefore needed several on the roster - now teams rarely roster more than four. So demand has dropped sharply, far more than the supply has. It took awhile for the trend to be established in the draft, though, because even though teams rostered less, the top guy on each team handled the ball 20-25% of the team's offensive plays, far and away the most usage after QB. Higher supply, less demand = lower draft position. Also, with teams now returning to more dual-RB workloads (by design, as opposed to being forced to by injury or ineffectiveness) the demand for the everydown workhorse is being pushed down further. More and more resources are being poured into the receiver positions (WR, TE), and that roster space has been taken from the backfield positions. (Interesting ... very few teams have enough quality players to fill the top 3-4 receiving positions, suggesting that the supply is much lower than the current demand. One of the reasons I think this coming season is going to showcase a significant increase in rushing attempts league-wide.)

  16. Anonymous says:

    Running back seems to be a disposable position. Even bad teams can stop the run if they load the box, giving more big bodies to crush a running back increasing chance of injury. Any definitive stats on the higher reward for passing over running? Regardless, A predominantly run first team in the NFL would be an anachronism.

  17. Independent George says:

    Can we call this the 'Shannahan Effect'? The years don't align exactly, but the Champ Bailey-Clinton Portis trade seems to illustrate the reasoning behind the shift better than any particular draft result.

  18. Blake W. says:

    As for OL v DL, don't DL have shorter careers? If so they would need to be replaced more frequently.

  19. Blahblah says:

    I think a lot of it is that we're seeing what is more mature market for football players. These days we are less likely to see a player who is just head and shoulders better than everyone else on the field. You see that all the time in high school and to a lesser extent in college. Coaches put those players at QB and RB. At each step, the RB becomes less important because the players around him are comparatively better (tha gap caused by a lack of quality players shrinks)

    Over time, the same things happen. There are more football players, more specialization, better training, etc that make it less likely that a guy will be surrounded by markedly inferior players. That's going to make the individual less important and the team more important. QB is an obvious counterexample, but I think that is just because of a number of factors about the game that is making passing more important. That's a situation where I think maybe the position of QB is getting more important, not a change in the quality of the players themselves relative to other players

  20. Anonymous says:

    I think it would be nice to see the same graph by split by number of those positions on the field. OL will always be 5 and QB 1; the DL and LBs would need to use the average across the league. Overall I think it would be neater this way. At the moment the graph is showing (a lot of the time) simply the number that play that position

  21. j. says:

    this is the coolest site i have ever seen.

Leave a Reply

Note: Only a member of this blog may post a comment.