≡ Menu

An ordinary hit

An ordinary hit.

Just about every article about Trent Richardson references his unimpressive 3.6 yards per carry average from last season. That stat is pretty meaningless, in my opinion. I suppose if you took a random running back from NFL history, and his YPC average in one random season was 3.6, and I knew nothing else about the player, I suppose I would probably assume that the running back was not (or was not going to become) a star. But Richardson isn’t a random running back from a database, because…

  • His 3.6 YPC average came on 267 carries, which represented 77% of all carries by Cleveland running backs
  • He was a rookie last year
  • He was a high first round pick

Since 1970, only 13 first round rookies have recorded 70% of all running back carries by their team. Two of those players were Richardson and Tampa Bay’s Doug Martin last year. Of that group, Richardson did post the lowest YPC average, but he was within 0.1 YPC of LaDainian Tomlinson. The next two lowest averages belong to Robert Edwards and Emmitt Smith; the former suffered a career-debilitating injury in a beach football game after his rookie season, while the latter ran for the most yards in NFL history.

Yeah, Richardson’s yards per carry average was well below average. But the universe of first round running backs who became workhorses right away as rookies and had a low YPC average consists of a HOF running back, a future HOF running back, and a player who suffered the flukiest of injuries. Richardson has something else in common with Emmitt Smith: after both of their rookie seasons, Norv Turner came on board as offensive coordinator.

But let’s say you don’t want to give Richardson any credit for his draft status. And you’re not in the mood to give him a pass just because he was a rookie. OK. Since 1990, 48 running backs have averaged fewer than 3.8 yards per carry while recording at least 70% of all running back carries for their team. Twenty-six of those players were at least 27 years old, and on the back half of their careers. Here are the other 22 running backs:

2012Trent RichardsonCLE76.7222679503.5611513671
2009Matt ForteCHI78.9242589293.64574710
2003Ricky WilliamsMIA89.52639213723.59503511
2002Edgerrin JamesIND70.7242779893.572613541
2001LaDainian TomlinsonSDG91.42233912363.6510593670
1999Curtis EnisCHI84.2232879163.193453402
1998Jerome BettisPIT79.62631611853.75316900
1998Curtis MartinNYJ79.92536912873.498433651
1998Eddie GeorgeTEN92.12534812943.725373101
1997Jamal AndersonATL75.12529010023.467292843
1997Karim Abdul-JabbarMIA73.1232838923.1515292611
1996Curtis MartinNWE80.82331611523.6514463333
1996Karim Abdul-JabbarMIA70.92230711163.6411231390
1995Edgar BennettGNB882631610673.383616484
1995Ricky WattersPHI74.42633712733.7811624341
1995Errict RhettTAM91.52533212073.6411141100
1995Garrison HearstARI782428410703.771292431
1995Marshall FaulkIND71.52228910783.7311564753
1994Rodney HamptonNYG72.32532710753.296141030
1994Errict RhettTAM71.92428410113.567221190
1994Jerome BettisRAM90.62231910253.213312931
1992Reggie CobbTAM79.12431011713.789211560

That looks to me like a pretty impressive group of running backs, including some of the best runners of the last two decades. Note that this excludes Marshawn Lynch, who was the main running back for the Seahawks in 2010. He averaged 3.5 YPC, but since he was traded from Buffalo in mid-season, he didn’t hit the 70% threshold.

A player’s yards per carry average can be inflated by a few large runs, and it can be deflated by the lack thereof. Richardson’s longest run was 32 yards; his second longest was 26, and both of those went for touchdowns. Those were Richardson’s only two carries of 20+ yards last year, and a 19-yarder against the Ravens was his only other run that cracked 15 yards. Personally, I’m not worried about Richardson’s ability to hit the home run. I suspect Richardson will easily eclipse two 20+ yard runs and three 15+ yard runs in 2013.

What’s more important is consistent success in the running game. One thing I like to look at is 1st and 10 runs. There were 24 players with 100 rush attempts in that situation last year. On average, the group gained at least 4 yards on 46.0% of their carries. Richardson actually eclipsed the average there, picking up 4 yards on 47.3% of his runs.

Running backRush4+ ydsSucc Rate
Shonn Greene1507952.7
C.J. Spiller1166152.6
Alfred Morris1789352.2
Stevan Ridley1427250.7
Adrian Peterson1778950.3
Jamaal Charles1628150
Marshawn Lynch1487450
Reggie Bush1286450
Frank Gore1336548.9
Ahmad Bradshaw1135548.7
Trent Richardson1316247.3
Steven Jackson1426747.2
Mark Ingram1044947.1
Doug Martin1808446.7
Ray Rice1225545.1
Ryan Mathews1104944.5
Arian Foster1506644
Chris Johnson1536743.8
Mikel Leshoure1235141.5
Matt Forte1385741.3
DeAngelo Williams1064340.6
Vick Ballard1094440.4
Michael Turner1194739.5
BenJarvus Green-Ellis1566038.5
Darren McFadden1042726

Richardson played just fine as a rookie. His YPC average was poor, but he played on a bad offense and was banged up for much of the year. A few more big runs, and no one would be talking about his 3.6 average gain. If you liked Richardson last year, I see no reason not to like him this year.

  • Scott Tanner

    I think you’re largely right about Richardson. He was in an impossible situation, he’s still incredibly talented, and I think we can expect good things forward. But I don’t think that list you provided shows that YPC is a stupid or meaningless stat. It may be, for reasons independent of this. That list, however, just seems to show a bunch of other talented running backs in similarly crappy situations. If YPC was stupid/meaningless, then we wouldn’t necessarily expect to see the YPC of the players on that list (Curtis Martin, LT, Forte, George, etc.) improve dramatically, even though I’m pretty sure most of those guys enjoyed pretty dramatic improvements in YPC later on.

    The counter to that argument is probably something like “yeah, because their teams improved.” Which is probably true! But at that point it seems like we’re so far down the rabbit hole that it becomes difficult to assign value to any stat in a team sport. Receivers aren’t responsible for their yards per route because it’s so quarterback dependent, quarterbacks aren’t responsible for completion percentage because it’s so offensive line dependent, etc. This is obviously the argument stretched to an absurd extreme, but I’m not convinced these stats show the uselessness of YPC.

    You point out that Richardson didn’t have any long runs, which depresses his YPC. And you say that this is sort of flukey, and think he can improve next year. This very well may be! But it’s only really a point specific to Trent Richardson, not about YPC. I think factoring in long runs is a feature, not a bug for YPC. Chris Johnson’s ability to break off giant runs may inflate his YPC, but it’s also representing real and obvious value.

    I’m not saying YPC is perfect; it certainly isn’t. It can’t capture the value of a RB intentionally diving over a pile on 3rd and 1 and getting a really valuable first down but only getting 1 yard and deflating their YPC. But I haven’t seen an alternative formulation that does a much better job of valuing RBs. In general, I actually think the subjective eye test does a better job evaluating RBs than most other position, but that’s sort of a separate argument. I don’t mean to hijack the discussion about Richardson, especially because I actually agree with you about him. But I would be interested to see you flesh out the YPC argument further.

    • Chase Stuart

      Good stuff, Scott. I agree that digging deeper on YPC is a good idea — I have some posts in draft mode on the subject.

    • Neil

      I’m with you, Chase, on patience w/ Richardson, but I chuckled at the line about how he’d be viewed better if he wasn’t banged up for much of the year and had a few more big runs… Well YEAH, you could say the same about pretty much every RB in the league. 🙂 Isn’t one of the big tenets of Football Outsiders’ research that the ability to bust big runs is maybe *the* defining difference between good RBs and mediocre ones? The line pretty much does all the work for the first few yards; it’s a guy’s ability to get to the second level and beyond that determines whether he’s a good RB or not.

      • Mike

        Neil, that might be the single most insightful comment I have ever read on the internet. Bravo!

      • Chase Stuart

        Well, I think the issue is Richardson was injured and had few big runs at a disproportionate rate to the rest of the league.

        With big runs there are two issues at play, IMO. Luck and value.

        As far as value, every yard gained after a new first down is achieved is the least valuable type of yard around — it’s the same as a punt return yard. So the difference between a 24 yard run and a 4 yard run on 3rd and 3 is the same as 20 yards of special teams yards. They’re valuable, of course, but not nearly as valuable on a per-yard basis as the first four yards.

        According to Burke, 3rd and 3 from the 26 is worth +0.06 EP, while 1st and 10 from the 30 is worth +0.90 EP. That means on a 4-yard run on 3rd and 3, the back was picking up 0.21 EP/yard. If the RB instead got to the 50, that’s worth +2.04 EP, which means those last 20 yards were worth only 0.057 EP/yard. Valuable, but not nearly as valuable per yard. And that’s why YPC can be really misleading, since it treats all yards equally.

        As for the “luck” part of it, I suppose we need to split luck into two parts. Part of it is small sample size variance. If a RB gets a “big” run every 50 runs, you might not see his true value over 277 carries. Maybe instead of 5 or 6 big runs, he has only 2 some years. Other years he has 10. That wildly blows his YPC out of proportion.

        As for the second part of luck…. are big gains in general a matter of luck? Over a long sample size, no, I don’t think so. But things like missed tackles and other elements that contribute to a long run that are outside of a RB’s control can play a big part in getting big runs. This is an issue with all stats, but YPC is so sensitive to outliers, IMO, that it’s worth keeping in mind.

        • Neil

          Good point about the value of yards inside first-down range vs. additional yards beyond the sticks. There’s certainly a ton of value in consistently banging out 3-4 yards toward a first down on any given carry. If you had some kind of robo-RB who gained a guaranteed 2.5 YPC every time he touched the ball, his YPC would be horrible, among the worst in the league… and yet he’d easily be the most valuable player in the NFL.

          So I agree, it’s a combo of home runs and singles that makes a RB great. My only point is that it’s a little circular to say “his YPC would have been better if he’d only had a few more long gains”… Well, of course that’s true, but a few more long gains would also mean he’d have been a better RB, because the ability to break off long gains largely defines what it means to be a good RB.

          That’s the whole point behind FO’s “adjusted line yards”. If a play gets blown up in the backfield or chugs ahead for 3 yds and a cloud of dust, that’s almost 100% on the O-line. If a run goes for longer than that, the OL still gets credit for the 1st few yards, but everything else is on the RB. The only flaw in that way of thinking is when you have guys tiptoeing around looking for holes to hit a HR in — when they go boom, they get the credit, and when they go bust, they get none of the blame.

          • Chase Stuart

            I think FO actually caps long runs, which I agree with. I think part of the problem with YPC is the difference between a 60 yard run and a 30 yard run often has little to do with the talent of the RB, but 30 yards is one full tenth of a YPC for a 300 carry back.

            And again, I think the variance issue is at play here. If long runs are say, like INTs, than a full season won’t tell us much. But it will really throw off the YPC of a player.

            • sn0mm1s

              Yes, they cap long runs and I have argued against it from pretty much the beginning of their presence on the web (at least when judging the play of the individual RB). It is fine to do that when they are trying to correlate their DYAR/DVOA to team wins but I think it really sells good players short. In the DYAR/DVOA world a RB that has 3 consecutive 10 yard carries (moving his team from their own 20 to the 50) is worth more than another RB taking the ball 79 yards on the first carry and being tackled at the 1. The first scenario implies repeatability of a team’s ability to successfully run (which is why I have no problem with DVOA being a team stat). However, no one on this planet would trade those 3 10 yard runs for the 79 yarder.

              Another thing to point out is that Richardson’s YPC is low, but it is extremely low when compared to the league’s YPC. Over the past 10 years or so the leaguewide YPC for RBs is historically at its peak. The 1990s was the most difficult era to run in judged by YPC – this past decade or so has been by far the easiest. 2012 Richardson averaged 3.56 YPC , average carry for all RBs (excluding TRich) was 4.25, the top 31 RBs (assuming 31 other starters) ranked by carries (excluding TRich) was 4.42.

              Now, compare that to Emmitt Smith’s 3.89 YPC as a rookie. League average was 4.05. Top 27 RBs churned out 4.21.

              Tomlinson had 3.65 YPC, league average 4.05, top 31 4.12.

              I don’t think the future looks as rosy for TRich as you are making it out.

              • sn0mm1s

                It looks like in Burke’s world those 3 10 yard runs add 1.7 pts while the 79 yarder adds 5.62 pts. I find that a much more palatable result.

                • Chase Stuart

                  Note that ignoring YPC and just using yards gets you to this place, too. I would not argue that 3 10-yard runs is better than 1 79-yard run.

              • Chase Stuart

                Good stuff, especially about the era adjustments. One counter, though: the read option seems to be a big driver in the increase in YPC, and that wasn’t a part of the Cleveland offense.

                • sn0mm1s

                  That might factor in, but this increase in YPC has been occurring for longer than most teams were running a read option (which hasn’t been long and the teams running it are still few). I have suspected for a while that it is more due to many teams passing with greater efficiency. You no longer really build your defense around stopping the run – since the pass is so much more effective. I haven’t done the research but I would expect the stricter enforcement of the passing rules around 2004 has caused an uptick in rushing numbers.

                  Also, I used leading rushers (not carriers) for the top X comparisons by mistake – but I doubt it would change things much.

          • Chase Stuart

            I suppose now that we have PBP data going back to 2000, I could really dig into some studies on long runs. I’ll throw that on to the ever expanding to-do list.

          • Matt Marcinkiewicz

            Fred Jackson.

  • Wade Iuele

    I really like that table of 4+ yards gained on 1st and 10. Some interesting usage data for lots of RBs. Thank you!

  • That trend toward higher YPC for runners began with the re-emphasis of the 1978 defensive contact rules, in 2004.
    I have adapted an adjustment for comparing pre-2004 and post-2004 numbers for QBs, RBs, & WRs/TEs.
    1,200 yards is the new 1,000 yards rushing (+20%)/1,000 post-2004 yards rushing is the pre-2004 850 yards rushing (85%).
    Don’t ask me to explain the different percentages going up vs going backward because I can’t, it just seems to work out.
    Likewise I adjust QB passing yardage by 20% going up from pre-2004 numbers and cut by 15% when translating to pre-2004 numbers.
    Dan Marino’s 1984 record 5,084 yards would be an equivalent 6,101 yards in today’s pass-happy league & Drew Brees’ record breaking 5,476 yards passing would only be worth 4,655 yards back in the day.

    • sn0mm1s

      I don’t think that is the case for YPC. The 1990s was the most difficult decade to run in and of the great RBs Barry Sanders ran against the toughest NFL judged by YPC. The average RB while Barry played was 3.92 excluding his carries.

  • Damon Thurman

    Great article and comments. Very interesting and insightful. I guess the jury is still out on Richardson. About YPC, I was recently looking at a high mileage car that was supposedly mostly highway miles and I asked a mechanic friend of mine if he put any weight in highway miles being easier on a car. He said to him miles are miles, period. We may be over thinking this. YPC is YPC. Diving for a 1 yard first down or ripping off a long run should balance itself out over the course of a season. I think 3.6 YPC is a valid concern. Tell me the last top five running back that averaged less than four yards per carry? You can name many big names who averaged less than 4 yards per carry at some point in their career but I doubt they were a top five back that year. Bottom line, Richardson must up his YPC considerably to be an elite back.

  • Dave

    A few thoughts:

    – Yes, that top list includes some good RB’s, but it also includes many mediocre ones (Curtis Enis, Rodney Hampton, Errict Rhett, Karim Abdul-Jabbar, Edgar Bennett, etc.);

    – As for the second list, I am dubious about any measure of RB “success” that puts Shonn Greene at the top of the list;

    – Richardson also is bringing value in the passing game (51 catches for 357 yards in 2012), which shouldn’t be overlooked;

    – It is almost impossible statistically to separate the success of an RB from the quality of his offensive line. YPC certainly doesn’t;

    – It’s also difficult to separate it from the quality of a team’s passing game, though Adrian Peterson is a notable example of an RB who has succeeded even when his team has a weak passing offense.

  • Adam

    It isn’t about numbers. It’s about gamefilm. If you watch Trent Richardson run, he doesn’t hit the holes. Sure there may not have been very many holes in 2012 Cleveland, but there were some monster holes that Trent didn’t hit hard enough and got stopped for little or no gain. There were other times where he would have a hole, not recognize it and bounce to the outside, losing 2 yards in the process. He has no vision for a RB and you can’t coach that.

  • Hi there! I know this is somewhat off topic but I was wondering which blog platform are you using for this
    site? I’m getting fed up of WordPress because I’ve had problems with hackers and I’m looking at options
    for another platform. I would be great if you could point me in the
    direction of a good platform.

  • GPJ

    Existentialist football writer says “eeeeeh, fumbles, 0.3 YPC…nothing matters.”