≡ Menu
Bettis ran for only five yards on this play

Bettis ran for only five yards on this play

Congrats to the newest members of the Pro Football Hall of Fame. You can read my thoughts on the candidates here; while this class is not exactly the one I would have picked, Jerome Bettis, Tim Brown, Charles Haley, Junior Seau, Will Shields, and Mick Tingelhoff were all outstanding players. In addition, Bill Polian and Ron Wolf were the inaugural selections for the Contributors spots, so congratulations to them as well.

The Bettis candidacy is an interesting one. Many want to focus on his underwhelming 3.9 career yards per carry average. But as I have written many times, I am not keen on putting much weight on YPC as a statistic. Brian Burke has also written about how coaches don’t view running backs in terms of yards per carry, but rather by success rate (which correlates poorly with yards per carry). Danny Tuccitto calls yards per carry essentially “a bunkum stat.”

The counter to the argument that Bettis was just an average back is that he ranks 5th on the career rushing list. I’m not too swayed by that, either, as career [any statistic] is rarely the best way to grade players. One thing I’ve done in the past is calculate career rushing yards, but only excluding the first 50 rushing yards in each game.1 So if a running back goes for 80 rushing yards, 45, and 60 in three straight games, he is credited with 40 rushing yards over 50 (i.e., 30 + 0 + 10). This, I think, does a nice job of not giving too much credit to compilers.

For example, if a player sticks around and rushes for 600 yards in a season at age 33, well, he might well end up with just a few dozen yards over 50 yards per game. That means this statistic is much less sensitive to mediocre seasons, which I think is a very good thing. As it turns out, Bettis still fares very well in this metric; only Eric Dickerson passes him, a fact which I assume everyone can agree is appropriate.

Here’s how to read the table below. Note that for all purposes in this article, all rushing performances prior to 1960 have been excluded.2 Players like Jim Brown who qualify based on their post-1959 production are included in the table, but only their statistics since 1960 are shown. Here’s how to read the table: Bettis ranks 7th in this metric, and 6th otherwise in total rushing yards since 1960 with 13,662. Bettis had 5,816 rushing yards over 50, and that translates to 42.6% of his career total.3

RkPlayerTot Rsh RkTot Rush YdYards Over 50Perc
1Emmitt Smith118355827045.1%
2Barry Sanders315269804352.7%
3Walter Payton216726799247.8%
4Eric Dickerson713259670250.5%
5Curtis Martin414101635245%
6LaDainian Tomlinson513684607244.4%
7Jerome Bettis613662581642.6%
8Edgerrin James1012246565146.1%
9O.J. Simpson1811236535947.7%
10Adrian Peterson2710190531152.1%
11Fred Taylor1411695520544.5%
12Tony Dorsett812739507539.8%
13Marshall Faulk912279506041.2%
14Corey Dillon1711241478742.6%
15Thurman Thomas1312074474039.3%
16Clinton Portis299923469247.3%
17Jamal Lewis2210607466143.9%
18Jim Brown368514463254.4%
19Franco Harris1212120461238.1%
20Frank Gore1911073459341.5%
21Tiki Barber2410449453043.4%
22Earl Campbell319407452248.1%
23Shaun Alexander309453442846.8%
24Steven Jackson1511388433838.1%
25Eddie George2510441426840.9%
26Ricky Watters2110643411338.6%
27Ottis Anderson2610273407639.7%
28John Riggins1611352407335.9%
29Ricky Williams2810009398639.8%
30Chris Johnson348628391545.4%
31Terrell Davis527607391151.4%
32Priest Holmes428172384047%
33Ahman Green329205382141.5%
34Thomas Jones2310591378735.8%
35Warrick Dunn2010967365133.3%
36Marcus Allen1112243358429.3%
37Marshawn Lynch338695356941%
38Gerald Riggs418188339041.4%
39Stephen Davis468052326640.6%
40Maurice Jones-Drew438167324039.7%
41Willis McGahee378474319437.7%
42Arian Foster736309318250.4%
43Jamaal Charles606856314845.9%
44George Rogers567176312743.6%
45Jim Taylor497898312539.6%
46Terry Allen358614311036.1%
47Larry Johnson746223303948.8%
48Robert Smith636818300244%
49Michael Turner547338299140.8%
50Matt Forte507704294638.2%
51Garrison Hearst477966289536.3%
52LeSean McCoy646792287342.3%
53Chris Warren517696282436.7%
54Freeman McNeil458074281634.9%
55Leroy Kelly557274275137.8%
56Curt Warner626844272639.8%
57Wilbert Montgomery656789266239.2%
58Herschel Walker398225261731.8%
59Deuce McAllister796096258042.3%
60Travis Henry806086256142.1%
61Lawrence McCutcheon686578254138.6%
62Rudi Johnson865979251542.1%

In my mind, this means Bettis wasn’t much of a compiler. Who was? How about Marcus Allen, who has just 3,584 yards over 50 but 12,243 career rushing yards. Of course, in Allen’s case it may not have been his fault that he was given few carries during his prime years, but the point here isn’t that Allen was bad: just that his 12,243 yards is a bit misleading (of course, his 3,584 yards may be misleading as a measure of how talented he was, too).

Only three players who are not active had a number over 50% in that final column: Brown, Barry Sanders, and Terrell Davis.4 And if you include the playoffs, Davis looks even better, jumping up to #21 when I ran those numbers two years ago.

Bettis was somewhat of a compiler, of course: you have to go down to #12 to find a player who had a lower percentage of “yards over 50” to career yards. And that’s when you get to Tony Dorsett, who was definitely a compiler of sorts. And if you raise the bar from 50 yards to a higher number, Bettis begins to look much worse.5 He is certainly not one of the best running backs in NFL history, and he may even be in the bottom half of running backs in the Hall of Fame. But once you look past his YPC average, he remains one of the most productive running backs (if not necessarily most dominant) in NFL history, at least when it comes to pure rushing.

  1. I had actually forgot about that post before I finished this article, but searching my archive, I found it! There is at least one difference between that post and this post that you might find meaningful: there I included postseason games, while I excluded them here. []
  2. This also applies to the “rank” information in the table. []
  3. I’m not quite so sure how useful this last column is, but I figured I might as well include it. []
  4. In case you’re curious, Gale Sayers rushed for 2,125 yards over 50, which was 43% of his 4,956 career rushing yards. []
  5. Note: At the end of my original post, I changed the cut-offs to 75 and 99 yards, as well. []
  • sn0mm1s

    YPC may not be the end-all be-all of stats but when you are below league average regarding YPC for your career you aren’t a HOFer.

    • Chase Stuart

      For what it’s worth, Bettis was above-average in YPC until the last 4 years of his career:

      Year  team  rsh   rshyd  YPC    NFLAvg  Diff
      1993  ram   294   1429   4.86   3.89    0.97
      1994  ram   319   1025   3.21   3.72   (0.51)
      1995  ram   183    637   3.48   3.93   (0.45)
      1996  pit   320   1431   4.47   3.85    0.62
      1997  pit   375   1665   4.44   3.98    0.46
      1998  pit   316   1185   3.75   3.99   (0.24)
      1999  pit   299   1091   3.65   3.90   (0.25)
      2000  pit   355   1341   3.78   4.08   (0.30)
      2001  pit   225   1072   4.76   4.06    0.71
      2002  pit   187    666   3.56   4.22   (0.65)
      2003  pit   246    811   3.30   4.16   (0.86)
      2004  pit   250    941   3.76   4.14   (0.37)
      2005  pit   110    368   3.35   4.01   (0.66)
  • sacramento gold miners

    Interesting article, and it was refreshing to see Bettis, Brown, and Haley finally get their due. I should add the photo of Bettis running over future HOF linebacker Brian Urlacher was a key game in the regular season for the 2005 Steelers. Jerome gashed the tough Bears defense for over 100 yards in the victory. Talking about yards per carry with Bettis is the same as discussing scrambling yards for Dan Marino, it just wasn’t his game. It’s important to note the Steelers were playoff teams the last four years of Bettis’ career, so we don’t want to overemphasize league averages. Bettis was dealing with injury issues, and the natural wear and tear of his position. Compilers just don’t lead 15-1 teams in rushing the way he did in 2004.

    No one is suggesting Bettis is an inner circle HOF RB, just like no one thinks Tommy McDonald is an inner circle WR. There’s simply no rigid guidelines for determining who receives the games highest individual honor. Bettis wasn’t going to beat you with speed and chunks of yardage, he just physically wore down teams, and was a great closer. And if stats were kept for third and short conversions, Bettis was like Mariano Riviera.

    Clearly, this wasn’t a case like Vinny Testaverde, who complied mostly mediocre seasons, and hung on too long. Ranking top five or six in NFL rushing history will always be extremely impressive. We may never see a player of this size excel the way Bettis did, and the voters understood the evidence. Some are upset about their favorite player being left out this time, but it’s the process we have. The Steelers will have three other strong candidates coming up from the 2005 team in the future, and I’m looking forward to the discussion.

    • Richie

      Vinny Testaverde, who complied mostly mediocre seasons, and hung on too long

      Hung on too long according to whom?

      • sacramento gold miners

        Richie, Vinny Testaverde was definitely the anti-Bettis. A long career, but only a few good seasons, and his teams rarely won. The main reason Testaverde is around better QBs in the all time rankings is because he played so long, his last three seasons were painful to watch. Three different teams, and that’s not how you want an overall number one pick to exit the NFL.

        Considering the physical tools coming out of Miami, and how highly he was touted entering the NFL, his career was disappointing.

        • Richie

          I don’t think there’s anything wrong with a player sticking around as long as teams are willing to have them.

          NFL careers are short. Most of these guys’ lives have revolved around football since they were kids. Having that ends, sucks. I say more power to them, if they want to keep playing as long as possible.

        • Jeremy Crowhurst

          I think Testaverde only hung around “too long” if you’re talking about him in any kind of Hall of Fame discussion, and let’s face it, nobody is talking about him in that context. As for how his career went overall, compared to the other #1 overall QBs in the last 30-40-50 years, most people would put him in the middle, almost regardless of the criteria that you’re using.

  • David Norris

    I am old enough to see almost all of these people play. Jim Brown was the most dominant running back in the game. he had it all: speed, power, balance, and determination. You have him 18? Any analysis that says that is, obviously, wrong.

    • I am young enough to know to read entire articles before posting silly comments.

      “Note that for all purposes in this article, all rushing performances prior to 1960 have been excluded.2 Players like Jim Brown who qualify based on their post-1959 production are included in the table, but only their statistics since 1960 are shown.”

      • Arif Hasan

        And he leads the metric in percentage of yards over 50, so from the “have to make Jim Brown awesome” perspective, it does a great job.

  • Rob Pitzer

    This overstates his value (I think) because receiving yards aren’t included. Would be interesting to see this using Yards from Scrimmage and a higher cutoff.

  • J.D.

    This is a very interesting way of looking at RB production, and I’m surprised that Bettis wasn’t hurt more by this re-evaluation.

    Another thing I’m curious about is how moving the threshold would change things (not just for Bettis, but for all RBs). Would ranking RBs by, say, yards over 75 dramatically change any player’s place on the list? How about yards over 25?

  • sacramento gold miners

    Marcus Allen rates low in this particular metric, but there are three factors to consider when doing the overall evaluation of his great career. Allen almost certainly would have finished with higher career yards had Al Davis not benched him for significant periods of time late with the Raiders. We also have to mention Allen’s superb TD rushing total, many of which were of the short yardage variety, thus holding down his yards per carry. Lastly, we do have to talk about winning. Allen turned in a spectacular performance on the biggest stage for the Raiders, and the Chiefs haven’t been to the AFC Title Game since he played for KC.

  • Tim Truemper

    I have a little difficulty accepting the Compiler label for certain players. For one, Bettis was a valuable member of a SB contending team, and given that value, was used. Thus, he earned some yardage that naturally adds to his career stats. Additionally, I think it says something about a player’s ability to be able to compete and succeed to a certain degree at an advanced time in their playing careers. If Bettis was on the Steeler roster as an emergency vet back-up, and then compiled rushing yards during garbage time of certain games when the score was out of hand, then I might say, “yeah, looks like compiling.” If I am misunderstanding the conceptual nature of the term, I am welcome to being informed on this.

  • Tim Truemper

    BTWE I forgot to acknowledge that you did not see Bettis as much of a compiler. My statement is more about the appropriate use of the term beyond just looking at the numbers- that applies to any player.

  • krl97a

    For the record, Tony Dorsett also had limited carries (17 per game, one of the lowest career totals on this list), and wasn’t a “compiler”, if by that you mean an average back getting stats mostly anyone would have gotten but playing a long time for whatever reason. Watching him play quickly dispels that notion. In fact a frequent criticism of the team from the late 70s through the mid 80s when he was splitting time with Herschel Walker was that Dorsett wasn’t used enough. The counterpoint was that Landry was extending the career of a somewhat undersized back, but either way Dorsett had maybe the fastest acceleration to top speed in NFL history, possessed deceptive power, and is rightly considered one of the greatest running backs of all time.

    That said, I like this analysis. This just shows that every stat has its limitations in describing reality. There is no statistical magic bullet. One area I haven’t seen much investigation into yet but would like to is standard deviation of RB gains within a game, a season, and/or a career. I’d do it myself but I don’t have full play by play data. Consistency is one of the most undervalued stats; probably more important than YPC since it sustains drives, controls clock, and avoids punting.

  • Since I wrote this post, Frank Gore (in 2015, 2016, and the first three weeks of 2017) rushed for 2,137 yards, although only 528 of those yards came after already rushing for 50 yards in that game.

    So for his career, Gore (through 3 games of 2017) is at 13,210 career rushing yards, 5,121 rushing yards over 50 yards, and a 38.8% rate. He’s 7th all-time in career rushing yards, but the 5,121 would put him “only” at 12th all-time behind Fred Taylor (without updating for anyone else since this post, but a quick scan doesn’t identify any clear risers). And his sub-40% rate is also pretty compiler-y: In the top 25, only Franco Harris and Steven Jackson are lower, and only Thurman Thomas is below 40%.