All Hail Total Rushing Yards!

Even league insiders are fooled by the simplistic mirage of total yards. I'm doing a salary analysis of RBs right now, similar to what I just did with safeties. It's striking how RBs are paid. They're paid for past total rushing yards, not YPC, not EPA or WPA or SR. More on that in another post, but here is a case study in point.

On Sunday the Dolphins beat the Bills 30-23. C.J. Spiller had a great day for the losing team, with 0.05 WPA, 7.9 EPA, 67% SR, and 12 carries at a 7.6 YPC clip. Plus, he caught 9 passes for 76 yds. He totaled 1 rushing TD and 1 receiving TD.

Reggie Bush notched -0.03 WPA, 4.0 EPA, 52% SR, and had 25 carries at an 8.1 YPC pace. He caught 1 pass for 6 yds and had a single rushing TD. But Bush lost a fumble that led to a Bills TD, putting his Dolphins in a 1st quarter hole. On the next drive he couldn't convert a 3rd and 1, leading to a punt that resulted in a 0.25 WP low-point for his team.

Who gets named AFC Player of the Week? Bush, with his 203 total rushing yards. That clearly trumps Spiller's 91 total rushing yards. But Spiller had the decidedly superior total impact, thanks to his receiving success and the fact he lost no fumbles.

You're probably thinking that you have to win the game to get one of those weekly awards. It's not a preoccupation with total rushing yards. Instead, you might say it's a preoccupation with who had the best day on a winning team. I can buy that, until you look at Donald Brown's day on Sunday.


Brown led his Colts to their first and possibly only win of the year. He compiled 0.10 WPA, 5.9 EPA, 41% SR with 10.1 YPC average. He totaled 161 rushing yards with no turnovers.

Chris Johnson is gunning for 1,000 total rushing yards for the season--as if that means anything. Johnson, despite his overhyped barely-average performance a couple weeks ago, is forcing his team to endure one of the worst, most overpriced RB seasons of all time. And yet, 1,000 total yards is still in play. To get 1,000 yds, a running back needs only 63 yds per game. If he averaged a dreadful league-worst 3.0 YPC, he'd need to average 21 attempts per game to get to 1,000 total yards.

I don't expect anyone outside of the stathead community to be aware of WPA or SR or anything of that stuff. But I would think that they could somehow account for turnovers, receiving yards, or critical 3rd downs. "Football people" want to tell you that they ignore stats and just look at the game. They want us to believe they have a near mystical sense of how to recognize good play. The fact is they do rely on stats, just the wrong ones.

Forget total yards.

  • 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

31 Responses to “All Hail Total Rushing Yards!”

  1. Topher Doll says:

    I do think you are venting a bit Brian, which is fine, but keep in mind the nature of WPA and EPA, both of which I like and approve of, but aren't always perfect. I just rewatched that game, and if you take Bush out of that game, the Dolphins lose, he played very well, though his fumble was killer. Compare that to Spiller, a player who, while key to that offense, wasn't as important as Bush was. Bush was a bigger part of the offense, was more key, and while WPA and EPA are fantastic metrics, I don't believe they really showed either players value in this game when you compare them. You put Bush's production on the Bills, I predict the Bills are closer to a win, not a lock, but closer, while putting Spiller's production on the Dolphins doesn't change much.

    Both RB's played good games, and the media has a different take on game recap than those who delve into the game a bit more, mostly because big numbers appeal to the average fan. Because of that, nuances are lost. I don't like the dumbing down football for people, but that's how it is.

  2. Brian Burke says:

    Topher, that makes no sense.

  3. Jeff Fogle says:

    BB, is it possible for you to post the positives and negatives in your metrics for these particular plays?

    *Bush's fumble was negative xxx
    *Bush's 76-yard TD run was positive xxx

    *Bush's 3rd and 1 failure from the first quarter you cited was negative xxx
    *Bush's 3rd and 2 run for 28 yards from the Buffalo 40 in the 2nd quarter was postive xxx

    Thanks in advance. Might give us a sense of how different perspectives are evaluating Bush's two biggest yardage plays vs. what look to be the two biggest negative plays.

  4. Brian Burke says:

    The WP graphs show the WPA of each play.

    The point isn't just about Bush vs. Spiller. It's about the uselessness of total yards as a measure of value. Isolating the couple biggest plays for each player doesn't tell you about the value of one stat over another. RTFP.

  5. Jeff Fogle says:

    You said:
    "But Spiller had the decidedly superior total impact, thanks to his receiving success and the fact he lost no fumbles."

    Sorry for not specifically referencing that sentence in my first comment after I had RTFP. Thought it would be clear, but I guess it wasn't.

    Bush did have a fumble around his own 40-yard line. But, having a 76-yard TD run may have trumped that in the mind people watching the game or using other metrics.

    Bush did fail to convert the third down you mentioned. But, he also rushed for 28 yards on a different third down conversion that you failed to mention that moved the ball deep into Buffalo territory.

    Was hoping a quick posting of the numbers in the categories you listed would help me and others who had RTFP understand why you thought Spiller had a decidedly superior total impact. And if there was a possibility that weighting impact based on whether or not a player is facing a softer defense when playing from double digits behind could explain part of the difference between how people might percieve Spiller vs. Bush in that game (More than half of Spiller's yards came when Buffalo was down by more than one score. Bush was never down by more than one score, so never faced any variation of a prevent).

  6. jditoro says:

    Brian, I'm familiar with EPA, WPA and SR, but can you clarify the meaning of RTFP? :)

  7. jditoro says:

    seriously, i think brian was just using cj and bush as illustrative examples to make a more general point about total yards being a stat which gets a disproportionate amount of importance attached to it. then you guys nag and nag on the relatively less important details. brian, stop posting this stuff and use it to assemble a more competitive team somewhere! you are like a finance professor posting statistically significant trading advantages instead of profiting from them, and then all they do is nag you!

  8. Brian Burke says:

    RTFP = Read The Full Post. It's a very advanced concept.

  9. James says:

    That's not what I thought the F stood for...

  10. jditoro says:

    me either... :)

  11. Jeff Fogle says:

    Nitpicking is at the heart of science. That's how stuff gets figured out...

    BB said:
    "I would think that they could somehow account for turnovers, receiving yards, or critical 3rd downs. "Football people" want to tell you that they ignore stats and just look at the game."

    Let's do Bush and Brown.

    *Turnovers: Bush had a fumble and Brown didn't.

    *Recieving Yards: Brown only had one catch for two yards. Bush also only had one catch.

    *Critical Third Downs: Brown didn't convert any, and was barely involved on third down plays in a game where Indy was 3 of 12 as a team. Bush converted a couple, including a 28-yard run that moved the ball deep into Buffalo territory. Not a big edge for Bush, but an edge.

    So, if you stop at those three, one might argue Brown had the more impactful day, depending on how negatively you weigh fumbles.

    But...we also have:

    *Each guy had a long TD run (76 yards for Bush, 80 yards for Brown)...those cancel out.

    *Each had an additional breakaway run that set up a field goal (that 28-yarder mentioned earlier or Bush, and a 39-yarder for Brown...those cancel out in terms of scoreboard impact.

    *On other touches:
    Bush 105 yards on 24 for 4.4 yards-per-touch
    Brown 44 yards on 15 for 2.9 yards-per-touch

    Bush had 35% of his touches go for 2 yds or less
    Brown had 47% of his touches go for 2 yds or less

    There are some methodologies where the plusses from Bush on those other plays...and the avoidance of minuses on plays of 2 yards or less in comparison, would outweigh the loss of the fumble. And, people eyeballing the games on satellite (the football people who tell you they just look at the game) would have seen a lot more productive plays from Bush than Brown (almost half of Brown's plays were 2 yards or less, just over a third for Bush).

    There's an assumption being made that selectors ("football people") were focusing on total yardage when there are other reasonable reasons to pick Bush over Brown for AFC player of the week. One could "forget total yards" and still pick Bush.

    For Bush over Spiller...viewers could reasonably decide that the 76-yard TD run made up for the fumble (that could have just as easily led to a FG for Buffalo rather than a TD). Then, you have a game where Spiller got more than half of his production when his team trailed by double digits in the second half...while Bush didn't get any yardage against a defense that had softened with a big lead.

    In the article, an assessment was "assigned" to football people without asking any what they actually thought. Then, the conclusion was an admonition to football people, scolding them for the assessment they had been assigned.

    JDI said of BB:
    "you are like a finance professor posting statistically significant trading advantages instead of profiting from them"

    This is not a generally held view within the legal betting markets. The most successful investors have their own methodologies they're pretty happy with. It would be cool if BB entered the markets and was able to profit from his approach. Best wishes to him if he decides to do so.

  12. Jim Glass says:

    I'd think that as to paying FA running backs, past fumble record should be discounted, as lost fumbles are so highly random.

    I'd also discount 76-yard runs. They are very random too (you aren't going to get one from the other team's 20) and if a smart play call/bulldozer blocking puts a RB in a position where he has a straight run to the end zone, what's the big credit to him? But a few lucky dramatic runs like that can make the fans forget many other weaknesses in the RB's game, and inflate his ypc to make him appear much better than he really is.

    I'd put the solid premium on success rate. RBs aren't about getting big yardage plays -- passing is *always* better for that. They are about controlling the game: converting short first downs, controlling the clock, helping control the line, blocking, establishing physical control on offense, catching short "check down" passes, diverting the defenders from pass defense.

    I'll repeat a couple things I've said before:

    1) The list of great break-away big-play RBs who played on terrible offenses that never won anything year-after-year goes down your arm. OJ Simpons (whose teams averaged 4-10 for his career), Gale Sayers, Walter Payton, Barry Sanders...

    I suspect having one of these guys often actually is a curse that hurts a team, because management feels obliged to weaken the passing game by using what resources it has to build the offense around the famous "big play" RB, when passing is always better for big plays. Call it Simpson-Sanders Syndrome.

    2) Vince Lombardi's Packers were legendary for destroying opposing Ds with their running game to set up opportunistic passing. But in their two Super Bowl seasons they were smack at the bottom of the league in yards-per-carry and long-gainers. In fact, to the modern eye the coach of that team looks like a moron for being at the top of the league in rushing attempts with such poor ypc, and at the bottom of the league in pass attempts in spite of league-top passing efficiency. How dim was that?

    But Lombardi wasn't a moron. Neither were the opponents who reported being destroyed by the Packer O those years. The Packer Sweep was designed in depth specifically to get 3-4 yards each play, every play, not to break long ones. And it didn't. But I'd wager that if the play-by-play were obtained for those years its Success Rate would prove to be pretty darn good.

  13. Dave says:

    Aside from Bush's turnover, I think what hurts his WPA is that his 76 yard TD run came late in the game when the Dolphins pretty much had it wrapped up. If he would have made the run right after the Bills scored off of his turnover I think it would have had more impact on his WPA.

    Even so, I was a bit surprised to see his WPA after watching the game and thinking he had a good game. I figured that he before his long TD run he had done enough to make up for the fumble. And while the long TD run wasn't all that important it did decidedly put the game out of reach, which was comforting at the time haven seen a few games early in the season in which the Dolphins blew late leads.

  14. Topher Doll says:

    I agree Brian, total yards is a useless way to measure a players complete production, but that wasn't the point I was making. The point I was making was the the point of the award appeals to the casual football fan, and as such, you don't want your award to have a multi-page explanation attached to it, that doesn't attract people. Add in that Bush had a good game in a number of other areas, and was a huge reason why the Dolphins won. N

    No one will disagree that Total Yards is a bad way to judge a running back, but when you have a small point tied into a rant, it's hard to get that point. Your post starts with Bush vs. Spiller, it's body is RB vs RB vs RB, it's conclusion is muddled and you are surprised when people think your point is comparing RB's? I think 100% of people agree, Total Yards bad, but when you try and compare Bush and Spiller, two RB's who had very good games, as an example, it doesn't convey your point well. I think I seem to speak for most of the comments so far, it seems you are taking yours, and most peoples, hate of Total Yards out on a RB who had a pretty great game, rather than writing up an analytical article laying out the data like you usually do. I love this site and your writing, but this feels like you read the Bush article and just wrote up this right afterwards, it doesn't feel like your most put together work.

    In the end, I think Jeff said it best, you could toss out Total Yards and still pick Bush.

  15. Anonymous says:

    Some (a lot) of a RBs success is due to the offensive line. A RB who is tackled behind the line of scrimage might not be at fault. The offensive lineman got beat.
    All I'm saying is a rucial 3rd and 1 where someone like Troy Polamalu or Ray Lewis gets a great jump and busts a play in the backfield shouldn't count negatively against a RB.
    Countering that, I've seen some elite RBs like LeSean McCoy turn a busted play in the backfield to a 7 yard gain by some shifty cutback.

    In anycase, I agree with teh idea that "Tatal X" is usually not that good a metric, compared to "Average X" or success rate or EPA or WPA.

    Brandon

  16. Brian Burke says:

    Jeff Fogle- Nitpicking relevant information might be the heart of science, but derailing the discussion onto irrelevant trivia isn't. Besides, those nits were picked a long time ago.

    All 500 words of your comment above can be summed up into the players' respective EPA and WPA stats. That's what they're for--so we don't have to play your pointless game.

    Do you really want to defend the notion that RB total rushing yards is meaningful? Otherwise, what's your point, other than to be a petty contrarian?

  17. Brian Burke says:

    Now there's two good comments! ^

    Relevant. Thoughtful. Critical thinking.

  18. slushhead says:

    Brian,

    Have you ever done a study on the variance in individual running backs' performances? I couldn't find one anywhere on the site. One of the problems I've always had with total yards is that 15 carries for no gain followed by an 80 yard run is far worse than 16 5-yard carries. It seems to me that some runners have a large variance in terms of success on each play, where as others are more consistent. As variance is one of the most overlooked (although not on this site) considerations with regards to football strategy, this might be interesting.

  19. Jared Doom says:

    One could argue RBs with more touches (correlates with total yards) are handicapped because of both the expectations of opposing defenses and fatigue.

    I compiled the 2011 game logs of the current top 6 RBs in running yards. Yards Per Carry (YPC) has a -0.15 correlation with Carries (I noticed this holds for individual games but not for the whole season). For the whole season, carries / game has a -0.16 correlation with SR.

    So if you're evaluating the season SR or single-game YPC of two RBs, it seems the one with fewer carries may have an unfair advantage.

  20. jditoro says:

    slushhead: SR speaks to that in the most meaningful way. because yo could also say that 2 5 yard rushes could be very different. 5 yards on 3rd and 6 doesn't help, while 5 yards on 3rd and 4 is huge. 5 yards on 1st and 10 is also great because there is a higher chance of a first down from 2nd and 5 than from 1st and 10. so SR (which I believe measures whether the play improved the chances of attaining the next first down and the next score or not) speaks to that point. Thus, a back with a low SR but high total yards would have gotten there through one or two meaningless big runs coupled with getting stuffed multiple times. WPA would further examine how much those runs and stuffs improved or decreased his team's chances of winning the game. This is why Brian says that most of the nuances are captured in the advanced stats, thus making them, well, advanced :)

  21. Will says:

    SR is a good way to get at the variance in a player's performance but it doesn't tell the whole story (and it tells a different story, the situational one). I would love to see EPA and WPA reported with a +/- standard dev or variance. This would be good for comparing players or types of plays.

  22. Jeff Fogle says:

    Appreciate you taking time to comment BB. We'll have to disagree on what's trivial, and what's pointless.

    There is a prevailing view elsewhere in analytics that what you think your numbers are capturing isn't as accurate as you seem to think (same for DVOA across the street, and that variation of football feng shui). Those doing analytics in the markets don't publish their work though...so you'd have to jump into the markets more directly to get a sense of that I'm afraid. I hope you'll consider doing that at some point down the road...

  23. Brian Burke says:

    Jeff,

    That just it. This has nothing to do with betting markets or DVOA. This has to do with people who say they ignore "stats" but who obviously cling tightly to less meaningful stats.

    Every one of your comments is wildly off point. You either do not understand what is written or you are intentionally misleading readers, compulsively criticizing, and trumpeting your own bizarre view of the world. Whatever it is, I do not care for it.

    I don't care about your esteemed betting markets. Beating them over the long term is child's play. My silly efficiency model does nearly as well before you even make the most obvious adjustments for QB injuries or playoff-seed locks. But I have better things to do with my kids' college funds.

    My question to you is, what does any of this have to do with total rushing yards?

    Go comment somewhere else. You are literally draining the intelligence from these comment threads.

  24. Tom says:

    Beating the markets in actuality over long periods is a very difficult task, Brian. To call it child's play is both arrogant and naive.

  25. Topher Doll says:

    Brian I think if one person misunderstood your post (I won't venture to call it an article) it would be something to ignore, but the fact that everyone seems to "not be getting your point" says something about how the post was written, not how the article is being misunderstood.

    - One person "misunderstanding" is a on the reader
    - Everyone "misunderstanding" is on the writer

  26. Brian Burke says:

    No Topher. The 1000 people who got my point didn't leave stupid comments. Commenters are a tiny fraction of the readership, maybe .1%.

    Tom--false. It's not hard. It is hard to beat the vig, though. But hey, don't let that stop you from publicly calling a stranger arrogant and naive.

  27. Tom says:

    No, Brian, it is hard. Picking straight-up winners slightly better than the market is not too difficult, but beating the precision of the market is very hard. No professional gambler is just picking winners.

    I think we've all got naivety in us, so don't take that as some sort of huge slight, I'm not just trying to ruffle your feathers.

  28. Jonathan says:

    A monkey would "beat the market" over the long haul 50% of the time.

    Anyway...

    I really don't understand the confusion. The post title is "all hail total yards!" The topic sentence is about total yards. The closing sentence is "Forget total yards." Even if you did a tl;dr on this, I don't understand how people can walk away with this idea that he was talking about anything else.

  29. Anonymous says:

    Not sure why you are being so defensive, Brian. At a high level, the guys 'disagreeing' above are simply saying that, although they agree with you that total yards is inappropriately used by the media as the primary measure of an RB, they disagree with the example you are citing.

    Further, I must say that I have found more value in the comments above than within the actual post itself (although the post is obviously necessary to lead the comments). In a weird way, I actually prefer your quality posts combined with your 'naive and arrogant' comments. So thanks, I guess!

    CR

  30. Brian Burke says:

    CR-not really. I've been doing this for a long time. Most guys above just want to find an excuse to be critical. They fully understand the post. It's just that this site attracts commenters with certain kind of personality.

    By the way, the example is air tight. Anyone who watched Reggie Bush that day saw his big run was a joke. He had a hole 10-yds wide to run through and only 1 safety to beat. The safety made a bad lunge on a soaked field and Bush was off to the races on a practically meaningless TD. The nitwit comments above don't care about that. They just want to read their own words on the internet.

  31. FrequentBird.com says:

    This site yields some surprising findings, all of which I am inclined to believe, as I am a huge sabermetrics fan. Could you please explain exactly why Spiller's EPA total is higher than Adrian Peterson's? Peterson's MVP credentials are being overly hyped, but no one has even suggested that he was not even best runningback or Peterson (Patrick) this season.

    Spiller had one fewer fumble, more receiving yards, and greater per-play productivity, but was that enough to overcome the sheer absolute yardage differential?

Leave a Reply

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