What's new
Fantasy Football - Footballguys Forums

Welcome to Our Forums. Once you've registered and logged in, you're primed to talk football, among other topics, with the sharpest and most experienced fantasy players on the internet.

Questioning Conventional "Wisdom" (Part 2): (1 Viewer)

SSOG

Moderator
This is part 2 in a series of posts looking at "conventional wisdom" through the prism of historical results. For those that are unfamiliar with me and my style, I tend to be one of the biggest proponents of going deeper than the counting metrics and prizing things like efficiency. I'm also a huge fan of using historical models to predict future trends, which is what this post will be all about. I say this will be a "series", but to be honest, I'm not going into this with any preconceived notions about how many entries there will be. It could be 2, it could be 20- we'll see how the offseason goes. If you're curious, you can read part 1 here.

Anyway, I've long been raised in the school of thought that RBs with insane workloads were at an increased risk of injury. I mean, it makes intuitive sense- every time you touch the ball is an opportunity for the defense to tackle you. Every tackle is an opportunity for injury. Shouldn't more touches result in more chances to get injured, then? While I was looking at some other results from last season, I happened to notice that the trend didn't hold at all... and now that Drinen's updated the Historical Data Dominator to spit out per-game stats, I can finally take a look at the actual data.

I fired up the Historical Data Dominator and set it to spit out a list of all RBs from 2004 to 2009 who posted at least 100 rushes. Why did I use those particular cutoffs? The 100 rush cutoff was chosen arbitrarily to prevent someone who played a single game and got 23 carries from showing up at the top of the list and screwing up our stats. I'm only looking for guys who actually performed at a workhorse level over multiple games. The 2004 cutoff was also pretty arbitrarily chosen. The Historical Data Dominator only spits out 100 results at a time, and going back to 2004 got as close as possible to giving us a 50/50 split between RBs with 20+ carries per game and RBs with <20 carries per game (44 backs with 20+, 56 backs with <20). Not very scientific, but I'm limited by the tools I have access to and the HDD doesn't let you establish minimums or maximums for a Stat1/Stat2 sort (at least, not yet- I'm looking at you, Drinen :thumbup: ). Obviously I chose 20 carries per game as my breakpoint because that's generally the casual definition of a "workhorse RB".

Set #1 consisted of 44 RBs. These RBs ranged from 20 carries per game (Chris Brown, 2004) all the way up to 26 carries per game (Larry Johnson, 2006). Set #2 consisted of 56 RBs. These RBs ranged from 19.9 carries per game (Lee Suggs, 2004) all the way down to 17 carries per game (several backs, most recently Adrian Peterson 2007). There's a little bit of noise in the data (for instance, Julius Jones made the list and missed 8 games in 2004... but those 8 games came at the beginning of the season and were therefore entirely unrelated to his workload), but I'm leaving it in under the assumption that it'll affect the top half and the bottom half of the list relatively equally.

The makeup of the two sets wound up being very similar. The average RB in set #1 was 26.36 years old and had 5.07 years experience. The average RB in set #2 was 26.64 years old and had 5.21 years experience. Obviously the backs in the first set were much more productive, averaging 0.19 more yards per carry (4.36 vs. 4.17) and scoring 62 more points per season (240.5 fantasy points vs. 178.3 fantasy points). This makes perfect sense- the better a player is, the more likely a team is to give him the ball, which means the guys who get the ball the most tend to be better than the guys who don't. I'm not telling anyone anything they don't already know... yet.

What's potentially surprising about the data, though, is that the "workhorse" RBs played, on average, a FULL GAME MORE PER SEASON than the "near-workhorses". The workhorses played 14.80 games per year compared to just 13.82 games for the near-workhorses. Now, there's a few minor causal factors at play- for one thing, getting injured early in the game will lower your per-game carry average. Think of Frank Gore last year, who had one game with just one carry because he got injured right away. His carries per game were 16.4 if you include that game, and 17.5 if you don't. Still, even that extreme example (getting hurt on your first carry) only made a difference of 1 carry per game over a full year- the difference between the RBs in set #1 and the RBs in set #2 is 3.44 carries per game on average. The "injured early in the game" phenomenon doesn't come close to describing the difference in health between workhorses and near-workhorses over the last 6 years.

To take it one step further... Drinen did a study in 2001 that found that RBs who have played in at least 32 consecutive games coming into a season will still only average 14 games played in that season. These workhorses averaged 14.8 games played, which is substantially higher. In other words, RBs getting a very heavy workload get injured at a lower rate than RBs who are considered "iron men" (i.e. haven't missed a game in at least 2 years)... which is pretty much the opposite of the expected effect.

So what does all this mean? What's behind this phenomenon? What applications does it have for fantasy football? Here's one man's take on the subject:

First of all, do I think this evidence shows that more carries makes an RB less likely to get injured? No, I do not. I think the logic behind the more carries = more chance of injury hypothesis is bulletproof. So why then does the data not support the hypothesis? Personally, I believe it's because there's a strong selection bias at work. At the end of the day, it gets down to one inescapable truth: coaches are smart. They get paid millions of dollars because they know what they're doing. If they are giving an RB 20+ carries, odds are it's because, in their extensive evaluation of said RB, they have determined that he is more durable than average and he can therefore withstand the pounding. The result is that the only RBs that get 20 carries a game are the RBs who are built exceptionally enough to withstand the punishment of getting 20+ carries a game. If an RB isn't built to withstand the rigors, then coaches won't use him in that manner. This also goes a long way towards explaining why guys like Tatum Bell, Jerious Norwood, and Mewelde Moore- guys who always outperform their teammates with limited touches- are never given a bigger workload. It's because coaches are smart, and they're very good at identifying just how much workload an RB can really take.

As for what applications it has for fantasy football... that's pretty clear to me. When you have a guy who has been used in a limited role, and coaches are suddenly talking about making him a workhorse, that shouldn't be a red flag. Look at Maurice Jones-Drew last year. There was a lot of concern over whether he could handle 300+ carries. At the end of the day, though, the only evidence we need that MJD can handle 300+ carries is the fact that the coaching staff is giving him 300+ carries. Coaches are smart. They know what they're doing. If a player is in line for a dramatically increased workload, then we're not in any position to be questioning whether he can handle that workload.

The other application falls on the other end of the spectrum. I already mentioned Jerious Norwood and Mewelde Moore as guys where fantasy owners have been screaming for years about how they need to get a bigger workload. Lots of people have bought those two RBs "on the cheap" with visions of what they'd do with 250+ carries dancing through their heads. At the end of the day, though, coaches are smart. If we've noticed how good those players have been on limited carries, I guarantee you the coaches have as well. If the coaches still haven't increased their workload, perhaps it's time for us to question whether or not there's something going on behind the scenes, some information that the coaches have access to that we don't, something that is preventing those guys from getting a bigger workload. In hindsight, those guys who were "buying low" on Darren Sproles and Leon Washington last offseason were really just acquiring them on the flawed assumption that touches will follow production.

 
great post, very interesting and i appreciate the insight...OTOH (and not to discredit your entire analysis/findings, which look valid to my untrained eye) the immediate counter-example to your MJD that popped into my head was Marion Barber...a guy succeeding big-time in an RBBC who they tried to make the workhorse that broke down and hasn't really been the same since.

Granted that's one example (and judging by the statistics you've got there, likely a relatively isolated one).

point being that the question of whether a back can handle an increased workload still is one that poses some increased risk.

 
Last edited by a moderator:
So from your research can you say that RBs that perform at a high elite level will be in the league longer?

or simply, more touches = more seasons?

and for the record, I totally agree you...its no wonder Emmitt Smith had 13 High performance seasons in a row, Curtis Martin with 10, Eddie George with 8, LT with 7, etc. Basically all without injuries that effected their performance.

 
So from your research can you say that RBs that perform at a high elite level will be in the league longer? or simply, more touches = more seasons?
I didn't look at it on anything other than a single-season basis. The intention wasn't to say that guys who get a lot of carries are going to stick around for years to come (that'd need a completely different dataset), it's just to show that workhorses aren't really any more likely to break down than guys who are "1A in an RBBC" situation.
 
Great analysis, I always like your work (we agreed on SJax last year for this reason). I guess it begs the question, who will be the guy to break through and get a high workload this year? Should probably starting listening to coach speak, or at least pay attention to what they say and determine if it's bull or worth buying into.

 

Users who are viewing this thread

Top