The League

Smarter Stats

The Curse of 370, Part 2


I wanted to follow up this week's Smarter Stats column about the "Curse of 370" with answers to some very solid questions, and a few more thoughts about running back production versus overuse as they pertain to the one guy in Washington everyone can agree on right now.

1. What's so special about 370? Don't running backs with 365, or 360, or 350 carries run into the same problems?

The number 370 in and of itself is not an absolute, i.e., "If you carry the ball 370 times in a regular season, your head will explode." It's a simple barometer of use and overuse not completely unlike 100 pitches in a game, though the correlation between workload amounts and overuse might be simpler to tie together in football than in baseball due to much smaller sample size (except for Erik Bedard of my hometown Seattle Mariners, whose head apparently will explode if he throws that hundredth pitch.) What we found through historical analysis is that 370 is a starting point for a higher predominance of injury and production falloff. It's not the be-all, end-all, though it is a pretty good indicator.

2. Does correlation really equal causation in this case? Doesn't every running back fall into one of three categories: Backs who get hurt, backs who don't, and backs who are Eric Dickerson?

Well, let's expand the workload discussion a bit to include total running back touches (carries and catches), not just carries. In 2006, Aaron Schatz ran the numbers from 1978 through 2004 and found that the correlation between workload and carries, as measured by yards the next season, was twice as high as it was between overall touches and the workload indicators we've discussed. When you move to receptions and eliminate carries, the correlations actually swing to the positive side.

This makes sense when you consider the on-field reality that it's a lot easier on the body to take a bubble screen and fight it out with an outside linebacker or cornerback then it is to shoot up the middle and deal with those behemoths inside. 40 or 50 fewer car accidents in short yardage mean a lot when you multiply them by X number of seasons. On 105 different occasions in NFL history, a running back has combined catches and carries to equal 370. Here are the ten highest and lowest percentages of catches in those totals:

Lowest Catch Percentage, 370+ Touches

View image

Highest Catch Percentage, 370+ Touches

View image

What I discovered in the research for an April, 2007 article I wrote for Rotoworld's annual fantasy football guide was that the top ten backs in carry percentage saw their totals decrease the next season more than the top ten backs in catch percentage:

View image

3. What about backs like John Riggins, Emmitt Smith, and LaDanian Tomlinson, who went over 370 carries and survived to enjoy further 1,000-yard seasons? Is this kind of workload something Emmitt would just call a "rite of patches?"

Indeed he would, and it's not a fait accompli that every back who tiptoes over that line will find his career "debacled" the next season. Riggins was an interesting case because he went over 370 carries for the first time in his twelfth season -- by far the latest entry (Walter Payton was next, in his ninth season), and left the game two seasons later. Before 1983, he had never rushed more than 260 times in a season -- in fact, Riggins totaled 372 carries in 1981 and 1982 combined. In his case, it may have been the combination of judicious use of his talents prior to 1983, his own power, and the fact that he benefited from a tremendous offensive line.

Smith also benefited from a great line, and his case is interesting because he went over 370 carries twice -- in 1993 and 1995. 1994 was a down year for him production-wise, but that was also the year he held out the first two games, and his yards per carry actually went up from 4.59 to 5.25. However, the career drop after 1995 was significant. From 1990 through 1995, his first six seasons, Smith averaged 335 carries for 1,493 yards, a 4.46 yards per carry average and 16 rushing touchdowns per season. From 1996 through 2001, his second six seasons, Smith averaged 299 carries for 1,205 yards and 8.7 touchdowns per season. He actually had one more 1,000-yard season in those second six years, but the pre-play productivity was way down for a number of reasons, one of which was that the tires were getting worn.

Tomlinson went over 370 carries once, with 372 in his second season, and he's never come closer than 348 since. Of his 2,823 regular-season touches through 2007, 16 percent have been catches. Again, productivity without overuse.

4. With Clinton Portis on pace for 373 carries, is reducing his time on the field the best (and only) way to go?

Not at all. Portis has exceeded 370 touches (though never 370 carries) in three seasons: 383 in 2004 (10.44% catches), 382 in 2005 (7.85% catches) and 372 in 2007 (12.63% catches). The problem this year is that not only is he on pace for 373 carries in the regular season alone, he's only caught 10 passes in 173 total touches -- a percentage of 5.78. This is baffling especially in an offense with a quarterback who is learning the intricacies of the West Coast system, and the little dump-offs that are such a part of it. Then again, when you consider that new Portis backup Shaun Alexander's receptions dropped every year in Jim Zorn's former home of Seattle from 2001 through 2006, maybe it's not so baffling after all.

So, to review -- 370 is not an absolute brick wall that no back can get past, but it is a fairly reliable indicator of overuse in a sport that regularly chews its backs up with half a care. Clinton Portis has been the Redskins' best player this season, and the Redskins can do better by him without losing a bit of his productivity.

By Doug Farrar  |  October 24, 2008; 11:52 AM ET  | Category:  Statistics Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati  
Previous: The Curse of 370 | Next: Seven Week 7 Awards


Please email us to report offensive comments.

I think Jim Zorn is well aware of Clinton's workload, which is why in 5 of the Redskins' 7 games this year, he was held under 25 carries.

The Redskins have found themselves in two games this year where running the ball with their best back was absolutely critical to winning. Obviously, these games won't happen very often. The Redskins are currently a 5-2 team who hasn't won a game by a margin greater than a touchdown.

Zorn's smart enough to not have Portis carrying the ball in situations where the game is decided, something that Herm Edwards struggled with twice, and Shaun Alexander had done to him in 2005. The signing of Alexander is indicative of the desire to have someone else take the punishment at the end of games, and I think Portis' final carry total will end up around 340.

It's a high workload to be sure, but Portis has been over 300 three times in his career and is still running strong.

Posted by: Greg Trippiedi | October 24, 2008 3:25 PM

"Based on the so-called 370 curse, I suppose as a head coach I should sideline the NFL's leading running back before I make the playoffs, because I'm worried about next season..?...?:

Not at all, but Coach might want to call a few more screens if the trend has been for the back to catch the ball out of the backfield with success as well. One thing we do know is that a more integrated mixture of catches and carries (the "injury benefit" line seems to be about 85/15) does help in many cases. Especially with a player like Portis, as keen as he is to pass block and run inside when necessary.

Posted by: Doug Farrar | October 24, 2008 2:10 PM

Yes, injuries will result from additional use, but you can also get injured playing special teams in a pre-season game. Just ask Clinton.

It's just as logical to assume that teams with back that have 370 carries have a strong ground game and that teams start to key on the running backs much more once the trend has been established.

Based on the so-called 370 curse, I suppose as a head coach I should sideline the NFL's leading running back before I make the playoffs, because I'm worried about next season..?...?

Maybe not.

Posted by: stumped | October 24, 2008 1:30 PM

To the critics of this research -

Carrying the ball into line 350 times in a year sure wears down the body. Just try it sometime. Tough to do. Tough to maintain. Football is a hard game.

Ask the hundreds of former players with blown out knees, arthritis, back problems, etc.

Posted by: charlie | October 24, 2008 1:20 PM

This is a tortured excuse for bad research. Why don't you guys just admit you were wrong?

1. Half of all RBs will have fall-offs in production from one year to the next.

2. Another quarter or so will be injured in any given season.

3. Discovering that high-carry backs usually either have a fall-off or are injured doesn't tell us anything. It's not a "fairly reliable indicator" of anything except the normal variations in the NFL.

Posted by: Ted | October 24, 2008 1:12 PM

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company