December 6, 2016

Pass-Catching Backs, Part II

By Bill Barnwell, Football Outsiders – special to BSMW Patriots Game Day

Last week’s column, looking at reception-crazy running backs, noted that teams that used “playmaker” running backs who led their teams in both rushing attempts and receptions won fewer games than the average team. Now, this would seemingly make the Arizona Cardinals, New Orleans Saints, and right analog stick in Madden 2004 all quite unhappy, but let’s not confuse correlation with causation and say that having a LaDainian Tomlinson or Tiki Barber on your team makes them worse. Well, not yet.

I found last week that teams won, on average, 7.73 games when a running back led the team in receptions. Drilling down some, I found that teams who employed a secondary running back to lead them in receptions won 8.34 games, over a half win better than the average. I mentioned that playmaker-led teams performed slightly worse.

There have been 79 instances since 1978 where a running back led his team in both rushing attempts and receptions. Those teams won 7.25 games per season, a full-game worse than the John D. Williams and Dave Meggett-style offenses, and more than a half-win below the average NFL team. How could that be? These teams had players good enough to be dynamic threats both behind the line of scrimmage and out in the flat, and they were somehow still crummy?

I was still skeptical. There’s gotta be some advantage to having a playmaker on your team. I thought about LaMont Jordan, who’d been close to the team lead in receptions for the Raiders last season. He had 70 receptions while Jerry Porter had 76, but had a below-average DVOA (Football Outsiders’ metric measuring Jordan’s performance versus the league-average in the same situations — for further explanation, please read our Methods page) because he wasn’t doing anything with them; a lot of the time, he was catching meaningless dumpoffs on third and long and going a few yards, and he rarely, if ever, broke a long gain, leading to a low 8.0 yards per catch average. (For reference, the average running back who led his team in receptions averaged 8.89 yards per catch over the course of the study.) I wondered whether there were certain running backs who were just padding their stats with dumpoffs and other assorted flotsam, while running backs who were breaking big plays shouldn’t be associated with them. I took the ten running backs with the highest yards per catch average and the ten lowest and compare them below.

yards_per_catch.jpg

There are definitely sample size issues here, but a three-win-per-season difference is pretty gigantic. I found that yards per reception correlated with team wins for these 79 players at .29, a decent-sized correlation when it comes to football data; for the larger group of running backs who led their teams in receptions, that number went down to .21. I still wasn’t convinced I’d found the solution, though.

Since the numbers being analyzed here aren’t too complicated — yards, receptions, touchdowns, and carries — I decided to pull out the fantasy point metric again to see if it revealed anything. Just as a reminder, the formula for fantasy points is that used in most leagues I’ve seen and participated in: (((Receiving Yards + Rushing Yards) /10) + ((Rushing TDs + Receiving TDs) * 6) + Receptions). LaDainian Tomlinson’s brilliance on the waggle, sadly, will not be included for study.

The correlation between the 79 running backs’ fantasy points for the season and their team’s number of wins was .38, much stronger than the yards per catch correlation of .21. So now, something had been revealed that seems somewhat obvious: if a player is going to lead your team in carries and receptions, he best be getting a lot of yards and touchdowns, or your team’s not going to do very well.

I wondered whether either was more important, and separated the fantasy points into fantasy rushing points and fantasy receiving points. Fantasy rushing points correlated to wins (.34) higher than receiving points (.21), but that would also intrinsically follow — after all, teams that are losing throw the ball more than they run it, while teams that are winning do the opposite (that is, unless you’re Philadelphia). That being said, I checked whether carries correlated to wins, and for the 79 running backs who led their teams in carries, the correlation was only .21, so it was more so what the running backs were doing with their carries than the fact that they were getting them at all.

As I said at the beginning of this piece, it’s very important to not confuse correlation with causation. With that in mind, I think what history is telling us here is that having a playmaker on your team isn’t a bad thing — you just have to make sure that he’s a good one.

Leave a Reply