Print and Go Back ESPN.com: NFL [Print without images]

Saturday, December 22, 2007
Before the Turner frenzy begins, consider these cautionary RB tales

By Aaron Schatz
Football Outsiders.com

San Diego running back Michael Turner probably won't play against the Denver Broncos on "Monday Night Football" (ESPN, 8 ET) because of a shoulder injury.

Michael Turner
Michael Turner probably will be leaving San Diego for a bigger spotlight and riches elsewhere.

In fact, it is possible he has played his final game for the Chargers. Last offseason, no team was willing to give up a first- and third-round pick to sign Turner as a restricted free agent. The Chargers decided not to trade him so they'd have an insurance policy in case LaDainian Tomlinson got injured.

Next offseason, Turner is an unrestricted free agent, and San Diego won't be able to keep him around this time. A number of teams will be willing to give Turner starters' money and the job.

Most fans know that Turner has been a good backup, but they might not realize how good. Turner has averaged 5.7 yards per carry over his entire career, and that doesn't come from just one or two long runs. At Football Outsiders, we use a stat called defense-adjusted value over average, which looks at success on each play based on the down-and-distance, adjusted for situation and opponent (explained further here.) In 2005 and 2006, Turner had the highest DVOA of any running back with at least 50 carries.

Clearly, Turner's numbers benefit from a small sample size. He never has had more than 80 carries in a season. Small sample size can lead to both very good numbers and very bad ones, and Turner actually has bad ratings this year. Unlike in past years, he is getting most of his yardage off just a couple of runs. His three longest runs of the year came the first time San Diego played Denver, including a 74-yard touchdown. In San Diego's other 13 games, Turner is averaging just 2.7 yards per carry.

Michael Turner
Year DVOA Rank*
2005 27.9% 1
2006 46.5% 1
2007 -23.5% 61
* Rank of all RBs with 50 or more carries

Tomlinson is a steady, consistent back who can carry the ball 350 times a year. There's no question he's more valuable than Turner. However, even when we include this year's poor performance, Turner has been better than Tomlinson on a per-play basis.

There's a second set of reasons for that, none of which is easily quantifiable or even provable. Turner gets more rest. He doesn't start, coming in when the defense is tired. Defenses don't prepare for him the way they do for Tomlinson, and they don't adjust accordingly. There's probably some truth in those things, but there's every reason to think Turner's a pretty good running back.

LaDainian Tomlinson
Year DVOA Rank*
2005 14.5% 13
2006 24.0% 4
2007 15.8% 13
* Rank of all RBs with 50 or more carries

There also are reasons to think he's not, however. One of these reasons is named Richard Huntley.

Huntley was Jerome Bettis' backup for the late 1990s Steelers. In 1998, Huntley had 55 carries for 242 yards; decent numbers, but a product of context and full of lots of third-and-long useless yardage, as was reflected in his minus-35.2 percent DVOA. The next year, Huntley surprised everyone with a breakout season. He gained 6.1 yards per carry on 93 carries, became a solid receiver in the rushing game, and put up a 21.9 percent rushing DVOA, which ranked fourth in the league

Miami tried to sign him as a restricted free agent, but the Steelers gave him a three-year, $4 million contract and a chance to compete with Bettis for the starting job in training camp. Bettis won out, but Huntley once again had the fourth-best DVOA in the league in 2000.

In 2001, as an unrestricted free agent, Huntley signed in Carolina, where he would challenge oft-injured Tim Biakabutuka for the starting job. Biakabutuka got hurt, and Huntley was awful.

Huntley averaged less than 4.0 yards per carry with a DVOA of minus-26.5 percent, which ranked 40th of the 43 running backs with at least 100 carries. He had only three more carries before he was out of football.

Huntley's failure is an example of something that DVOA can't account for but a scout or an observer with access to DVOA can: the rest of the offense.

As useful as DVOA is, it can't account for the quality of an offensive line or a quarterback's ability to put a running back in good situations. It's just impossible to get a first down on third-and-8 running the ball every time.

Richard Huntley
Former Steelers' tailback Richard Huntley (above) parlayed stardom as Jerome Bettis' backup into a big role with the Carolina Panthers. He did not prove to be a bargain.

Huntley played with a very good offensive line in Pittsburgh. He certainly did not in Carolina.

With the Steelers, his quarterback was Kordell Stewart, and as you might remember from an article I wrote a few weeks ago, a scrambling quarterback opens up lanes for the running back.

With the Panthers, Huntley's quarterback was Chris Weinke, who is about as mobile as a turnip. The situation in Pittsburgh was just more conducive to being a successful running back than the one in Carolina. Advanced statistical analysis can help strip some of that information out of the numbers, but not all of it.

Huntley presents the cautionary tale for teams that think signing a successful backup from a pretty good rushing team is a good way to upgrade your running back spot. Of course, for every Richard Huntley, there's also a Priest Holmes.

Recently retired Holmes probably is the most successful back ever to switch from backup on one team to starter on another.

The first season Baltimore gave him any playing time, he ran for 1,008 yards and gained 4.3 yards per carry. The next season, 1999, Holmes was hurt and played only eight games, but he averaged a ridiculous 5.6 yards per carry, good for a 22.9 percent DVOA, second in the league.

The Ravens drafted Jamal Lewis to replace Holmes as the starter and won the Super Bowl, but backup Holmes still ranked fourth in the league in DVOA. If Football Outsiders had been around then, we would have championed Holmes the way we champion Jerious Norwood now.

Kansas City signed Holmes in 2001, and he gave the Chiefs 3½ years of MVP-level performance. The 2001 Ravens, on the other hand, lived through San Diego's nightmare. Once their quality backup was gone, Lewis tore his ACL, and the Ravens were stuck with grizzled veteran Terry Allen as their starter.

Perhaps the player most similar to Turner is LaMont Jordan. Jordan had extreme DVOA ratings with the New York Jets in very limited playing time: an awesome 46.6 percent on 39 carries in 2001, a hideous minus-40.0 percent on 84 carries in 2002, and another awesome 36.6 percent with 93 carries in 2004. Like Turner, Jordan backed up a veteran who led the league in value according to our DPAR (defense-adjusted points above replacement) stats (Curtis Martin).

The salary cap-strapped Jets had no way to keep Jordan to back up Martin, even though Martin was old and likely to break down the next year (which he did). Jordan moved on to Oakland, which gave him a five-year, $27.5 million deal.

With a small sample size as a backup, Jordan's performance swung wildly from good to bad. With a large sample size as the starter, he was almost exactly league-average, which was still a big improvement on what Amos Zereoue had done for Oakland the year before. That was really Jordan's only full season as a starter; he has missed a lot of time since then with injuries.

Huntley, Holmes and Jordan were three players who, like Turner, ranked among the best of the NFL backups. These three players went to three new teams. The three situations were very different, particularly when it came to the quality of the offensive lines, and the three running backs played accordingly. Turner's career -- and his future as a star back -- depends on the same thing: which team he signs with and the quality of that team's passing game and offensive line.

Aaron Schatz is president of Football Outsiders Inc. and the lead author of Pro Football Prospectus 2007 and 2008.