When the Titans signed Shonn Greene this week, the key number that was bandied about as to why the signing was a great move was Shonn Greene was incredibly successful in 2012 on converting third-and-short into first downs. It's true, he was. Eleven times Shonn Greene ran the ball on third down and three or less, and 11 times that resulted in a first down. It's also true that the Titans struggled at times in 2012 to run the ball in short yardage. Like Shonn Greene, the Titans converted 11 first downs on designed runs on third down and three or less, though it took them 16 attempts to get there rather than Greene's 11. Ergo, instant upgrade, right?
If only the story was quite that simple... let's compare how two players did in third and short.
Those numbers look pretty close to me. Player A comes out slightly better in percentage terms. Over pretty modest sample sizes like that one less conversion for Player A or one more conversion for Player B basically eliminates the difference between the two of them.
Let's compare how two different players performed in third and short.
If we're comparing apples to apples, Player A and Player B are both clearly better than both Player C and Player D. At the same time, there's better reason to believe Player C is better than Player D. For him to drop down to Player D's conversion rate would have required him to convert five fewer times. Player D has a much smaller sample size, but the difference is still enough one extra conversion on the sample size wouldn't eliminate it.
Player A is Shonn Greene, not just in 2012 but over all three seasons he was a featured part of the Jets' running game. Player B is the other Jets' designed runs on third-and-short the past three seasons, excluding quarterback sneaks and runs by wide receivers. It was not just Shonn Greene that was really good at converting third-and-short the past three years. It was instead everybody on the Jets. In this broader context, it's not clear that Shonn Greene is especially good on third downs. Instead, he played for a team whose offensive line was particularly good at converting third downs.
Player C is Chris Johnson, not just in 2012 but over the past three seasons. That's not every year he's been a featured part of the Titans' running game, but I chose it to match Greene's tenure. Looking at his years in more detail, he struggled a lot in 2010 but has done a much better job of converting third-and-shorts the past two years. This shouldn't come as too much of a surprise, since I covered it last offseason. Player D is every other Titans designed run on third-and-short over the past three seasons, similarly excluding quarterback sneaks and runs by wide receivers. In this context, Chris Johnson actually looks like a very good third down back, one who converted at a much higher rate.
So, what does this mean? Let's run through some possible implications and objections.
PROPOSITION: Shonn Greene converted a higher percentage of third-and-short runs than Chris Johnson did, so Greene is a better third-and-short back than Johnson.
EVALUATION: Possibly true. The first part is definitely correct, but the reason I was inspired to write this post is I do not believe the second part automatically follows.
PROPOSITION: Chris Johnson converted a higher percentage of third-and-short runs relative to his alternatives than Shonn Greene did, so Johnson is a better third-and-short back than Greene.
EVALUATION: Possibly true. The first part is definitely correct, but as with the first proposition it is not clear that the second part automatically follows.
PROPOSITION: Shonn Greene only looks like a bad third-down runner because the Jets' other backs were also very good third-and-short runners.
EVALUATION: Doubtful, for several reasons. First, the other third-and-short carries went to players like a washed-up LaDainian Tomlinson, who was on the whole a less productive back than Greene, and Bilal Powell, who didn't see the field as a rookie and has yet to establish himself as more than just a guy. These are not world-beaters. Second, it wasn't just Tomlinson or Powell or any individual player that matched Greene's production. It was instead every Jets back, as none of the individuals stood as either less or more productive than Greene. That strongly suggests to me that third-and-short success was a matter of institutional factors rather than Greene being a particularly good third-down runner.
PROPOSITION: Chris Johnson only looks like a good third-down runner because the Titans' other backs were terrible third-and-short runners.
EVALUATION: Possibly true. I have said a lot of not very complimentary things about the play of Javon Ringer and Jamie Harper. Ringer was a useful third-down back, but I have never thought he was a good NFL runner. I think even less of Harper (my comparison of him to the better version of LenDale White is not really a compliment).
PROPOSITION: The difference between a back who may be okay on third downs and a good one makes a big difference in an NFL team's fate.
EVALUATION: False. Over the past two seasons, the Titans have converted 16 of 27 third-and-short rushes. Let's assume that Greene gets 20 of those rushes and converts at his average, and Chris Johnson gets the other 7, converting at his average. In that case, the Titans would have converted for 22 instead of 16 third downs. That's six extra first downs over two years, or three per season.
PROPOSITION: Three extra first down conversions per season on third-and-short is really valuable.
EVALUATION: False. How much exactly an extra third-and-short conversion is worth is hard to determine. The maximum value is probably 4 points, the difference between fourth-and-goal from the one and a touchdown. That sets a maximum aggregate value of three extra third-and-short conversions at 12 points. The Titans were outscored by 141 points in 2012. 141 >>>>>>>>>>>> 12, and that's assuming the maximum point expectancy value for all the extra third-and-short conversions. The real effect of three extra third-and-short conversions is probably more like 6 points than 12. Now it's possible third-and-short conversions in the right situation can be really, really valuable, but the flipside of that is third-and-short conversions can also be pretty worthless.
PROPOSITION: The Andy Levitre signing matters more for how good the Titans are at converting third-and-short than the Shonn Greene signing.
EVALUATION: Possibly true. While I believe backs can definitely influence how an offensive line looks, I definitely believe that the single biggest factor in how good teams are at converting third-and-short is the quality of the offensive line. The Jets were really good at converting third-and-short because they had a very good run-blocking offensive line. The Titans definitely upgraded their offensive line with the Levitre signing, including in the run game even though he's a better pass protector than run blocker.
PROPOSITION: Tom hates the Shonn Greene signing.
PROPOSITION: You hate how much money they gave him, and the whole point of this post is Shonn Greene isn't as useful as the press makes it sound.
EVALUATION: Definitely true in terms of money, and in why I wrote this post, but that doesn't mean I hate the signing. If you read what I wrote up of Greene's signing, you'll note I didn't say anything about how he did on third-and-short, or his numbers with the Jets, or anything like that. Instead, I wrote about his traits as a runner. He definitely has a different style than any back the Titans have spent resources on in the past couple seasons, and I believe having a back with that style is a valuable thing. I just hate the contract, and that he's in line to be the Titans' lead back in 2014 after they cut CJ.