Keenan Allen, Tavon Austin, and the Most Overrated Combine Measures
One of the misconceptions about stat heads or anyone who places heavy emphasis on data in football/fantasy football is that we care about all numbers the same.
The truth is that most people who use analytics in fantasy football reject most stats, labeling them as explanatory rather than predictive, i.e. for prediction purposes, they’re worthless. In many ways, stat geeks are just more discriminating in regards to which numbers they’ll use; they use fewer stats than the general public, just better ones.
For example, prior to the 2013 NFL Draft, I mentioned that wide receiver Keenan Allen moved up my board quite a bit. I got a few responses to the tune of “But Allen’s 40 suks n thats all u care about. Your a f’ing moran.”
I might be a “moran,” but the stats I care about most are those that are going to help me make better predictions and win fantasy football leagues. More speed is better for every position, but it’s not all that vital for wide receivers, who thrive on size. So us stat geeks aren’t just sitting around and jacking each other off every time someone runs a 4.3. I care a whole lot about straight-line speed for running backs, but for the majority of positions, the 40-yard dash isn’t terribly important.
Predictive of Success
Measurables are important for fantasy owners because it’s really easy to create models, form hypotheses, and most important, test our theories to improve our approach to the game. I’d argue a stat geek can give you more insights (and accuracy) into a particular draft class in a few hours with a spreadsheet than all of the cumulative hours put in by some team’s scouts over the course of a year. If that weren’t true, we probably wouldn’t see Trent Richardson drafted in the top five, Tavon Austin in the top 10, or Mark Ingram in the first round.
The most important thing we need from a measurable is that it can help us make better predictions regarding a player’s future. One of the common qualms with the 40-yard dash is that “players almost never run 40 yards in a straight line during games.”
To which I respond: Whooooo carrrrrrrreeeeessss?
Employers don’t ask current workers to take IQ tests as part of the job, but they sure can help predict who’s best for it. The 40-yard dash and other measurables are like IQ tests; the performance is representative of something that might or might not help an employer. For certain positions, a particular measurable might not matter as much as for other positions, much like an IQ test is probably a better predictor of success for a chemist than a cashier.
The Most Overrated and Underrated Measurables
By looking at a particular measurable and stacking it up against NFL production, we can test how strongly the two have been correlated in the past. In doing that, it’s clear that the short shuttle is one of the most overrated measurables out there.
I charted the strength of the correlation between the short shuttle and NFL production (in terms of approximate value) for the running back, wide receiver, and tight end positions. To give you an idea of the weakness of the relationship, I charted it next to the broad jump, which I believe is one of the most underrated metrics.
For the short shuttle, a negative correlation means that as the time goes down (meaning it gets better), NFL production increases. Shockingly, the short shuttle correlation is negative only for running backs, meaning that as times for wide receivers and tight ends have increased (gotten worse), NFL production has actually improved! That suggests the short shuttle has no predictive ability for pass-catchers.
Further, the strength of the correlation between short shuttle times and running back success is just (-.09), which is almost nothing. How many times have you heard a TV analyst say that a running back needs to be “quicker than fast?” I think Mike Mayock even has a “QTF” tattoo on his lower back. Unfortunately, running backs don’t need to be “quicker than fast.” They need explosiveness, represented in both straight-line speed and the broad jump.
Like the 40-yard dash, the broad jump is a maneuver that you’ll never see an NFL player complete on the field during a game. As much as I’d like to see Eli Manning forced to broad jump for the win, it’s not happening.
And again, whoooooooo carrrrreeeeessss? The broad jump is very representative of explosiveness. Because of that, there’s a strong correlation between a player’s broad jump and 40-yard dash. Both metrics are most important at positions that require the greatest degree of explosiveness. For so long, we thought wide receivers needed explosiveness more than just about any other position. It’s far more important for running backs, though, as evidenced by the graph.
As I’ve mentioned, measurables have less and less impact on player projections as the players gain more NFL experience. I care a whole lot about a running back’s 40-yard dash when he’s a rookie, a little less after he has a season’s worth of NFL touches, a little less still after two seasons, and so on.
Still, there are so few NFL games in a season and players are so dependent on one another for success that we can really use measurables to project performance even after a player has been in the league a handful of years.
For example, assume you’re trying to predict the play of a 25-year old running back who was really successful in one city but signed as a free agent elsewhere. Since running back play is so dependent on the offensive line, it’s not out of the question for a back to see sustained success in spite of lackluster talent. If his offensive line situation changes and you need to assess his true skill level, the 40-yard dash and broad jump can help do that.
Measurables are important, but we need to understand how to best utilize them; different metrics are valuable for different positions, just as is the case with other jobs. The 40-yard dash, broad jump, vertical, and so on can all help us more accurately forecast NFL player performance for different positions.
Regardless of the position, though, it appears as though the short shuttle is not very representative of future NFL play. Even if there is a relationship there, it would be difficult to acquire value by emphasizing it since the majority of NFL teams factor it into their rankings to a degree that’s stronger than the measurable’s actual worth. And since your fantasy draft will be a reflection of the NFL draft when it comes to rookies, that means that you, too, won’t find value by emphasizing the short shuttle and, perhaps, could even acquire value by intentionally seeking players who performed poorly in the drill but showed explosiveness elsewhere.