During the offseason, if you're like me, you have a certain routine. You travel to a foreign country, leaving behind your expensive computer, and start compiling and analyzing football data.
Okay, maybe you don't. But here's something I feel is somewhat of a shared experience. We've all seen tipped balls in football, and fumbles, and interceptions. Some of those have not gone our way. And we've all stared at the screen in awe, possibly screaming or throwing remotes, wondering - and I'm paraphrasing - "You're telling me this guy runs a 4.4 and he couldn't beat (seemingly slower opposing player) to the ball?!"
The rational beings hidden deep within us understand that a sprint timed in your underwear isn't going to forecast your ability to react to the unexpected. The unexpected involves a change of direction. The change of direction drill - the short shuttle - doesn't account for longer distances, though. (Imagine a player of your choice shuffling laterally towards a fumble...)
Because of this, the combine numbers have always left me feeling a bit under-served. Why provide us with figures that represent situations that we will never encounter in the game? There has to be some way to interpret these seemingly arbitrary drills and translate them to a game situation.
And then the 'ah-ha!' moment came. The short shuttle involves two changes of direction, two five yard sprints, and a ten yard sprint. The 40-yard dash timing includes a split at 10 yards (most often used when grading linemen). Subtracting a players timed 10-yard split from their 20-yard short shuttle leaves you with the time for two changes of direction and two five yard sprints. Halving this number yields what I call the "5 yard average" - essentially the time it takes an athlete to cover 5 yards in a direction opposite his current momentum.
Allow me to better explain. Below is the short shuttle description from the National Athletic Testing System official website:
Short Shuttle - Right:
Athlete begins in a 3 point stance with left hand on start line.
Run directly to the right 5 yards and touch line with right hand.
Reverse direction and run directly to the left for 10 yards and touch line with left hand.
Reverse direction again and run 5 yards through the start/finish line.
And also a picture I drew (disclaimer: I'm left-handed, but use a right-handed mouse):
The bolded section (the middle line in the picture) is being removed because we know, from the first 10 yards of the 40, how quickly a player completes that portion. What's left is run 5 yards, reverse direction, reverse direction, run 5 yards. We then divide that in half, and we're left with the time it takes to change direction and run 5 yards.
Then, using basic algebraic modelling (for those who care, I assumed the acceleration curve over the first ten yards is comparable to y = sqrt(x) ), I reintegrated the 5-yard average to the 10, 20, and 40 yard dash times, giving me the approximate time it takes these athletes to change direction and cover that distance. I've decided to call those values "maximum times," given that they represent a complete 180 degree change in momentum. This is how linemen operate in space, how receivers run routes, and how corners break on balls. This is also, unfortunately, very data dependent. Players missing full 40 breakdowns or short shuttle times cannot be calculated.
Finally, let's have some examples, shall we? We'll go through the Cowboys' 2012 draft class and see how they stack up.
Name | 40 Time | 10 Yard Split | 20 Yard Split | Short Shuttle | Short Shuttle - 10 Yard Split | 5 Yard Average | Max Time 10 yds | Max Time 20 yds | Max Time 40y ds |
---|---|---|---|---|---|---|---|---|---|
Claiborne, Morris | 4.43 | 1.56 | 2.58 | 4.12 | 2.56 | 1.28 | 1.74 | 2.76 | 4.61 |
Crawford, Tyrone | 4.78 | 1.69 | 2.80 | 4.44 | 2.75 | 1.375 | 1.87 | 2.98 | 4.96 |
Wilber, Kyle | 4.64 | 1.65 | 2.65 | 4.31 | 2.66 | 1.33 | 1.81 | 2.81 | 4.80 |
Johnson, Matt | 4.52 | 1.52 | 2.6 | 4.07 | 2.55 | 1.275 | 1.72 | 2.80 | 4.72 |
Coale, Danny | 4.37 | 1.53 | 2.59 | 4.09 | 2.56 | 1.28 | 1.73 | 2.79 | 4.57 |
Hanna, James | 4.49 | 1.52 | - - | 4.11 | 2.59 | 1.295 | 1.74 | - - | 4.71 |
McSurdy, Caleb | 4.75 | 1.68 | 2.70 | 4.28 | 2.60 | 1.30 | 1.79 | 2.81 | 4.86 |
The numbers to focus on are the rightmost four columns. 5-yard average, as I explained, represents the ability to reverse direction and travel 5 yards. By this measure, Matt Johnson is narrowly the quickest of the Cowboys' seven selections. Overall, the entire group is capable of very rapid changes of direction, which may indicate that it is something the scouting department highly values. For comparison purposes, the calculated average for the "5-yard average" for all defensive backs in the 2012 rookie class was 1.31 seconds, a mark bested by even 7th-rounder Caleb McSurdy. The average for all defensive linemen (including pass rushers) was 1.37 seconds.
When you want to measure the effect of this number on positions that require greater range, the next three numbers come into play. You'll notice a slight shift in the rankings in the 10-yard column, due to certain players having better acceleration through their first 10 yards than others. The 20-yard column sees Morris Claiborne take the lead, showcasing his higher mid-range acceleration, but he's finally overtaken in the 40-yard column by rookie receiver Danny Coale.
How do these numbers match up with the rest of our roster? Who are the quickest players on the team? What about our UDFAs? Watch out for the next installment in which I break down the Cowboys' Defensive Backs. In the meantime, if you want the numbers for any of the 2012 rookies (the entire class, not just ours), leave a comment and I'll reply with the data.