by Tony Villiotti

February 20, 02013

Which Combine drills should draftniks be watching most closely during this year’s event?

DRAFTMETRICS sets out to answer that question in its final pre-Combine article. This subject was discussed by playing position in the Combine Chronicles series of articles, but this article takes a much broader look at the issue.

It should be noted that both quarterbacks and fullbacks are excluded from this article. Fullbacks are excluded due to the small number of showcase participants. Quarterbacks are excluded because of DRAFTMETRICS belief that the Combine drills are largely unimportant for the position.

For the purposes of this article, DRAFTMETRICS developed a Multiplier, described below, that facilitates an objective appraisal of three issues:

**• What are the most important drills for each playing position?
• How does the importance of the drills vary by playing position?**

The Multiplier is based on the impact of the results of each Combine drill (by playing position) on the probability of becoming a starter. If the difference in Combine results has little of no effect on the probability of starting, that drill is unimportant in predicting future performance. A large difference in probability indicates that the drill could be (emphasis on “could”) an important predictor of success.

The Multiplier is based on the average of two calculations:

• The percentage of players with results in the 80th percentile or better who become starters divided by the percentage of players with results below the 50th percentile who become starters

• The percentage of players with results in the 50th percentile or better who become starters divided by the percentage of players with results below the 50th percentile who become starters

Described another way, the Multiplier represents how many players with better than average results in a particular drill become starters for every starter with below average results. So a Multiplier of 2.00 means that twice as many players with “good” Combine drill results become starters as players with average or below average Combine results.

The mechanics of calculating the Multiplier is best illustrated using the actual calculation for cornerbacks participating in the 40-yard dash. 51% of all cornerbacks with Combine results in the 80th

percentile or better became starters; 42% of all cornerbacks with Combine results in the 50th

percentile or better became starters; and 22% of all cornerbacks with results below the 50th

percentile became starters. The calculation is as follows:

**• 51% (probability of starting for players in 80th percentile or better) divided by 22% (probability of starting for players below the 50th percentile), or 2.32
• 40% (probability of starting for players in 50th percentile or better) divided by 22% (probability of starting for players below the 50th percentile), or 1.82**

There are several things to bear in mind regarding the Multiplier:

**• A high Multiplier indicates that the drill is potentially significant for that playing position
• A low Multiplier indicates that the drill is probably not that significant for that playing position**

Two tables are presented at the end of this article, one that reflects the Multipliers for players who started at least one season and the second for players who started at least three seasons. In reading the tables please remember the phrase, “there are lies, damned lies and statistics”. DRAFTMETRICS certainly believes there is valuable information to be gained from the Combine results, but some of the correlations are somewhat dubious.

The tables, for example, show that the flying 20 is an important drill for offensive tackles. This doesn’t pass the smell test as tackles seldom run very far downfield on a running play. It could be interpreted, though, that the drill results are simply a measure of athleticism for offensive tackles and, in that context, may make sense. Whatever one’s interpretation, the message isn’t that the information is of no value, just that you can’t apply it blindly.

The highlights from reviewing these tables are as follows:

It is interesting to note that the Multipliers for those who started at least three years were considerably higher than the players who started for at least one season. There were, for example, 39 instances of Multipliers of 2.00 or more for three-year starters and only 11 instances for one-year starters. This would seem to indicate a stronger correlation between Combine results and starting for the longer term starters.

When reading the following tables please note the following:

**• Reading down each column, the Multiplier for each drill across all playing positions is shown
• Reading across each row, the Multiplier for all drills for each playing position is shown
• Numbers in bold indicate the highest Multiplier for that position
• “Max Data Points” represents the maximum number of data points for that playing position**

Tony is the founder of DRAFTMETRICS.COM can be e-mailed at draftmetrics@gmail.com and followed on Twitter @draftmetrics