Dynasty MetriX score, or DMX for short, is a quantitative measure of an NFL rookie draft prospects potential. It is a combination of 23 different metrics that are statistically calibrated and combined to provide one overall metric. When we calculate this metric going back to 2001 and compare this to actual fantasy production we can correlate the two and determine how often a prospect “hits” given their DMX Score. So how do we come to this score? First we gather information. There are three basic types of quantitative data for each prospect, college production, draft position and athleticism. Let’s explain each briefly and the value they bring to the overall analysis.
Age Weighted College Production – It’s hard to compare player production across different levels of competition, offensive systems and usage within those systems. What we can do is compare players final year of production to the production of their entire team to understand the value of that player to the team. We look at rushing and receiving yardage and TDs in this manner and then weight the results by age to create a production measure that takes into account the value of the prospect to their team and the age of that prospect when the production occurred. The younger a prospect produces great yardage and TDs, the better of course.
Draft Position – There are a couple of reasons why I use draft position in this measure. First, the draft, at its essence, is a market. It’s a market much like the stock market. NFL teams go to extreme lengths to vet these players and understand and rank their value to the team and then choose accordingly. Embedded in that choice (and therefore that draft position) is both a very real approximation of the value of the player and a ton of research that goes well beyond the layperson’s capabilities. NFL teams have armies of security people, scouts, coaches, player personnel directors scouring every piece of information available for each player in the draft. Just like the stock market, these professionals sometimes get it wrong and you have a draft bust. Most if the time, they get it right and that logic is a critical part of our evaluation. Second, incorporating draft position allows us to evaluate all players together, no matter whether they are large school, small school, Canadian, German or whatever. When evaluating age weighted production, you have to account for the fact that an RB that produced 1,500 yards at Alabama performed against a different level of competition than an RB that produced 1,500 yards at Tiffin University. If, for some reason, the market likes the Tiffin RB better than the Alabama RB and picks the Tiffin RB first, that information should be taken into account. In our metric it is.
Athleticism – This is where the combine and pro day processes play out. Between their final game and the NFL draft, all prospects go through athletic tests to compare their athleticism to their peers. We do the same thing in the DMX score. We combine similar measures in order to break it into four categories, speed, agility, power and strength. Those then get combined into one athleticism score. For speed we use standard Speed Score measure, which incorporates weight into the calculation of 40-time which is a much better relative measure. Agility combines the shuttle and 3 cone scores. Power combines the vertical jump and broad jump and strength relates to bench press.
Once we compile all of these measures, we create a z-score which is a statistical way to measure a metric against it’s average. So, using z-score if your metric is above average it would be a positive z-score, if it is below average you get a negative z-score. This makes comparing scores across prospects and across history for each position easy. The more positive a DMX score the better and the more negative the worse the prospect. All of this statistical analysis is done by position, so that the similarities in position are being compared. Each RB is compared to the average RB score for each metric in order to calculate the z-scores and calculate the DMX score. This makes is possible to compare RBs not only within their class but among history.
As an example, below is the top rated RB DMX score that I’ve calculated in the 16 year history that I have this measure. As you can see it’s LaDainian Tomlinson with a DMX score of 1.72. This sheet below shows how the raw score at the bottom gets z-scored and aggregated at the next level, combining into a score each for athleticism, draft position and age weighted production, then combined into one DMX score.