The draft: When 13=1
Many fans tend to think of the draft as an orderly process where a drafted player has less chance of success than the players picked before him and more chance of success than the players selected after him. While there is some truth in this, I have a different view and think the data shows that there are really ranges of draft choices that have approximately equal value.
While the cut-off points for these draft choice ranges are largely a subjective determination, many objective factors were considered in determining the end point of each draft range. These factors include the number of career starts, the number of players who were selected for Pro Bowls, the number of players who received All Pro recognition, the number of players who were starters for certain period of time, the number of players whose careers lasted a certain period of time and the number rookie starters.
Based on a review of the data from the last 20 drafts (1994 through 2013) here are the draft ranges I am espousing:
1. Selections 1-13
2. Selections 14-24
3. Selections 25-46
4. Selections 47-73
5. Selections 74-114
6. Selections 115-187
7. Selections 188 and after
The assertion here is that there is very little, if any, advantage to moving up within any of the groups (e.g., from 13 to 1) because, historically, results have been about the same for all draft positions in that range. To illustrate this point, here is a summary comparison of the outcomes for selections one and 13.
The only caveat is that it may make sense for a team to move up within a range to take a player at a high value position, such as quarterback, if demand exceeds supply. This is what the Redskins did when they move up to take RG3 in the 2012. Whether that move made sense or not, depends on whether you asked the question at the end of the 2012 season or the end of the 2013 season.
In this analysis, a player receives credit for a Pro Bowl appearance only if he was an original selectee, regardless of whether he played in the game or bowed out due to “injury”. Alternates and other substitutes do not receive credit for a Pro Bowl appearance.
The next table illustrates the degree of difference among the draft choice ranges. In order to demonstrate the point in a concise fashion, the table focuses only on Career Starts and Pro Bowl selections. For each of the two measures and for each range of draft choices the table shows the average for the entire group and then the best and worst scores (along with the draft position posting those scores).
This table also shows pretty significant differences in the best and worst measurements within each draft choice range. This illustrates the “choppiness” of the data and draft results. The following table further illustrates this point by showing number of career starts by draft position for position one through position 13.
This table reinforces the conclusion that the establishment of the draft ranges is more art than science.
My next article will focus on the probability of selected outcomes by draft range. For that analysis I will get away from using the whole 20-year period and use time periods that make the most sense for the selected metric. For example, a player selected in the 2013 draft cannot possibility have been a five-year starter or have a five-year career because he has not been in the league long enough, so a shorter period will be used for that metric.
Follow Tony on Twitter @draftmetrics