by Tony Villiotti
April 16, 02014
A few articles ago, I wrote about variability among draft classes in an article entitled “How Deep is a Draft”. A reader asked whether I had ever been looked at variability for first round selections only. I had not previously done that, so it sounded like a good idea for an article. I reviewed first round selections from two perspectives. First, the classes were reviewed for variability, as suggested by the reader. Second, I thought it would be fun to also rank the draft classes from best to worst.
But why the 1999 through 2008 draft classes? My thought was that I wanted to include the number of five-year starters because I think it is one of the most important metrics in evaluating draft success. In order to include five-year starters, and consistent with past articles, it made sense to end the study with the 2008 draft class. This allows a player six years to achieve five-year starter status. So, if I wanted to use a 10-year study period, 1999 to 2008 was it.
Six metrics were considered both in the review of the variability and the ranking of the draft classes. A table at the end of this articles shows the metrics for each draft class.The metrics used were:
-The number of draftees that played five or more years
-The number of draftees that started for three or more years
-This means a player started at eight games in each of at least three seasons
-The number of draftees that started for five or more years
-The number of games started by the draft class, as adjusted for two circumstances
-One adjustment was for the number of selections in the round (some years only had 31)
-The second, and more significant, adjustment was for differences in the number of years since the draft class was selected
-For example, history tells us that about 72% of the starts in a draft class occur by the end of the seventh season, so the actual number of starts by the 2007 class had to be divided by .72 to calculate the number of adjusted starts.
-The number of draftees selected for at least one Pro Bowl
-The number of draftees selected as an All Pro at least once
My conclusion regarding variability among the first-round classes is that, while there are year-to-year fluctuations, the overall degree of variability is pretty modest. As would be expected, there is more variation for the metrics that are most difficult to attain (e.g., All Pro selection), with some that difference due to “outliers”. Comments on each metric are as follows:
Play five or more years
-High was 32 (2006 draft class) and low was 26 (1999 and 2000, years with 31 selections)
-Six of the 10 classes had either 26 or 27 players with careers of five years or more
Start three years or more
-High was 27 (2004) and low was 21 (2008)
-Six of the 10 classes had 23-25 players start three or more years
Start five years or more
-High was 23 (2006) and low was 14 (2008)
-It is highly likely that the 2008 class will add one or more five-year starters as careers progress
-If 2007 and 2008 are excluded the range is 18-23
-Five of the 10 classes had 20-22 five-year starters
-High was 3416 (2001) and low was 2643 (1999)
-There are significant projections used in the later draft classes to arrive at the adjusted starts, so that is something that must be considered
-Seven of the 10 classes were within 10% of the average number of adjusted starts
Pro Bowl Selections
-High was 15 (2001) and low was 9 (2008)
-It is possible that the more recent draft classes will add to their totals
-Six of the ten draft classes had 10-12 Pro Bowl selections and eight had at least 10 players selected to a Pro Bowl
All Pro Selections
-High was 8 (1999, 2003 and 2007) and low was 3 (2004)
-Eight of the 10 classes had 6-8 All Pro Selections
The relatively low level of variability for the draft classes makes it difficult to rank them. Even under the best of circumstances, a ranking is difficult because it depends on whether the person doing the ranking places a higher value on post-season honors, the number of five-year starters or the depth of the class. The table at the end of this article shows the metrics plus my judgment regarding the best and worst selection in each draft class.
I considered all of the metrics and ranked the classes in the following order: 2001, 2006, 2005, 2007, 2004, 2002, 2003, 2008, 1999, 2000. How much do your rankings vary from mine?
My rationale for the top three selections is as follows:
-The 2001 draft class has (1) the highest number of starts (nearly all of which were “real” and not projected), (2) the most players selected to a Pro Bowl and (3) 29 of the 31 draftees played five years or more
-The downside of this class is that it did not rank at the top in starters
-The 2006 draft class had all of its 32 members play at least five years, had the most five-year starters, produced the second highest number of projected starts and had the second highest number of All Pro selections
-The 2005 draft class ranked high in pretty much all of the metrics except for number of adjusted starts.
(1) New England forfeited a first round selection in the Spygate fiasco