This piece was originally published on fantasydouche.com in April of 2012.
In an earlier post I wrote that the Jimmy Johnson Chart is not consistent with a reasonable expectation of the range of human abilities. The Jimmy Johnson Chart presupposes a group of super-human football players at the top of the draft each year, when those football players don’t exist in reality. I used the example of wide receiver performance, and the fact that experts often disagree as to the best prospect in a draft, to illustrate this point.
But because people often associate the very top of the draft with quarterbacks, I figured it would be useful to look at quarterback results as well. We’ll be looking at each quarterback taken at each pick in the draft, and how that quarterback performs relative to the other QBs in career yards thrown. To put it in terms that can be compared with the Jimmy Johnson Chart, we’ll convert those performances to Standard Deviations Above Average.
Here is a graph which shows how each pick in the draft has fared in terms of standard deviations above average. The comparison group is each other quarterback in the draft. For instance, Peyton Manning is 2 standard deviations above the average for his draft class. I do that and then average out every pick in the draft and create a trend line.
The first thing to note is that the points are scattershot. That’s consistent with my argument that experts often disagree as to who is good and who isn’t good. Talent evaluation is hard. The scattershot nature of the graph above reflects how difficult talent evaluation is. The second thing to note is that the trend is fairly flat. It is not steep like the JJ Chart.
We can compare that trend against the Jimmy Johnson Chart to see if the real world of quarterback abilities is actually consistent with the theoretical world of abilities presupposed by the JJ Chart. On the graph below I show the trend for QBs and WRs, along with the Jimmy Johnson values. The JJ Chart starts at 6 SDs above average. Real world results suggest a more reasonable expectation is 2 SDs above average.
I’ve never met a point I didn’t beat into the ground and this one is no different, so I am going to say it again. The Jimmy Johnson Chart offers a view of football talent that doesn’t exist in reality. The super-human football players that the JJ Chart assumes are available, aren’t actually available.
All of my analysis has been based on averages because that’s really the way you have to look at it. You have to throw Peyton Manning, Sam Bradford, David Carr, Jamarcus Russell, Eli Manning, and Matt Stafford in an imaginary bag and pretend that you’re going to reach into the bag and pull out a name. But even if you didn’t look at it that way, even if you assumed that you could pick with total clairvoyance, your picks would not be as unique as the JJ Chart says they would be. No quarterback in the last 20 years is 6 standard deviations above the rest of the quarterbacks in his draft class.