Brett Favre – Greatest Ever?

“No data have meaning apart from their context.”
– Donald Wheeler

Unless you live under a rock (or perhaps outside of the USA) you have likely heard that Brett Favre has decided to retire from the NFL.

Like him or not, no one can deny Favre’s greatness.  He played the game the way it is meant to be played.  He was tough, competitive, and left everything on the field.  Personally, while not a Packer fan, I always enjoyed watching him play.  To be sure, I will miss him.

Greatest Ever?

The sports radio shows are all debating where to place Favre on the list of all time best QB’s.  Most have him top 5 or 10.

Since I am into data, I thought I would do a little research myself.  While I was playing around with the data it dawned on me just how badly most people screw this type of analysis up.

MS Excel Fun

Typically, if someone wanted to see how consistent Favre was over his career, say for total passing yards, they would most likely find the data and dump it into MS Excel.  Then they would get busy with the infamous chart wizard finishing with a graph similar to the one to the left (click to enlarge).

From here the debate would really begin.  Favre haters would explain how inconsistent he was.  They would say, “Just look at the data man!  It’s all over the place.”  Meanwhile, Favre fans would explain how these haters need to get their eyes checked.  After five minutes of debating the two would eventually agree to disagree.

Adding Context to the Analysis

It doesn’t have to be this way.  Instead of creating an old fashioned line chart as shown above, the two friends would be better served by adding some context to the analysis.

The easiest way to accomplish this is via a standard control chart.  The simplest choice, and my personal favorite, is to create an Individuals chart as shown to the left (click to enlarge).  The “context” is created by the addition of control limits and a measure of central tendency, namely the mean.

Once we have this graph in front of us it is pretty obvious that Brett Favre was one consistent passing machine!  All the data points fall in between the control limits demonstrating good old common cause variation.

In this example, the Favre hater would likely walk away mumbling something under their breath as the Favre fan would sit back, smile, and remember just how much they loved statistics… and Brett Favre!

If you enjoyed this article please subscribe to LSS Academy via our RSS feed.

---