Second Round: QB Analysis

Last week saw the beginning of Easy Mode’s preseason analysis of fantasy football. My co-author kicked off the first round properly with a look at the running back. And now, we move on to my quarterback rankings.


I’m not going to write nearly as much about my rankings as I don’t have very much faith in my abilities as a football scout. There are experts out there who spend far more time than I pondering the subject of who is the fifteenth best quarterback in the NFL. There is a fairly high bar to becoming an acknowledged fantasy football expert, and while that bar may consist in part of an ability to market yourself as an expert, it is the only workable foundation I can come up with to determine a player ranking system. I would prefer to do an analysis of the past success of each expert’s picks, but that seems like a bit much work and it’s probably far too difficult to find the historical data, anyway. The next best thing to using the best expert (well, I can think of a few different methods if I had reliable data on past success rates) is to use the entire group of experts. Most people probably do something similar by employing a variety of experts’ lists, but I’ve gone ahead and averaged out the picks from a group of experts.

I picked my experts by trying to use the more popular fantasy football websites that already have QB preseason rankings up. It would be nice to employ statistical weights based on site popularity, but my lack of both motivation and knowledge of statistics once again rears its ugly head to foil my plans. So, this is just a simple averaging of eleven experts (technically, I used yahoo’s combined rankings which is the average of four experts plus public voting, I had no real reason not to use all four experts’ picks separately but that seemed like I would be favoring Yahoo unduly, whatever). I realize that a lot of people prefer close examinations of various stats and such, but like I said before I have no confidence in my ability to predict the future. Furthermore, I don’t know of any particular performance prediction method which is superior to others, and I don’t trust my ability to recognize such a method in any event.


Without further ado, here are my rankings of the top 30 quarterbacks as voted by eleven experts from such websites as, and The average ranking is the first number after the player’s name and the standard deviation is the second. I realize this is very much amateur hour at the statistical analysis of fantasy sports table, much less the statistics table.

1. Manning 1.00 0

2. Palmer 2.55 0.82
3. Brees 3.27 1.10
4. Brady 3.82 0.98
5. Bulger 4.82 1.17

6. McNabb 6.82 2.68

7. Young 8.55 2.38
8. Kitna 8.91 2.77
9. Romo 9.64 1.86
10. Hasselbeck 9.73 1.62
11. Vick 10.64 3.20

12. Leinart 12.45 2.66
13. E. Manning 13.73* 2.83
14. Cutler 13.73* 4.43
15. Rivers 13.91 2.39

16. Roethlisberger 15.45 2.11
17. Favre 16.09 1.51

18. Smith 19.73 1.95
19. Losman 19.82 3.16
20. Grossman 20.91 3.88
21. Delhomme 21.00 1.48
22. Pennington 22.00 2.72
23. Schaub 22.09 3.45
24. McNair 22.20 2.49

25. Campbell 24.50 2.22
26. Green 25.80 2.94

27. Leftwich 27.60 2.17
28. Jackson 28.20 2.94
29. Garcia 29.20 3.89
30. Russell 30.78 2.39

*The tie for 13th place went to Eli Manning because his standard deviation was lower. One website ( ranked Cutler at 24th. I try to be risk averse in my draft picks, so I went with the safer of the two. If you don’t know what standard deviation is, my hazy recollection/wikipedia lookup definition is that it’s a measure of how closely the data points are grouped around the mean. So, a lower standard deviation means that the rankings were more closely grouped around the average ranking for that individual player.

The spacing represents some tiers that I think exist in the list, although I’m probably forcing things a bit. One thing that I can say without reservation is that every expert that I’ve come across puts Peyton at number one. The next four picks, tier 1.5, are fairly evenly spaced out. No one in the top five received a lower ranking than six. McNabb (tier 1.75) would continue the trend of evenly spaced out picks, but two websites ranked him at twelve. He’s an injury risk and maybe even a personality risk, what with the dog fighting and whatnot. Seven through eleven, tier 2, have regular spacing and not too many rankings that diverge significantly from the norm. One guy ranked Vince Young low, two guys did the same to Kitna, and Vick has the highest standard deviation of the bunch with experts either loving or hating him, relatively speaking. The next significant jump in rankings comes up number twelve with Matt Leinart. Leinart through Rivers seem to be tier 3, although I could be reading my data incorrectly here. Cutler is the biggest gamble in that group with picks ranging from 9th to 24th. Without that 24th pick, Cutler would move up a pick and be just behind Leinart. Roethlisberger and Favre are probably the entirety of tier 4. You can see the rest of the tiers just as easily as I can write them out here, so I’ll leave that to you, gentle reader. Hopefully, you don’t find yourself trying to decide between Pennington and Schaub in any event.


As the rankings get lower, the list becomes increasingly dodgy. Not every website ranked every player, so I had to finagle things a bit. In fact, I have a total of 47 quarterbacks ranked using this method, but the bottom of the list probably leaves out some players due to my laziness. As a point of interest, Kerry Collins is the worst ranked player for whom I wrote down rankings data. He just barely edged out Trent Dilfer for this honor.

I have a bit more analysis of the less useful variety. Of the websites that I used, probably had the rankings closest to the average. Technically, Yahoo won this award but their rankings were already an average and they only 22 picks, so they got voided. For the most divergent rankings, ESPN tied with for the wackiest/ballsiest quarterback rankings. ESPN apparently has high hopes for Vick and Rivers. Jamarcus Russell easily beat out pretty boy Brady Quinn with Quinn coming in a distant 36 or so to Russell’s 30 spot. With that, my dispassionate look at quarterback rankings is complete. Good luck to you all, unless your in my league.


Edit: Today’s announcement of Vick’s indictment might have some effect on his fantasy ranking.

Published in: on July 17, 2007 at 6:39 pm  Comments (1)  

The URI to TrackBack this entry is:

RSS feed for comments on this post.

One CommentLeave a comment

  1. […] the best system that I can (or will) come up with to rank players for a fantasy draft. You can click here read my previous post for a bit more explanation if you’re interested. My framework was […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: