About/Contact
Football Perspective covers the history and statistical side of football. My contact information is at chase [at] footballperspective [dot] com. All stats come courtesy of Pro-Football-Reference.com.
I’ve been writing about football for the past decade, at footballguys.com, pro-football-reference.com, the Fifth Down blog at the New York Times, and smartfootball.com. So what will I be doing here? I’ll be blogging about everything football-related, from Jerry Rice to Bobby Douglass, and from the 1978 Patriots to who is the greatest quarterback of all time.
Hopefully you can learn something in at least a few of these posts. Welcome and be sure to comment!
{ 12 comments… read them below or add one }
Chase, it was great to see an updated version of your GQBOAT series. I have a few questions/concerns:
1) Are your season/career rankings still adjusted for schedule and weather? If so, is there any chance you could publish a list of each quarterback’s adjustments, or simply include them as an extra column with the rankings?
2) How do you incorporate rushing yards into the CYP metric? I understand the method of only counting yards above 4 YPC, but I can’t figure out how you translate that into the context of a QB’s overall per play average.
3) You noted not having sack data for playoff games before 2008. I’m sure you already know this, but complete playoff sack data can be found at nfl.com, so why aren’t you using it? The same questions applies to playoff fumbles, as well.
4) In the same vein as the previous question, is there a reason you don’t adjust playoff games for schedule and weather? If anything, those would be the most important games to adjust, since you give them extra weight.
5) I remember for the 2009 rankings you combined regular season and playoff value into one overall career value chart. Are you planning on doing that for the 2012 rankings? That would be very helpful.
6) Since you have connections with PFR, would you guys ever consider listing season-by-season value and career value on PFR’s individual quarterback pages? I noticed Scott Kacsmar’s 4QC and GWD stats are now listed, and I think your QB Value stats would also be a great addition.
7) In my opinion, interceptions are overpenalized in the ANY/A formula, and thus in your Value rankings as well. I understand that the figure of -45 comes from empirical research, but interceptions are almost entirely random from year-to-year (as you’ve written about), so it seems illogical to place such a heavy penalty on an outcome that can mostly be attributed to luck. Even in retrodictive studies like GQBOAT, do interception numbers really tell us anything about a quarterback’s skill? I remember reading somewhere that it takes 5,000 pass attempts for a QB’s interception % to represent even HALF skill.
I know that’s a lot of questions, but I love your work and I want more!
Thanks, Adam. These are all great questions, but I suspect unsatisfying answers.
1) They are not. Just a timing issue here. They absolutely should be included, and I hope to have them in the next iteration. For reference, I often still check the old results, which are still up at the PFR blog.
2) Insight question: the answer, I don’t. It’s a fudge, and not included in the per play average.
But let’s say a QB has a 4000 converted yards on 500 attempts. He is averaging 8 CY/A. If he also had 400 rushing yards on 50 carries, he gets a bonus of +200. So if the league average CY/A was 6.0, he would get a Value of 1200, based on being 2 CY/A over average on 500 attempts (+1000) and an additional 200 yards of rushing value.
I suppose you could then just say he has 4200 converted yards on 500 attempts, for a CY/A of 8.4, but that will change as the league average changes. If the league average was 5.0, it would be a different story.
The moral is I still don’t know what to do with rushing data. The big problem for me is that rushing attempts are tricky; they include kneels and sneaks and three yard scrambles for first downs, which will really drop your CY/A but shouldn’t.
3) Easy but unsatisfying answer: My database doesn’t have it. Thanks for the heads up on NFL.com having it, but until PFR incorporates the data, I won’t have it.
4) Agreed that some adjustments should take place. Remember with weather, though, it is “normalized” and not actual weather. Again, at some point, we hope to get all of this 100% correct.
5) Yes, that would be helpful. Again, just didn’t get around to it yet. I suppose I still might.
6) We have batted this idea around before. At one point we said we would, but I think it got lost in the shuffle.
7) I hear you. And I don’t disagree. But to me, -45 just feels right for this system. It would seem odd to have a bunch of high INT seasons creep up the rankings. But I agree with your sentiment.
Thanks for answering my questions so quickly, Chase. Believe me, I understand you’re a busy man, and you don’t have time to worry about every little detail in every article.
I’ve experimented with a similar QB Value system to the one you’re using, except I give only a 10 yard bonus for TD’s and a 20 yard penalty for INT’s and fumbles. Admittedly, it does overrate high YPA/high INT seasons like Carson Palmer’s 2011 or Kurt Warner’s 2000. But if someone has to be overrated by the system, I’d prefer it to be the gunslinger types, not the conservative checkdown guys. IMO, most metrics vastly overrate seasons like Damon Huard’s 2006, David Garrard’s 2007, and even Tom Brady’s 2010. They put up decent YPA numbers, but had obscenely low INT rates, which were never duplicated again in any of their other seasons, indicating it was more luck than skill. But I guess the importance of each stat just comes down to personal preference.
For playoff weather adjustments, I’ve been using a formula of (50 – wind chill) * .015 for games played under 50 degrees. For example, if a game is played at a wind chill of 25, the quarterbacks’ NY/A baselines would be lowered by .375 yards per play. I also raise the baseline by .5 for dome games on turf, and .3 for dome games on grass. Those are all completely arbitrary numbers, but they seem intuitive to me.
I couldn’t find your email address anywhere on this site, which is why I used the comments forum. Do you have a contact email I could use in the future?
I do find the irony in you posting a comment asking for my contact info on the one page of the site that lists my contact info.
Wow I feel like an idiot now.
Hi Chase,
I am regular reader of your blog and thoroughly enjoy reading it. I wanted to know if you had written any article or a blog post regarding a team going for a 2pt conversion when up 5. I was wondering which is a better proposition for a winning team when up 5 with about 10 minutes left in regulation, going for XP or a 2 pt conversion.
I think going for 2 is pretty obvious there, subject to the caveat that teams have a roughly 50/50 shot at converting.
Here is one thing I wrote about a similar situation: http://www.footballperspective.com/chan-gailey-wisely-went-for-1-up-5-against-tennessee/
Do you have a list of answers to your “Super Bowl XLVII Prop Bets”
Hey Chase,
I was wondering if we could talk over email about my possible contribution to the site. I love the quantitative depth at which you and many others analyze the NFL. I have constructed a few models of my own at school. I am an economics major, but also work for the Daily Caller as a journalist and the business column editor. I think both my writing and analytic skills allow me to write solid articles. I could send you a piece and let you tell me if it is worthy of publishing. I am really interested in just getting my stuff out there and getting feedback as well.
Regards,
Andrew
Thanks Andrew. I tried e-mailing you with the address you registered with when posting, but the e-mail bounced. Please e-mail me if you would like to discuss.
Don’t know where else to post this, but will there be an item on the passing of George Sauer?
Good question, Tim. I’d like to, but those types of articles tend to consume the most amount of time, a resource I’m low on right now.
{ 1 trackback }