When it comes to statistical analysis of football, no one does it better than Football Outsiders. For the past four years, in addition to their work on the internet, they’ve also published the Pro Football Prospectus. This year however, their publication is titled the Football Outsiders Almanac 2009. So what happened?
From the introduction of the book:
So why the name change, and why aren’t we in bookstores?
For those who don’t know, our frst four books were published through an agreement with Prospectus Entertainment Ventures, the company that owns Baseball Prospectus (as well as the expansion projects Basketball Prospectus and Puck Prospectus). It was PEV that had the publishing contract (first with Workman, then Plume). This year, for various reasons, Plume decided they no longer wanted to publish books related to other sports besides baseball. Other publishers were interested in doing our book, but by the time Plume made their decision, it was too late to get on the publication schedule for 2009.
Thus, because of this mess, they decided to go the self-publishing route.
The head honcho of Football Outsiders is Aaron Schatz, he’s a Patriots fan and New England native. He and I go back a few years, so I asked Aaron to answer a few questions and give us a taste of what to expect in this year’s edition.
I’m not sure what we’re going to do for next year yet. We have a couple of publishers who are interested in putting out Football Outsiders Almanac 2010 as a standard book. Once we get into October and the book is done selling for this year, we’ll sit down and figure out whether it makes sense to go back to a regular publishing format. We definitely lose a lot of the “promotional value” of the book by doing it ourselves, since we’re not in bookstores to catch the eye of casual readers who may not know about our website. However, there are also significant advantages to producing a book online. We keep a larger share of the gross sales. Also, the previous books were written and edited under a completely ridiculous rushed schedule, where we basically had to do the entire thing in about six weeks after the NFL draft. By doing it ourselves this year, we had an extra month. I think it meant a lot fewer errors in the text, not to mention a huge heaping helping of sanity for me and the other writers. My wife definitely prefers the self-publishing schedule because she didn’t have to be a single parent for six weeks.
2) How have you improved the DVOA in version 6.0?
DVOA, for those who don’t know, is our main statistic. It stands for “Defense-adjusted Value Over Average.” We take the success of every single play during the season and compare to other plays based on situation and opponent. We’ve updated constantly since I started doing this back in 2003. The biggest change this year is that we’re now considering the baseline differently for offense and defense. That allows us to better measure some plays that are really the responsibility of the offense only, where the defense has no effect: false starts, delay of game, and aborted snaps. We’ve also improved the way we adjust for teams playing from behind or with a lead in the fourth quarter.
3) I see you’re also doing more College Football in this publication, tell us about that…
Well, as a Bostonian I don’t really follow college football. The line I usually give to people around the country is that if it doesn’t have to do with Doug Flutie, nobody around here really cares. But I also know that college football has a huge following in other parts of the country, and there’s no reason why we can’t provide the same kind of intelligent analysis for the college game. There’s also the benefit of eventually being able to make better projections which NFL draft picks will succeed once we have more data on the college game. So I went out and looked for people who loved college football as much as I love the NFL, who write well, and who had the same outlook on doing advanced stats, and I found Brian Fremeau and Bill Connelly. We’ve been doing their stats on FO for a couple years now, and this year I wanted to expand that with a full college preview. So the book has about 90 pages of college football in addition to all the NFL material. There are stats tables and writeups for every team in the six BCS conferences, plus a handful of the top independents and mid-major teams. The goal of our college content is the same as the pro content– we want to go beyond just ranking the teams 1-120 and really look at WHY teams won or lost last year and why we can expect certain teams to improve or decline this year. College fans will really enjoy it and people like me who know nothing about college football can learn what they should be looking for on Saturdays in preparation for the 2010 draft.
By the way, we have BC projected 14th in the nation, for those people who do care…
4) What are some of the key differences between the Football Outsiders Game Charting Project and the official boxscores and play-by-play.
Oh, we measure all kinds of things in the Game Charting Project, adding detail to the play-by-play so we can better analyze teams and players. The biggest item is probably defensive coverage — measuring defensive backs by how they do in coverage rather than just when they make tackles. We mark the formation on every play. We mark the number
of pass rushers and blockers so we can see which teams do the best when blitzing or not blitzing. We track why passes are incomplete, so we know which quarterbacks tend to overthrow their guys, or who suffers from the most dropped passes. We count quarterback hurries by defenders, dropped interceptions, and a number of other things.
5) Since we’re dealing mostly with a Patriots fanbase audience here at BSMW, tell us something surprising about the Patriots that we’ll learn in the book…
Here are five fun tidbits.
- Last year, the Brady-less Patriots actually had the best offense in the NFL from Week 9 onwards, according to our DVOA stats.
- The Patriots led the league with an average of 6.2 yards after catch; the next highest team, New Orleans, averaged just 5.6 yards after catch.
- The Patriots ran WR or TE screens 30 percent more often than any other offense.
- Showing the weakness of last year’s secondary, the Patriots didn’t have a good pass defense even when they hurried the quarterback. Only New Orleans and Detroit were worse when there was a quarterback hurry. The Patriots allowed a league-high average of 7.3 yards after catch on plays where they hurried the quarterback.
- The Patriots’ 47-7 snowstorm blowout of Arizona was the second-most impressive game played by any team since 1994, according to single-game DVOA ratings. The only team to score higher in one game was the 1994 Philadelphia Eagles, when they beat eventual Super Bowl champion San Francisco 40-8 in Week 5.