Late last month, the folks at Football Outsiders released the 2012 edition of their Football Outsiders Almanac. Although I sometimes disagree with what they have to say, I'm always intrigued by the angles they take on breaking down the NFL. And of course it doesn't hurt that they have former NN writer (Florida) Danny Tuccitto working as an associate editor.
The release of the 2012 FOA means it is time for another Q&A, and naturally we're chatting with Danny. You can swing over to FOA to pick up a PDF copy of the Almanac for $12.50.
Niners Nation: The big question concerns the mean projection of 7.2 wins and the "plexiglass principle" you mentioned in the chapter. How do you asess this in light of the coaching staff entering their second season and last year's lockout removing offseason workouts? How reasonable or unreasonable is it for 49ers fans to think having a full offseason can be a difference-maker? Or is there simply not enough data on this kind of topic to assess it?
Danny Tuccitto: I've got a piece on this very issue going up on FO later in the week, so stay tuned. For now, I'll say this. I think it's perfectly reasonable for a 49ers fan to believe that the switch from Singletary to Harbaugh makes it less likely that San Francisco will come back to the pack this season like the typical "plexiglass" team. At the end of the day, that might go down as one of the biggest bad-coaching reversals of all time. Off the top of my head, Bill Parcells replacing Rich Kotite for the Jets in 1997 seems more drastic, but it's a short list.
I don't think having a full offseason is going to matter much, though. Yeah, there's nowhere near enough teams that would qualify for any kind of statistical test simply because work stoppages hardly ever happen. To boot, the 1982 strike happened during the season, so even there you had a full offseason. Incidentally, the 1980-1982 49ers were the personification of plexiglass: from a 6-10 non-playoff year to a 13-3 championship year to a 3-6 non-playoff year despite the 1982 playoffs being a 16-team tournament.
But I digress. The point is that there's no way to do a study on this. Thinking it won't matter is just my opinion, and honestly I'm not even sure how you'd objectively assess whether I'm right or wrong after the season. If San Francisco ends up going 8-8, did the full offseason prevent 6-10? If they end up going 13-3 again, did the full offseason prevent 11-5? Did it not matter at all? Heck, if they go 4-12, did it prevent 8-8? After all, their seven-win jump came after an abbreviated offseason, so maybe that's better for them? (I'm half-joking on that one). There's just no way of knowing.
NN: One big discussion point for the team has been the incredible turnover differential. On the defensive side of the ball, how sustainable is that differential?
Well, there are a couple of ways to look at this: one big picture, one arthouse film. In terms of the big picture, I'll direct you to this excellent Yahoo! piece by Joe Fortenbaugh of National Football Post. Basically, any way you slice the 49ers' overall giveaway/takeaway ratio last season, historical precedent predicts a huge fall this season. The exact numbers depend on how you characterize "49ers-like" teams, but we're talking somewhere between a 15-to-20-turnover swing in the opposite direction this year.
Now, when it comes to turnovers, FO likes to look at the trees in addition to the forest. We approach it from a play-by-play angle, not just a season-total angle. We care about what happened on a play with a potential turnover, not just the actual turnovers that count in turnover differential. In other words, did Alex Smith's pass get flat-out dropped by a defender? Did Michael Crabtree volleyball set the pass to a defender who caught it? Was an interception (either way) just a desperation heave at the end of a half? Did a Jeremy Maclin fumble bounce happily into the awaiting arms of Dashon Goldson to seal an increasingly unlikely victory? Basically, what we're saying is that throwing an accurate pass is a skill, forcing a fumble is a skill. Tip drills, Hail Marys, and random bounces off of the turf are not skills.
From this perspective, 2012 isn't as pessimistic looking. For Smith, we have a stat called Adjusted Interceptions, which removes Hail Marys, late-game, fourth-down desperation heaves, and receiver tip drills from the interception total, but adds in potential interceptions dropped by defenders (which we keep track as part of our game charting project). Last season, Smith's actual interception rate (i.e., interceptions divided by attempts) was 1.1%, which was the third-best in league history among quarterbacks who threw over 400 passes. History just screams that he's due for a regression to the league average (2.9%) this year. However, his adjusted interception rate was 1.3%, second-best in the NFL behind Aaron Rodgers (0.8%), which means his inevitable regression won't be because he was getting really lucky; just skillfully throwing accurate passes away from defenders. Compare that with someone like Sam Bradford, who had an actual interception rate of 1.7%, but an adjusted interception rate of 3.4%. Defenders dropped as many interceptions against him as they caught (6).
What's more, although we need a few more years of data before we write this up definitively, it looks like that difference between actual and adjusted interception rate is a good leading indicator for the following year's stats. Guys like Bradford (i.e., actual <<< adjusted) see their actual interception rate explode the following season, whereas guys like Smith (i.e., actual < adjusted) don't suffer as much. Same goes in the opposite direction, which Smith was an example of last year. In 2010, his actual interception rate (2.7%) was slightly higher than his adjusted interception rate (2.5%), and so his actual interception rate decreased slightly in 2011. Like I said, need more data, but it's just one more thing in favor of the idea that Smith probably won't turn into the second coming of Vinny Testaverde this season.
In terms of fumbles, the main thing we look at is fumble recovery rates, which are almost entirely dumb luck. Teams should recover 50% of fumbles, so a real outlying recovery rate means they were unusually lucky or unusually unlucky, which should reverse itself the following year. In 2011, San Francisco's offense recovered eight of their 13 fumbles. There's about a 30% chance of that happening randomly, which isn't enough to say they were unusually lucky. Any poker player will tell you a 30% chance might as well be 50/50. On defense, they recovered 13 of their opponents' 27 fumbles, which is almost exactly 50/50, so no luck there either.
The only part of the equation we don't really have a handle on is the predictability of defensive interceptions, but we're working on it. Another kind of preliminary quirk we've found, but requires a much deeper look than just noticing patterns anecdotally, is that defensive backs with a lot of passes defensed, but not a lot of interceptions, tend to have more interceptions the following year, and vice versa. Carlos Rogers was a great example of this last year. As I mentioned in the book, he had only four interceptions from 2008 to 2010, but led the league with 24 passes defensed (on 97 targets) in 2008, and averaged 16 per season (on 72 targets) over the three-year period (seventh-most among corners).
NN: No Q&A would be complete without an Alex Smith question. Your KUBIAK system projects out some modest improvements for Alex Smith (albeit with more interceptions). This question is somewhat similar to the first question, but Smith specific: does the second year of the Harbaugh system provide the potential for greater upside? Can FO's advanced stats assess this or is the Alex Smith career path sufficiently bizarre enough to make that difficult?
Guess I should have mentioned this a minute ago, but, yeah, we project Smith to have a 2.2% interception rate this year; slightly higher than last year. Anyway, Harbaugh (and Greg Roman) had almost everything to do with his improvement in 2011, so I don't see why more tutelage wouldn't translate to more improvement. If we're talking purely about fantasy numbers, I think any improvement there will be due to what Jerry Rice recently mused about, i.e., taking Smith's diapers off. A team doesn't go out and get the kind of receiving talent they did, and add a Darren-Sproles-type player in the draft if they're going to stick with the ground-and-pound of the previous year. That's not to say we're going to see a Packers/Patriots/Saints offense stats-wise, but it's almost inconceivable that they finished 31st again in pass attempts.
Specifically, I think you'll see a lot more deep passes because of Randy Moss, which actually plays into one of Smith's strengths. According to our numbers, he was the ninth-best quarterback last season on passes that traveled 16 or more yards in the air. Just the mere presence of more talent on the outside should help insofar as all-of-a-sudden Vernon Davis isn't getting triple-teamed in the middle of the field.
NN: How does FO view the addition of Randy Moss. As a guy coming off sitting out a year and currently 35 years old, what is the thought on his potential impact with the 49ers?
Expanding on what I just said, I don't think it takes a genius to realize that Moss gives San Francisco a deep threat they haven't had in years. However, studying up on Harbaugh as much as I have, I think the main reason they signed Moss -- aside from the bargain basement price -- is tactical. It opens up all kinds of new wrinkles in play design just having him out there. What does a safety do when Moss and Davis are running simultaneous go routes at him? What happens to a defense's run-focused game plan after the first time Moss beats them deep, regardless of whether the pass is actually completed? Their play-action passing game immediately gets even more efficient than it was last season (fifth-best). For an X's-and-O's junkie like Harbaugh, it really must have been like Christmas morning the day they signed Da Real Otis.
With all of that said, putting a definitive fantasy projection on a guy like Moss is basically impossible. KUBIAK has him at 34 catches for 494 yards and 5 touchdowns, but that's because we only see him getting 57 targets (read: decoy). I can already see the comments section striking down upon me with great vengeance and furious anger for that one. But let's keep in mind the fact that, after Crabtree's 114 targets and Davis's 95, the next-most-targeted San Francisco pass catcher -- not wide receiver, pass catcher -- had only 36 targets (Delanie Walker). Granted, that's because they had a bunch of stiffs playing musical caskets at split end. However, those stiffs combined for 73 targets, which was only 16.5% of the offense's total targets. Let's say opening up the offense means 500 total targets this year. That would mean 83 targets for Moss provided a full season and no siphoning of targets towards, say, VD. Maybe Moss hits that, maybe he doesn't. The point is that the possibilities are all over the place with him. At the high end, we have 83. At the low end, he could totally flame out, and end up on the waiver wire by Week 5. (Hey, it happened in 2010.) Somewhere in the middle is 57.
NN: The assessment of Frank Gore points to last season as the beginning of the end for Gore. How do you guys view LaMichael James, Kendall Hunter and Brandon Jacobs as guys who could potentially carry the load alongside Gore?
Hey, we all should hope our last best year of work is as good as Gore's was in 2011. Yeah, we project him to decline considerably this season for various reasons. First, he's 29 years old, and backs historically tend to decline after age 28. Second, we foresee a big decline in workload: about 50 touches. One of the things about Gore that went totally unnoticed last year was that he was an afterthought in Harbaugh's passing game. In the previous two seasons, Gore averaged 74 targets. Last year, he had 31. That's a lot of production put out to pasture.
Like their receiving corps, the usage rates for San Francisco's running backs heading into 2012 are a total enigma at this point. As is the case with adding Moss and Mario Manningham (and to a much lesser extent, A.J. Jenkins), teams don't bring in Brandon Jacobs and LaMichael James to star on the scout team, especially with an aging starting running back. The theme seems to be lessening Gore's workload, so the other backs will most likely serve as situational vultures. Jacobs will handle the short-yardage and goal line stuff. Kendall Hunter and James will handle passing downs. All three will get token touches on running downs to keep defenses honest. Gore will take care of the rest.
One thing I'll say about James, who fans are (understandably) expecting big things from, is to be patient. As I noted in the book, he has a better Speed Score than Sproles, which bodes well. However, he had practically no involvement in the passing game at Oregon, and there's no way that Master of the Max Protect, Jim Harbaugh, is going to put him out there on some kind of Alex Smith fratricide mission.
NN: On the offensive line, the chapter discusses the 49ers struggles in power success rate and stuffed rate, in spite of the use of numerous jumbo packages. "Four starters return (Staley, Iupati, Goodwin, and Davis), so it's unclear how any of this will change outside of added experience and the benefit of a full offseason." Once more on the full offseason, for Iupati and Davis entering their third year, is there reason for optimism for either or both?
I'm no offensive line guru, but our Ben Muth (formerly of Stanford's offensive line under Harbaugh) is. For the unaware, I'd suggest checking out Ben's scouting report on the Niners line last season. He's downright bullish on Iupati, and mentions how the full offseason will help some of the (few) underwhelming nuances in his technique. Davis is another matter, although Ben thinks he's not lacking in physical talent. The issue with Davis for the upcoming season is that he gave up the most blown-block sacks of any lineman in 2010 (12.0), and followed that up with the sixth-most in 2011 (8.5). Given all this talk of more passing in 2012, the first half of San Francisco's schedule might turn out to be absolutely brutal for Davis. He'll be matched up against Clay Matthews, Jr., Cliff Avril, Mario Williams, Justin Tuck, Red Bryant, and Chris Long in the first nine games. Things get much easier in the second half of the season, though.
NN: What are your thoughts on the 49ers secondary? The team was more well known for their dominant rush defense, but given how much of a passing league this is, it seems like further improvement is needed from the secondary. Specifically, does FO have a specific view on Dashon Goldson pre-all-22 film release?
I already talked about Rogers in terms of interceptions, but I'll add that he was once again among the league leaders in passes defensed last year. Along with Tarell Brown, and Chris Culliver, an honorable mention in our top 25 prospects chapter, the Niners are set at cornerback. Safety is another matter altogether, especially when it comes to depth. Thus far in his career, Dashon Goldson seems to have this pattern where he alternates good seasons and bad seasons, but that might just be background noise. Donte Whitner is a nice player who turned the Saints' bounty program on its head when he knocked Pierre Thomas out. Behind them, though, are C.J. Spillman, Trenton -- we'll call him Trent -- Robinson, and Colin Jones. I know NN's James "Ninjames" Brady has a tingle down his leg for Spillman like he once did about Dominique Zeigler, but we can't overlook the fact that the three backup safeties have basically zero NFL experience at the position. The 49ers' benefited from fifth-