2009 DVOA Rankings: 49ers through Week 2

Fooch's Note: Florida Danny put this together but had to run out so I'm posting it for him. Thanks to Danny for getting this together. He'll touch it up later this afternoon/evening.

AUTHOR’S NOTE: For the Niners Nation newcomers – say that 5 times fast – I’m our resident stats guy. In general, my role on here is to write up Niner-related articles from a statistical perspective. Most of my posts come during the season, as I write up a statistical preview and review of each game. For the vets who are wondering, I haven’t done so thus far this season because the matchup stats aren’t all that meaningful yet (for the stat-savvy folks out there, I’m talking about inadequate sample size). Aside from these game previews/reviews, I post stat-based articles periodically when the specific situation warrants it. For instance, my first post ever on NN was a statistical indictment of Mike Nolan as it regarded the bullsh*t he was feeding us during his weekly press conference. So, if you like stats or, more importantly, if you’d like to augment your 49er fandom with some information that you might not get elsewhere, look out each week (beginning after the 3rd game) for my stat previews/reviews of each 49ers game, as well as my other periodic stats articles.

I know, I know. I haven’t posted in a really long time. It’s been a confluence of factors, of which many of you would be familiar: graduate school, full-time work, preparing for fantasy football, etc. It didn’t help that the offseason was pretty boring, and – oh yeah – the human anti-stat, Mike Nolan, is no longer the 49ers’ head coach (aka no longer my go-to post topic).

So, with that said, today we’re introducing a new weekly post. Basically, it’s going to be a power rankings post not unlike what you see around the internet. However, the twist will be that, rather than telling you where the 49ers rank according to the inner thoughts of Pete Prisco, I’ll be telling you where the 49ers rank on a whole host of objective measures published by Football Outsiders (FO).

For those who don’t know about FO, first things first, go check them out, especially if you’re into the statistical aspects of the NFL game. We’ve had many battles here on NN about the value of FO’s stats, mostly due to a misunderstanding of statistical interpretation in general. Neither FO nor I pretend that statistics can tell you everything there is to know about NFL football. They’re simply another piece of the puzzle.

The fact of the matter is that, unless you’re a die-hard "football stats are meaningless" kind of fan, then you’re already being inundated by, and using, NFL stats. So, if you accept that you’re going to pay attention to any NFL stats at all as part of being a fan, then the important thing is to pay attention to the most valid stats, i.e., the ones that give the best information.

This is where FO comes in. Put simply – and this is coming from a statistical measurement specialist not being paid by FO to hock their products – they put out the best NFL stats you can find in the public domain. Why? Because they take your basic NFL yardage stats, and adjust them for how difficult it was for the team/player to gain those yards on a given play. For instance, gaining yards against DET’s defense last year was just a tad easier than gaining yards against PIT’s defense. Similarly, it’s a lot easier to gain yards when you’re down by 30 than it is to gain yards when you’re down by 3, and it’s much more difficult to rush for 3 yards on 3rd and 2 than it is to rush for 3 yards on 3rd and 20. Even if you knew nothing about statistical methodology, you’d likely compare things like opponents, game situations, and downs/distances when deciding which team is better than another. So if we agree that these factors matter in ranking NFL teams/players, wouldn’t it be a better idea to rely on stats that incorporate them rather than on stats that totally ignore them? Obviously, this is a rhetorical question. The main idea here is that I use FO’s stats because the stats you find on NFL.com don’t account for any of the factors that impact the relatively difficulty of gaining a yard. In other words, NFL stats assume that every yard is created equally, which we know can’t be true, so that tends to make an ass out of you and me when we rely on them to argue with our friends about teams and players.

OK, so now you know what I’m going to be presenting in these weekly posts, and why I’m going to be presenting it. Therefore, without further ado, let the stat geekery begin!

After the jump, I’ll grab a pen from my pocket protector, and detail the 49ers’ statistical rankings through their first 2 games…

OVERALL RANKINGS

The primary statistic by which FO ranks NFL teams is defense-adjusted value over average (DVOA). In English, that name means FO looks at the yardage gained on every NFL play, adjusts it to account for the defense it was gained against (as well as other factors), determines whether that yardage represented a successful play, and then compares that success rate to the league average. Therefore, DVOA is expressed as a percentage, and is interpreted as offensive success rate per play. For offensive (OFF) DVOA stats, positive percentages are desirable because a high offensive success rate is a good thing. In contrast, negative percentages are desirable for defensive (DEF) DVOA stats because a high offensive success rate is a bad thing. Finally, although special teams (ST) DVOA is calculated a little differently (see here for more details), positive percentages are what we want to see. Below are the 49ers’ team DVOA rankings through Week 2 (bold = top 8 in the NFL; italics = bottom 8):

As you can see, the 49ers are the 18thbest team in the league overall according to FO’s rankings. With respect to interpretation, you should read the percentage in the "Total" column to mean that the 49ers are 1.0% more efficient overall than the average NFL team (FYI…TOT DVOA = OFF DVOA + ST DVOA – DEF DVOA).

Two points are worth making here. First, the Niners have the worst TOT DVOA among the NFL teams that are 2-0. The next worst 2-0 team is the Giants, who have a 14th-ranked 14.9% TOT DVOA. This shouldn’t be taken as a bad sign, however, because the DVOA stats above, along with all stats I’m going to present here before Week 4, have not been adjusted for opponents. This isn’t to say they’re meaningless, though, because the stats have been adjusted for all other important factors, including the ones I discussed earlier. So, for instance, the 2-0 Broncos’ 7th-ranked TOT DVOA of 28.8% is a mirage given that they’ve played CIN and CLE. It’s likely that their TOT DVOA will drop precipitously after the opponent adjustments are introduced after Week 3. Nevertheless, the fact that the 49ers are basically an average team thus far, despite playing two NFC West teams, isn’t exactly something we should be celebrating.

Second, and more optimistically, the Niners haven’t finished a season with a positive TOT DVOA since 2002, and have not ranked higher than 25th at the end of a season since 2003 (they were 18th). Thus, it seems the stats confirm what our eyes are seeing. Namely, that the 49ers are an improved team this season.

When looking at the 49ers rankings for the three team units, you see that the DEF and ST have been carrying them. The ST prowess is nothing new: SF finished 3rd in ST DVOA last year and 4th in 2007. The DEF, however, appears to have improved considerably (Remember: Negative percentages are good for DEF DVOA). Specifically, they’re 17.2% more efficient through Week 2 of 2009 than they were at the end of 2008 (7.3%; ranked 18th). Even more impressive is the fact that an above average DEF is something that hasn’t been seen by Niner fans since 2001, a season in which their DEF DVOA was -3.6%. And if you want to find a 49ers full-season DEF DVOA better than the one they have right now, you’d have to go back to 1997 (-25.8%).

The offense is another story altogether. Continuing the trend of recent 49er offenses, this year’s version is less efficient than the NFL average through 2 games. For the sake of comparison, the average NFL offense so far this season is that juggernaut in KC (-0.1%). Um yeah, wow, the Niners’ offense is 15% worse than the Chiefs’, who actually had Brady Croyle start 1 of their games. Soooo not awesome. Don’t fret, though. As you’ll see shortly, there are some specific aspects of the 49ers’ offense that are actually pretty good.

DOWN AND DISTANCE RANKINGS

Just a quick FYI for this section: "Short" means 0-3 yards, "Mid" means 4-6 yards, and "Long" means 7 or more yards. Also, because teams punt on 4th down the vast majority of the time, 3rd down DVOA stats actually include plays in which the team went for it on 4th down. Below are the 49ers’ down and distance DVOA rankings on both OFF and DEF (bold = top 8 in the NFL; italics = bottom 8):

Lots to talk about here. Through 2 games, it’s obvious where the 49ers’ OFF is committing an epic fail: 2nd down. Because their overall 2nd down DVOA is closer to their DVOAs for 2nd & Mid and 2nd & Long, we can infer that the Niners’ OFF is finding itself in a lot of these unfavorable 2nd down distances. However, that’s not to say that they’ve been inefficient on 1st down. Rather, it just means that the 49ers’ OFF on 1st down has been kind of boom or bust. Basically, they either pick up another 1st down or they go nowhere and end up in something like 2nd and 7. Now, you might think a 3-yard gain on 1st down is efficient. NFL history says otherwise. That’s because, in coming up with their success criteria for each down, FO used a long-since-discovered statistical fact that 5 yards (not 7) is the 2nd down distance at which OFF’s have a better-than-50% chance of picking up a 1st down. In other words, on 2^nd & 7, the OFF is at a tactical disadvantage vis-à-vis the DEF.

On 3rd down, the Niners’ OFF isn’t doing all that bad when you consider most teams, regardless of the opponent, are pretty bad on 3rd and Long. Also, as you’ll see in the next section, one aspect of the offense is actually pretty darn good on 3rd down, whereas the other is horrifically bad. Feel free to take a guess.

As far as the DEF goes, they’re in the top half of the NFL on each down. What’s interesting, though, is how there’s a kind of snowball effect for DEF success on each subsequent down. Specifically, the Niners’ DEF seems to be putting teams in a lot of 2nd & Long holes given that their overall 2nd down DVOA is closest to this split. In other words, much of the DEF’s 1st down success translates into further success on 2nd & Long. Similarly, their success on 2nd and Long creates the very advantageous situation in which the 49ers’ DEF can attack on 3rd and 4+ yards. (i.e., 3rd & Mid or 3rd & Long).

One more thing that I have to mention here is how much of an improvement the Niners’ DEF is showing on 3rd down over last year’s unit. Specifically, the 2008 DEF had an 11.2% overall 3rd Down DVOA (ranked 21st), and was ranked 30th in the NFL with a 75.7% DVOA on 3rd and Long. You want to know how to lose games, demoralize fans, and tire out your DEF players? Coach 3rd down defense like Mike Nolan. (Aside: Yes, that was my one shot at Nolan for the day.) But why the improvement on 3rd & Long? Must be the pass DEF right? Was that Nolan’s fault too? Answer in a moment.

DOWN AND TYPE-OF-PLAY RANKINGS

Here they are (Same thing applies as before about 3^rd/4^th Down):

For the sake of comparison, which will become relevant in a second, here again are the overall down-by-down DVOAs from the previous table:

As they relate to the OFF’s ineptitude on 2nd down, we can see from the first table that the stats indicate inefficiency in both the run and the pass. However, as the Overall 2nd Down DVOA is closer to the 2nd Down Pass DVOA, we can infer that the pass OFF is more to blame. This makes sense given what I said in the last section about the OFF finding itself in a lot of Mid to Long situations on 2nd down.

For 3rd down, I asked you to guess which part of the OFF is better. Based on the first table, the correct answer – surprisingly – is the pass OFF. It turns out that Shaun Hill and the pass OFF are over 50% better than the league average on 3rd down, while Frank Gore and the run OFF are over 125% worse than average. What’s even more remarkable is that this is a total flip-flop from last season, when the Niners’ 3rd down pass OFF ranked 23rd at -6.1% while their 3rd down run OFF ranked 13th at 23.5%. Keep in mind that FO’s success threshold for 3rd (and 4th) down is getting a 1st down because, obviously, anything less than a 1st down means punting or turning the ball over on downs. So, with that in mind, what we basically have through 2 games this season is an OFF that is actually good at passing its way out of 3rd down situations. I’d say that’s a promising development, and a strange one given the philosophical differences between Jimmy Raye and Mike Martz. Perhaps it’s the lack of Martz’s signature 15-step drops on 3rd & Long that’s making the difference here? Time will tell.

Speaking of 3rd down passing, and going back to the question I posed in the previous section, yes, the 49ers’ pass DEF on 3rd down is the reason they’ve improved so much overall on 3rd down (at least through 2 games). As the first table shows, the Niners’ 3rd Down Pass DEF DVOA is over 30% better than the NFL average, which puts it in the top quartile of the league. Keep in mind that this has come against two teams, ARI and SEA, who are much better at passing (ARI pass DVOA = 39.5%; SEA = 21.7%) than running (ARI run DVOA = -31.1%; SEA = -3.0%) on offense. Therefore, the Niners’ 3rd Down pass DEF DVOA is most likely to get even better once FO incorporates opponent adjustments in a couple of weeks.

Deeper question: Why has the pass DEF improved on 3rd down? I offer two explanations. First, Nolan’s hybrid vanilla cow pie defense has been replaced by the more aggressive attack-style defense preferred by Manusky and Singletary. Aggression on 3rd down passes is a good thing, just ask the Steelers. Second, the passive pass-defending duo of Walt Harris and Mark Roman has been replaced by the more aggressive Shawntae Spencer and Dashon Goldson. Combine these 2 players with the hyperagressive Nate Clements, and you have the makings of something the 49ers haven’t had in a while: a turnover-creating secondary. So, essentially, what you have here is a situation whereby Singletary’s personnel, scheme, and philosophy changes are being vindicated by the statistics – through 2 games at least. It’s hard to argue against Singletary’s changes when they’ve immediately resulted in a 135% improvement over last season in 3rd & Long Pass DEF DVOA.

One last thing I’ll say about the down-by-down DVOAs. Except for 2nd Down OFF, the 49ers are exhibiting a situational pattern on both sides of the ball that is exactly what we want to see. Namely, as 1st and 2nd down tend to favor run-heavy play-calling, we’d like to see the Niners’ OFF and DEF having success in the run game on these downs. Likewise, as 3rd down tends to favor pass-heavy play-calling, we’d like to see the two units having success in the passing game on 3rd down. Well, that’s exactly what we see from the Niners through 2 games. They’ve been good in the run game when running plays are more likely, and they’ve been good in the pass game when passing plays are more likely. That, my fellow Niner fans, is called situational efficiency.

FIELD ZONE RANKINGS

For the OFF rankings in the table below, "Deep" means inside SF’s 20 yard line, "Back" means between SF’s 20 and 39, "Midfield" means between SF’s 40 and their opponents’ 40, "Front" means between their opponents’ 39 and 20, "Red Zone" means inside their opponents’ 20, and "Goal-to-Go" means inside their opponents’ 10. These are reversed for DEF rankings such that, for example, "Red Zone" means the opponent has the ball inside SF’s 20 yard line. Here are the rankings:

From this table, you can see that both sides of the ball are having their biggest successes inside their opponents’ 10 yard line: The OFF’s best Field Zone DVOA has come in goal-to-go situations, while the DEF’s best Field Zone DVOA has come when the opponent is pinned deep in their own territory. Obviously, this has directly impacted the 49ers’ 2-0 start in two ways. First, the OFF hasn’t wasted their scoring opportunities, and, after all, the object of the game is to outscore your opponent. Second, the DEF has helped SF maintain a consistent field position advantage by not letting opponents move out of the shadow of their own goalline.

SPECIAL TEAMS RANKINGS

The way FO comes up with their ST DVOAs is different from their methods for OFF and DEF DVOA. It’s pretty intuitive though. Basically, the way it works is that each yard line is associated with a different probability of scoring points. For example, a team is more likely to score a TD when starting from their opponents’ 20 than from their own 20. Similarly, a 30-yard FG attempt is more likely to be made than a 50-yard FG attempt. For kicks and punts, efficiency is determined by the expected point difference between the yard line from which the ball is kicked and the yard line to which the ball is returned. For example, the highest efficiency punt would be one that’s kicked from your own 1 yard line (i.e., a place where the opponent is most likely to score) to your opponent’s 1 yard line (i.e., a place where the opponent is least likely to score). For returns, efficiency is determined in basically the same way, except in reverse. For FGs, efficiency depends on the length of the kick.

So, when you read the table below, understand that each stat represents the number of expected points of field position each special teams unit creates compared to the NFL average for similar kicks, returns, or FGs. Therefore, in this interpretation, positive numbers are good. For instance, the 49ers’ FG stat of 2.4 net points means that Joe Nedney has given the Niners 2.4 points more than what the average NFL team would expect to score at the distances from which he’s kicked his FGs. If you want more details about ST DVOA, see here. And below are the ST DVOA stats:

As you can see from the table, Nedney has been a beast, the punt team has been above average, and the return teams need improvement. These are pretty much the conclusions you would reach just watching the first 2 games. When Nedney and Andy Lee are on the field, I’m confident. When Allen Rossum is on the field, I’m taking a bathroom break because either (a) he’s not returning it very far, or (b) he’s going to hand it off to Clements for a 10-yard loss.

BOTTOM LINE

OK, so through 2 games, we can draw the following conclusions about the 2009 49ers based on their team statistics thus far:

1. They’re an average team that is currently being carried by its DEF and ST.

2. In stark contrast to the Nolan Era, they’ve become a very situationally effective team. Their statistical strengths so far seem to be 3rd down pass OFF, 3rd down pass DEF, goal-to-go OFF, long-field DEF, and field position. It’s really important to understand that these are the game situations that correlate the most with winning. It’s no wonder, then, that the Niners are 2-0.

3. Their statistical weaknesses so far are 2nd down OFF and returning kicks/punts. These are relatively minor things, so it’s actually nice to finally engage in nitpicking when it comes to the Niners.

Remember, though, the stats I’ve presented in this article have not been adjusted for opponent. We’ll have a much better idea about how efficiently the 49ers have been playing once FO starts incorporating opponent DVOA adjustments in a couple of weeks. Nevertheless, based on the stats, it’s a lot more fun to be a Niner fan these days.

**DVOA statistics used to produce this article were obtained from Football Outsiders.