The 49ers and DVOA: You might want to shield your eyes
EDITOR'S NOTE: For a whole now, I've been looking for somebody to do a more detailed statistical look at the 49ers. After his piece on Mike Nolan's denial issues, I think I've found our guy in Florida Danny. So, going forward, he'll be providing a statistical look at the 49ers matchup each week. Whether you're a reader of Football Outsiders or not, these analyses will provide an alternative look at what might turn out to be the key in any given game. Since the Giants game is fast approaching, he emailed me his piece and I'll post it under my name. Beginning next week, he'll be posting under his name on the front page. Also, if you like what Danny has to say in this article, make sure and Buzz and Digg it up to get it.
__________________________________________________
I have to admit. In my gut, this week’s game against the Giants has the feel of a blowout loss for our 49ers. In fact, the more and more I’ve researched this matchup, the more and more my gut has felt like I ate bad sushi. What do I mean? Well, when I reminisced not-so fondly about blowout losses of Nolan-era past, a few games immediately came to mind: 41-10 at Chicago in 2006, 42-3 at Philadelphia in 2005, and 52-17 at Washington in 2005. What immediately jumped out about those three games was that they were road games east of the Mississippi River. So that made me wonder, “What is Mike Nolan’s record east of the Mississippi River?” Well, for your vomiting pleasure, below is every 49er game played east of the Mississippi during the Nolan era (Note: New Orleans, St. Louis, and Kansas City are omitted because I’m not going to split hairs about being five feet east or five feet west):
2007 @ Cleveland – L, 7-20 (-43.7%)
2007 @ Carolina – L, 14-31 (-21.1%)
2007 @ Atlanta – L, 16-20 (-14.8%)
2007 @ New York Giants – L, 15-33 (-37.5%)
2007 @ Pittsburgh – L, 16-37 (-55.2%)
2007 – 0-5, average score of 13.6-28.2 (-34.5%)
2006 @ Detroit – W, 19-13 (+7.5%)
2006 @ Chicago – L, 10-41 (-43.8%)
2006 – 1-1, average score of 14.5-27.0 (-36.3%)
2005 @ Jacksonville – L, 9-10 (-73.6%)
2005 @ Tennessee – L, 22-33 (-38.0%)
2005 @ Chicago – L, 9-17 (-54.3%)
2005 @ Washington – L, 17-52 (-77.5%)
2005 @ Philadelphia – L, 3-42 (-51.9%)
2005 – 0-5, average score of 12.0-30.8 (-59.1%)
2005-2007 – 1-11, average score of 13.1-29.1 (-42.0%)
You read that right. In games east of the Mississippi the last three seasons, Nolan’s 49ers were 1-11, with an average loss of 16 points. Also notice that they gave up 30+ points seven times. That’s not a good sign given what the last 3 weeks have looked like in terms of scoring defense. However, to really get a good understanding of what the above stats mean for this week’s game against the Giants, we have to take a look at the numbers in parentheses, which equal the difference in defense-adjusted value over average, or DVOA, between the 49ers and their opponent during that season.
After the jump, Danny looks at some of the past East-of-the-Mississippi matchups and corresponding DVOAs. He wraps it up with a prediction for this game. It's like watching a car crash. It's a bad thing but you can't help but look...

Fooch eloquently introduced DVOA to Niners Nation in a post last week about how the 49ers rush defense is the driving force behind their overall defensive woes this season. Although it’s a highly sophisticated statistic that requires a huge effort to calculate—and kudos to Football Outsiders for putting forth that effort—the beauty of DVOA is that it’s meaning is relatively easy to understand. Basically, DVOA tells you whether a team performs above or below the league average on a play-by-play basis: positive values mean above average situational performance overall and on offense, while negative values mean above average situational performance on defensive. For example, the 49ers overall DVOA through Week 6, -8.1%, tells you that they are 8.1% worse than average. Now, let’s go back to the stats.
One thing you’ll notice about the DVOA differences listed above is that the 49ers beat the only east-of-the-Mississippi road opponent that they were better than (Detroit in 2006), and lost to the 11 opponents that were better than them. How does that relate to this week’s game? Well, the Giants current overall DVOA, +34.6%, makes them the 2nd best team in the league, while the 49ers overall DVOA, -8.1%, earns them the honor of 23rd place in a 32-team league. That’s pretty solid evidence in favor of, at the very least, a 49er loss this week. It gets even worse, though, if you take into account how the actual difference between their DVOAs. Simply subtracting the two values shows that the 49ers are 42.7% worse than the Giants! Aside from this being an awful DVOA matchup for the 49ers in general, the fact that it is one being played east of the Mississippi makes it even more ripe for a blowout. Below I’ve ranked the previous game results in decreasing order of DVOA difference:
2005 @ Washington – L, 17-52 (-77.5%)
2005 @ Jacksonville – L, 9-10 (-73.6%)
2007 @ Pittsburgh – L, 16-37 (-55.2%)
2005 @ Chicago – L, 9-17 (-54.3%)
2005 @ Philadelphia – L, 3-42 (-51.9%)
2006 @ Chicago – L, 10-41 (-43.8%)
2007 @ Cleveland – L, 7-20 (-43.7%)
2005 @ Tennessee – L, 22-33 (-38.0%)
2007 @ New York Giants – L, 15-33 (-37.5%)
2007 @ Carolina – L, 14-31 (-21.1%)
2007 @ Atlanta – L, 16-20 (-14.8%)
2006 @ Detroit – W, 19-13 (+7.5%)
If you look at the four games that immediately surround this week’s 42.7% DVOA difference, the 49ers were 0-4, losing by an average score of 31.8-11.0. Clearly, these stats do not bode well for our beloved 49ers in their game against the Giants this week.
In addition to where the game was played, another interesting trend related to east-of-the-Mississippi DVOA differences has to do with when the game was played. In the first list I presented, the games for each season are in reverse-week order: from latest in the season to earliest in the season. Take a look at the first game the 49ers played east of the Mississippi each year. In these games, the 49ers lost by 39 at Philadelphia in 2005, by 31 at Chicago in 2006, and by 21 at Pittsburgh in 2007. Against three teams with, on average, a 50.3% DVOA advantage, an advantage similar to the Giants’ advantage this week, the 49ers lost by an average score of 40.0-9.7. Now look at the games in the list that were played this season. Don’t see any? That’s because this week’s game against the Giants is the first one of the season. Uh oh!
Finally, the eternal optimists out there might say, “Well, maybe this season the Giants aren’t as good at home as they are overall, and the 49ers aren’t as bad on the road as they are overall.” Aside from the fact that you eternal optimists have to know the 49ers suck on the road, here are the stats. The Giants’ overall DVOA at home is +41.4%, or 6.8% better than their overall all-venue DVOA. In contrast, the 49ers’ overall DVOA on the road is -16.2%, or -8.1% worse than their overall-venue DVOA. So basically, the Giants are even better at home, and the 49ers are even worse on the road. Again, it’s not looking good.
To recap, the relevant statistics for this week’s east-of-the-Mississippi matchup are as follows:
(1) The 49ers have gone 1-11, losing by an average of 16 points, in such games under Nolan.
(2) The 49ers are 0-11 in such games when they are at a DVOA disadvantage.
(3) The 49ers are 0-4, losing by an average of 20 points, in such games under Nolan when they are at a DVOA disadvantage similar to the one they encounter this week.
(4) The 49ers are 0-3, losing by an average of 30 points, in the first of such games each season under Nolan.
Given the above, here’s my prediction for this week’s game:
| San Francisco 49ers | 13 |
| New York Giants | 37 |
Fun Coaching Numbers of the Week
Mike Nolan’s success rate on replay challenges: 40.0%
Tom Coughlin’s success rate on replay challenges: 40.0%
Number of timeouts called by Mike Nolan outside the last five minutes of a half: 14 in 6 games (2.3/G).
Number of timeouts called by Tom Coughlin outside the last five minutes of a half: 8 in 5 games (1.6/G).
Comments
The perfect time
At the beginning of the year I said that either the Niners were going to get better or the Yorks would finally fire Nolan.
Here’s the perfect opportunity. If the Niners can possibly go into the Meadowlands and take a game then they’re 3-4 and facing a team at home that they’ve already beaten on the road. They could be 4-4 going into the bye.
If, as many here suspect (and given the statistical breakdown, who wouldn’t), the red and gold get creamed back East on Sunday morning before we here on the West Coast break out the dip and chips, then Monday would be the perfect day to replace Nolan. There’s little reason to wait for the bye.
I heard the replay of most of Mike Martz’s interview on KNBR this a.m. I really don’t know if his offense is the answer here or if he’s the permanent answer for head coach, but he does a better interview than Mike Nolan’s miles of denials. I’d give him the nod to finish out the season. Just my very humble opinion.
by Bob On The Coast on
Oct 17, 2008 6:59 AM PDT
reply
actions
0 recs
martz interview
I missed that interview. If you recall much of it, would you be able to throw up a FanPost about it?
Niners Nation - The premier 49ers blog on the Internet!
by Fooch on
Oct 17, 2008 7:39 AM PDT
up
reply
actions
0 recs
Buzz/Digg
Just a reminder to Buzz and Digg this story if you particularly enjoy it.
Niners Nation - The premier 49ers blog on the Internet!
by Fooch on
Oct 17, 2008 7:47 AM PDT
reply
actions
0 recs
sad
well thats sad…this morning listening to Ronnie Lott he got me all jacked up for an upset but this def brought me down.
DVOA sounds interesting but i like the simple stats example:
if gore touches ball 24+ times were 2-0
if gore touches ball 24- times were 0-4
by cazzuno on
Oct 17, 2008 9:44 AM PDT
reply
actions
0 recs
actually...
it’s funny that you should bring that up…
football outsiders’ first ever stat analyses showed that the “team x is w-l when running back y touches the ball z+ times” mantra is a myth because it’s got the chicken and egg thing wrong. running back y getting z+ touches during a game team x wins is just a byproduct of the fact that teams run the ball when they’re ahead and pass the ball when they’re behind. it’s as simple as that. what matters more than the number of running back touches is WHEN those touches occur. for a perfect example of this, you need not look any farther than last week’s eagles game. rather than giving gore the ball to preserve the lead, martz went against time-tested, clock-killing, lead-preserving play-calling. what happened? they didn’t win and gore didn’t reach your magic number. basically, it’s more important that gore is getting touches at the end of a game than at the beginning of a game because it means that the niners are ahead in the 4th quarter, which is a pretty solid predictor of actually winning the game. the 24+ carries thing will take car of itself.
here’s the link to football outsiders’ analysis:
http://www.footballoutsiders.com/stat-analysis/establishment-clause
by Florida Danny on
Oct 17, 2008 11:02 AM PDT
up
reply
actions
0 recs
THANK YOU OH MY GOD THANK YOU I CAN ONLY EXPLAIN THIS SO MUCH TO PEOPLE.
My Dave Righetti is better than your Dave Righetti.
by howtheyscored on
Oct 17, 2008 12:20 PM PDT
up
reply
actions
0 recs
understood, it was more of a joke but i agree with the whole if your up you run if your down you throw thing
point was more that gore through 3 quarters had what 20 carries for about 100 yds and the niners were winning
thx anyway
by cazzuno on
Oct 17, 2008 1:48 PM PDT
up
reply
actions
0 recs
yeah...
i agree with that last part…that’s what made it so frustrating to watch…it was almost hostler-esque…dude’s half of your offense through 3 quarters, and he’s running at a 5 yard-per-carry clip, and you stop giving him the ball when that’s the time in the game that you’re supposed to give it to him? it was just mindboggling.
thanks for commenting.
by Florida Danny on
Oct 17, 2008 2:03 PM PDT
up
reply
actions
0 recs
Haha, sorry for coming off like a jerk… there are just phrases that flip switches in my brain that make me start thinking in all caps.
I agree with you about last week. It was awful. It was stupid. it was damn near unforgivable that Gore didn’t end up with at least 25 carries with the way he was eating yards and time on the ground. 25 carries wouldn’t have put us in a lock for the win, but they certainly would have been a far more effective way to preserve, and even extend, the lead that we did have.
My Dave Righetti is better than your Dave Righetti.
by howtheyscored on
Oct 17, 2008 2:44 PM PDT
up
reply
actions
0 recs
Great analysis...
While this is all very interesting, I can give a more simplistic version of why the 49ers are going to lose this game:
Head Coach
———————————
San Francisco: Mike Nolan
New York: Not Mike Nolan
-- Jay T.
by jaytierney on
Oct 18, 2008 8:57 AM PDT
reply
actions
0 recs
agreed...
…look for Mike Nolan to waste at least 2 timeouts before there are 5 minutes left in the half/game.
by Florida Danny on
Oct 18, 2008 10:22 AM PDT
reply
actions
0 recs






















