Football Outsiders and the 49ers: Team Stats through Week 4
AUTHOR'S NOTE: If you're interested in a discussion of the 49ers' #1 ranking in Red Zone Pass Offense DVOA, mosey on over to Fooch's post from earlier today. In my post, I'm just rubbernecking it. His post stops and stares.
Welcome back for this week's look at how the 49ers' team stats rank in the NFL according to Football Outsiders (FO). Now that the season is 25% done, there are 2 changes that affect the stats and rankings in this post:
- All FO stats are now adjusted for opponent. So beginning this week, and for the rest of the season, all of the 49ers' overall and situational stats should be interpreted as if they've played a league-average schedule. In other words, FO has removed the effect that playing good teams (e.g., the Vikings) or bad teams (e.g., the Rams) has had on the 49ers' overall and situational performance.
- I've included two new sets of stats because their sample sizes are now large enough for them to actually give us valid information: (a) DVOA by quarters and halves, and (b) DVOA in the shotgun formation.
OK, so with that housekeeping taken care of, it's onward ho!
TEAM RANKINGS
Below are the 49ers' overall team DVOAs and rankings (See here for my explanation of FO team stats). Based on the incorporation of opponent adjustments, I've added three new stats to the table:
- VOA, which is DVOA without the opponent adjustments. I present this so we can see the effect that the 49ers' schedule has had on their performance.
- SOS, which is strength of schedule according to DVOA. I present this for the same reason as VOA.
- Variance, which is the consistency of DVOA from week to week. I present this to show how consistently (or inconsistently) the 49ers have performed from week to week. Lower percentages mean more consistency, and teams are ranked from most to least consistent.
Here's the new table (bold = top 8 in the NFL; italics = bottom 8):
|
Total |
Rank |
Offense |
Rank |
Defense |
Rank |
Special Teams |
Rank |
|
1.4% |
15 |
-13.0% |
24 |
-16.6% |
6 |
-2.1% |
16 |
|
Total VOA |
Rank |
SOS |
Rank |
Variance |
Rank |
|
|
|
4.7% |
17 |
-7.6% |
24 |
8.0% |
11 |
|
|
After the jump, I'll (a) discuss this table; (b) break down the offensive, defensive, and special teams rankings; and (c) unveil a new feature that pays homage to the world's most famous motivational speaker (no, I don't mean Mike Singletary)...
Last week, the Niners were 22nd in Total (TOT) DVOA, 27th in Offense (OFF) DVOA, 10th in Defense (DEF) DVOA, and 20th in Special Teams (ST) DVOA. So, essentially, the STL game and the incorporation of opponent adjustments resulted in improvements across the board. Shutting out a team 35-0 will do that, though, regardless of how bad that team is. Two other things I'll mention about the 49ers' TOT DVOA are that they're ranked 6th out of the 8 one-loss teams, and that their opponent this week, the Falcons, is the team immediately ahead of them in the rankings (3.3%, 14th).
Going a little deeper into the effect of opponent adjustments, the 49ers' 24th-ranked SOS and the fact that their VOA is better than their DVOA both suggest that they're not quite as good as what we've been seing from them through 4 games. This isn't to say that they're not good. On the contrary, they're a slightly above average team regardless of their opponents thus far. I simply bring this up to make the point that they have benefitted to some degree from an easier-than-average schedule.
Finally, from the 49ers' Variance statistic, it looks like they're one of the more consistent teams in the league from week to week. What's important to realize, however, is the difference between this year's Variance and those of previous seasons. Specifically, during the Mike Nolan Era, they went from 32nd in 2005 to 23rd in 2006 to 15th in 2007 to 2nd in 2008. In other words, they went from bad and inconsistent to just plain bad, which is pretty much the normal trajectory of a rebuilding team - except for the "still bad after all these years" part. This season, though, they're good and consistent. Now that's more like it. Thank you, Mike Singletary.
OFFENSIVE RANKINGS - OVERALL
Here' how the offensive rankings look overall and by type of play (bold = top 8 in the NFL; italics = bottom 8):
|
OFF |
Rank |
Pass |
Rank |
Run |
Rank |
Variance |
Rank |
|
-13.0% |
24 |
6.7% |
19 |
-24.7% |
30 |
2.8% |
9 |
|
OFF VOA |
Rank |
Pass VOA |
Rank |
Run VOA |
Rank |
|
|
|
-10.4% |
25 |
11.7% |
19 |
-24.2% |
30 |
|
|
From the above breakdown, we can draw several conclusions:
- Overall, the OFF has been pretty inefficient, but very consistently so (sounds Nolan-esque to me). The Niners' Variance statistic suggests that their OFF has generally fluctuated between 20% and 5% below average from week to week this season.
- The running game is definitely anchoring the OFF, but, by "anchoring," I mean holding it down. Whereas the pass OFF is actually above average compared to the rest of the league, the run OFF is 3rd-worst in the league. Like I said last week, if you're going to have a run-first mentality on OFF, at least be good at it.
- Whereas the passing game has benefitted from going up against weaker pass DEFs, the running game has been horrible regardless of the opponent. Indeed, except for their game against the Vikings, the Niners have played the 18th, 22nd, and 28th ranked teams in Pass DEF VOA (average = 23.4%). In contrast, they've played two top-6 run DEFs (i.e., the Cardinals & MIN; average = -28.5%), but also two bottom-11 run DEFs (i.e., the Seahawks & Rams; average = 2.5%). So, whereas the OFF's best 3 passing games have come against bad pass DEFs, their run OFF has been really good against one bad run DEF (i.e., SEA), but also really bad against another bad run DEF (i.e., STL); in addition to being understandably bad against 2 good run DEFs.
OFFENSIVE RANKINGS - SITUATIONAL SPLITS
So we know that the pass OFF has been good and the run OFF has been really bad. Maybe the situational stats have something to say about, "Why?"
Here are the OFF's pass and run DVOAs by down:
|
Down |
OFF |
Rank |
Pass OFF |
Rank |
Run OFF |
Rank |
|
1st |
0.4% |
19 |
5.8% |
21 |
1.2% |
14 |
|
2nd |
-42.9% |
31 |
-33.5% |
27 |
-41.4% |
30 |
|
3rd |
6.6% |
17 |
42.6% |
10 |
-101.8% |
31 |
As this table shows, the pass OFF plays efficiently on 1st and - especially -- 3rd downs, but sucks on 2nd down. In contrast, the run OFF gets progressively worse with each subsequent down, culminating in that horrendously bad 3rd Down Run OFF DVOA. I have a strange feeling that some of you are going to use these 3rd-down splits to reiterate your displeasure with Jimmy Raye's call for a running play on 3rd & 6 @ MIN.
But, of course, you might say that distances matter. Here are the OFF's down splits by distance (Short = 0-3 yards; Mid = 4-6 yards; Long = 7+ yards):
|
Down |
OFF |
Rank |
Short |
Rank |
Mid |
Rank |
Long |
Rank |
|
1st |
0.4% |
19 |
N/A |
|||||
|
2nd |
-42.9% |
31 |
-33.8% |
28 |
-29.6% |
22 |
-52.0% |
31 |
|
3rd |
6.6% |
17 |
1.9% |
18 |
2.7% |
13 |
14.7% |
21 |
Based on this table, as well as the previous one, it's apparent that the 49ers throw the ball a lot more on 3rd down than they run it, and that they're not encountering many 3rd & short situations. I don't need to consult the play-by-play to figure this out. Rather, if the 49ers were running the ball a lot on 3rd down, their overall 3rd-down split wouldn't be above average because that -101.8% 3rd Down Run OFF DVOA would be spread like a cancer over more plays. Similarly, if the 49ers were encountering a lot of 3rd & short situations, their 3rd-down split wouldn't be above average because 3rd & short is a high-frequency running down, again meaning that their -101.8% 3rd Down Run OFF DVOA would infect their 3rd down split more. So, all in all, just pray the Niners continue to throw the ball on 3rd down.
Another argument for more passing comes from the formation splits. Here's a table showing the Niners' OFF DVOAs based on whether they're in the shotgun formation or not:
|
% of Plays in Shotgun |
Rank |
Shotgun DVOA |
Rank |
Non-Shotgun DVOA |
Rank |
Difference |
Rank |
|
25.8% |
29 |
19.8% |
16 |
-21.2% |
29 |
41.0% |
10 |
Shaun Hill takes a snap in shotgun once every 4 plays, a ratio that's 4th-lowest in the league. But notice, the Niners' OFF is over 40% more efficient when Hill is in shotgun than when he's not; a difference in efficiency that ranks in the top 10. For sure, part of this has to do with FO's general finding that OFFs are more efficient in shotgun formations. Indeed, only 10 OFFs in the league right now are better when they're not in shotgun. However, the average difference in the NFL right now is 18.3%, meaning that shotgun plays benefit the 49er OFF over twice as much as they do for the average NFL team.
Now, I'm certainly not saying the Niners need to become the Patriots or anything (i.e., go shotgun almost 60% of the time), but these stats don't lie. Working in a few more plays per game from the shotgun would most likely help the OFF play more efficiently; especially given the fact that the 49ers' OL can't seem to pass block worth a damn (I'll get to that in my Saturday player rankings post).
OFFENSIVE RANKINGS - "TONY ROBBINS" SPLITS
Two sets of situational splits provided by FO are something akin to psychological indicators: field zone DVOAs and quarter/half DVOAs. Kind of like what hustle stats tell you about NBA teams, these stats tell you whether NFL teams are playing motivated football. Besides physically assessing the 49ers' motivation during games, what better (and easier) way to see if their collective heads are in the game than looking at whether they play smart football inside the 20s, come out of the locker room hot, and finish games well. After all, do you think it's any coincidence that motivational guru Mike Singletary includes "create great field position," "total ball security," "execute," and "finish" in his formula for success? A team that doesn't play well inside the 20s is not creating great field position. When is total ball security most important? Inside the 20s. What does coming out of the locker room flat look like? Poor execution in the 1st quarter. And finally, what is finishing if not performing well in the 4th quarter? Oh, and if that didn't convince you, then also know that FO has found red zone and 1st quarter efficiency to be 2 of the 3 stats most highly correlated with winning.
So, rather than bombarding you (aka making you even dizzier) with all field zone and quarter/half DVOAs, I'll just give you what I'll call FO's "Tony Robbins" DVOA splits (TR splits for short). Here they are for the 49ers' OFF:
|
Inside SF 20 |
Rank |
Red Zone |
Rank |
1st Quarter |
Rank |
4th Quarter |
Rank |
|
-41.1% |
28 |
39.6% |
5 |
-29.9% |
23 |
15.3% |
16 |
Clearly, we can see why Singletary believes his OFF has a long way to go as it relates to his formula for success. Based on these stats, the OFF is great in one TR split, middling in another, below average in another, and horrible in another. Taken together, that's pretty average. However, viewed from an optimist's perspective - I mean, we're talking about Tony Robbins here after all - these stats suggest that the sky's the limit for this team if the OFF ever gets their collective heads in the game.
DEFENSIVE RANKINGS - OVERALL
Here' how the defensive rankings look overall and by type of play (bold = top 8 in the NFL; italics = bottom 8):
|
DEF |
Rank |
Pass |
Rank |
Run |
Rank |
Variance |
Rank |
|
-16.6% |
6 |
-5.0% |
9 |
-35.9% |
1 |
4.0% |
19 |
|
DEF VOA |
Rank |
Pass VOA |
Rank |
Run VOA |
Rank |
|
|
|
-16.2% |
7 |
-1.4% |
10 |
-40.7% |
1 |
|
|
All I can say is, "ZOMG!!!!!! NINERZZ D ROOLZZ!!!" That's right, the San Francisco 49ers have the most efficient run DEF in the NFL right now. Consider the Nolan Era dead and buried (thankfully). Also consider the fact that none of the DVOAs in the table seem to have been affected by SOS, i.e., their DVOAs are not that much different from the non-opponent-adjusted VOAs. Finally, I should point out that, according to official NFL stats (i.e., yards allowed per game), the run DEF is ranked 4th. So thank FO's superior stat methods for allowing us to bask in the glory of a "We're #1" ranking.
DEFENSIVE RANKINGS - SITUATIONAL SPLITS
So we know that the DEF is pretty awesome all-around. But how awesome (or not) are they on specific downs?
Here are the DEF's pass and run DVOAs by down:
|
Down |
DEF |
Rank |
Pass DEF |
Rank |
Run DEF |
Rank |
|
1st |
-16.2% |
5 |
4.8% |
12 |
-40.8% |
2 |
|
2nd |
-29.2% |
5 |
-20.2% |
6 |
-40.2% |
4 |
|
3rd |
-2.4% |
13 |
-3.6% |
11 |
4.4% |
21 |
I'd say that's pretty awesome. What I like most about the stats in this table is that the Niners' DEF is great both on high-frequency running downs (i.e., 1st & 2nd), as well as high-frequency passing downs (2nd & 3rd). To me, the fact that they're ranked 21st in defending 3rd-down runs is totally inconsequential. Teams pass the ball the vast majority of the time on 3rd down, so I'd rather they be above average against the pass - which they are - than above average against the run - which they're not. Basically, what you see above is a near-spotless record of situational efficiency.
Here are the DEF's down splits by distance (Short = 0-3 yards; Mid = 4-6 yards; Long = 7+ yards):
|
Down |
DEF |
Rank |
Short |
Rank |
Mid |
Rank |
Long |
Rank |
|
1st |
-16.2% |
5 |
N/A |
|||||
|
2nd |
-29.2% |
5 |
0.7% |
19 |
94.1% |
31 |
-70.7% |
1 |
|
3rd |
-2.4% |
13 |
16.6% |
20 |
-37.3% |
12 |
10.8% |
16 |
There are two take-home messages from these stats. First, the most important (only?) area where the DEF seems to need improvement is on 3rd & short. Giving up 1st downs on 3rd & short can lead to a downward spiral in hurry. I'm not too worried about this, though, because of Singletary's "no downward spirals" coaching style, and the fact that their 1st and 2nd down performances seem to be creating very few 3rd & short situations.
Second, I've never seen one yard mean so much to a down-and-distance split like it does to the Niners' DEF on 2nd down. Specifically, they're 164.8% better (!!!) when their opponent has 7+ yards to go on 2nd down than when their opponent has 4-6 yards to go. Their overall prowess on DEF seems to hinge on stopping 1st down plays for minimal gain, thereby putting the opposing OFF in a 2nd & long situation. Therefore, in order for the DEF to keep up their recent domination, they're going to have to continue to be stout on 1st down.
Another indicator of their across-the-board awesomeness is the DEF's shotgun vs. non-shotgun splits. Here they are:
|
% of Plays in Shotgun |
Rank |
Shotgun DVOA |
Rank |
Non-Shotgun DVOA |
Rank |
Difference |
Rank |
|
42.6% |
10 |
-16.3% |
5 |
-16.8% |
7 |
0.5% |
7 |
The DEF has been facing a fair amount of shotgun so far this season, which is to be expected given that (a) shotgun is essentially ARI's base offense, and (b) teams like STL tend to be in shotgun a lot when they're on the wrong end of a blowout. As you can see though, facing more shotgun than normal hasn't affected the DEF's DVOA. Specifically, they're in the top quarter of the league regardless of whether the opposing OFF is in shotgun or not, and their DVOA difference is infinitesimally small.
DEFENSIVE RANKINGS - TR SPLITS
We saw earlier that the OFF needs to get their head in the game a little bit more. Now it's time to check out the DEF's TR splits:
|
Red Zone |
Rank |
Inside OPP 20 |
Rank |
1st Quarter |
Rank |
4th Quarter |
Rank |
|
8.2% |
20 |
-45.8% |
5 |
-23.2% |
6 |
9.2% |
19 |
If the OFF was 1/4 of the way to Singletary's motivation destination, then the DEF is twice as far along in the journey. Namely, they come out of the locker room on fire (mabe Singletary learned firewalking as a motivational tactic from Tony Robbins?), and they're great at keeping teams pinned deep inside their own territory. The DEF still has plenty of room for improvement with respect to the other TR splits; but, unlike the OFF, they're not atrociously bad in either of those two aspects of motivated football.
SPECIAL TEAMS RANKINGS
Below are the Niners' ST DVOA stats broken down by unit:
|
ST |
Rank |
FG/XP |
Rank |
Kickoff |
Rank |
Punt |
Rank |
|
-2.1% |
16 |
0.4 |
8 |
-3.6 |
30 |
6.8 |
4 |
|
ST VOA |
Rank |
|
|
Kickoff Return |
Rank |
Punt Return |
Rank |
|
-1.1% |
18 |
|
|
-1.2 |
15 |
-5.5 |
30 |
Before I discuss the table, I'd like to clear up something about the interpretation of ST DVOA. To be honest with you, I was a little confused about whether the net points stat for each specific ST unit is supposed to be interpreted on a per-kick/punt, per-game, or year-to-date basis. In other words, even I didn't know the answer to, "Compared to the league average, has the kickoff coverage team cost SF 3.6 points per kickoff, 3.6 points per game, or 3.6 total points so far this season?"
Well, I had a little e-mail convo with Aaron at FO, and he was kind enough to clear this up for me. Basically, net points in the above table should be interpreted in a year-to-date fashion. So, going back to my previous example, the correct interpretation is that, compared to the league average, SF's kickoff coverage unit has cost the team 3.6 total points so far this season. Incidentally, Aaron also told me they're going to be converting net points into a per-kick stat at some point in the near future, which I think is a good idea from an interpretation perspective. Anyway, just wanted to clear that up. Back to the stats...
Unlike last week, there actually are a few things to discuss regarding ST. First, Andy Lee and company have reclaimed their spot in the top 8, netting the Niners over a touchdown's worth of points so far this season. Second, Joe Nedney is 1 of only 8 Ks in the NFL this season who is netting his team more FG points than the league average. Finally, when comparing the Niners' ST DVOA to their ST VOA, we see that they've faced a relatively weak ST schedule thus far. Indeed, aside from MIN, who has the #3 ST VOA, they've faced STs ranked 13th (SEA), 22nd (ARI), and 28th (STL). And even in the MIN game, the 49ers' ST played far worse (game-specific ST DVOA = -19.8%) than how the average team would be expected to play against the Vikings (ST DVOA = -8.4%), all else being equal.
Oh, and, by the way, Allen Rossum or Arnaz Battle or whichever Player X is back there this week still sucks at returning punts.
BOTTOM LINE
OK, so through 4 games, we can draw the following conclusions about the 2009 49ers:
- They're an above-average team that plays consistently from week to week.
- Their schedule has been relatively soft so far this season.
- Their OFF would really benefit from passing the ball more, especially out of the shotgun.
- Their OFF needs to start games better and do a better job deep in their own territory.
- Their DEF is a juggernaut for sure, but it needs to finish games better (ala the STL game) and do a better job when other teams enter the red zone - that is, if other teams ever do enter the red zone.
- As expected, Andy Lee and Joe Nedney are having yet another great season. Punt returner X...not so much.
On Saturday, I'll break down the player stats and rankings. See you then.
**DVOA statistics used to produce this article were obtained from Football Outsiders.
2 recs |
101 comments
|
Comments
As always, a great read.
Thanks again, Danny, for being so awesome by bringing this stuff and breaking it down for us.
My pessimism goes to the point of suspecting the sincerity of the pessimists.
by shlecko on Oct 8, 2009 6:32 PM PDT reply actions 0 recs
that's why...
…i get paid the big bucks…oh wait, no i don’t ;-) in any case, you’re welcome.
by Florida Danny on Oct 8, 2009 11:09 PM PDT up reply actions 0 recs
Seriously
This is great stuff Danny.
by Brendan Scolari on Oct 9, 2009 12:07 AM PDT up reply actions 0 recs
You're getting rewarded...
Labor’s of love are always more fulfilling than plain labor. Remember, you’re building something. Your work ain’t near done yet.
And we thank you for letting us be a part of it, as well as elevating and educating us.
Well, we're waiting....
by drummer on Oct 9, 2009 12:30 AM PDT up reply actions 0 recs
wow
we can’t run the ball – and yet we want to run it 60% of the time. The OL better get their stuff together and we need to get Gore back.
A hearty thank you to Rich Aurilia for all the good memories, and to the Niners for finally getting the uni's (mostly) right.
by wjackalope on Oct 8, 2009 6:34 PM PDT reply actions 0 recs
Singletary
recently stated he wanted a 50/50 offense. Guess they’re taking notice of the inability to run. And I love the Shotgun stats… Shows we can trust Hill to step up and make some throws.
by mountaindew77 on Oct 8, 2009 6:50 PM PDT up reply actions 0 recs
Surprised by average 3rd & long defense.
I had thought our dime package was pretty darn good, but the numbers say +10% and ranked 16th. I’m making the assumption our dime is on the field in 3rd & long situations as offenses typically use 3-4 WR sets on those downs.
Is there any way to extract meaningful numbers on our dime/nickel packages from other FO stats?
But otherwise, this is saying we’re consistently average because we’re consistently amazing on defense and we consistently suck on offense. And they also show what we all already knew – sometimes a team has to throw to open up the run. But I think the best we can hope for right now is some better OL play and more diversified play calling like last week. I think the shotgun formation may be seen by Sing as conflicting with his goal of total ball security.
Thanks again for all the time and effort – this is not a trivial effort and we all appreciate it.
by MontanaPass on Oct 8, 2009 7:20 PM PDT reply actions 0 recs
#1 Against the Run
That’s something we can definitely hang our hats on.
"We Deserve"
by YaHeard on Oct 8, 2009 7:47 PM PDT reply actions 0 recs
Not voting on the poll
Which do I think is more vital to the team’s success?
a) #1 in red zone pass offense
b) #1 in run defense
That is like picking which butt cheek on J-Lo you like best.
by bignerd on Oct 8, 2009 7:51 PM PDT reply actions 1 recs
I wonder how you can poll that...
Would the results be on how many red hand prints on a particular cheek?
Well, we're waiting....
by drummer on Oct 9, 2009 12:36 AM PDT up reply actions 0 recs
Run defense is more important than red zone passing offense
Because the red zone passing offense is almost assuredly unsustainable. Red zone stats have a very small sample size so there’s a lot of variance, but usually a team’s regular passing offense should be the same as their passing offense in the red zone. In other words, the Niners have the #19 passing offense so far, I’d expect that to be about how well the red zone passing offense performs for the rest of the season.
The run defense on the other hand could very well be the team team’s true talent level. The front seven has certainly looked outstanding so far.
by Brendan Scolari on Oct 8, 2009 8:08 PM PDT reply actions 0 recs
I would say red zone scoring instead of redzone passing.
I think this is more important than our d being number 1 vs the run. Whats the point of stopping people if you cant put up points
"Optimist Prime"
Come along and ride the muthaship!
by rlott#42 on Oct 9, 2009 10:02 AM PDT up reply actions 0 recs
Obviously red zone scoring is important
But my point is that it’s not as sustainable over the long run. There’s no reason to expect our red zone offense to be significantly better than our offense anywhere else on the field going forward.
by Brendan Scolari on Oct 9, 2009 11:18 AM PDT up reply actions 0 recs
Any chance of seeing the Falcon's numbers?
I’d love to see the breakdowns on their defensive units to see how we can run on them and see how well those seam passes to VD will work.
by MontanaPass on Oct 8, 2009 8:35 PM PDT reply actions 0 recs
ask me for something specific...
…and i’ll be more than happy to provide the info.
by Florida Danny on Oct 8, 2009 10:00 PM PDT up reply actions 0 recs
I know it's a lot of work...
… but it’d be kind of cool to see how the opponents stack up against the 49ers’ weaknesses (i.e. opponent run defense and pass offense).
by sfgfan on Oct 9, 2009 9:38 AM PDT up reply actions 0 recs
here you go...
as far as i can tell, the niners have 3 glaring weaknesses: run offense, 2nd down offense, OL play. here’s ATL’s defensive stats for those 3 things:
run defense: 3.8% (25th)
2nd down defense: 14.1% (24th)
defensive front 7 ALY: 4.78 (28th)
defensive front 7 ASR: 4.3% (28th)
based on this, you’d think that, if the offense’s weaknesses are ever going to get healthy, it’d be against ATL’s defense.
by Florida Danny on Oct 9, 2009 8:23 PM PDT up reply actions 0 recs
Breakdowns of run defense by down and hole.
I want to see where they’re weak against the run and on which down, if that’s possible. How Ryan does on 3rd & long is important, too.
by MontanaPass on Oct 9, 2009 12:31 PM PDT up reply actions 0 recs
i think yours overlaps somewhat with the above, but...
here’s ATL’s defensive front 7 ALY by run direction (from the perspective of the opposing offense):
left end: 5.45 (27th)
left tackle: 4.97 (27th)
center/guard: 4.94 (29th)
right tackle: 2.39 (3rd)
right end: 4.64 (22nd)
their opponents have run predominately to left end and behind center/guard. the niners run — as you’re well aware — predominately behind center/guard and right tackle.
their run defense by down is
1st: -2.6% (19th)
2nd: 3.3% (20th)
3rd: 27.8% (27th)
finally, ATL’s 3rd & Long offense DVOA is 57.5% (9th)
so i’d say that ATL’s run defense presents the worthless niner run offense with an opportunity. also, 3rd & Long when ATL has the ball is going to be a key situation to watch. the niner defense isn’t that great at it so far this season, but ATL’s offense is really good at it.
by Florida Danny on Oct 9, 2009 8:35 PM PDT up reply actions 0 recs
Not scoring the red zone is morale killer, something the stats don't reflect.
by Vertigo on Oct 8, 2009 8:48 PM PDT reply actions 0 recs
I was hoping all the comments wouldn’t be like “yeah, stats are awesome” now that there’s a stats thing that actually tells us things that are positive. We would have been a bunch of hypocrites.
Context, people. More context is good. Less context is bad. If you're willing to be reductive, then you're willing to be wrong.
by howtheyscored on Oct 8, 2009 10:15 PM PDT up reply actions 0 recs
i actually had the same thought...
while writing this. was totally expecting fewer comments generally, but also fewer “FO’s stats are BS” comments specifically. turns out i/we were pretty much on the money. it’s not altogether surprising though. human beings have a tendency — refined through evolution — to internalize success and externalize failures. if the stats say the niners suck, well, the stats themselves have to be the problem. if the stats say the niners are awesome, well, it’s because, duh, the NINERZ R AWSUMZZZ!!!
by Florida Danny on Oct 8, 2009 10:21 PM PDT up reply actions 1 recs
enough with the refined through evolution
by goatfather on Oct 8, 2009 11:01 PM PDT up reply actions 0 recs
haha...
i’m just being provocative on purpose. a lot fewer comments now that the stats say the niners are good. figured “evoltion vs. creation” might get people’s dander up enough to actually say something on here.
by Florida Danny on Oct 8, 2009 11:03 PM PDT up reply actions 0 recs
TRY EVOLVING INTO SOMEONE WHO KNOWS HOW TO SPELL
Context, people. More context is good. Less context is bad. If you're willing to be reductive, then you're willing to be wrong.
by howtheyscored on Oct 8, 2009 11:04 PM PDT up reply actions 0 recs
LOLZ!!!!
first i have to devolve into someone who SPELLZZ!
by Florida Danny on Oct 8, 2009 11:07 PM PDT up reply actions 0 recs
Nice.
Context, people. More context is good. Less context is bad. If you're willing to be reductive, then you're willing to be wrong.
by howtheyscored on Oct 9, 2009 9:51 AM PDT up reply actions 0 recs
Oh ok but can I just say
The evo vs creation arguement gives an open ended license to faulty conceptions of neo-Darwinism. Darwin had a lot more to say!!
by goatfather on Oct 9, 2009 7:57 AM PDT up reply actions 0 recs
You should also credit yourself (and howey)
For framing the debate the past few weeks. At this point I think most people either love the stats and look forward to them or don’t trust/like them but have gotten themselves on record as believing such. I bet people still read your post but no longer feel the need to either discredit or defend your work but accept it for what it is and perhaps learn something at the same time.
No matter how seriously anyone takes the info, you deserve credit for doing good detailed work and bringing additional analysis to the table.
You gotta bring ass to get ass.
by SpurredOn on Oct 9, 2009 1:45 AM PDT up reply actions 0 recs
Defense
While the red zone stats are impressive, i think our run defense stats are amazing. The only problem i have with our red zone stats is that out offense has, at times, struggled to get into the red zone. so the defense means more to the team, because you know they are bustin heads over the entire 100 yards of the field, not just in a 20 yard stretch that we struggle to get to sometimes because of our offense; even though it is improving every week and i believe could finish in the top half of the league if things start clicking. But thats just my two cents
by Camraman926 on Oct 8, 2009 9:04 PM PDT reply actions 0 recs
D averages
Altho Sing said Niners weren’t in prevent D, they played soft in the final Minn drive. This padded the yards & scoring quite a bit.
Hopefully it was a learning experience: Take a chance & if you fail, at least you have some clock left to force overtime or better.
by BobE on Oct 8, 2009 10:16 PM PDT reply actions 0 recs
I'll say it again
They got to the 32 yard line with 12 seconds remaining. The D did their job until then. Why they had 5 guys guarding nobody between the 0 and 20 yard lines, I don’t know. It was a fluke play. The strategy was sound (tackle them in bounds). We got Favred. It’s not gonna happen again.
You have been DFiBrillated.
by Dubs fan in Boston on Oct 9, 2009 4:27 PM PDT up reply actions 0 recs
I thought they were supposed to be good at running the football?
Imagine if how much better the team would be if they could run the ball effectively……
What we've got here is a failure to communicate.
by chikmagnet_565 on Oct 8, 2009 10:43 PM PDT reply actions 0 recs
I think we get going vs. ATL
ATL’s run D is awful up the middle — right where we like to run. If Chilo Richal pulls his head out, Coffee may run for 130 yards this week.
by nickbradley on Oct 9, 2009 10:37 AM PDT up reply actions 0 recs
Maybe we ought to have a Frank Gore Approval poll?
We’ve had 2 on Jimmy Raye.
We’ve had about 2 dozen on Shaun Hill
We’ve had at least 1 on VD.
Maybe it’s time to get to the real heart of the offense’s problem . . . Frank Gore and his 30th ranked rushing attack. F.O. are always right! Frank cannot hide any more.
BTW, I knew we’d get a big bump in the rankings this week. After all we just crushed F.O. preseason juggernaut in the Rams.
by bignerd on Oct 8, 2009 10:59 PM PDT reply actions 0 recs
don't think...
FO’s ever gonna live that one down.
by Florida Danny on Oct 8, 2009 11:00 PM PDT up reply actions 0 recs
this has to be a chasm
right?
Sharlon Schoop - de favoriete Nederlandse honkbalspeler van McCovey Chronicles.
You always have to be one step ahead of your drunk friends
--Daisy Owl
by Viliphied on Oct 8, 2009 11:08 PM PDT up reply actions 0 recs
Red zone passing
Could the big difference there be that Shaun Hill’s weaknesses are less important in the red zone? The knock on him is that his arm isn’t very strong (for an NFL quarterback), but he’s generally pretty accurate and he avoids bad decisions. It seems like the red zone is exactly the place that would play to his strengths.
GROUGTHINK ALERT
The first Chester Arthur fanboy ever.
by groug on Oct 8, 2009 11:23 PM PDT reply actions 0 recs
Quite possible
And defenses are already playing us tight to seeing a compacted defense near the goal line is nothing new. I also think the development of a TE makes any offense better in the red zone.
You gotta bring ass to get ass.
by SpurredOn on Oct 9, 2009 1:48 AM PDT up reply actions 0 recs
yeah
VD has been responsible for most of HIll’s red zone success. I’m impressed with Hill’s play and have a lot of confidence in him. However, every time I watch the film there’s someone WIDE OPEN that he fails to see.
by goatfather on Oct 9, 2009 8:00 AM PDT up reply actions 0 recs
All good points from everybody in this nest...
You have been DFiBrillated.
by Dubs fan in Boston on Oct 9, 2009 4:28 PM PDT up reply actions 0 recs
Sorry Danny, I just cannot find any hidden gems to discuss. The stats pretty much reflect what we’ve been seeing.
The only odd number is the offense’s abysmal production on 2nd down. Horrible in both running and passing, I cannot think of a football explanation. Do the run and pass plays differ on 2nd down? 49ers are average running ball on 1st and average passing the ball on 3rd. Sorta takes away the argument the team is ineffective because the play calling is predictable.
by bignerd on Oct 9, 2009 12:04 AM PDT reply actions 0 recs
I wonder if opponents blitz us more on 2nd down
So we’re always running into a blitz or play-action passing with extra people coming, thus Hill is being sacked or rushing a throw. Then boom, it’s 3rd and long.
You gotta bring ass to get ass.
by SpurredOn on Oct 9, 2009 1:49 AM PDT up reply actions 0 recs
Danny -- what do you think about FO's Field Position Stats for this week's game?
Atlanta is crushing the rest of the league in average field position — both #1 in avg. starting field position and field position allowed.
http://www.footballoutsiders.com/stats/drivestats
According to the stats, we lose a half yard of field position every time there’s a change of possession, while Atlanta gains 5 3/4 yards (all data averaged together). The stats are even worse for us when you adjust for strength of opponent (13 and a half net yards for Atlanta per COP, 3 1/4 for us).
Based on those numbers, we’re looking like we should expect to lose about 100 yards of field position in the game — more than enough to lose a close game.
Thoughts?
by nickbradley on Oct 9, 2009 10:33 AM PDT reply actions 0 recs
OK...
first, let’s clear up the fact that FO says their drive stats aren’t adjusted for opponent. therefore, ATL’s 13.64 net and SF’s 3.19 net are unadjusted.
all’s not lost though. let’s adjusted them based on opponent and turn them into a “above NFL average” type of stat ala DVOA, DYAR, etc. basically, i’m going to create a stat called defense-adjusted net line of scrimmage above average (DLOSOA for short):
ATL has played
MIA, net = -5.73
CAR, net = -8.06
NE, net = 1.96
avg opp net = -3.94
SF has played
ARI, net = 0.51
SEA, net = -0.98
MIN, net = 9.62
STL, net = -8.78
avg opp net = 0.09
now, the adjusting and standardizing steps take a bit of hardcore stats, so i won’t bore you with that….instead, i’ll just tell you that this season’s correlation between net LOS/Dr and opponents’ avg net LOS/Dr is -.741, which is massive. also, a team gains 1.75 net LOS yards for every net LOS yard worse that their opponent is. anyway, so after doing the adjusting and standardizing, here’s what you find according to my DLOSOA:
SF = -0.09 (17th)
ATL = 6.32 (3rd)
so the gap narrows a bit when you adjust for opponent and compare with the NFL average, but ATL still has a clear advantage: we’d expect the niners to lose about 6.5 yards of field position for every change of possession against the falcons. so according to your 10 change-of-possession standard, i’d expect the niners to lose 65 yards of field position, not 100.
i think that being able to stop the falcons running game is going to be a bigger predictor of success in this game than the field position thing. field position will take care of itself if the niners aren’t allowing ATL to move the ball on the ground.
FYI…here are the DLOSOA’s i came up with for all 32 teams:
Rank Team DLOSOA
1 WAS 9.32
2 NO 7.14
3 ATL 6.32
4 CHI 5.22
5 HOU 4.71
6 MIN 4.29
7 PHI 4.21
8 PIT 3.79
9 DEN 3.62
10 JAC 3.16
11 SD 2.65
12 IND 1.28
13 NYG 1.20
14 BAL 0.92
15 DAL 0.51
16 NYJ 0.04
17 SF -0.09
18 TB -0.42
19 BUF -0.57
20 KC -0.71
21 OAK -1.32
22 STL -1.32
23 SEA -1.71
24 CIN -2.48
25 GB -3.03
26 MIA -3.39
27 ARI -4.51
28 CLE -5.56
29 TEN -6.99
30 CAR -7.85
31 NE -8.58
32 DET -10.01
by Florida Danny on Oct 9, 2009 12:33 PM PDT up reply actions 0 recs
Thanks Danny...
Those Numbers for DLOSOA are close I got when I
1. Multiplied all team possessions by average starting field possession
2. Multiplied all opponent possessions by average starting field possession
3. Subtracted #2 from #1
4. Divided result by total changes of possession
The gap between us and Atlanta is 6 1/4 yards per C.O.P.
There are an average of 15 C.O.Ps per game — not 10. That’s how I came up w/ 100 yards.
100 yards of field possession is huge in an evenly-matched game. I went back and looked at the ’08 game against Miami and it came down to field possession — they out-positioned us by 126 net yards! The rest of the game was completely even (no mistakes, no turnovers). We stalled midfield twice and were forced to go for the TD at the end of the game. If we had better field position, we would have gotten a FG on one of those drives and could have kicked a FG to win the game.
The first time was when Delanie Walker muffed the kickoff in the second quarter and we ended up on the 11 — we drove to the Miami 41 and punted through the end zone (21 yard net punt). If it had simply been a touchback and we had the same drive, that would have been a 49-yard FG att for Nedney.
The second time was in the third quarter, where we stalled midfield and punted into the end zone.
If either one of those drives resulted in a field goal, we would have been able to kick a GW FG at the end instead of stalling out in the red zone.
(http://www.nfl.com/gamecenter/2008121408/2008/REG15/49ers@dolphins#tab:analyze/analyze-channels:cat-post-playbyplay).
http://www.49erswebzone.com/forum/thread.php?num=131070&highlight=outsider
by nickbradley on Oct 9, 2009 2:17 PM PDT up reply actions 0 recs
Any chance Rossum steps up his game this week because he is playing his former team?
I really don’t know much about how these stats are calculated but how much does Harvin’s TD return have a baring on our statistic. Singletary also elected to go to our 2nd line kick/punt returner squad to get his OLB some rests against the Rams.
Atlanta has been good at specials teams for about as long as I remember. I do think the 49ers are better than what their current statistics indicate, except for questions about the kickoff coverage team.
by bignerd on Oct 9, 2009 2:22 PM PDT up reply actions 0 recs
There's a Good Chance Rossum has a good week because...
SOMEBODY is going to lose their roster spot after they bye week (Crabtree). Rossum is on the bubble.
by nickbradley on Oct 9, 2009 2:47 PM PDT up reply actions 0 recs
Doubtful
Unless Rossum absolutely has a bad day, I doubt the cut will come from anywhere other than WR. For a team that doesn’t spread the defense out, keeping 7 WRs (Bruce, Morgan, Battle, Jones, Hill, Spurlock, and Crabtree) is utterly ridiculous and a poor way to use roster space.
by sfgfan on Oct 9, 2009 3:02 PM PDT up reply actions 0 recs
Agreed
There’s no sense havign 3 inactive WR’s every week.
by Brendan Scolari on Oct 9, 2009 4:44 PM PDT up reply actions 0 recs
it's gotta be
“super size me” getting cut

MORGAN SPURLOCK
by Florida Danny on Oct 9, 2009 8:10 PM PDT up reply actions 0 recs
p.s.
thanks for giving me this idea…i’m going to start including DLOSOA in my tony robbins splits for the team rankings since it’s a good measure of “create great field position.”
by Florida Danny on Oct 9, 2009 12:45 PM PDT up reply actions 0 recs
p.p.s.
well, technically, it’d be called defense-adjusted net line of scrimmage per drive above average…which, when shortened is DANLOSDOA…so i’ll short it even further and call it DAN….that’s easy enough. :-)
by Florida Danny on Oct 9, 2009 12:49 PM PDT up reply actions 0 recs
DANLOLSDOA
Context, people. More context is good. Less context is bad. If you're willing to be reductive, then you're willing to be wrong.
by howtheyscored on Oct 9, 2009 1:23 PM PDT up reply actions 0 recs
wait a sec...
oops…i was looking at, and gave you, the wrong list…here are the correct DANLOLS stats:
Rank Team DANLOLS
1 NE 9.21
2 ATL 6.40
3 CLE 5.49
4 NYJ 5.47
5 MIN 4.83
6 ARI 4.57
7 NYG 3.88
8 GB 3.66
9 CHI 3.52
10 SF 3.01
11 BAL 2.47
12 HOU 1.17
13 SEA 0.54
14 JAC 0.36
15 DAL 0.24
16 TB 0.14
17 CAR -0.33
18 CIN -0.79
19 IND -0.80
20 NO -1.23
21 PIT -1.23
22 DEN -1.71
23 MIA -2.38
24 SD -2.74
25 KC -3.67
26 DET -3.95
27 OAK -4.67
28 PHI -4.87
29 BUF -5.26
30 TEN -6.43
31 STL -7.36
32 WAS -7.54
so ATL only has a 3.39 net LOS yard advantage over the niners. that means about 30 yards of field position for a 10-drive game, not 100. most of the difference between the two teams is due to differences in schedule strength.
by Florida Danny on Oct 9, 2009 2:22 PM PDT up reply actions 0 recs
What's the 3rd
stat that is highly correlated with winning (besides 1st Quarter and Red Zone efficiency)? And what’s the order? And is this offense or defense or both?
by rdub49erfn on Oct 9, 2009 10:43 AM PDT reply actions 0 recs
the 3 are...
red zone defense, 1st quarter offense, and overall 2nd half efficiency in close games
by Florida Danny on Oct 9, 2009 11:04 AM PDT up reply actions 0 recs
There may be correlation
But do you honestly think that means anything? Why would 1st quarter offense be any more important than 2nd quarter offense?
by Brendan Scolari on Oct 9, 2009 11:19 AM PDT up reply actions 0 recs
well, one obvious reason...
is that better teams get early leads. early leads come from 1st quarter offense. it may just be a correlation, but it’s a perfectly explainable one. don’t you remember how the 49ers used to be up 14-0 at the end of the 1st quarter every game back in the walsh/seifert days?
also, if you want to get real technical, 1st quarter offense is the result of scripted plays, and scripted plays are the result of gameplanning, and gameplanning is the result of film study and knowledge of tendencies. so 1st quarter offense is an indicator of how good the coaching staff is when it comes to preparation from a strategic standpoint. strategic expertise is one of the 5 facets of expertise, so it makes perfect sense that better teams would have better coaching staffs when it comes to strategy. 1st quarter offense is a measure of that.
by Florida Danny on Oct 9, 2009 11:25 AM PDT up reply actions 0 recs
i should add...
that my 2nd explanation there introduces a timeline, and therefore turns a correlation into a causation…you know, technically.
by Florida Danny on Oct 9, 2009 11:26 AM PDT up reply actions 0 recs
that's post hoc
though.
I mean, isn’t point differential more highly correlated? That’s highly causative too (duh) you score more points, you win.
The question is whether or not red zone defense, 1st Q offense, 2nd half efficiency in close games are actual (teamwide) skills.
As discussed here
FIRE BRIAN SABEAN... UNLESS HE KEEPS DRAFTING WELL. .. AND SIGNS UNDERRATED PLAYERS LIKE AFFELDT OR PHELPS. .. OR ALRIGHT WHO'S PLAYING WITH THE ALIEN MIND-SWITCHING RAY?
-------
PARPG- Indy post-apocalyptic roleplaying game currently in early planning stages.
by zenbitz on Oct 9, 2009 1:08 PM PDT up reply actions 0 recs
I mean, isn’t point differential more highly correlated? That’s highly causative too (duh) you score more points, you win.
The question is whether or not red zone defense, 1st Q offense, 2nd half efficiency in close games are actual (teamwide) skills.
Exactly. Give me a team with a good offense, not a team that happens to be performing way over their heads in one particular situation because of a small sample size.
by Brendan Scolari on Oct 9, 2009 1:40 PM PDT up reply actions 0 recs
It's not always luck in the red zone
The red zone success percentage the 49ers currently have is product of small sample size.
However teams with good running games and/or a TE’s appear to better equipped to succeed in the red zone. A new development has been the WR who can win a jump ball to the corner of the end zone.
VD appears to be a legitimate red zone weapon which appears to explain their early success. I agree that red zone success is dependent upon a bit of luck but I don’t think it’s a flip of the coin scenario.
by bignerd on Oct 9, 2009 2:17 PM PDT up reply actions 0 recs
No your right
It’s not completely luck in the red zone, there is some skill there (running game, good TE’s, big WR’s like you said) but for the most part a team’s offense shouid be consistent across the field.
Like if the Niners offense was the 20th best (on a talent basis) I could see them being the 15th best in the red zone, but I can’t see them being top 5.
by Brendan Scolari on Oct 9, 2009 2:23 PM PDT up reply actions 0 recs
what do you mean by...
for the most part a team’s offense shouid be consistent across the field
“should”? i’d say most NFL teams are pretty inconsistent across the field. in fact, i’d love to find out how many teams are actually consistent across the field according to some minimum “consistency” threshhold. the really bad teams might be bad across the field, and the really good teams might be good across the field….but that’s only 1/3 to 1/2 of the league. i’d bet that, for most of the league, offense is wildly inconsistent (for whatever reason) across different field zones.
by Florida Danny on Oct 9, 2009 2:50 PM PDT up reply actions 0 recs
I'd bet they are
Because of small sample size, but i think their true talent level is consistent across the field. How well would those stats correlate from year-to-year? I’m guessing they wouldn’t correlate well at all.
by Brendan Scolari on Oct 9, 2009 4:44 PM PDT up reply actions 0 recs
i don't think we're...
talking about between-year consistency here…in the context of the 49ers, we’re arguing about whether the 49ers will display any kind of within-year consistency in the red zone, and whether that consistency (or inconsistency) is affected by their between-field-zone consistency.
in any event, i’m confident not too many team stats autocorrelate from year to year (just based on what i’ve noticed casually during other analyses). but even if they do, conclusions aren’t that valid what with all the qualitative differences between this year’s version of a team and next year’s version.
by Florida Danny on Oct 9, 2009 7:36 PM PDT up reply actions 0 recs
Different areas of the field require different skill sets...
Taking it to the extreme, inside either 5 yard line, rushing dominates. That both red zones are made up of 25% of these zones should not be overlooked. In your opponents red zone, the field is compressed, yadda yadda yadda. It’s not straight apples to oranges here.
You have been DFiBrillated.
by Dubs fan in Boston on Oct 9, 2009 4:47 PM PDT up reply actions 0 recs
...
by the end of a season, a team will have, oh, i don’t know, about 700 or so plays in the 2nd half as a team, about 250 or so on offense in the 1st quarter, and about 150 plays in the red zone on defense. while there are definitely valid points to raise here, small sample size is not one of them…as it relates to the correlation because DVOA is a per-play stat, so the unit of measurement is plays not games or drives or seasons or whatever else would make the sample size small.
but also, you say the following:
happens to be performing way over their heads in one particular situation because of a small sample size.
so to argue against one causal inference, you opine that SF is playing way over their heads (not sure how you know that factually) and that their stats are so good because of small sample size. how can you possibly know for a fact that SF’s red zone stats are caused by small sample size? either their RZ stats are accurate reflections of their true (i.e., correct) RZ ability or they aren’t. there’s nothing inherent to sample size that would cause anything related to their stats. over a sample size of 1 play, we can get an accurate reflection of their ability if we’re lucky enough to see the 1 play that they performed at their average. point is, sample size doesn’t cause their stats, the niners cause their stats. if their stats are a mirage, it’s likely due to some disproportionate sampling issue (aka facing weak RZ defenses so far) that’s merely reflected in, not caused by, small sample size.
by Florida Danny on Oct 9, 2009 2:45 PM PDT up reply actions 0 recs
What is the 49ers red zone sample size?
Aren’t the 49ers almost perfect at scoring TD’s when inside the opponent’s 20?
by bignerd on Oct 9, 2009 2:51 PM PDT up reply actions 0 recs
the raw number of plays
does not, in and of itself, indicate whether or not the sample size is adequate. That would depend on the variance of the quantity measures.
I would venture a guess that 150 plays is no where NEAR a big enough sample size – if you consider all the possible opponents, game situations, personnel issues, weather, etc. etc.
FIRE BRIAN SABEAN... UNLESS HE KEEPS DRAFTING WELL. .. AND SIGNS UNDERRATED PLAYERS LIKE AFFELDT OR PHELPS. .. OR ALRIGHT WHO'S PLAYING WITH THE ALIEN MIND-SWITCHING RAY?
-------
PARPG- Indy post-apocalyptic roleplaying game currently in early planning stages.
by zenbitz on Oct 9, 2009 3:05 PM PDT up reply actions 0 recs
I think the magic evaluation number in statistics is 30. It could be 30 red zone possessions provides a comprehensive sample size instead of red zone snaps.
by bignerd on Oct 9, 2009 3:12 PM PDT up reply actions 0 recs
i get where you're going with this...
but we’re talking about football here, not curing cancer. if we’re only going to draw any conclusions about football if we have a double-blind, placebo-controlled, longitudinal study, then we might as well put the stats away, and go have a beer.
by Florida Danny on Oct 9, 2009 6:37 PM PDT up reply actions 0 recs
So you think the Niners will continue having one of the best red zone offenses in the NFL this year?
Frankly, I find that idea pretty absurd. The offense is norwhere near the level of the Colts, Saints, etc. but all the sudden they are just as good because they are near the goalline? FO has done lots of research about how red zone offense is very inconsistent because of the sample size, just like 3rd down offense. I’m surprised your arguing this point.
by Brendan Scolari on Oct 9, 2009 4:47 PM PDT up reply actions 0 recs
no...
i don’t think anything either way…they might continue to be really good, they might not…small sample size makes the variability of the RZ efficiency estimate huge right now…based on this “sample-size-caused” variability their RZ performance can go in a wide range of directions the rest of the year, only one of which is them having some spectacular decline. you’re ignoring the rest of the variability, even though it is just as “caused” by small sample size as spectacular decline is. in other words, as i said, small sample size doesn’t “cause” them to decline; small sample size causes increased variability, which includes both decline, plateau, or improvement.
by Florida Danny on Oct 9, 2009 6:28 PM PDT up reply actions 0 recs
Ok, what I'm saying is
The offense, in a much larger sample, has been fairly poor. That is likely their true talent level, and that’s the level I expect the red zone offense to regress to. Is that a fair assessment?
by Brendan Scolari on Oct 9, 2009 6:47 PM PDT up reply actions 0 recs
yeah...that's fair...
at least you’re basing it on something evidentiary.
just a few general points though (not trying to argue for the sake of arguing…just putting this out there):
1) with DVOA, red zone offense and total offense are ALWAYS going to be highly correlated because offense DVOA is a weighted average wherein red zone offense gets a large weight. i’d be shocked to ever find a situation in which they’re not correlated. that would be really problematic for FO’s weighting system.
2) remember that the 49ers are playing lights out right now only in red zone passing offense (DVOA = 109.1%). their red zone run offense, although also much better than their overall run offense, is only 4.3% right now, which doesn’t strike me as some wild statistic like 109.1%. as the season progresses, the red zone run offense can get better just as easily as the red zone pass offense comes back to earth; which essentially washes out the whole regression to the mean thing when it comes to their overall red zone offense (aka run and pass).
3) in a related point, their overall red zone offense DVOA is 39.6%, which suggests that they’re passing the ball a lot more than running it in the red zone. given that their overall pass offense (6.7%) is way better than their run offense (-24.7%), it’s less likely that the red zone pass offense DVOA is going to nosedive if they continue this play-calling imbalance. in other words, their red zone offense DVOA — in a way — already reflects their overall offense DVOA, i.e., they’re a much better passing team than running team.
4) likewise, running the ball more in the red zone is just as likely an explanation for any drop in their red zone offense DVOA as is some variation of regression to the mean. in other words, what you might interpret as regression to the mean might simply be them running more, thereby infecting their red zone offense DVOA with that inefficient overall run offense.
their overall red zone DVOA can just as easily drop because they start running it more in the red zone
by Florida Danny on Oct 9, 2009 8:02 PM PDT up reply actions 0 recs
oops...cut and paste fail
their overall red zone DVOA can just as easily drop because they start running it more in the red zone
by Florida Danny on Oct 9, 2009 8:03 PM PDT up reply actions 0 recs
No I don't remember
I was 7 when Seifert left the Niners and I never saw a game Walsh coached.
But I’m still not buying your reasoning. Sure, 1st quarter offense will have a correlation with winning, but only because having a good first quarter offense means you are more likely to have a good offense in general. Bad offenses generally fare poorly in the 1st quarter because they fare poorly in all quarters, and good teams fare well in the 1st quarter because they tend to fare well in all quarters. But giving one quarter more weight than any other quarter makes zero sense to me.
It’s a bit like looking at OPS with runners in scoring position in baseball. It’s a small sample size so it can tend to fluctuate and teams can over or underperform their regular OPS, but in general a team with a good OPS with runners in scoring position will win more than a team with a poor OPS. This is not because some teams are better on offense with runners in scoring position than normally, it’s simply because good offenses are more likely to succeed in all situations if that makes sense.
As long as the Niners offense continues to be poor I don’t think we should count on them being good in any particular quarter. While a small portion of the success could be due to gameplanning, conditioning, etc, most of the difference is probably just due to small sample size.
by Brendan Scolari on Oct 9, 2009 1:39 PM PDT up reply actions 0 recs
It could also be run to pass, pass to run situation too
A passing team wants to jump out to a significant early lead to get their opponent out of the run game. A running team won’t mind a contested 1st Qtr because they believe their running game will open the offense up later in the game. Off course a great passing team or great running team will likely be successful in most 1st quarters like they will be successful in most quarters.
by bignerd on Oct 9, 2009 1:48 PM PDT up reply actions 0 recs
Possibly
And i do think part of it is that Raye gameplans to run a lot but then has been forced to throw more as the game goes on (at least in the Arizona and Minnesota games) but I still don’t think having success in a certain quarter portends future success more than having success at any other point in the game would.
Thinking about the small sample thing more, the reason the 1st quarter DVOA is so low is probably almost solely because of the Viking game. Hill was sacked twice and fumbled and we didn’t get a first down the whole quarter. The offense was not very good early on in the Rams game either, but they were fine in the Cards and Seahawks games.
by Brendan Scolari on Oct 9, 2009 2:15 PM PDT up reply actions 0 recs
A passing team wants to jump out to a significant early lead to get their opponent out of the run game.
Just because they want to doesn’t mean they can. A running offense can just as easily put up some extra points in the first quarter. I think Danny is right on the money here. Three major reasons:
1. First quarter starts at 0-0. Once you get past that, if you’re winning by two TDs, you’re often playing to have the game end quicker.
2. Scripted plays cannot be overlooked. They are a result of good coaching. Thus the performance of the scripted plays are not in and of themselves completely indicative of a good team, but they are indicative of good coaching, which is in turn indicative of a good team.
3. A team that does not come out ready for the first quarter puts itself at a significant disadvantage for the rest of the game, and is indicative of a team that is not mentally ready to execute. I think this is more a case of “if perform poorly in the first quarter, then you’re a bad team” than the reverse, but it’s mutually exclusive.
You have been DFiBrillated.
by Dubs fan in Boston on Oct 9, 2009 4:44 PM PDT up reply actions 0 recs
Obviously they were'mt alwasy up 14-0
But they did have a great habit of getting a score on their opening drive and usually a TD somewhere in the 1st quarter. The ability to score early usually means a good drive with a number of plays so, much like a leadoff hitter making a pitcher show his pitches, you get many looks at what the defense is trying to do. If you consistently move the ball early you also assure that even if you don’t score, you at least don’t open with many 3 and outs, thus you can flip field position.
You gotta bring ass to get ass.
by SpurredOn on Oct 9, 2009 2:24 PM PDT up reply actions 0 recs
Early leads
Scoring first is important. Something I always pay attention to, how many possessions does a team have when they are in the lead vs trailing. If you take the lead but your defense gives it right back, you once again have possession while trailing. Having the lead and the ball places pressure on the defense to not fall further behind. Just think of the momentum with scoring going into the half and leading by 7 points, then receiving the 2nd half KO and scoring again. A tie game very quickly becomes a 14 point game.
The Niners never trailed vs Seattle or STL. They lead the majority of the 2nd half in Minnesota and for all but one drive at AZ. It allows a team to run their game plan and not fear making a mistake.
You gotta bring ass to get ass.
by SpurredOn on Oct 9, 2009 2:20 PM PDT up reply actions 0 recs
I actually doubt this is the case
I would guess that a lead has to be 14 points or so (maybe 10) before it matters early in the game.
Do you have any numbers to back up your proposition? It’s not unreasonable… but I would like to see some numbers on it.
Actually… just to back you up… I looked at some WPA graphs for niners games this year, and it does seem like the first TD is worth 25% of a win (going up 7-0). But it doesn’t seem to matter if it occurs at the beginning or end of the half. Going up 3-0 appears to be a much weaker effect (<1/2).
I guess what you would have to demonstrate is something like if it’s better to go up 7-0 in the first quarter or 14-7 in the second. My guess is that the 2nd would be better, but only by a tiny amount (because of less fewer possesions remaining)… mathematically, it probably matters a great deal what fraction of drives result in TDs (points) for the two teams.
FIRE BRIAN SABEAN... UNLESS HE KEEPS DRAFTING WELL. .. AND SIGNS UNDERRATED PLAYERS LIKE AFFELDT OR PHELPS. .. OR ALRIGHT WHO'S PLAYING WITH THE ALIEN MIND-SWITCHING RAY?
-------
PARPG- Indy post-apocalyptic roleplaying game currently in early planning stages.
by zenbitz on Oct 9, 2009 3:19 PM PDT up reply actions 0 recs
An all time great tennis coach (I read his book, forget who though) always said
The most important point is the point that sets up your break point (i.e. the competition to get to the threat zone). In this case, if you’ve got the ball with a 7 point lead, that’s the most important place to score/prevent scoing because it is where you can break a game open or keep it in the balance. One interesting study would be to look at maximum lead each team has in each game and see how often winning occurs (i.e. how often does somebody come back from a 10 point deficit).
You have been DFiBrillated.
by Dubs fan in Boston on Oct 9, 2009 4:52 PM PDT up reply actions 0 recs
No numbers just something I follow
I think play calling from coaches changes based on having the lead. Most coaches become more cautious/conservative, but I’ve heard Steve Young many times say that he loved Shanahan’s willingness to go for the throat when the defense was trailing and getting desperate. I’ve noticed tackling gets sloppier when trailing in the 4th quarter because players start going for the fumble strip and/or over-pursue in hopes of making a big play. Same as QBs start taking chances when trailing, or pass rushers know it’s more likely the opponent will pass when trailing. Even crowds become less racous when their team is behind.
Obviously the initial focus was on scoring first and there is no guarantee that means you’ll keep the lead. But if you spend more time ahead you can dictate play calling and make your opponent more predictable.
You gotta bring ass to get ass.
by SpurredOn on Oct 9, 2009 5:13 PM PDT up reply actions 0 recs
2nd down offense
is probably just a sample size issue….but assuming it’ not.
1) Assumption: The Niners run game sucks.
2) Given: The Niners like to run (50% of the time, anyway).
So, on first down – they either run or pass. When they run, they are about average…
Never mind. Danny – these tables are meaningless without the number of situations in each box. Sorry, this is otherwise very nice…
FIRE BRIAN SABEAN... UNLESS HE KEEPS DRAFTING WELL. .. AND SIGNS UNDERRATED PLAYERS LIKE AFFELDT OR PHELPS. .. OR ALRIGHT WHO'S PLAYING WITH THE ALIEN MIND-SWITCHING RAY?
-------
PARPG- Indy post-apocalyptic roleplaying game currently in early planning stages.
by zenbitz on Oct 9, 2009 1:30 PM PDT reply actions 0 recs

by 





















