Week 1 DVOA Ratings
by Aaron Schatz
You love them when your team is high! You hate them when your team is low! Once again, the famous Football Outsiders DVOA and DAVE ratings return for 2016.
Most of you are familiar with DVOA, but we probably have some newer readers who have never met our good friend DAVE. Well, DAVE is our rating that combines the preseason projection with the results of early games to give us a better prediction of how each team will rank at the end of the year. For those who don't know the story, this metric is called DAVE as a reaction to criticism that our stats are too much alphabet soup. I mean, who can argue with a guy named Dave? (Technically, it stands for DVOA Adjusted for Variation Early.) In this week's DAVE ratings, the preseason projection counts for 90 percent, and the current DVOA counts for 10 percent. The value of the preseason projection changes each week until we are only using current-year data after Week 8.
Of course, though I'm calling our main metric DVOA here, it is actually VOA because there are no opponent adjustments right now. We do not apply opponent adjustments until after Week 4, so in Weeks 1-3 DVOA and VOA are the same thing. Please don't get all nitpicky about it. Most readers know what's up, and if you don't, I just told you!
The fact that DAVE currently is still counting our preseason projection as 90 percent of its estimate of team quality is another element of the "don't overreact" mentality that leads Football Outsiders to call the first few days after Week 1 "National Jump to Conclusions Week." If we only look at Week 1 performance, the best team in football right now is the San Francisco 49ers. It's actually not that strange to say that the San Francisco 49ers are the best team in the NFL according to DVOA. This happens a lot when I go back to break down past seasons to add the 1980s and 1990s to our database. What's strange is saying that "a team quarterbacked by Blaine Gabbert is currently the best team in the NFL according to DVOA." Again, don't jump to conclusions after one week. Just ask last year's No. 1 team after Week 1, the Tennessee Titans.
If you click on that link, you will see that last year after Week 1, we had a perfect symmetry of 1-0 and 0-1 teams, with no team that won its opener ranked lower than a team that lost its opener. That was very unusual, the first time it had ever happened going back to 1989. This year, things went strongly in the other direction. When you have 11 different games decided by one score, you're going to end up with some games where our ratings don't quite sync up with the final result. There were four of those games in Week 1. Which game you feel is most surprising depends on whether you are more surprised by teams ranking higher or teams ranking lower.
- At first glance, the most surprising DVOA result from Week 1 is the game we covered in Any Given Sunday, Handsome Jimmy G leading the Patriots to a surprising road win in Arizona. The Patriots are all the way down at 25th in our ratings after Week 1, while Arizona is up at No. 13. I'm surprised how big the gap is, but not that it is there. Remember, DVOA is measuring efficiency, and the Cardinals averaged more yards per play (5.8) than the Patriots (5.5). More importantly, the Cardinals were keeping on schedule down by down, successful on 51 percent of their plays, while the Patriots were successful on just 39 percent of their plays but were saved by a few big third-and-long conversions.
- However, the biggest gap in our ratings where the Week 1 loser has a higher rating than the winner is between Dallas (No. 9) and the New York Giants (No. 22). Those two teams have a DVOA gap of 29.5%, compared to 28.8% for the gap between Arizona and New England. And this gap is really strange because the Giants (5.9 yards per play) actually were much more efficient than Dallas (4.4 yards per play). The Giants had some longer gains, but the two teams were similar in success rate, and thus similar in offense and defense. Most of this gap is special teams.
- There's a smaller but still significant gap between Carolina (No. 11) and Denver (No. 19).
- The smallest "reverse gap" from Week 1 is less than 10 percentage points of DVOA, as both San Diego (No. 10) and Kansas City (No. 15) ended up with positive ratings after their overtime thriller on Sunday.
Win-loss records are scrambled even more in this week's ratings because both Detroit and Indianapolis (barely) had positive DVOA ratings for Week 1, while both Tennessee and Minnesota (barely) had negative DVOA ratings.
[ad placeholder 3]
We all should look at these Week 1 ratings with a healthy dose of skepticism. The method we use for normalizing each season's DVOA to 0% can lead to somewhat hinky results with a small sample, and even just two games on Monday night will move those baselines significantly. In addition, special teams ratings are going to be a little awkward for a few weeks until we can figure out just what the new kickoff touchback rule is doing to average kickoff distance and value.
Most importantly, just because we don't do opponent adjustments in Weeks 1-3 doesn't mean you can't add common-sense opponent adjustments in your brain when looking at the tables. San Francisco is No. 1, but we know how badly the Rams played on Monday night. Philadelphia is No. 2, but we had Cleveland projected as one of the two worst teams in the league. By the way, this common-sense adjustment also applies to DVOA's favorite team, the Seattle Seahawks. The Seahawks ended up third for the week even though they barely beat the Miami Dolphins. Yes, DVOA suggests that the Seahawks played better than the final score would indicate... but against the Miami Dolphins, another team we had projected to do poorly in 2016.
(For those who might ask: Yes, we've considered doing adjustments in the early part of the season based on our preseason projections. Trying to build and test that is on our to-do list. As I've mentioned numerous times, it's a really long to-do list. However, we are finally getting to some of the things on it! Exhibit A: The arrival of Football Outsiders Premium Charting Data. For those of you who purchased this subscription, we should have 2015 stats listed on the pages by Wednesday, and 2016 stats will begin to get updated on those pages after Week 2.)
The common-sense opponent adjustment should also apply to how you see Carolina and Denver after their Super Bowl rematch, in particular the rankings for the Carolina and Denver defenses. That being said, each year we point out that there's a difference between using "National Jump to Conclusions Week" to declare that a week ago we were all wrong about how good or bad some team is going to be this year, and jumping to conclusions about what a single win or loss means for that team's playoff chances. As you will see if you click over to our playoff odds report, the Panthers are this year's best example of the importance of a single win when you only get 16 chances. Carolina's DAVE rating is still the highest in the NFC South, but an 0-1 start combined with a road win for Tampa Bay means that the Bucs and Panthers are now effectively tied in mean wins in our simulations, and the Panthers only have a slight advantage over the Bucs when it comes to playoff odds, 45.4 percent to 44.0 percent.
* * * * *
Once again this season, we have teamed up with EA Sports to bring Football Outsiders-branded player content to Madden 17 Ultimate Team. Each week, we'll be picking out a handful of players who starred in that week's games. Some of them will be well-known players who stood out in DVOA and DYAR. Others will be under-the-radar players who only stood out with advanced stats. We'll announce the players each Tuesday in the DVOA commentary article, and the players will be available in Madden Ultimate Team packs the following weekend, beginning at 11am Eastern on Friday. We will also tweet out images of these players from the @fboutsiders Twitter account on most Fridays. The best player of each week, the Football Outsiders Hero, will require you to collect a set of the other four Football Outsiders players that week.
The Football Outsiders stars for Week 1 are:
- QB Matthew Stafford, DET (FOOTBALL OUTSIDERS HERO): Led all QB with 197 DYAR in Week 1 (31-for-39, 340 yards, 3 TD, 0 INT).
- ROLB Lavonte David, TB: Led NFL with 5 defeats this week: 3 TFL plus 2 tackles that stopped third-down receptions short of conversion.
- RG Laurent Duvernay-Tardif, KC: No sacks allowed; Chiefs RB gained 33 yards on 4 carries to the right and 69 yards on 14 carries overall.
- SS Tony Jefferson, ARI: Led team with 10 tackles, including 5 run tackles for 2 yards or less.
- WR Willie Snead, NO: Led all WR with 73 DYAR in Week 1 (9-for-9, 172 yards, TD).
* * * * *
Most stats pages are now updated with 2016 data. OFFENSIVE LINE and DEFENSIVE LINE will begin updating after Week 2. Snap counts will be updated later this evening, and drive stats and pace stats should be as well. Football Outsiders Premium (both the standard DVOA database and the new game charting data) will also be updated beginning after Week 2.
* * * * *
[ad placeholder 4]
TThese are the Football Outsiders team efficiency ratings through one week of 2016, measured by our proprietary Defense-adjusted Value Over Average (DVOA) system that breaks down every single play and compares a team's performance to the league average based on situation in order to determine value over average. (Explained further here.)
OFFENSE and DEFENSE DVOA are adjusted to consider all fumbles, kept or lost, as equal value. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season. As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.
Please note that there are no opponent adjustments in DVOA until after Week 4. (It's still listed as DVOA instead of VOA because I don't feel like going through and changing all the tables manually.) In addition, our second weekly table which includes schedule strength, variation, and Estimated Wins will appear beginning after Week 4.
DAVE is a formula which combines our preseason projection with current DVOA to get a more accurate forecast of how a team will play the rest of the season. Right now, the preseason projection makes up 90 percent of DAVE.
To save people some time, please use the following format for all complaints:
<team> is clearly ranked <too high/too low> because <reason unrelated to DVOA>. <subjective ranking system> is way better than this. <unrelated team-supporting or -denigrating comment, preferably with poor spelling and/or chat-acceptable spelling>
94 comments, Last at 17 Sep 2016, 4:45pm
#87 by Bobman // Sep 17, 2016 - 1:38am
It sure does feel like old times again.
As soon as McAfee, Vinny, and Quan Bray get arrested for transporting Lithuanian midgets across state lines for immoral purposes (with hydrogen peroxide, some bleach, and a sh!tload of lime, to quote Matthew McConaughey), the ST rank will end up where it used to reside as well.
I suspect when opponent adjustments kick in, Detroit will be relinquishing their #1 offensive ranking.
#65 by Dave Bernreuther // Sep 14, 2016 - 8:06pm
I know I'll be accused of being a Grigson apologist, but I'm not that upset* at how bad this team's defense is this year (and is bound to continue to be). It has long been the case that they need a way to get a real pass rush on a real QB with 4 men, without which they're screwed. While it's very easy to just yell from the rooftops "he hasn't addressed [insert position group here]", it's not like the 2015-16 offseason really gave him all that much of an opportunity to address their pass rush using sane means. (See: Olivier Vernon's contract.)
The best pass rushers were gone in the draft, there weren't any reasonable options in FA, so the decision to roll the dice on coaching and development of Parry/Anderson... while not the world's greatest gamble, it's not what I'd call terribly negligent either.
(Same can be said of the O Line. Luck is going to take plenty of hits of his own accord, and all this "he didn't address the OL!" outcry is overblown, given that there's very little OL talent, even the Broncos, Seahawks, Panthers, and Patriots had glaring OL issues last year and did just fine... and even if he'd been given the chance to draft a universally heralded "stud" OT along the way, there's no sure thing. Take, for instance, the 2013 draft, for which he was pilloried for taking Bjoern Werner late in the first. Picks 1 and 2 were consensus top choice material. Offensive Tackles. Can't-miss. And where are they now? Can you even imagine how much shit Ryan Grigson would be taking if he had chosen one of those guys and gotten these exact same results?
(All that said... The Richardson trade alone is a fire-able offense. Arrogance and tantruming is a fire-able offense. And this defense, especially now that they're already 9 deep in the secondary, literally hurts my eyes to watch, and I have no optimism whatsoever that it will get any better. I'm looking forward to the Thanksgiving night game against the Steelers purely for the spectacle; I HOPE Ben breaks Tarkenton's passing yardage record.
I just think people sometimes go a little bit overboard with the piling on. The D sucks without a pass rush. They did OK with blitzes against bad QBs, Manusky had two hands behind his back and got scapegoated unfairly, but there wasn't really an opportunity to change anything but the coach. A new GM? What would you have had him do differently this offseason?)
*- OK, I'm upset. It is sickeningly bad with little hope of improving. The only way I'd pay to see it is if I was rooting against it. I'm just, realistically, not entirely sure what opportunity there was to improve it.
Now watch. They'll go out and somehow totally shut down CJ Anderson and the Broncos and win next week. (Although having superior talent and coaching yourself out of a win against the Colts is kind of a Kubiak staple, so none of us should be surprised...)
#68 by theslothook // Sep 15, 2016 - 1:43am
My biggest problem with grigson has not been the draft necessarily. They've been picking high generally since the day Luck came onto the team and even last year - they were still picking in the late teens; so naturally its tough to acquire top shelf talent that way.
My biggest problem is how hes approached free agency. He's essentially gone after mid tier veterans year after year. It starts with people like d'qwell jackson, laron landry, gosder cherilus, trent cole, arthur jones, eric walden etc etc.
Its not that teams shouldn't go after mid tier veterans. NE, Seattle, and Arizona have all been successful at it. But it needs to have a strong core for those mid tier guys to really shine. The colts defense lacks that core. I would have much rather paid a huge amount of money to go after a prized free agent than a medley of middle tier veterans. Now you might say there were no veterans to acquire. I would have been happy to see the colts try for malik jackson. I would love to have seen them try to acquire josh norman.
And of course, trent richardson. I sort of agree with Will. Its a pretty big lack of awareness about today's game to really consider trading a first rounder for richardson. It also should have sent a signal that the browns were willing to shop the third overall pick one year after he was drafted.
#76 by Dave Bernreuther // Sep 15, 2016 - 12:26pm
With regard to the mid-tier stuff, I think that's largely part of a cap management strategy. If there was a perfect fit (if they thought that Malik Jackson had been that), I don't doubt they'd be willing to shell out. They paid quite handsomely for Cherilus, which probably wouldn't have been a disaster if he was healthy, and all the other ones were low-risk contracts. I'm not remotely upset that guys like Trent Cole and Andre Johnson and others suck because those contracts were cheap and easy to get out of. You could sort of see his evolution in that regard after the first year. Learning from his mistakes, so to speak.
(To me, the biggest bit of cap mismanagement was keeping Freeney and a $15m cap hit in a D that didn't suit him on a team not expected to contend The biggest non-TRich payment-to-talent mismatch strikes me as Jackson, who is for some reason STILL running around slowly out there completely unable to cover anybody... I don't care how good a leader he is or how well he knows the D, he's a liability.)
I think Jones, post-suspension, actually can be kind of a difference maker. I disagree on Norman being the right fit in this type of D. (I think there's a reason the Panthers were totally fine letting him walk.)
No argument here on Richardson.
#88 by Bobman // Sep 17, 2016 - 1:51am
The team has been horribly unlucky with injuries, which has not helped. Gosor the Gozarian Cherilus and Art Jones seemed like reasonable moves at the time. The CB who was a 3rd rounder in 2013 (name forgotten) could never stay healthy and was finally cut.... And who knows, Art Jones might still contribute if he can stay on the field, which is a huge IF. Then Parry and Anderson were both hurt last year, right? man, that DL is a medical Bermuda triangle.
What is ironic is that some of the old geezers on D have turned out to be their highest achieving performers--nobody expected great things out of Mike Adams, but he's a ray of sunshine back there and has been tough and durable. D'Quell Jackson? Same thing. Erik Walden, who destroyed Indy in one game vs Green Bay (like his only good game ever) was heralded as a run-stopper (warning: 1950's football ahead!) and the signing was ridiculed by smarter folks than me (Nate Dunlevy) but he's actually done okay--he's no sack-master but he shared the team lead last year (with fellow geriatric, Rob Mathis). So Walden, when he's not head-butting helmetless opponents, is doing okay--far better than the entire draft class of 2013.
And recent veteran contracts like Andre Johnson's, I agree, show some GM maturity. If it works, great, if not, it's a one-year rental.
Not sure I ever liked Grigson; same for Pagano. Is it too late to hire Bruce Arians back?
#85 by Winterguard78 // Sep 16, 2016 - 2:50am
Contrary to national media narrative, Eric Fisher is an ascending LT who is already at least league average (he was PFF[grain of salt]#1 rated OT week 1) and on a second contract paying him a little more than $10 million per year. That 2013 Draft class was brutal (especially the 1st round) but to say Grigson would be taking more heat if he somehow drafted Fisher than Bjorn Werner (who is literally a punchline) is kind of silly.
I think Pagano is a good coach, but from a few hundred miles away (although I worked in Naptown for a year in the early00's) even if him and Grigson were 100% on the same page I don't see how he's still employed. He's like the GM version of Jeff Fisher except Fisher actually did some nice things at one point (admittedly he's had longer to stumble into success than Grigson) You can't give him any credit for drafting Luck because it was so obvious Matt Millen couldn't have screwed it up and the best players he drafted that have made it to 2nd contracts got overpaid (I'm thinking mainly of Hilton and Dewayne Allen- Luck's worth $30 million a year if that's what it takes to keep him)
You would think he would have at least signed a few veteran minimum younger players after FA frenzy died down and worked the waiver wire like crazy for DB's and edge players, maybe tried a low risk trade like NE did with Mingo or talked to Paul Kruger or God forbid Greg Hardy. Anyhow, no animosity towards the Colts at least this year except when they play KC.
#3 by PaddyPat // Sep 13, 2016 - 7:59pm
These are some weird numbers, though I suppose they typically are after Week 1. One is struck by the enormous disparity in team quality when trying to evaluate some of them. The Seattle/Miami numbers jump out in particular.
#14 by Perfundle // Sep 13, 2016 - 9:51pm
The biggest difference in the Seattle Miami game can be seen in the third down stats. Seattle put itself into much better third down situations than Miami; it just couldn't take advantage of them. Seattle had less than 5 yards to go on 7 of their 16 third downs, while Miami had less than 5 yards to go on 0 of their 14 third downs.
The reason for this disparity is that Seattle massively outplayed Miami on second down. They were pretty even on first down, but Seattle averaged 4.78 yards on second downs compared to Miami's 0.65.
#7 by Perfundle // Sep 13, 2016 - 8:36pm
If you want to talk about being ranked too high, Gabbert's Total QBR against the Rams was a freaking 91.1. I want to know who watched him constantly missing his receivers or staring them down and nearly getting intercepted a few times and thought that he played nearly a perfect game.
#51 by Mountainhawk // Sep 14, 2016 - 1:37pm
They mark throws as overthrown, underthrown, deflected, or dropped, which isn't really all that subjective except on very edge cases. They aren't grading the execution of a play, just what happened on the pass so that every incompletion isn't scored the exact same.
#54 by deus01 // Sep 14, 2016 - 2:55pm
I don't think they have humans grading plays anymore but instead link each play to historical analysis of similar situations. Brian Burke was explaining some of this and ESPN posted an article describing it and the changes.
I don't think it's really much different than DVOA in the sense that we know the same inputs (results of plays) that go into a black box and sometimes spit out crazy numbers.
#8 by theslothook // Sep 13, 2016 - 8:37pm
I'm trying to remember a worse week 1 performance I've seen. Maybe that jets pats in 07, but the pats were one the greatest teams of all time that year. Does anyone have any contenders they can remember off the top of their heads?
#10 by Perfundle // Sep 13, 2016 - 9:06pm
Blaine Gabbert can, because he was on the losing end of it. The Chiefs beat Gabbert's Jaguars 28-2, and Gabbert played nearly identically to Keenum last night.
Gabbert 2013: 16-35 for 121 yards, 0 TDs and 2 ints
Keenum 2016: 17-35 for 130 yards, 0 TDs and 2 ints
Also, here's a comparison of their drives, ignoring end-of-half drives:
Jaguars 2013: punt, punt, punt, int, punt, punt, punt, punt, punt, punt, int, punt, punt, downs
Rams 2016: punt, punt, punt, int, punt, punt, punt, punt, punt, punt, int, punt, downs
The Jaguars had a VOA of -118.2% in that game, which is considerably worse than that of the Rams.
#90 by Bobman // Sep 17, 2016 - 2:00am
The short answer?
The season fan who just spent a bundle to watch either team, the one led by Gabbert or the one led by the guy who got beat out by Gabbert.
Because being Gabbert or the other guy, you're still raking in good money, you have minor sponsorship deals, and the adoration of dozens. But that poor fan is paying for the privilege of having his hopes and dreams crapped on weekly.
#9 by Denverite // Sep 13, 2016 - 8:45pm
Curious (not in a passive-aggressive sense) why Carolina's offense grades so much more efficient than Denver's (and ditto re: defense). Denver outgained them per play (5.4 to 4.9) and per possession (34.1 to 33.3).
#71 by Aaron Brooks G… // Sep 15, 2016 - 8:32am
Even if it did, it would be a small effect.
DVOA, for better or worse, is like an integral of your success on a play-by-play basis. It will tend to filter out brief noise blips. Even the difference between stellar and incompetent clock-management doesn't yield a large whole-game difference in play-successes.
#27 by Eisenlords // Sep 14, 2016 - 9:11am
About the patriots vs Cardinals game.
Cardinals outperformed the patriots on both offense and defense. But patriots dominated special teams. I would agree with all 3 of these statements, but not to the degree that DVOA states.
Offense - Ari 10th, NE 22nd 30.6% in favor of Arizona
Defense Ari 17th NE 26th 23.3% in favor of Arizona
Special Ari 32nd NE 2nd. 25.1% in favor of New England
I feel that the system is underating the importance of special teams, at least in this one instance, or the opposite overrating the gap in offense and defense between the two teams. The patriots were able to win this game because of the net yard advantage from pinning Arizona deep.
Patriots kick from the 35, high kick not in the end zone - Arizona has to return the ball. Arizona starts at the 8 due to a penalty. That's a 57 yard gain.
Arizona kicks from the 50 (Collins penalty on previous play) they kick the ball 20 rows high into the stands. Patriots start at the 25. That's a 25 yard gain.
This sort of thing happened all game long and is reflected in the special teams rating (plus a missed FG).
The issue is that in this particular case DVOA seems to think patriots fumbles on offense, and allowing touchdowns to Larry fitz on defense was more costly and less efficient than the amazing field position produced by special teams.
Will be interesting to see in the future if special teams deserves a larger peice of the pie, with the new kick off rules.
Also curious, how does this 25.1% advantage on special teams compare historically? It has to be up there as one of the most lopsided special teams scores.
#29 by Digit // Sep 14, 2016 - 9:26am
The new kickoff rules that puts a touchback at the 25 yard lines was really taken advantage of by Belichick with Gostkowski at his disposal. Instead of booming touchback after touchback, Gostkowski aimed his kicks around the 1-5 yards area, encouraging more returns. This also resulted in ST putting Arizona at an average of the 18 yard line - or an average of 7 yard difference compared to where the new touchback would place the ball.
Since Belichick didn't do this when touchbacks were at the 20 yard line, I'm guessing that DVOA is going to be a little wonky if you're trying to compare this historically.
#39 by Rick_and_Roll // Sep 14, 2016 - 11:05am
Does the new touchback rule mitigate or even reverse the kicking advantage that factors into kick-offs in Denver? It seems it may be harder to keep the ball from being a touchback, especially for road kickers.
#43 by Travis // Sep 14, 2016 - 11:52am
The new kickoff rules that puts a touchback at the 25 yard lines was really taken advantage of by Belichick with Gostkowski at his disposal. Instead of booming touchback after touchback, Gostkowski aimed his kicks around the 1-5 yards area, encouraging more returns.
In contrast, the Colts, kicking off from the 50 with 37 seconds left, chose to kick the ball into the stands, saving the Lions time and (probably) yardage.
#45 by Tundrapaddy // Sep 14, 2016 - 12:07pm
I noticed that, too. After hearing all preseason about kickers apparently playing the revised rules to kick 'higher', thus providing the coverage unit more time to get under the ball, the Colts opted instead to boom it well through the end zone.
No time off the clock, ball at the 25. Could've had 1-2 seconds off, and ball somewhere inside the 15. That seemed like a massive oversight.
#61 by jds // Sep 14, 2016 - 4:50pm
So moving the kickoff to the 35 to increase touchbacks (and kill kickoff coverage), sort of worked, until teams started bringing it out from the endzone. So in an effort to encourage not bringing it out from the endzone, the league moves the spot to the 25. With the result that we will now have higher, shorter kicks, and therefor more (and better) kickoff coverage. So no rule changes are going to eliminate the kickoff, so if the NFL truly wants to kill the kickoff, they are going to have to expressly do that.
#62 by Digit // Sep 14, 2016 - 5:30pm
At some point, really, they should just put Bill Belichick on the competition committee and ask a simple question: "What would you do if this rule came into existence?"
And why do I think this particular idea about kickoffs to the 25 was met with great enthusiasm by Jeff Fisher (who actually -is- on the competition committee)? It's not like his team would be scoring enough to know how to take advantage of this...
#72 by Aaron Brooks G… // Sep 15, 2016 - 8:34am
Frankly, they should ask all the coaches that.
Ideally, they should really wargame the various rules. Set up different situations, and have the different staffs play through how they would handle it. Now, this can be gamed too, but it would at least weed out the most obvious flaws.
#66 by Dave Bernreuther // Sep 14, 2016 - 8:14pm
A lot of people gave Pagano shit for using a timeout a few plays before the score when:
- Caldwell could easily have been the bigger fool for not using his.
- you can't assume the score
- apparently they did have an issue with personnel/communication
- 37 seconds is still not very much time
(still not saying he should've called it. But it's not exactly Caldwell in 2010 either. There's a chance that they actually did need it.)
And yet almost nobody gave him shit for having a freakish athlete of a punter/kicking specialist that has been known to routinely beat QBs in throwing accuracy contests and could drop a punt in a bucket from 60 yards... whose only job all summer was literally to practice kicking and know how and where balls might land after he kicks them... just blast it out of the end zone and let the Lions start a drive 35 yards away from a make-able indoor field goal without costing them any time.
Squib the damn thing and cover the kick. Your special teams are about an order of magnitude more reliable than your defense. Trust them, not guys you probably can't even name on D.
To me, that was lunacy. What's the absolute worst that could happen on a slow squibber... it goes into the end zone? Hmm.
#77 by Dave Bernreuther // Sep 15, 2016 - 12:28pm
I'd say that that's not exactly the kind of thing you should ever plan on being possible, but then again, this is the Colts we're talking about.
Meanwhile, on the other end of the coaching spectrum...
Hmm, not like there were any similarities there in the game situations. Nope, none at all...
#91 by Bobman // Sep 17, 2016 - 2:04am
@ 44 and 46, You can't simply teach coaching skills like that. Either you got 'em or you... uh, hmmm, what was I saying again?
Oh yeah: Colts Mantra since Grigson and Pagano arrived: Run the Ball. Stop the run. Neither of which they do competently, but at least it's easy to remember when the press and players ask.
Their ST strategy appears to be equally relevant to today's game.
#31 by nat // Sep 14, 2016 - 9:55am
I think you're missing an important characteristic of VOA: its per-play nature.
Per play, the Arizona offense did much better than New England. But if you look at the drive stats, New England had the advantage in yards per drive, and was as close to equal as you could get on the DSR (moving the chains) stat. It was only in turnovers that Arizona had a meaningful advantage. Meanwhile, it was Arizona with the better average starting position for its drives, thanks to the fumbles.
You're not really wrong about the importance of special teams. You're not really right about it either. Most of the disparity between VOA and the game's result was due to the Patriots managing to have effective drives despite being less efficient on the early downs and thus less efficient per play overall.
In the end, the game was decided by a missed field goal from a distance that had historically been uncomfortable for the kicker. The game could have gone either way. And that's pretty much what the three major drive stats (yds/drive, DSR, and TO/drive) plus the kick and punt results would have led you to expect.
#33 by Eisenlords // Sep 14, 2016 - 10:18am
I guess my point is that overall DVOA or VOA is misleading (when I say misleading I mean it was not a good indicator of who would win) in this game when offense and defense have large margins of 30% and 20%, in favor of the cardinals, and then when you look at it a little closer you can say that in actually Arizona wasn't THAT much more efficient than New England, at least to the point where the discrepancy wasn't large enough to win them the game.
Currenty, this game is an outlier, but it is one point of data saying that per-play efficiency of offenses and defenses is not as strong of an indicator of who will win the game as one might think. Or, maybe it's just that special teams should be weighed heavier against these factors to better reflect the actual game result.
Overall it will be interesting to see if we find out this year that DVOA should be weighting special teams efficiency heavier in the formula than it currently is.
#34 by Thomas_beardown // Sep 14, 2016 - 10:21am
VOA isn't really designed to be useful as a single game descriptive stat. It's designed as a predictive stat (with some description) to determine the "true" quality of a team over hundreds of plays.
#35 by nat // Sep 14, 2016 - 10:33am
You need to be careful about comparing VOAs. If you compare both offensive and defensive VOAs in a single game in the way you did, you are effectively double-counting every play from scrimmage. It's not quite so bad for special teams, since much of the action there is counted just one way or the other.
It's subtle. Having both offensive and defensive DVOA works the way it does because it's handy for comparing how teams did in different games, where you wouldn't be double-counting anything.
A better way is to say that, per-play, Arizona was better with the ball than New England by about 23% to 31%. That's comparing either offensive or defensive VOA, but not both. Much of that was negated by special teams play. Much of it was rendered moot by New England's strong third down offensive play. Thus, a close game decided on a field goal attempt from an iffy distance.
#28 by Aaron Brooks G… // Sep 14, 2016 - 9:22am
How many FGs and XPs did MIN miss, and they still aren't last in ST?
A modified version of that argument is also true for DET. 6th in ST, even though a shanked XP nearly cost them the game? That's a far cry from the exploits of RoboKicker.
#74 by JoeyHarringtonsPiano // Sep 15, 2016 - 10:49am
"A modified version of that argument is also true for DET. 6th in ST, even though a shanked XP nearly cost them the game? That's a far cry from the exploits of RoboKicker."
Sam Martin had an awesome game, and the coverage units were solid. The Colts had lousy field position most of the game....which is why I'm surprised Detroit's defense is 28th, not 31st or 32nd.
#92 by Bobman // Sep 17, 2016 - 2:09am
JHP, Anyone who watched that game surely knows why Detroit's D was not 32nd at least. That spot was claimed and reserved already.
I'm truly hoping that if Indy gets a healthy secondary back they can crawl out of the cellar, but for the time being... well let's just say, the utilities are in their name at the bar in the man-cave has been stocked for them.
#38 by Matt_Dodges_Di… // Sep 14, 2016 - 11:03am
So I'm a bit confused here. I would expect that, given there's no opponent adjustments here, the offense/defense for opponents would either be identical (if team A's off was 5%, team B's def would be 5% as well) or that there would be some relatively fixed difference between the two.
I understand the per-play nature of this whole thing, but why then wouldn't the value be the same for both on each play?
#44 by nat // Sep 14, 2016 - 11:52am
Just to add to that point:
This can affect both the success attributed to a unit in a particular game and to the average VOA compares it to. So corresponding offensive and defensive VOAs won't exactly match even if there are no examples of these special penalties in the game in question.
#48 by hrudey // Sep 14, 2016 - 12:34pm
Jets are ranked at #25 (32.2%) having surrendered 23/30 for 366 with 1 TD and 1 INT, but registered 7 sacks.
Jaguars are ranked at #26 (34.6%) having given up 20/34 for 199 with 2 TD and 0 INT, and only 1 sack.
Are sacks really this big a thing? Apologies for not formatting this inquiry in the zlionsfan format, but I actually am more curious than anything.
#64 by Perfundle // Sep 14, 2016 - 6:26pm
I imagine sacks on third down aren't as bad as sacks on first and second down, since the drive is usually over with a sack or an incompletion, and only one of Cincinnati's seven sacks was on third down. Also, and this is looking at their offense as a whole, Cincinnati was very boom-or-bust on the first two downs. Of their eleven third downs, They had only one short one (less than five yards), and seven long ones (more than ten yards).
#55 by Raiderjoe // Sep 14, 2016 - 3:28pm
no way are Raiders 16th best team. went into neow Orleans and beat drew brees-led tema. imprerssive win.
my official power ranks abfter week 1 are:
6-30 orther teams
ramns right now equivalent of dying hyena with mange. Hyenas ugly enough when healthy like rams are crappy enough at best of times, btu right now with sideways tossing c. keenum in charge tema is like diseased hyena.
#57 by theslothook // Sep 14, 2016 - 3:47pm
I get that it was against No in the dome, but the raiders D looked horrible. They could not get any pressure and all parts of their pass D looked awful. As long as you can block Mack, you can throw on the Raiders.
#58 by dmstorm22 // Sep 14, 2016 - 4:08pm
To be fair, the Saints at home is a tough task.
I expect their defense to get better over the course of the year as well. Their schedule also seems rather easy when looking at the opposing offenses they may face.
#69 by theslothook // Sep 15, 2016 - 1:45am
I'm not convinced they have much talent outside of mack. Maybe one of their opposing rushing ends will work out and I think irvin is a solid player. But their linebackers and corners are especially weak.
#70 by panthersnbraves // Sep 15, 2016 - 3:17am
Why exactly is Carolina's ST so low? They had the missed FG at the end after getting iced, and the one KO that was returned was poor, but that was more due to a great kick than a poor return. On the good side, Lee launched a 76 punt that flipped the field. I sat out Sunday's games (because someone was going to graze a QB's helmet after a throw, get a penalty, and I was going to end up buying a new TV) - but I would have figure that for 22nd, not 31st...
#79 by Matt_Dodges_Di… // Sep 15, 2016 - 2:48pm
Another generic DVOA question: on the "FO basics" page, it says that a team's value is basically 4 parts off, 3 parts def, and 1 part ST. Why isn't total DVOA reported that way instead of straight addition between the three?
#80 by dmstorm22 // Sep 15, 2016 - 3:09pm
What they are really saying is most teams have their spread between offensive DVOA, defensive DVOA and ST DVOA grade out that way.
It's why around the end of the season, you'll often see the best teams at something like +20% offense, -15% defense, 5% ST, or something to that effect.
My understanding is it is not part of the formula to account for this - there is no actual weighting.
#81 by Thomas_beardown // Sep 15, 2016 - 3:55pm
No, I'm pretty sure total dvoa is (4x(offense) + 3x(defense) + 1x(st))/7.
I remember lengthy discussions about the possible under/over ratedness of Lovie Smith Bears who were so good at Def St and not so good at Off. Also the ridiculous Norv Chargers year when they managed to sink their season entirely because of St.
(end result is that I don't think DVOA should be treated linearly).
#86 by nat // Sep 16, 2016 - 10:20am
For special teams, it's as simple as "They designed it that way". ST DVOA is calculated differently from the offense and defense, and then converted into a percentage equivalent such that it can be added without any scaling factor.
For Offense and Defense DVOA, they represent opposite sides of the same coin. Logically, you'd expect them to be on the same scale, and that's how total DVOA treats them.
The 4-3 ratio we hear about comes from studies that did a regression using offense and opposing defense season average points per game as inputs and offense points scored in each game as the output. I wonder how much of the effect was simply that offenses have more impact on how many drives there are in a game.
Looking at 2015 regular season drive stats, it looks like offensive ToP per drive is a bigger determiner of how many drives/game a team has than defensive ToP per drive. This effect is more than just the offensive quality drive stats (Yds/Dr, TO/Dr, DSR). It's a matter of play calling and pace as well. Those are things that offenses can control.
If anything, it looks more like defensive quality drive stats (Yds/Dr, TOs/Dr, and DSR again) have a bit more to do with points/drive surrendered than offensive quality drive stats have to do with points/drive scored.
This could use some closer study with more data, or using DVOA instead of drive stats. I suspect that we would find that offensive and defense DVOA have about equal impact on per drive scoring, and that the 4-3 ratio is an artifact of using game stats instead of drive stats in those studies.
I'd be happy to hear otherwise. Or to have my ideas confirmed. It is a curious thing.