Can Seahawks Keep Up High Offensive DVOA?

Seahawks QB Geno Smith and RB Rashaad Penny
Seahawks QB Geno Smith and RB Rashaad Penny
Photo: USA Today Sports Images

NFL Week 4 - The Buffalo Bills remain No. 1 in DVOA after Week 4 with a win over the Baltimore Ravens. It was a close win, so the Baltimore Ravens remain at No. 3. In between them are the Philadelphia Eagles at No. 2. The Eagles switch places with the team they beat this week, Jacksonville, which is now fourth. There's a pack of teams nearly tied there from fourth through seventh: Jacksonville, Tampa Bay, Dallas, and San Francisco. Then Kansas City is eighth.

The placement of Tampa Bay and Kansas City is a bit unexpected after the Chiefs looked to be dominating the game between the two teams this week. Surprisingly, Tampa Bay actually ended up with the higher single-game DVOA in this one, 29.3% to 9.5%. This is where we get into the usual arguments about garbage time, but our research has showed that garbage time performance, although often down-weighted, is still predictive. The Chiefs had the higher Post-Game Win Expectancy (59%) because they ran more plays, but the Buccaneers were actually more efficient. (Tampa Bay ended up with 6.4 yards per play compared to just 5.4 yards per play for the Chiefs.)

There weren't a lot of really huge wins in Week 4. Every game in the NFL except for Buccaneers-Chiefs was within one possession at some point in the fourth quarter. Philadelphia was the only team to put up DVOA over 50% this week, with 63.7% against the Jaguars. However, there's still plenty of movement in the middle of the DVOA rankings for a couple of reasons. First, because teams are so close together, so small changes in the ratings mean larger changes in the rankings. Second, because this week we introduce opponent adjustments for the first time this year.

For example, Cleveland falls from seventh to 12th not only because the Browns lost to Atlanta but also because of adjustments for playing this year's third-easiest schedule so far. Denver, Pittsburgh, and Detroit each drop five spots. Houston drops six spots from 25th to 31st. Carolina, whose schedule so far ranks 29th, falls all the way from 18th to 28th after a sizeable loss to the Arizona Cardinals. The Panthers are the only team under -50% DVOA for the week.

Likewise, there are some big moves upward. The Chargers move from 15th into the top 10 after beating the Texans. Tennessee moves up six spots to 20th after beating the Colts. The Saints lost to Minnesota but have the highest DVOA for that game, 22.4% to -18.1%, and move up five spots to 22nd.

Is Seahawks Offense Real?

The biggest move up this week belongs to the Seattle Seahawks, who climb eight spots from 21st to 13th. Some of that is their win over Detroit, and some of it is the effect of a top-10 schedule now being added into adjustments. The Seahawks are shocking everyone with their powerful offense so far this season, which is basically tied for second in DVOA behind Baltimore. (Technically third, as changes in opponent adjustments based on how San Francisco played last night knocked them a rounding error behind Miami.) Seattle also ranks 31st in defensive DVOA. We knew the Seahawks defense was going to be bad this year. We thought their offense was going to be bad too. It's not, at least not so far, and that puts Seattle 13th overall going into Week 5.

Is there any way that Geno Smith and the Seahawks can keep this up? I think it's pretty unlikely. To compare Seattle to other recent teams in a similar situation, I looked at our preseason projections going back to 2010 and compared each year's projections to how good teams actually were in DVOA through four games.

Now, there's two ways you can do this. You can look at the difference between actual DVOA and projected DVOA. The problem with that is that some of the biggest "overachievers" will be teams that were already projected to be good and instead were phenomenal early. For example, one of the biggest differences between early-season DVOA and projected offense belongs to the 2013 Denver Broncos in Peyton Manning's first year there because they were over 50% through four games! But that's not really a team that "overachieved" in quite the same way as the 2022 Seahawks.

So instead, I looked at the difference between actual offensive DVOA rank and projected rank. Based on that measure, the 2022 Seattle Seahawks are tied with the 2011 Buffalo Bills as the biggest early-season offensive overachievers in recent years. Here's a list of those biggest overachievers, along with the starting quarterback through four games and how well that offense did the rest of the year:

Biggest Offensive Improvements
from Preseason to First 4 Games, 2010-2022
Team Year QB Preseason
Projection
Rk Offense
after 4 G
Rk Offense
Rest of Year
Rk
SEA 2022 G.Smith -9.9% 30 20.6% 3 -- --
BUF 2011 R.Fitzpatrick -10.7% 30 32.3% 3 -10.3% 24
CLE 2014 B.Hoyer -12.2% 29 20.8% 4 -21.7% 29
DEN 2016 T.Siemian -10.0% 30 9.4% 6 -18.8% 29
OAK 2011 J.Campbell -9.7% 28 26.6% 4 -5.1% 17
BUF 2015 T.Taylor -12.4% 30 10.7% 7 10.6% 8
NYJ 2010 M.Sanchez -8.6% 28 24.3% 5 -5.3% 23
BUF 2020 J.Allen -8.2% 27 19.5% 5 12.5% 5
JAX 2020 G.Minshew -8.6% 28 16.9% 6 -22.9% 32
LAR 2017 J.Goff -8.3% 27 14.8% 6 10.5% 6
PHI 2016 C.Wentz -6.8% 26 10.1% 5 -8.0% 22
PHI 2013 M.Vick -8.4% 25 19.0% 4 24.3% 2

So the question we have to ask is whether the Seattle Seahawks are going to keep this up. Looking at this table, I think it is unlikely. Yes, the film shows that Smith is playing very well so far. But for the most part, the teams here that are similar to this year's Seahawks -- with a journeyman former backup quarterback pressed into duty as the starter -- reverted to expectations for the rest of the season. This is also true of some of the teams that would come next on this table, including the 2018 Denver Broncos (with Case Keenum). The teams that tended to keep their early-season offensive power for the entire season are the teams with young developing quarterbacks: not just the 2020 Bills and Josh Allen, but the 2011 Panthers (Cam Newton), 2019 Ravens (Lamar Jackson), and 2018 Chiefs (Patrick Mahomes) are just below the teams on this list. You also get some teams with snazzy new schemes, such as the Chip Kelly 2013 Eagles, or teams with both young quarterbacks and new schemes, such as the 2017 Rams.

There's one team that makes me think there might be some hope for Smith and the Seahawks, and that's the 2015 Buffalo Bills. Tyrod Taylor wasn't quite the same kind of journeyman as Smith, as he never had the chance to start in Baltimore, but he is an example of a veteran quarterback (fifth season) with very low expectations who had a surprisingly good offensive start -- and then kept it up all year.

The reason I don't think that Geno Smith is going to do what Tyrod Taylor did in 2015 is that we've already seen what Geno Smith does in the Seattle offense. Last year, with the same receivers and same offensive coordinator, Geno Smith had -3.2% passing DVOA in three starts. Yes, the offensive line is much improved this season, but Geno Smith is still the same guy he was a year ago. Now, that Geno Smith of a year ago was better than expected, so perhaps Seattle is better on offense than we had projected before the season. But I really don't think they're this much better, or even close to it. I expect their offensive DVOA to end the year ranked maybe in the high 20s, similar to the 2011 Bills.

No team this year has fallen short of its preseason offensive projection to an extent anywhere near how much Seattle is outperforming its projection. The closest teams are Cincinnati (projected 11th, currently 27th) and, ironically, Denver (projected ninth, currently 24th).

Denver is also the team that is outperforming its defensive projection by the biggest gap in rank. We had them projected all the way down at 29th and they are currently eighth in defensive DVOA. In the other direction, the biggest gaps between defensive projection and actual early defense belong to the New York Jets (projected seventh, currently 25th) and Cleveland (projected 12th, currently 30th).

Early Opponent Adjustments

As noted above, this is the first week we are including opponent adjustments in our ratings. They're still only at 40% strength, since at this point a large part of the opponent adjustment is how good you are yourself. (After all, you are one of the four opponents played so far by each of your opponents!) The Washington Commanders have played the NFL's toughest schedule through four games, including three teams currently in DVOA's top six (Philly, Jacksonville, and Dallas). The Miami Dolphins are right behind them, having of course beaten two of our top three teams (Buffalo and Baltimore).

The easiest schedule so far belongs to the Houston Texans, but the average DVOA of Houston's opponents is nowhere near as negative as the average DVOA of Washington and Miami's opponents is positive. Arizona, Cleveland, and Carolina have also all had easy schedules so far, but their schedules get a lot harder after this. All three teams rank in the top 7 for the average DVOA of their remaining opponents.

How do opponent adjustments affect the player stats so far?

  • Carson Wentz has played the most difficult schedule of pass defenses, with Philadelphia, Dallas, and Jacksonville all in the top seven. Applying opponent adjustments takes his DYAR from horrible (30th without adjustments) to slightly less horrible (27th). Matthew Stafford has also played a tough schedule of pass defenses so far.
  • Josh Allen has played the easiest schedule, which is why he is currently only eighth in passing DYAR. He would be fourth without the opponent adjustments. Baker Mayfield has also played an easy schedule, which is a little scary... he's actually been worse than it has looked.
  • James Conner of Arizona gets the biggest DYAR boost from his schedule, with three games so far against top-10 run defenses (Rams, Chiefs, and Panthers). Joe Mixon and James Robinson also get bumped up.
  • Rashaad Penny and the two Green Bay running backs have the biggest DYAR penalty due to schedule. Penny has played the best (San Francisco) and worst (Detroit) run defenses so far this season. All four of Green Bay's' opponents so far have below-average run defense, in particular the New England Patriots who are 10th in pass defense but 31st in run defense right now.
  • The biggest DYAR boost for wide receivers belongs to Tyreek Hill while the biggest DYAR penalty from opponent adjustments belongs to Stefon Diggs. Diggs is No. 1 in receiving DYAR anyway. Hill is seventh.

* * * * *

Football Outsiders playoff odds, snap counts, and the FO+ database are now all updated through Week 4.

A reminder that all our free stats pages, including DVOA and player position stats, require registration to view. This is not a paywall*! You only need to register (for free) and then log in to the site to view these pages. While you're at it, you can get a seven-day trial of FO+ and check out the FO+ features like a deeper DVOA database, weekly fantasy projections, fantasy football research tools, Derrik Klassen's new All-32 game preview column, and picks against the spread.

*New: It's not a paywall, except on Mondays! As of this week, we are now posting DVOA data on Monday instead of waiting until Tuesday as in the past. If you want to see data from the current season on Monday, including all of Sunday's games, you will need to be an FO+ subscriber. On Tuesday morning when we post the update with Monday Night Football added in, all of the free stats pages become free again. The exception will be snap counts, which will still be available to everyone Monday but now will be updated earlier.

* * * * *

This is the Football Outsiders Top 16 through four weeks of 2022, 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). Click here for the full table.

OFFENSE and DEFENSE DVOA are adjusted for opponent and performance indoors and consider all fumbles, kept or lost, as equal value. Opponent adjustments are currently at 40% strength. 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.

DAVE is a formula which combines our preseason forecast with current DVOA to get a more accurate projection of how a team will play the rest of the season. DAVE is currently 70% preseason forecast and 30% actual performance. DAVE ratings are adjusted for various quarterback injuries based on an estimate of how much of the season will be played with backup quarterbacks.

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>

RK TEAM TOTAL
DVOA
LAST
WEEK
TOTAL
DAVE
RANK W-L OFF.
DVOA
OFF.
RANK
DEF.
DVOA
DEF.
RANK
S.T.
DVOA
S.T.
RANK
1 BUF 42.1% 1 23.9% 1 3-1 10.0% 10 -28.0% 2 4.1% 3
2 PHI 38.2% 4 17.8% 2 4-0 17.9% 4 -24.8% 3 -4.5% 30
3 BAL 30.1% 3 16.8% 3 2-2 23.1% 1 2.7% 16 9.7% 1
4 JAX 20.2% 2 0.2% 18 2-2 5.7% 12 -14.3% 6 0.2% 16
5 TB 20.0% 5 12.6% 4 2-2 1.7% 15 -20.8% 4 -2.5% 27
6 DAL 19.6% 6 5.0% 9 3-1 0.7% 17 -15.3% 5 3.5% 6
7 SF 19.2% 9 6.0% 8 2-2 -7.4% 21 -29.5% 1 -2.9% 28
8 KC 13.9% 8 10.5% 5 3-1 17.6% 5 2.0% 14 -1.7% 24
9 GB 9.7% 12 8.4% 7 3-1 13.3% 8 2.9% 17 -0.8% 21
10 LAC 7.0% 15 9.1% 6 2-2 7.9% 11 -1.5% 11 -2.3% 26
11 ATL 6.7% 14 -3.7% 20 2-2 11.0% 9 8.1% 22 3.9% 4
12 CLE 5.7% 7 2.1% 14 2-2 16.3% 6 11.4% 30 0.8% 13
13 SEA 4.7% 21 -4.3% 22 2-2 20.6% 3 17.5% 31 1.6% 10
14 MIA 3.8% 11 0.7% 16 3-1 20.6% 2 10.7% 28 -6.1% 31
15 DEN -0.4% 10 1.0% 15 2-2 -10.3% 24 -10.9% 8 -1.0% 22
16 CIN -1.2% 20 3.1% 11 2-2 -14.5% 27 -12.1% 7 1.2% 11

Comments

116 comments, Last at 07 Oct 2022, 5:28pm

#1 by Dr. Gamera // Oct 04, 2022 - 3:53pm

The Miami Dolphins are right behind them, having of course eaten two of our top three teams (Buffalo and Baltimore).

I don’t doubt that dolphins could eat buffaloes and ravens if given the opportunity, but land and sea creatures typically don’t interact in that way.  (Yes, I know about the orca-moose exception.)

Points: 0

#6 by Aaron Brooks G… // Oct 04, 2022 - 4:24pm

Boy, do I have the link for you!

\you lose nine times out of ten
\\you seem really neat

Points: 0

#16 by Mike B. In Va // Oct 04, 2022 - 6:09pm

This, my friends, is the quality content I come here for.

Points: 0

#102 by DGL // Oct 06, 2022 - 10:57am

I would Like this post, but, well...

Points: 0

#2 by jacobstevens // Oct 04, 2022 - 4:08pm

Tyler Lockett isn't getting much use and Geno's ALEX rating is still mediocre (but even that's better than I expected). His play is still very Alex Smith-like and with that defense you just can't get too far. Penny's playing nicely so they can contend for a dance ticket on a weaker schedule but yeah, really that's about the most you can reasonably expect. 

Points: 0

#8 by Aaron Brooks G… // Oct 04, 2022 - 4:28pm

ALEX doesn't correlate to whether or not you are good.

Bad Wentz and Bad Allen had high ALEX. Brees was usually on the bottom of the ALEX list. Brady is usually middle-ish.

ALEX just measures aggression. There's good aggression and bad aggression. Fitzpatrick would give you both, varying from pass to pass.

Points: 0

#21 by kamso // Oct 04, 2022 - 10:01pm

He has career highs in yards per game and receptions per game so far this year. He’s being used more than ever and it will probably continue. He’s the most underrated receiver around.

Points: 0

#3 by KnotMe // Oct 04, 2022 - 4:12pm

Is there any particular reason to only put the top half of the table at the end? Esp since you often end up talking about the bottom half. 

Points: 0

#75 by jheidelberg // Oct 05, 2022 - 4:07pm

Go to top of screen DVOA & Statistics and all 32 teams show up.  Also, if you simply search for "2022 Team DVOA Ratings" you will get the entire table.

Points: 0

#79 by Vincent Verhei // Oct 05, 2022 - 4:46pm

We found a lot of readers weren't even aware those stat pages existed. Inviting them to click on the stat pages to see the full table will hopefully encourage them to stick around and explore for a while.

Points: 0

#114 by LionInAZ // Oct 07, 2022 - 2:47am

Still, before the last couple of years FO provided the full table of DVOA rankings for all 32 teams. Is it that difficult to do it now? We can still go to the stat pages for the detailed breakdowns. 

Just don't understand the choice to degrade this particular analysis page.

Points: 0

#115 by KnotMe // Oct 07, 2022 - 9:28am

If that is the goal, I would stick a link to the full table at least. Still, I like that the site is generally easy to navigate with everything nicely up at the top. Bit worse on mobile but still pretty good. 

Points: 0

#4 by Aaron Brooks G… // Oct 04, 2022 - 4:14pm

This is where we get into the usual arguments about garbage time, but our research has showed that garbage time performance, although often down-weighted, is still predictive. 

This should probably be interpreted as: DVOA can't tell the difference between teams that stop trying to give up yards and teams that can no longer prevent yards. When a team keeps trying and keeps stopping yards, DVOA can see and account for that, so it's more predictive.

Points: 0

#5 by Boots Day // Oct 04, 2022 - 4:22pm

In Quick reads, Vince noted that Patrick Mahomes' late interception was terrible for his DVOA. I assume the same carries over for the entire Chiefs' offense. It wasn't quite garbage time at that point, but if Mahomes had completed that pass, we would have entered garbage time.

Of course, a team that's playing with a big lead all game is also more likely to have fewer yards per play, since they're not likely to be taking a lot of deep shots and are more interested in killing some clock.

Points: 0

#15 by nsheahon // Oct 04, 2022 - 6:04pm

He threw that pick with under 5 minutes left in the game and up 3 scores. That is definitely garbage time.

Hard to imagine where watching that game you would think “Yeah, TB definitely looked like the better team.”

Points: 0

#31 by ih8california // Oct 05, 2022 - 11:58am

The people who run DVOA are completely blinded by their system. They refuse to believe anything can be wrong with it despite situations like KC/TB where KC clearly dominated the game (if you watched it) yet DVOA says TB was 20 pts better, I could see if it said KC/TB were even or close, but to have such a disparity given the game results is ridiculous.

They are the real-life Principal Skinner meme. 

 

Points: 0

#33 by colonialbob // Oct 05, 2022 - 12:17pm

This is where we get into the usual arguments about garbage time, but our research has showed that garbage time performance, although often down-weighted, is still predictive.

Hmm. 

Points: 0

#37 by Pat // Oct 05, 2022 - 12:51pm

The comedic part of the previous comment was that Aaron's tried, over and over, to adjust things to downweight those situations. It just doesn't work. It still means something.

It's a lot like preseason games before they became totally pointless in the 2010s since no one plays. Back in the 2000s you could dismiss them all you want, but the first halves of those games were just as predictive as regular season games. It's easy to point to KC/TB and say "oh, that doesn't mean anything." Really? Look at the teams that couldn't do anything even though they were being blown out. Indianapolis. Tennessee. Those teams have bad offenses. It's not surprising they continued being bad.

Plus, why are we treating that game as if it was like, 70-0? The Bucs had basically exactly half of their yards in the first half, which ended 28-17. They were down 2 scores at the beginning of the 4th quarter. That's not garbage time by any stretch of the imagination.

I seriously don't get it. There's like, two drives I'd label garbage time in that game. Maybe 10 plays.

Points: 0

#39 by ChiefsCML // Oct 05, 2022 - 12:55pm

I think the point of contention people have is that it was quite clear the Chiefs weren't putting as much into the game after a certain point, as they felt they had it in the bag. And this is quite typical of Andy Reid teams. Regardless of whether or not you'd label it garbage time, the Chiefs clearly felt it was in the bag. I doubt they'd do that in a championship game.

Points: 0

#41 by colonialbob // Oct 05, 2022 - 1:24pm

A few points in response to this:

  • "it was quite clear the Chiefs weren't putting as much into the game" - I think this is very easy to say but much harder to actually suss out. Were they not putting as much into the game, or did TB just figure some stuff out a bit too late to actually matter? There are ways to get at that question (e.g. "they went from running the ball on 3/10 first downs in the first half to 7/10 in the second half") but frankly I'm not gonna trust the eye test of the average fan, including myself, to really be able to judge that.
  • Even if we assume the above is true, that doesn't mean it's a good thing. If they go into a shell and start giving up big yards to an opponent, that obviously is worse for their chances of winning that game than if they keep up the domination. At some point it obviously doesn't matter, but I think defining that point isn't straightforward. And while it's easy to say "they wouldn't do this in a championship game", the Chiefs have indeed done this type of thing in a championship game very recently. Certainly not the exact same situation, but they've given us reason to doubt their ability to hold a lead before.
  • Finally, as Pat mentions, TB was able to do well, and that's obviously better than if they hadn't been able to do so. Similarly, if the Chiefs could've "not put as much into the game" and still stopped TB, that's obviously a positive sign. So even if "garbage time" has less weight than "regular time", it's still perfectly logical that it has some weight.

It's easy to say "well, this game is different because X, Y, and Z", and that might even be true. But if you're creating a model, the last thing you want to do is include a "special pleading" variable where you go in and tinker with the results of individual games because of what you think might be happening.

Points: 0

#46 by Theo // Oct 05, 2022 - 1:36pm

When a defense is up a certain amount of points it just starts to give less. 

While the opposing offense is giving it 100% to catch up the defense knows it will get away with giving it 90 or 85%. Its just the nature of the beast. 

 

Points: 0

#51 by Pat // Oct 05, 2022 - 1:54pm

When a defense is up a certain amount of points it just starts to give less. 

When the other team subs in a backup QB, yeah, then you might sub out your starters for backups on defense and obviously at that point things matter less.

But this isn't even that. This isn't guys "giving 90 or 85%" or whatever. It's just prevent. Just watch it again. Late 3rd quarter, Chiefs up 21, the defense is still in a bog-standard alignment on 1st and 10. 10 guys on screen. 7 guys near the line. Same defense you'd see any other time.

Now move to late 4th quarter, say, after the pick. 1st and 10. Hey look! Linebackers are 5-6 yards off the line. Deep safeties are off screen. One of the 4 linemen drops into coverage to cover shallow. That spells "prevent," folks! And Brady does exactly what you're supposed to do against prevent. Move quick. High tempo. Screens and quick slants. High percentage. Go, go, go.

That's not "garbage time." Tampa did a great job there (helped by the turnover!) and then it was on the Chiefs offense to not give it back again, which they also did. It's just standard late-game stuff. It's not meaningless. Not everyone can move the ball like that against prevent! A misplaced slant, and a safety's off to the house.

Points: 0

#64 by colonialbob // Oct 05, 2022 - 2:38pm

The Bills/Titans game was a much better example of "garbage time" I think. Once they both had their backup QBs in, I'm sure there was a drop in effort by the defenses as well, and thus whatever happened after that is probably less predictive. Still, though, I don't think all that data is meaningless (having a good 4th/5th receiver helps, even if it doesn't help as much as having a good #1 receiver), and from what I understand the research FO has done backs that up.

Points: 0

#48 by Pat // Oct 05, 2022 - 1:42pm

I think this is very easy to say but much harder to actually suss out. 

It wouldn't be if we actually had pre-snap defensive alignments. It's pretty clear that at some point in the 4th, KC shifted to a soft zone, with linebackers 5+ yards off the line and not even bothering to rush 4. But that didn't happen until the 4th quarter, and I'd be surprised if they were really in prevent until that drive inside of 5 minutes.

Finally, as Pat mentions, TB was able to do well, and that's obviously better than if they hadn't been able to do so.

They got within 2 scores with 3:30 left in the game! A weird fluke turnover there and Tampa's got an honest shot, albeit with an onsides kick attempt. Yes, KC's going prevent, but they're not subbing in backups or anything. 

Points: 0

#50 by Aaron Brooks G… // Oct 05, 2022 - 1:50pm

Even if we assume the above is true, that doesn't mean it's a good thing. If they go into a shell and start giving up big yards to an opponent, that obviously is worse for their chances of winning that game than if they keep up the domination.

Careful there.

DVOA is mostly a predictive measure. It can retrospectively explain, but that's not its core concern. What it really wants to do is predict the next similar game setup.

EPA and Win% (really Win% Added) do a better job of explaining in-game decisions given a set of game parameters, because football is a game with a clock. There are backwards-seeking measures. DVOA hates running a game out. EPA thinks its fine, but doesn't love hemorrhaging yards. Win% *loves* running the game out and doesn't care about empty yards at all. If you're up 9 with 5 seconds left, it's basically impossible to have a bad play.

DVOA, for instance, doesn't understand kneeling out the clock. It just sees that as a QB scramble for a loss -- a hugely negative play. Win% sees that as the only sane choice. (EPA is basically neutral on it) Accordingly, a lot of game plays have to be hand-edited out of DVOA -- end of half Hails Mary, desperation late-game passing, kneeling, etc. DVOA doesn't know to ignore those, because DVOA doesn't know there is a clock.

Points: 0

#57 by Pat // Oct 05, 2022 - 2:14pm

Win% sees that as the only sane choice. 

There's a difference between "win% on offense" and "win% on defense," though. Running out the clock's an obvious win% advantage since you never give the ball back to your opponent and golly gee if you're winning, that always leads to you winning the game. So runs where you gain very little can still be significantly WPA positive.

But on defense you have to get to very late in the game to where you can give up yardage for clock freely, because you always want to get the ball back, since if the other team has the ball they can always win. Tampa's late drive to make it 41-31, for instance, was still WPA negative.

Win % thought processes also bug the heck out of me late in game because so much of it is contingent upon one play - that onsides kick. Yes, of course, it has an extremely low probability of happening, but when it does happen, it's exceptionally bad. So I tend to think more of "assume one onsides kick works" as the right decision making process.

Points: 0

#63 by colonialbob // Oct 05, 2022 - 2:34pm

True, and a very good point. That really gets at the point of what I was trying to express (poorly) - you can have things that don't make much of a difference by EPA/Win%, but impact DVOA, because they hold predictive power. You can also have plenty of things that don't make much of a difference by EPA/Win% and don't help DVOA because they're non-predictive - all of the situations you mention. 

What I was trying to say was that going into a prevent-style defense, either literally or just sitting back in a soft shell, may be a reasonable or optimal thing by EPA. But if you are able to go into that shell and not give up big chunks of yardage, that implies something about your defense being better than if you go into that shell and do give up that yardage, whether or not it actually affects the game you're playing.

Points: 0

#67 by ahmadrashad // Oct 05, 2022 - 2:49pm

"DVOA hates running a game out." -- Yep. There's probably some debate/confusion over the definition of "garbage time", because sometimes you will see teams start running the clock out in the second quarter. 

So I suspect people look up and see it's 38-17 in the third quarter and think "it's over, garbage time now", but it is understandable why DVOA doesn't see it that way. 

Points: 0

#43 by Pat // Oct 05, 2022 - 1:31pm

Yeah - and I'm saying, at best, that point is late in the 4th quarter. Even up 38-17 they weren't in a soft zone shell with minimal rush.

And so what if they're in a soft zone? They're still playing defense. You still have to complete those passes. Those coverages happen at other times in the game too. They're not standing around saying "whatever." Guys are still actively trying to rip balls out of people's hands.

Yes, late in the game, the linebackers were 5-6 yards off the line with 2 deep safeties and only 3 guys rushing. Yes, that makes it easier to complete underneath passes, which is what Brady did. But you still need to actually complete the pass, which isn't nothing. And again - that really only happened within 5 minutes left in the game. So like, 2 drives.

Points: 0

#47 by jheidelberg // Oct 05, 2022 - 1:39pm

You can not rest all of your starters.

Some of the backups in week 4 will play regularly later in the season replacing current starters due to injury or poor performance of the starters.

Everyone is still trying, fighting for jobs or stats.

 

Points: 0

#81 by ih8california // Oct 05, 2022 - 5:03pm

They got the ball on a fluke non-predictive play, but they were still on the 16 yd line. Surely TB's #1 overall DVOA defense should prevent them from scoring a TD on the 2 plays later.

Points: 0

#83 by ih8california // Oct 05, 2022 - 5:24pm

Plus, why are we treating that game as if it was like, 70-0? 

No one is, but DVOA had TB as +20 for the game, that is insanity, considering TB was never within 1 score from 4:01 in the 1ST QUARTER onward. 

Another thing I don't understand is that DVOA is supposed to take into account Defenses.  How is it TB gets more points from the system for scoring 31 points on an average Chiefs Defense, than the Chiefs do for scoring 41 points (12/17 on 3rd down) on the supposed #1 DOVA TB Defense.

Here are all the "stops" by the TB #1 DVOA defense:

  1. Downs. (2nd quarter, where CEH drops a pass that is a TD if he catches)
  2. Halftime (2nd quarter, 1-play drive where KC kneels and doesn't attempt to move the ball)
  3. INT (4th quarter, 4:42 left in the game down 17 or 3 scores)
  4. Punt. (4th quarter, TB forces its 1st Punt of the game with 0:45 left in the game down 10)

I'm saying that TB shouldn't have a high score for Offense in the game, what I'm saying is that there is NO WAY their Offense DVOA was better than KC's and their Defense DVOA should be significantly worse considering they were #1 in DVOA and they gave up more points. 

 

Points: 0

#100 by Pat // Oct 06, 2022 - 10:44am

that is insanity, considering TB was never within 1 score from 4:01 in the 1ST QUARTER onward.

Literally the first thing that happened in the game was a special-teams fumble that essentially handed the Chiefs a 7-0 lead, and another lost fumble deep in Kansas City's territory led to another score.

By expected points the Chiefs offense was +17.12 to the Bucs 14.15, but with the Bucs -5.44 of that was from the turnover: slicing that in half (since fumble recovery's random) basically puts the Bucs on even ground with the Chiefs. In terms of expected points/play the Bucs win out by a lot because they lost 2 drives due to the special teams fumble + first-play fumble (scaled up by play count the Bucs would've been +21.5).

Heck, adjusting just the fumbles shifts the game to only a ~4.6 point win by the Chiefs, and the play differential covers that entirely. 

It's the difference between looking at the game and looking at how the teams played. It's a game. Stuff happens. That's why it's fun to watch. The game got challenging for the Bucs very early, but they played really well. It's the same thing that happened with the Eagles and Lions. I mean, 25% of the time both of those fumbles get recovered by the Bucs and it's an entirely different game.

Points: 0

#103 by Aaron Brooks G… // Oct 06, 2022 - 11:32am

Game state matters to play decisions.

Points: 0

#104 by Pat // Oct 06, 2022 - 11:54am

The Chiefs weren't playing significantly different on defense until those last two drives. To be honest, it was a bit bizarre: Tampa basically hadn't run the ball for 2 quarters and they still had linebackers within a yard of the line on 1st and 10.

Points: 0

#105 by KnotMe // Oct 06, 2022 - 12:20pm

Well, they were still running RB out I think, so they probably had to consider the possibility of a run at least. 

Points: 0

#55 by Lost Ti-Cats Fan // Oct 05, 2022 - 2:13pm

If the people who run DVOA are completely blinded by their system, then they're taking the right approach.  Because the whole point of the exercise is to remove subjective analysis ("clearly dominated the game (if you watched it)") and replace that with objective analysis.

If the data is predictive, it's predictive.  It doesn't really matter whether this particular data point makes sense to you or not, go with what the numbers say.

Points: 0

#26 by luisguilherme // Oct 05, 2022 - 8:38am

That’s a limitation of DVOA and also a reason why Pete Carroll’s Seahawks are such overachievers in the metric, as they love to play close games, meaning explosive plays (which are DVOA gold) need to keep happening. 

Points: 0

#27 by cstoos // Oct 05, 2022 - 8:58am

Andy Reid teams always get smashed in DVOA when they are up because he goes to time wasting mode on offense AND defense. 

That's also why Andy Reid teams have given up some of the largest comebacks in history over the years. 

I would argue that while the reduction in DVOA doesn't necessarily represent the team's performance on a game by game basis as accurately because of this, at the end of the season it'll be pretty accurate because they'll inevitably blow a game that they were up 28 at the half because of this approach to clock management. 

Points: 0

#30 by Aaron Brooks G… // Oct 05, 2022 - 11:53am

Only great teams can blow historic leads.

The Lions are safe from that kind of record because they aren't good enough to have a huge lead in the first place.

Points: 0

#32 by jheidelberg // Oct 05, 2022 - 12:08pm

The Ravens have blown 2 leads of 17 or more points during the first 4 games, the only other team to do this in history is the Lions

Points: 0

#36 by Pat // Oct 05, 2022 - 12:25pm

Andy Reid teams always get smashed in DVOA

There are many, many things I've heard in the years I've been here, but this is without a doubt the first time I've heard that DVOA hurts Andy Reid teams.

Points: 0

#49 by cstoos // Oct 05, 2022 - 1:49pm

I didn't mean it as an overall thing.  I meant, because of how he manages the game with leads, DVOA sometimes doesn't match what we see with our eyes.  There have been plenty of games where KC dominates, but then gives up 2 late scores and the offense goes relatively stagnant once they have a big lead.

 

It feels like Reid would rather win by 8 than win by 28 based on how he runs games with a lead. 

Points: 0

#58 by Pat // Oct 05, 2022 - 2:16pm

Over the years, by far, people almost always say DVOA favors Andy Reid coached teams. Now, the reason they give varies, and it's usually contradictory. Because it's all silly.

There have been plenty of games where KC dominates, but then gives up 2 late scores and the offense goes relatively stagnant

That's not Andy Reid-specific. That's just prevent and running out the clock (and going conservative). You wanna talk about offenses going stagnant in the second half? I'm a Philly fan!

Points: 0

#78 by KnotMe // Oct 05, 2022 - 4:44pm

I've never seen a good explanation why teams would stop doing what works. (Every score you get is one more they have to get so....why not just keep doing what you have been and they can't stop?).  The only thing I can think of is balancing fatigue or something. 

Points: 0

#87 by Aaron Brooks G… // Oct 05, 2022 - 7:14pm

When should you stop doubling down on roulette is the same conceptual analysis.

Points: 0

#91 by Mike B. In Va // Oct 05, 2022 - 8:26pm

When adding points is less important than injury risk. It's a violent game even when you're leading by 20.

Points: 0

#101 by Pat // Oct 06, 2022 - 10:50am

It's just about reducing variance. Once you've got a significant lead, the only way you're going to lose is on high-leverage high-variance plays, so you give up the average to reduce the variance.

Points: 0

#7 by dmp // Oct 04, 2022 - 4:25pm

This is my most trusted source for analysis. Do you post rest of season strength of schedule?  I can grab the table needed, just wondering if it's already sitting somewhere. 

20-year reader here since college, very infrequent poster. No guff I love this site

Points: 0

#10 by Vincent Verhei // Oct 04, 2022 - 4:57pm

Yup. The Bengals have the most difficult future schedule, the Eagles have the easiest. 

Points: 0

#11 by Sportszilla // Oct 04, 2022 - 5:01pm

So here's where I think the stats from last year are somewhat misleading w/r/t Geno:

I know DVOA should capture a fair bit of this, but he faced three very good pass defenses (and the Jaguars) in his four games. The Seahawks offensive scheme last year was also designed (if you can call it that) around Russell Wilson and was fairly different from the scheme this year, which pretty clearly is centered more around Smith's strengths.

I don't know that Geno is really an above-average starting QB, but the talent around him on offense actually might be very high - two very good/elite receivers who compliment him well, several talented pass-catching tight ends, a very good pass-blocking line, and a dynamic running back.

We will see how it holds up against the Saints D this coming week, but I actually think it's plausible that this is a good offense, if not one of the best in the league.

Points: 0

#12 by Aaron Brooks G… // Oct 04, 2022 - 5:46pm

In reply to by Sportszilla

This is basically Smith's first season running his offense on a team not the Jets.

Points: 0

#13 by Sportszilla // Oct 04, 2022 - 5:50pm

It's obviously far more likely than not that Smith is at best a below-average starter, but given his draft status and the way his career progressed, it's also not like it's inconceivable that he's good now.

Points: 0

#23 by BJR // Oct 05, 2022 - 6:31am

In reply to by Sportszilla

We've already had a sample of what it might look like when the Seahawks offense goes against a likely top tier defense, in week 2 against the 49ers. They scored 0 offensive points.

No, I'm not drinking the Kool-aid just yet.

Points: 0

#14 by pm // Oct 04, 2022 - 6:04pm

I'd like to apologize to FO for mocking their prediction that Washington would be the worst offense in the league. I have to admit that watching Carson Wentz for 4 games has made me realize just how right Aaron was with that prediction. The worst offense in the league by far.

Points: 0

#24 by BJR // Oct 05, 2022 - 6:42am

Of all the QB trades this summer, this was the one that I'd confidently say looked bad at the time, not only with the benefit of hindsight. 

Wilson, Ryan, Mayfield are all off to poor starts (obviously no judgment on Watson yet), but I still think those moves were justifiable at the time. Wentz had just been run out of town two consecutive seasons for a mixture of atrocious play and (apparently) toxic attitude. I've no idea why Washington (of all teams) thought they could salvage him. I guess everybody is going to end up losing their job before the season is out, which is fully deserved after such a fiasco.

Points: 0

#29 by mehllageman56 // Oct 05, 2022 - 11:20am

(apparently) toxic attitude. I've no idea why Washington (of all teams) thought they could salvage him. 

The organization felt his attitude made him a perfect fit.

Points: 0

#25 by Pat // Oct 05, 2022 - 7:55am

I completely agree. At this point I've just accepted that Wentz doesn't realize his own issues. It's funny because I don't usually put much credit in off-field personality stuff but holy cow, does it fit with Wentz. I've never seen someone play so confidently with no basis in reality.

Points: 0

#17 by Pen // Oct 04, 2022 - 8:25pm

last year to judge "who is Geno Smith", then I might point out that I posted this on another website about a month before this season started:

Who’s stat line would you rather see vs NO, Pitt and Jax?

74/126... 58.7%.....784 yds.. 5TD.. 4 INT ..6.2 y/a
55/78..... 70.5%..... 571 yds.. 4 TD.. 0 INT ..7.3 y/a

The first QB is Josh Allen. That second is Geno Smith vs the same 3 opponents..

How about this guy, who played vs NO, Pitt, Jax (twice) and the Rams? The only other QB who faced every opponent Geno faced.

95/146... 65.1%.. 982 yds.. 4 TD.. 2 INT.. 6.7 y/a
65/95...... 68.4%.. 702 yds.. 5 TD.. 1 INT.. 7.4 y/a

The first QB is Ryan Tannehill, his stats boosted by two games vs Jax.

How about a guy who played against the Rams, NO and Pitt last year? Let’s take out that really good game against Jax Geno had.

63/109... 57.8%.....688 yds.. 4 TD.. 2 INT.. 6.3 y/a
45/71... 63.4%......507 yds ..3 TD.. 1 INT..7.1 y/a

The first QB is Aaron Rodgers. No Jax game here to boost Geno's numbers.

So, if we're going to use small sample sizes, best , perhaps, to put them into context. NO and Pitt were the 4th and 8th ranked pass defenses per DVOA last year. Geno did well compared to everyone else vs those teams.

It's like saying "well, we know what a bum Rodgers truly is because look how he did in these three games."

Truth is, Geno was a rookie on a horrific Jets team, then wasn't just ANY journeyman backup, he was back up to Rivers, Eli Manning and Russ Wilson. He was never going to win the job vs them no matter how much he improved.

We literally are seeing who Geno Smith evolved into right now.

Points: 0

#22 by KnotMe // Oct 04, 2022 - 11:24pm

 But him being decent would lend weight to the theory that there is a learning process between college and NFL and the drafting team matters. I.e. it took a while to get the Jets out. 

They also improved the o-linen so maybe the extra x second(s) matters. 

Points: 0

#18 by Kaepernicus // Oct 04, 2022 - 9:13pm

My pre-season call on 49ers and Seahawks being underrated is looking pretty good so far and that SEA over ticket for season wins is looking good too. Man was I wrong about the Eagles. Can't believe how much Jalen Hurts has improved as a passer. 49ers defense is actually better than I expected while missing both interior d linemen, 1 starting LB and Jimmie Ward at FS. The idea of that team with Armstead, Kinlaw, and Ward back is insane and may start approaching all-time levels with some injury luck. The sad side effect is there is almost no chance Ryans isn't poached as a HC next year. 

Also the conference differences are pretty stark. 4/5 top defenses in the NFC and 4/5 top offenses in the AFC. Another interesting one to track is how bad PHI, SF, and TB are at special teams. If it weren't for really good punting the 49ers would be on the bottom in those rankings for sure. I think people have spent too much time talking about the overloaded young QB differential between the conferences while ignoring just how heavily the top young defensive talent is also skewed to the NFC. Should be another defensive war in the NFC playoffs again

Points: 0

#28 by jheidelberg // Oct 05, 2022 - 9:17am

Three of the last four articles have been headlined by Geno Smith or the Seattle offense.

I am waiting for the inevitable regression toward the mean, whether it be the mean of QB’s or worse yet, the Geno Smith mean.

We will get our article’” The rise and fall of Geno Smith” and move on to other topics.

Points: 0

#35 by theslothook // Oct 05, 2022 - 12:23pm

I may need to give up watching the Colts mid week from now on. It's just too depressing. Gor those curious, I don't have a blog so this will be my outlet. 

I know Vegas had them at 9.5 and FO was bullish. I made it clear way before that I wasn't. Nate dunlevy wrote a blog assessing this and puts the blame overall at the general manager. 

I tend to disagree. Aside from the Wentz disaster, I don't think any of the individual moves themselves were so hard to justify. You can pick apart little things and say it adds up to a big thing, which Nate does, But I regard all of that as small potatoes for most franchises with good quarterback play. Imagine how much our opinions would change if Matt Ryan replaced Aaron Rodgers or Lamar Jackson; to say nothing about Stafford.

In reality, the team was forced to rebuild on the fly through the middle and embracing a season of pain was clearly not something either the owner or the front office or the coaching staff were willing to absorb. And I suspect being so competent in spite of the QB play kept the team focused on finding a good game manager to keep the ship afloat.

Heres the thing. When in the NFL has a team continued to succeed stringing together so many consecutive seasons with mediocrity at QB? The Chiefs kind of did it with Smith but they have Reid and even they had a season where they missed the playoffs. The Ravens did it for periods of time with even worse QB play than mediocrity, but they had Newsome and even they had awful seasons intermixed here and there.

When you zoom out, the result of this season isnt all that surprising. Every single team that gets even tier 3 play eventually collapses over time. The Colts with tier 3 play are probably a 8 win team in a decent division. When you give them league worst starting QB play, you get this. So even if Ryan was the QB they could have reasonably hoped for, the net result probably is unchanged by the optics look better.

In the end, I maintain this is probably the best overall outcome assuming the coaching staff and GM stay, because for all the small issues, they've done a good job under the circumstances and really have talent to support a young QB prospect. But firing everyone and rebooting is only likely to bring in worse alternatives.

Points: 0

#40 by Aaron Brooks G… // Oct 05, 2022 - 1:23pm

Heres the thing. When in the NFL has a team continued to succeed stringing together so many consecutive seasons with mediocrity at QB? The Chiefs kind of did it with Smith but they have Reid and even they had a season where they missed the playoffs. The Ravens did it for periods of time with even worse QB play than mediocrity, but they had Newsome and even they had awful seasons intermixed here and there.

The Teams Harbaugh. The Jimmy G 49ers. The Goff Rams. Reid pre-Mahomes. Game Manager Brady. Late Roethlisberger. Carroll Seahawks.

It can be done.

Points: 0

#44 by theslothook // Oct 05, 2022 - 1:32pm

Those teams all ended up missing the playoffs and disappointing at various points. The Rams saw the Goff end game and proactively ditched him.

Harbs the elder left just as the 49ers entered their dark ages.

Harbs the younger got rescued by Lamar.

And we shall see about Kyle Shanny but even his teams fell apart when Jimmy G was hurt, necessitating the whole Lance trade in the first place.

My basic point is akin to exhaustion. It's exhausting to carry your qb season after season.

 

 

 

Points: 0

#59 by Aaron Brooks G… // Oct 05, 2022 - 2:16pm

97% of teams have a failed season. Everyone misses the playoffs at some point. (Even Brady+Belichick)

Hell, just look at the Chargers for a cautionary tale about how mainstream roster construction can repeatedly fail catastrophically.

Points: 0

#61 by Mike B. In Va // Oct 05, 2022 - 2:27pm

As a fan of a team that went 20 years without a QB, that's pretty exhausting, too. Even though the owner was OK with it.

Points: 0

#68 by theslothook // Oct 05, 2022 - 2:59pm

10 years from now if it hasn't happened already, people are going to claim that they saw this career from Josh Allen coming. That somehow a prospect with his pedigree was destined for this kind of level of play. (By that I mean, having watched him a decent amount so far he's displaying a lot high level quarterback traits where the physical tool serve as a complement to them and not the other way around. This is sort of the difference right now between him and Mahomes, although Mahomes' off the cuff instincts are so ridiculous that I still put him in the pole position of top quarterbacks)

To me, there's nothing you can forecast about Josh Allen that would lead you to think he's going to be this good. And two years of bad quarterback play should have put his probabilities of reaching this at extremely low.

Much like how I refuse to credit anyone for Tom Brady or blast the draftnicks' myopia for missing some hidden trait that was buried beneath all of bad film tape, I don't give anyone credit for Josh Allen other than Josh Allen himself. For whatever reason, the game just clicked for him in a way that it didn't his first two years and never does for 90% of the quarterbacks out there

Points: 0

#92 by Mike B. In Va // Oct 05, 2022 - 8:30pm

Agreed. As a guy who watched the first two years with hope in my eyes, he did have that intangible "never give in" thing that made the fans love him, but that didn't make him not suck. He did that on his own.

Points: 0

#70 by Pen // Oct 05, 2022 - 3:02pm

Carroll's Seahawks had a HoF QB. Not sure why you include them in a list of teams with mediocre QB's.

Points: 0

#73 by Aaron Brooks G… // Oct 05, 2022 - 3:50pm

In reply to by Pen

<G>

The zeitgeist was Rivers was a HOF QB, too, until he wasn't.

Points: 0

#85 by gomer_rs // Oct 05, 2022 - 6:53pm

Over at Fieldgulls a poster by the handle of Fractalwalk did a Z test examination of Russel Wilson's EPA per play over his career.  Russ was right at or just above 1 standard deviation.  Which put him better then, but nearly equal to, Ben Rothelisberger by the same measure.

Points: 0

#86 by theslothook // Oct 05, 2022 - 7:07pm

I think a fair and even handed analysis of Russ's career would say it's debatable whether he is at the moment a hall of famer. I lean no but I tend to be a small hall guy.

On the other hand, I don't think he belongs in the Goff category of mediocrity that you have to carry every week and every season.

Points: 0

#88 by Aaron Brooks G… // Oct 05, 2022 - 7:16pm

The context around here for “mediocre” is “not a 1st ballot HoFer”. I use mediocre accordingly.

Points: 0

#89 by theslothook // Oct 05, 2022 - 7:47pm

When I use the word mediocre, it's really meant to imply below average or something that's rather easily replaced. 

So no I don't consider Stafford or Jimmy G or Kirk Cousins easily replaced because if that were true they wouldn't be starting and they wouldn't be getting big contracts.

I do consider someone like Alex Smith, i'm sorry to say mediocre because teams were always looking for ways to upgrade above his level. You can probably add Baker Mayfield to that list as well

Points: 0

#96 by Aaron Brooks G… // Oct 05, 2022 - 8:42pm

Mediocre really means ‘about average.’ Treating is as below average or replacement level is abusing the term.

Mediocre is hard to replace. That’s better than about half of all starters. Replacement-level is closer to the 65th-best for a QB. It’s a guy you can pick up off the street for the minimum.

A mediocre team is 8-8-1. A replacement level team loses to the Orlovski Lions or the Bradford Rams.

Points: 0

#90 by gomer_rs // Oct 05, 2022 - 8:01pm

What the EPA analysis suggests, which I know not DVOA or DYAR, is that for 9 years Russ's average performance was 'top 5' of 32 starting QBs.  (Z of 1 for 32 enterants would generally be the 4.8th best).

So, even in a small hall, that's pretty good.

Points: 0

#97 by theslothook // Oct 05, 2022 - 8:43pm

Russ' candidacy is problematic because he, like so many of his contemporaries, lacks the typical number of all pros and mvps - which have all been canabolized by the same small group of quarterbacks. 

Russ DVOA and DYAR numbers are bit too scattershot for me as well. Now granted, those are team metrics and Russ' context dependent individual play is hopelessly intertwined with the Seahawks offense. 

There isn't a clear standard for what a hall of fame QB is, but I do think if you are going to allow Russ to get in, you then can start making cases for other QBs who likely aren't going to get in - like Philip Rivers or Matt Ryan. 

Points: 0

#116 by gomer_rs // Oct 07, 2022 - 5:28pm

(1) Ryan and Romo would be hall locks if they had somewhat competent GMs during their careers.  Their numbers simply support it and they just dont have RINGZ!!!!

(2) Phillip Rivers is this generation's Boomer Esiason, the next guy that should get in if the hall were to expand by one guy at his position, but isn't REALLY good enough.  Especially having his INT rate as passing was getting easier is just inexcusable.

(3) Eli, has RINGZ, and that's all the good I can say about Eli.

 

Points: 0

#42 by colonialbob // Oct 05, 2022 - 1:30pm

I tend to disagree. Aside from the Wentz disaster, I don't think any of the individual moves themselves were so hard to justify. You can pick apart little things and say it adds up to a big thing, which Nate does, But I regard all of that as small potatoes for most franchises with good quarterback play. Imagine how much our opinions would change if Matt Ryan replaced Aaron Rodgers or Lamar Jackson; to say nothing about Stafford.

Isn't this on the GM, though? It's fine to believe he made tactically sound decisions, but I think what most people question with the Colts is the strategy itself. Sure, you can say they "weren't willing to absorb" a rebuilding year, but that feels like giving them an undeserved pass. They also haven't done anything to get themselves out of the rut of "acquire available (read: past their prime) veteran QB and run it back" every offseason, so even if you assume some "thou shalt not rebuild whatsoever" edict as beyond the GM's control, he still has committed the team completely to that path without enabling them to pursue any other options.

Points: 0

#45 by theslothook // Oct 05, 2022 - 1:36pm

What options are left to you besides stop gaps and over the hill veterans? Trusting the franchise to a lower round rookie QB is essentially embracing the very likely possibility of a lost season.

Yes, the constant musical chairs at QB is far more damaging long term. Better to take the medicine today and be better for it, but that's not how the incentives are aligned. You can credibly commit to that strategy only if you are a rookie head coach taking over a disaster or a veteran coach with a sterling resume and enough career equity to be given the green light. Every other front office and coaching staff is on a win immediately or else timeline. 

That's kind of why I put a lot of this on Irsay. Either decide right now that you don't want these people running your franchise long-term and get rid of them, or give them the vote of confidence to see out a longer-term vision without the ax of winning nine games this year or else hanging over them.

Points: 0

#52 by colonialbob // Oct 05, 2022 - 2:02pm

"besides the Wentz disaster" well sure if you take out his worst decision, of course his record looks better. :)

I'm sympathetic to the idea that a lot of this is due to Irsay, particularly the inability to really pursue a "we're gonna take our lumps for a year" season, but not only did the Wentz trade cost them a shot at Pickett this year, they also made the DeForest Buckner trade (next two QBs picked were Jordan Love and Jalen Hurts), and didn't look at a QB in the top few rounds (over the last three years that's Ridder, Willis, Hurts, Davis Mills, Kyle Trask, and Kellen Mond). Sure, obviously those are all individually defensible decisions, and there's no guarantee any of those QBs pan out - Mond has already apparently washed out, Mills certainly doesn't look like the answer, etc. But the Colts haven't even given themselves the chance of getting lucky with a Hurts (or Prescott, if we reach back a bit farther). 

Points: 0

#56 by theslothook // Oct 05, 2022 - 2:14pm

"besides the Wentz disaster" well sure if you take out his worst decision, of course his record looks better. :)

Of course. I happily acknowledge this was a major blunder. And of course, it played out in about the worst possible way imagineable - ie they missed the playoffs AND ended up forking over a 2nd and 1st for a 1 year rental. Just epically bad. But it sort of ends the discussion if we aren't willing to move past that. 

Look I agree with you. I would have tried my hand at hoped someone like one of the names you mentioned becomes Jalen Hurts or Dak Prescott or Tony Romo. I would. But if you are sitting in Ballard's seat and you are constantly facing a playoffs or else mandate, that's not the road you are going to take, which is  precisely why I was blaming Irsay. Can you really blame your underlings for making decisions that align with their incentives and not the franchise's? 

Points: 0

#66 by colonialbob // Oct 05, 2022 - 2:45pm

No, I get it. As a (mostly estranged these days, for various reasons) Texas Rangers fan, the whole "is it the GM being bad or just following a directive from ownership" debate has been going solid for the last half a decade at least. I certainly don't put all of the blame on Ballard, I just also don't think he gets to avoid it completely either. If he does get fired, though, I would certainly view him as a reasonable candidate for another job.

Points: 0

#60 by Mike B. In Va // Oct 05, 2022 - 2:25pm

So far the "bring in the possibly washed veteran guy" has worked 50% of the time, which is a lot higher than the odds on drafting a QB.

The jury is still out on the 3rd round of it, though it certainly doesn't look good at the moment. So while I don't agree with the strategy, looking at the results so far makes things a bit more palatable if the idea is to make the playoffs.

Points: 0

#62 by Pat // Oct 05, 2022 - 2:30pm

which is a lot higher than the odds on drafting a QB.

It's... basically exactly the same as the odds on drafting a first-round QB. I think first round QBs end up with a Pro Bowl nod like 40% of the time, and most of that is from before the "Pro Bowl alternate" era. It's generally safe to assume upper-half of first round is a 50% chance.

Points: 0

#95 by Mike B. In Va // Oct 05, 2022 - 8:40pm

If you swing for the fences with a high-drafted QB, "getting to a Pro Bowl" isn't really a good criteria. "Getting to a playoff game" is probably a better one, which means three players from the first round of 2018 (out of 5) have been successful enough to keep Ballard's job - and one of those is widely considered to be a failure and was kicked out of town for a serial sexual abuser, one teams wanted to switch to running back, and one is a unicorn.

2019 has none. 2020 has one.

I completely understand his decision making.

Points: 0

#98 by KnotMe // Oct 06, 2022 - 9:29am

If your going by "getting to a playoff game", 2019 has Murray,  2020 has Burrow and Hurts, so not sure where you were getting none and one.

It's pretty typical you get 1-2 above average QB per draft. And sometimes a couple more average ones. Ironicly, Geno Smith would being decent (not sold on that yet), would salvage 2013 which was the last year that was a total flop, although that was considered a weak class like last year. 
 

2017: Mahomes & Watson, with Trubisky being low average ish

2018: Lamar, Allen, Mayfield might be average.

2019: Murray

2020: Burrow, Herbert and Hurts and Tua are decent

2021: No elite yet, but Trevor Lawrence might make it and Mac Jones seems decent. 

2022: Weak class and no info. Pickett is probably the only one with much future unless Willis is on the Allen route. 

The hard part isn't so much drafting one. (you probably have a 20-30% chance if you have a high pick), it's the fact that you need 3-4 years to figure out what you got, so you don't actually pick that often. 

 

 

Points: 0

#107 by Mike B. In Va // Oct 06, 2022 - 12:59pm

My inability to count while posting during meetings blows up my own argument - somehow I blocked out Murray getting to a playoff game, and Hurts completely. ;) Herbert and Tua don't have a playoff appearance yet, and there's a slight chance Tua won't ever get one (although I really expect him to be playing next week.)

Still, though, you have to endure being *bad* to get there, then have a 3-4 year period before you figure out if the guy is actually worth keeping. Unless you get a rare guy higher that works out. The Bills weren't truly awful until the year they took Allen (they made the playoffs the year before, although by backing in), and the Ravens were never awful.

I seem to be killing my own argument at this point. Sorry for wasting everyone's time.

Points: 0

#65 by colonialbob // Oct 05, 2022 - 2:42pm

Those are two very different definitions of the word "worked", though. If bringing in the veteran guy works, you get 1-2 seasons of solid to good QB play before you have to find a new solution. If drafting a QB works, you get 2-4 years of solid to great QB play at a bargain price and then very likely several years of solid to great QB play at market price after that. The payoff for hitting on a QB in the draft is just so, so much higher.

Points: 0

#93 by Mike B. In Va // Oct 05, 2022 - 8:33pm

...but the downside is so much lower for someone in Ballard's position. I completely understand his position, since his ticket to the long term QB ride abruptly cashed himself out and left a team that can win with decent QB play.

Points: 0

#99 by colonialbob // Oct 06, 2022 - 10:32am

I mean.... is it? Sure, if you're assuming the Colts drafted somebody top 5 and turned the team over to him day 1, the downside of him failing is getting fired. But if they had drafted Kenny Pickett or (especially) a 2nd or 3rd rounder, they could still have started a veteran retread while at least giving themselves the potential for a QB of the future; you know, sort of like Tennessee is doing.

And this is all assuming he'd get fired after one year of missing the playoffs if he had picked a rookie instead of, you know, trading multiple draft picks for some veteran QB who misses the playoffs and also wears out his welcome after a single year. Hypothetically.

Points: 0

#106 by theslothook // Oct 06, 2022 - 12:42pm

Just by probability, If Ballard had gone with a lower round rookie instead of Philip Rivers and instead of Carson Wentz, I think both he and Frank Reich would be fired right now. 

Yes, there is a possibility that that low round rookie exceeds expectations wildly and now their careers are secured. 

From the point of view of a fan, If you're not winning the super bowl and you're going to end up in this hell hole situation either way, you might as well get it done sooner than later. 

From the point of view of Ballard, in one universe he's still employed and in another universe he's fired at least a year or two sooner. 

I know which decision I'm making if I'm Ballard

 

 

 

 

 

 

Points: 0

#108 by colonialbob // Oct 06, 2022 - 1:31pm

Why is it instead, though? Again, you've got tons of examples of taking a younger guy relatively high but not necessarily throwing him into the starting role right away. Packers with Love, Eagles with Hurts, Titans with Willis, Steelers with Pickett - all of them varying degrees of "sit the rookie behind a veteran QB". 

I also disagree with the certitude that Ballard/Reich would be fired. They specifically went out and traded good assets for Carson Wentz, a move that blew up in their faces by resulting in them doing worse than the previous year, missing the playoffs, and turning around and ditching him a year later. How is that not a move that gets a GM fired, but drafting a rookie with the idea of sitting him behind a Rivers or a Ryan type would?

Points: 0

#109 by theslothook // Oct 06, 2022 - 2:18pm

So let's run through your examples.

The Packers drafting love looks like a completely wasted pick and was suboptimal. The only reason Hurts ever saw the field was because Wentz imploded so badly that the team suffered a disaster season. And the coach got fired. Pickett fell to the Steelers for a reason and the fact that he was there at all is sort of a happy accident. And who knows what the Ritter pick will amount to. But taking this view seriously, it basically amounts to grooming a young prospect until he's ready to take over from the aged veteran. And I don't find much evidence for this either. Denver tried that with Brock Osweiler and it failed for Houston.

Really, the question is the following:

I have no answer at QB and I am facing a win now mandate(this is the assumption I am making). I can either hope one falls to me late in the first round (possible but hard to plan for), draft one in the second round or later and go with those options at starting QB. OR, I can sign a veteran who maybe has 2-3 years left, affords me more time to maybe pounce, Chiefs style, on a QB I want but not in a position of pure desperation. And never forget, sometimes you get lucky and you land a successful reclamation like Ryan Tannehill and your run continues for even longer than that.

Outside of the first round pick, going into the season with a low rounder as your planned starter usually leads to you getting fired. This is what happened to Rex Ryan via Geno, what happened to Fangio via Drew Lock, what should have happened to Hue Jackson via Deshone Kizer and on and on. 

The only times I can remember a team willingly went into a season with the planned starter being a low round rookie and it succeeded was Russell Wilson. Otherwise, I really can't think of any other examples.

 

 

Points: 0

#110 by Mike B. In Va // Oct 06, 2022 - 2:29pm

The only times I can remember a team willingly went into a season with the planned starter being a low round rookie and it succeeded was Russell Wilson. Otherwise, I really can't think of any other examples.

...and even then it was because he won the job outright over a couple of journeymen. They tried to hand it to someone else.

Points: 0

#111 by colonialbob // Oct 06, 2022 - 2:48pm

I have no answer at QB and I am facing a win now mandate(this is the assumption I am making). I can either hope one falls to me late in the first round (possible but hard to plan for), draft one in the second round or later and go with those options at starting QB. OR, I can sign a veteran who maybe has 2-3 years left, affords me more time to maybe pounce, Chiefs style, on a QB I want but not in a position of pure desperation

This is what I'm saying, though! Sure, you still acquire the veteran, but you also draft a potential replacement. Hurts only saw the field because Wentz imploded. Dak only saw the field because Romo got hurt. But they still ended up doing well! I don't know why drafting a Hurts or a Willis or a Lock or even a Pickett means that you have to go into the season with them as a starter. Would anybody have really taken issue if the Colts had drafted Pickett (or Desmond Ridder) and said "Matt Ryan is our quarterback for this year, and we feel this gives [rookie] a great chance to observe him and learn what it means to be an NFL player?" The 49ers did this last year with Trey Lance, and he was drafted 3rd overall! You're creating this false dichotomy between "pick a rookie" and "plan to start a veteran".

Points: 0

#112 by theslothook // Oct 06, 2022 - 3:15pm

I guess I am not sure I see much of a difference. The Colts have Elhiger. Before that they've had Eason, so its not like they didn't have some unknown rookie there just in case. It just didn't make any difference. So I don't see how Ballard didn't do what you are suggesting. Is Malik Willis really more likely to bea great player than Eason barely a round separates them? 

The argument I made from the begining was to avoid these veterans in the first place. Don't sign Ryan and design your offense with him in mind. Don't sign Flacco with the same plan in place. Go with a low round rookie from jump. If Dak Prescott can succeed coming in for a creaky veteran, he can also succeed coming in as the starter from jump. Or go with a reclamation hope out of someone like Trubisky or Mariotta. 

But that path taken is probably more variable than going with the slow steady creaky veteran who has some baseline level of play you can count on. 

 

Points: 0

#113 by KnotMe // Oct 06, 2022 - 4:26pm

It's possible that still is the plan and there wasn't a decent rookie available. None of the rest if league was impressed by last years QB class either. 

Not sure just throwing a guy out there who isn't ready is the best option. If nothing else, the bust rates of very high picks where this happens imply that it may be best to have rookie QB sit for a while while they get used to the NFL .

Points: 0

#69 by ahmadrashad // Oct 05, 2022 - 3:00pm

Not following the Colts too closely, it sounded like the GM/Coach were planning to stick with Wentz, but the Irsay told them either he goes or you all go. So it would be a real raw deal if the owner fired them for doing his bidding. 

The other thing I've noticed about the Colts is they are very willing to BS the fanbase to sell season tickets. e.g. pretending Manning could come back any week now, delaying announcement of Luck's retirement, etc.  So Ryan fits in with that as a plausible substitute, and would also explain the lack of other moves. 

Points: 0

#76 by jheidelberg // Oct 05, 2022 - 4:24pm

Yes, the constant musical chairs at QB is far more damaging long term.

For sure, the musical chairs is a major problem.  There are many good/excellent QB's that are not drafted early in the first round.  Of course Brady, Russell Wilson, but also Jalen Hurts, Lamar Jackson, Aaron Rodgers.  Please note, I left Geno Smith off of this list.  

Its better to have a good team and draft your QB so that he is not taking over a disaster.  I do not believe that this year's Colts other players constitute a good team, it sure appears that this is the worst Colts team in quite some time, regardless of the QB. 

The Colts win total is consistently boosted by playing in the AFC South (well they could conceivably win in Jacksonville some year).  Then there will always be other games such as tomorrow against Nathaniel, "I can devise a game play and make decisions to lose to the number 32 ranked DVOA team" Hackett, among others.

As you have said in a previous post, I agree with you that Frank Reich will unfortunately take the fall for this team and be dismissed.

 

Points: 0

#77 by theslothook // Oct 05, 2022 - 4:33pm

There are many good/excellent QB's that are not drafted early in the first round. 

This undoubtedly true. However, they are an even bigger crapshoot than first round QBs are. Pat mentioned above that QBs drafted in the top half are about 50 percent likely to make a probowl. I think the corollary is even more correct - 50 percent of QBs drafted high are going to end up busts. And the later you go, the further that probably rises. I wouldn't be surprised if drafting a Qb anywhere past the 2nd round has a conventional bust rate above 90%(ie - their mean value would look like Mitch Trubisky or worse). 

Again, I don't hate that strategy because its almost costless. If the guy works out, terrific! If he busts, you have clarity and likely a high pick. There is very little room for true middle grounds when it comes to low round rookie starters. But if my goal was self preservation, I certainly would not choose this route. After all, I am not going to be the one to coach the team and draft its players in the future.

The Colts have talent. Its not a lot and its not at impactful positions, but they aren't some rock bottom team completely bereft of building blocks the way Jacksonville was when they drafted Lawrence. I think right now they have the worst starting quarterback in football. And just as someone like Manning can make everyone look good doing their jobs, Ryan is probably making everything and everyone look incompetent. 

I feel bad for Ryan because whatever hall of fame good will he has engendered over his career; this horrendous season is going to be the lasting image people have. Its what happened with McNabb.

Points: 0

#80 by jheidelberg // Oct 05, 2022 - 4:51pm

I know that FO once did an average 1st round and 2nd round QB.  I need a writer to help me out, I believe that a first round average QB is Mark Sanchez and I believe that Geno Smith was the average second round QB.  If this is the case, FO may have to revise the numbers if Geno keeps dominating this season.

Yes drafting is a crapshoot, but you must roll the dice instead of accepting proven mediocre QB's.

Points: 0

#82 by KnotMe // Oct 05, 2022 - 5:07pm

The problem is, owners don't understand this and dumping an OKish guy early just gets the FO fired. (Even worse if they go somewhere else and improve). 

In terms of actually winning titles it's probably the best option, since a hit gives you a big window (although it's difficult to figure out who is mediocre and who isn't early enough). But it's not practical from a standpoint of ownership or selling tickets. 

Points: 0

#84 by jheidelberg // Oct 05, 2022 - 5:33pm

Yes, I agree with you.  As I have always stated, without being refuted, if all players were available in a draft, and you had the 15th pick, you would end up with the 15th QB as he is the 15th most valuable player in the league.  If your intent is to win now, and you only care about this season, you must take this guy, because if you do not, you will end up with another top player at another position and the 32nd best QB.

Good QB's do not grow on trees, and beyond the few teams that have an elite QB, everyone else is trying to win with what they have, whether that is Kirk Cousins, Jimmy G, Baker Mayfield, Daniel Jones etc., etc.

Points: 0

#38 by ChiefsCML // Oct 05, 2022 - 12:51pm

I don't know about other teams, but I genuinely don't think garbage time should matter much for Andy Reid coached teams. He's notorious for only revealing as much as he absolutely has to and keeping it vanilla even to the detriment of his team during regular season games. He goes full tilt during the postseason, though. Anyway, I would like to see how predictive garbage time is for Andy Reid teams in particular, because my intuition leads me to believe it wouldn't be very.

Points: 0

#72 by Tutenkharnage // Oct 05, 2022 - 3:30pm

He routinely rolls out triple-halfback double-option plays and other such wackiness on blowouts or meaningless NFC contests, doesn’t he? I can’t think of a time I’ve watched the Chiefs in the playoffs and thought, “Wow, they certainly surprised me with this week’s game plan!” I don’t watch them every week, but I watch them whenever I can, so maybe I’m just missing something. 

Points: 0

Save 10%
& Support Aaron
Support Football Outsiders' independent media and . Use promo code SCHATZ to save 10% on any FO+ membership and give half the cost of your membership to tip Aaron.