Thurman#1

Dude, I wasn't implying that or anything like that. You didn't just miss the point, you selected a new point out of thin air and assumed it was mine. Of course it considers passing, sacks AND scrambles. That's what I said. Take a look: Get it? Since passing DVOA already includes sacks ... when I was talking about their adjustments and said "if you include scrambles past the LOS" I was of course pointing out that the second stat includes passes, sacks and scrambles. So -for the second time - of course adding Tyrod's scrambles to his passing DVOA is going to make him look better than it makes most NFL QBs look. We all know that's the strong part of his game. Whyncha try reading carefully next time before you get your panties all sweaty and in a bunch.

Heh heh. If America eats less and exercises more, we gotta real shot at losing weight. Yeah, um.

Not really buying this much, Happy. Mostly I don't think these tell us anything we didn't already know. I'll try to be more specific. ARGUMENT ONE: Whether or not a sack destroys a drive is dependent on a lot of things that don't especially say much about a QB. A sack on first down, for instance is a lot less likely to destroy a drive than a sack on third down. Does it make a QB better that he got a sack on first down rather than third? A ten yard sack is more likely to destroy a drive than a one-yard sack. Does it say anything better about a QB that he gets sacked for one instead of ten yards? Or did he just have a more convenient escape route in that direction? I don't think that's even slightly clear. This article is slightly interesting for it's own sake - in terms of what Klassen is saying about sacks and how they affect drives - but it doesn't say much that helps evaluate QBs or boosts Tyrod's evaluation upwards either. I'm not a big Trapasso fan, honestly. But here's his conclusion to the whole article, "While Tyrod will probably always be near the top of the league in sacks, a proven track record of being able to rebound from those quarterback takedowns better than any team in the league is vital. Technically, it means a sack isn’t as damaging to the Bills as it is for other teams, which gives Buffalo’s offense a clear leg up on the rest of the NFL." Hunh? So Tyrod gets sacked near the top of the league and probably always will be up there? So his 42 sacks are OK, but we're supposed to be happy that the Bills very small percentage of converting on sacks is three or four plays higher than the other NFL teams? That's the big conclusion? What that is saying is that Tyrod gets sacked a lot more than other QBs but that the Bills do good things anyway on a larger percentage of those drives. And that gives us a leg up? That's very questionable. First, these stats were team stats, not QB stats and Tyrod's sack percentage (8.8%) was quite a bit worse than the percentage for Bills QBs (7.09%, 4th-worst in the league) and around 10% worse than any total team (ARIZ was worst with 8.08%). Put another way, Tyrod had 42 sacks while the 16th-rated QB had 33 and the 17th had 31. So Tyrod had about 9 more than average even though he didn't play all 16 games. Assuming the Bills had converted at a league average rate (16.01% according to the Klassen article) rather than our very fine rate, how many fewer conversions would we have had? Around nine. Not a lot positive there in terms of improvement of results. It's an interesting article about how sacks kill drives. Doesn't make Tyrod look better at all. ARGUMENT TWO: This seems to me to show that Tyrod was very good at running the ball. And that that is also true when he's under pressure. I would have guessed that. The stat addresses the fact that if you include scrambles past the LOS, the DVOA (really a whole offense stat, not an individual player stat, as Football Outsiders not just admits but publicizes) improves. Yeah, I would have expected that. I've never ever been concerned about his scrambles once he gets past the LOS. I'm worried about his pass game. Not Tyrod's run game. I know that's excellent, and I think we all do. This also doesn't change anything for me. ARGUMENT THREE: Yeah, it's a new article but that stat has been around a while now and we've been arguing it for months. It's great that he threw 10 TDs under pressure. But that means he only threw 7 when he wasn't. Seventeen total. It's great that he (the offense, really, in the scheme, with the play calls, yadda yadda yadda) performed so well when under pressure. But that's only part of the game. Agreed that RT was a real weakness, but these days very few teams have as many as four good OLs, as we do. I'm hoping that things improve at RT this year, however that happens. I don't see how this puts any dent whatsoever in the criticisms that running isn't as important as passing.

That sounds good, but there are plenty of exceptions. Belichick may be the best coach in NFL history, unfortunately. But he has a ton of turnover. He wants and gets a certain kind of player. The Pats went 5-11 in his first year though they'd gone 8-8 the year before under Carroll. He didn't coach guys who didn't fit his system so well that the team kicked butt. He brought in guys who fit his system. And it took him time. Pro Football Reference lists 55 guys on the roster of Carroll's last group, the '99 Pats. In '00 in Belichick's first year, they list 65. Guys gone from Carroll's roster: Terry Allen, Mike Bartrum, Terry Billups, Vincent Brisby, Chris Carter, Rico Clark, Ben Coates, Ferric Collons, Vernon Crawford, Damon Denson, Ed Ellis, Jerry Ellison, Heath Irwin, Steve Israel, Shawn Jefferson, Jeff Kopp, Bob Kuberski, Marty Moore, Sean Morey, Zefross Moss, Todd Rucci, Bernard Russ, Chris Sullivan, Lamont Warren, So 24 of Carroll's 55 were gone. And they won five games. 43.6% turnover and Belichick didn't coach the guys on the roster to greatness. He changed everything and input his systems. The next year, 2001, they list 61 guys. It's easier to count the guys who WERE on that 1999 Carroll roster than those who weren't. The survivors were: Drew Bledsoe, Troy Brown, Tedy Bruschi, Terry Glenn, Lee Johnson, Ted Johnson, Tebucky Jones, Ty Law, Willie McGinest, Lawyer Milloy, Brandon Mitchell, Marty Moore, Rod Rutledge, Adam Vinatieri, Damien Woody. Only 15 of 55 remaining. And only the highlighted guys were starters that year, though McGinest missed most of the year, maybe due to an injury or something. Belichick didn't so much coach the leftovers to greatness as he did bring in a group that fit his schemes. Not that I'm saying McDermott is the new Belichick. Just that even very good coaches sometimes have major turnover and don't do that well with guys left over from the last era.

The stats are two years old but still very relevant because they show that roster turnover is high for everyone. It's from a reddit: "Which teams have the longest average tenure? (League average is 2.263 seasons, between the Ravens and Colts.) Packers: 2.918 seasons Patriots: 2.882 Bengals: 2.861 Steelers: 2.628 Chargers: 2.506 49ers: 2.488 Falcons: 2.467 Cowboys: 2.455 Seahawks: 2.360 Vikings: 2.329 Saints: 2.329 Eagles: 2.324 Lions: 2.312 Ravens: 2.280 Colts: 2.241 Dolphins: 2.231 Texans: 2.227 Panthers: 2.211 Bears: 2.194 Jets: 2.192 Redskins: 2.176 Bills: 2.158 Cardinals: 2.156 Rams: 2.082 Chiefs: 2.071 Titans: 2.063 Broncos: 2.051 Raiders: 2.025 Giants: 2.011 Browns: 1.909 Jaguars: 1.853 Buccaneers: 1.568 All of my raw data comes from Pro-Football-Reference's yearly roster listings, which I think includes all players that were on a gameday active roster for at least one game in the season." Turnover is high for everyone. It's the nature of the game that guys at the bottom of the roster have always had extremely high turnover and it's harder to hold on to guys at the top of the roster in the salary cap era. Especially when you're close to the cap. Gilmore's a good example. I always remember the joke from Seinfeld, "you're basically rooting for the clothes." It's more so for teams with new coaches and schemes but it's really true for everybody. We won't know how much turnover there's been till the roster firms up after camp, but we're likely to be somewhere around 70th percentile this year with the cap issues and the new regime.

OK, if that's your point, I withdraw that particular objection. But I'm afraid I have a different objection. Which is this: there's a reason they rank teams offensively by how many yards they get rather than how many points they score. And the Bills were 16th in offensive yards. They did not provide the defense with good field position. Last year's offense made the defense look worse and the defense made the offense look better. The offense had the 11th best average drive start field position in the league while the defense had the 23rd best. Yards far better separate offensive performance from the defensive and STs performance. Whereas points have a much larger proportion of responsibility for the whole team. Both yards and points are important but yards better isolate each unit from the others. Oh, and I'd also say, "good enough"? Good enough for what? Good enough to make the playoffs? Yeah, probably, as a fodder team. But to be competitive for a Super Bowl victory? I don't think the offense was good enough. To win a Super Bowl with that offense, I'd argue you would have to have an absolutely sensational defense and a lot of luck and good timing besides. I have no objection to calling the defense bad. They were. But the offense wasn't as good as some Bills fans believe. Ask around the league where they rank and you'll find tend to find people ranking them 16th, not 7th, and thinking that's a pretty reasonable representation. We fans who watched the games might argue we know better and crank them up a few spots but I don't think too many reasonable non-Bills-fan observers would say they were the 7th best offense in the league.

I'm not arguing that Tyrod wasn't a contributor. He certainly was. But the bottom line was that the run game was the best in the league and the pass game was sub-mediocre. That's not all on Tyrod, but a lot is. If you're going to give the QB credit for running yards, you can't give him credit for all the passing yards. It's certainly not 100% Tyrod's yards when he throws, say a screen pass to Gillislee who fakes a guy out of his jock and scores a 45 yard touchdown. So I'd say that stat you're trying to use is also a bit simplistic. He's not responsible for 63.6% of their yards, not unless you're willing to say that Sammy Watkins is responsible for zero percent of their yards. Why wouldn't we run in the red zone? I agree. And that's the point. The run game was really good and the pass game wasn't. Why not run in the red zone and out of it? And you're assuming that all the running TDs came in the red zone, and that's not true. Oh, and a QB generally ought to have a higher QB rating in the red zone because it's gonna tend to be easier to get TDs there than elsewhere on the field. It's also true that in the red zone he completed 60% (40.9% from inside the 10) and had a 4.8 YPA.(2.2 inside the 10). And the nature of red zone passing attempts surely affects those numbers too. Not sure that particular stat (and I like passer rating) should be used in that kind of a split. People keep wanting to say that we had a lot of offensive TDs and therefore the passing game was good. And this doesn't logically folow, especially when you have such a terrific run game.

If Gilmore were the only argument here, it would make sense, as it's hard to say whether they wanted to pay him that much. It's not only Gilmore. It's Gillislee and Zach Brown and Robert Woods and Robey-Coleman and depth guys like Douzable and others on a team with very little depth. They needed to bring in safeties as they just didn't have anyone there. That used up most of the cap space they felt comfortable with using. Now what happens if we lose an LB to injury? They can bring in a journeyman but don't have anyone as good as Brown and aren't likely to get one. They're hovering close to the area they want to be in to have money available to bring in a cheap guy or two before camp if someone disappoints or when the cuts happen and still have enough left to keep their usual amount for in-season injury replacements. How can you say absolutely none of those decisions was primarily about cap? Those are exactly the difficult choices being cap hamstrung forces you into. You keep one guy so you have to let go one or two more. Even if you have poor depth and would like to keep him. I don't pretend to know how much Overdorf was responsible for, though my guess is it's more than most think. Fitting stuff under the cap if you know the salary range to shoot for isn't all that difficult. It's nothing someone with a bit of business or law background couldn't pick up with a few months to study. The difficult part I'd assume is the negotiations and probably other parts of the job I know nothing about. I have bad feelings about the guy but may be totally wrong. I'm willing to believe McDermott and Beane know enough that if they keep him around after a while to watch his performance it may be because he does a good job.

Well, first, they're not top seven, they're in a three-way tie for seventh. But that's a quibble. Here's the main point. For the third time now, I understand that they're top 7 in offensive touchdowns. And that's huge if you're looking at the performance of ... wait for it ... the offense, the whole offense. See how that works? Offensive touchdowns are produced by the whole offense. Whereas .. and here's what you missed the first two times ... when you are trying to look at how good the pass game is ... you look at what the pass game produced. Not what the run game produced. See how it's kind of an equivalence? The offense scored a lot of TDs. Because the run game was terrific and scored a bunch of TDs. Whereas the pass game was substandard and did not score a lot of TDs. And now the architect of that terrific run game, Roman, has left and is in Baltimore. This is cause for worry that the run game might not be as good next year. All is said to pee in your Cheerios was thank goodness for the run game that scored all those points. They covered up the poor performance of the pass game. I didn't say anything implying that the whole offense didn't score a lot of TDs. I understand that it did. I merely pointed out that it was the run game's doing, that the run game scored almost 2/3rds of the Bills TDs and that no other team had less than 50% of their TDs scored by the passing game. Don't know why stone cold facts like this would make you angry. Unless of course you're trying to use a measure of the performance of the whole offense to come to unwarranted conclusions about a mere part of the offense. So again, the offense scored a lot. Can't argue with that. But it was overwhelmingly the extremely good run game (scoring 29 of our 46 offensive TDs, 63% when no other team was above 50%. Can't argue with that either.

espn.com Look for the splits. I don't know. They just were. Ask the coaches, maybe. They thought it was their best chance with the personnel they had, maybe? The reasons don't matter to me, personally. What matters is that their percentages for whatever reason were very close to the same across the four quarters.

He did get them to play hard and smart, I think. I didn't like his extreme conservatism on things like punting and game management, but yeah, they played hard. But while you could say they were rebuilding, it's a stretch, as they were in years four and five of that rebuild. And what McDermott is doing isn't rebuilding, it's reloading. If it were a rebuild they'd have gotten rid of McCoy, Kyle Williams, Tyrod, Incognito and any other guys who are too old to be around and contributing three and four years down the road. That doesn't mean it's not a major project they're working on; it is. But it's not a rebuild. I wouldn't be so quick to judge on Bortles, but I agree that what Marrone said here was totally reasonable, particularly for a new coach trying to set a tone.

Yeah, but come on. The reasons that these two teams ended up with high run percentages was widely different. Here are how the two teams looked at run / pass percentages, divided by quarter. Falcs, by quarter 1st Quarter - Runs 94, Passes 131 … 42% runs 2nd Quarter - Runs 102, Passes 161 … 39% runs 3rd Quarter - Runs 109, Passes 139 … 44% runs 4th Quarter - Runs 113, Passes 102 … 53% runs OT - Runs 3, Passes 4 Bills, by quarter 1st Quarter - Runs 101, Passes 97 … 51% runs 2nd Quarter - Runs 129, Passes 146 … 47% runs 3rd Quarter - Runs 107, Passes 106 … 50% runs 4th Quarter - Runs 144, Passes 119 … 54% runs OT - Runs 11, Passes 6 The Bills were roughly the same quarter to quarter. They simply wanted to run more. Whereas the Falcons passed at much higher percentages through the first three quarters and then burnt clock in the fourth quarter because they were way ahead. The Bills didn't have the problem of being way ahead in a lot of games last year. And this is why. Good teams tend to be ahead and run more to run out the clock in the fourth quarter, which lowers their passing percentage from it's natural early-game numbers. That's not how it was for the Bills. And the four teams that made the NFC and AFC championship games were ranked 4th, 13th, 14th and the Falcons were 27th and I showed why. On the other hand, the four teams that made the playoffs with high run percentages (excepting Atlanta) were Miami, Dallas, Kansas City, and Seattle. One playoff win, Seattle over Detroit. It might not fall that evenly most years but without research I'd guess that it would tend to fall that way consistently, though with exceptions. And building a team on a template that tends to on the ceiling have one-and-done as opposed to a template that had all four of the top teams is not a wise move, I would argue.

Yup. Thank goodness for the run game that scored almost 2/3 of those TDs. And again, not a single other team scored less than 50% of their TDs in the pass game. It may indeed reflect on Tyrod. Though the coaches may well have something to do with that. But perhaps that was in understanding of who their QB was. Beside the point, really. It really is very rare for teams ranking so low in the league in run percentage to win titles. It happens, but pretty much absolutely everything else has to fall perfectly. If he'd argued that we had to be in the top three teams in the league in attempts ... but he didn't. When teams rank that high it's generally because they're way behind a lot and trying to catch up. I very much agree with you that you don't have to throw it all the time. So, we're never going to talk again about any statistics of any sort except for Ws and Ls? Is that correct? People tend to make this argument when they'd rather not talk about the stat being discussed.

You mean that the Bills D was bad? You get no argument from me there. But some people want to put all the blame on the defense and none on our below average pass game. Woh, nice pickup. I looked at that page and totally missed that. Holy cow, that's a huge difference.

It shouldn't be considered a negative. But there's a reason that we were only 16th in total yards. And that helps to explain the sad facts that while our offense received the 11th best average drive start position in the NFL, they handed back to the defense the 23rd best average drive start position. Field position matters and our offense didn't deliver. If we can win (a Super Bowl) with this amount of runs, I'd be perfectly happy. But to do so you pretty much need the best defense in the league and a good passing QB. I'm not worried about this year. We're almost certainly going to be fairly bad. I'm worried about developing in the right direction in the long term.

https://www.teamrankings.com/nfl/stat/yards-per-game

That's not the way I saw it. He had some really below average QBing while he was here. Manuel / Thad Lewis / Tuel his first year and Manuel / Orton his second and he got more out of Orton than expected. The 2013 offensive starters looked like this: Manuel/Lewis/Tuel, Spiller, Frank Summers, Stevie Johnson, Robert Woods, Lee Smtih, Cordy Glenn, Doug Legursky, Eric Wood, Urbik and Erik Pears. In 2014: Manuel/Orton, Fred Jackson, Watkins, Robert Woods, Lee Smith Cordy Glenn, Urbik, Wood, Pears, Seantrel Henderson. I don't see coaching as being the problem nearly as much as personnel. I don't know if he'll be a really good coach. Wouldn't surprise me either way. I didn't like the way he left, but he appears to have had insoluble problems with Whaley, and isn't the only coach who felt that way when here.

5.7 YPC, no he almost certainly won't. But will he be a valued and successful rotational contributor who will deserve the salary he's getting? Yeah, probably. Yeah. And the end result was that the run game was very very good and the pass game was below average. Yup, it still totals the same way, but the run game simply looked terrific and the pass game just didn't. You tried to use the total offensive stats to prove the pass game didn't need much improvement. What that actually proves is that the run game didn't need any and the offensive totals were solid too, basically because the run game (with a run-game genius coach who is now gone) was superb. Doesn't mean the pass game wasn't sub-standard, though. It was.

Yup. Top ten in offensive touchdowns. And 37% of those TDs came on passes. 17 out of 46. So yeah, we need a better passing game. To a very large extent. By the way, know how many other NFL teams scored less than 50% of their TDs on running plays? Zero. Our passing game was simply below par. Having lost Roman and Gillislee (Roman being the key factor), we're unlikely to do as well this year running. Last year's 29 running TDs wasn't just best in the league it was far and away the best. The second-best team had 24 and the two teams tied for third had 20. We're not likely to get 29 running TDs again. The pass game is going to have to step it up bigtime for this team to be competitive. That's part of the reason why the consensus among the pundits of progressive prognostication is maybe five to six wins this year.

Gillislee got more per play than McCoy. Yeah the play designs helped. But that wasn't why Gillislee was good. Losing him hurts. Gillislee was already on the team and looking excellent when we cut Karlos Williams. In 2015, Gillislee had averaged 5.7 YPG, and that's why there wasn't much of an uproar except maybe sadness about losing Karlos. This year we haven't got someone ready to move in and take over who has already looked very good the way that Gillislee had in 2015.

Dak = Dez Tyrod Taylor = the best running game in the league, including LeSean McCoy, Mike Gillislee and an offensive line that was sensational at run blocking. All that helps a QB an absolute ton. Oh, and Charles Clay who was open constantly. Having said that, not a QB in the league would choose Tyrod's offensive cast over Prescott's, but the Bills run game would have been a huge help to any QB and Tyrod fans want to talk only about WRs and ignore everything else. It will indeed be interesting to see how this all pans out. My guess is that Dak is going to come back to earth a bit as teams start to figure out how best to defend him. As a Cowboys hater I hope I'm right about that, it would be awesome to see.

I don't think looking at this one measure you've created and calculated "clearly settles" anything whatsoever, much less whether or not Flacco is better than Tyrod. It ain't "the analytics" saying "that they are essentially the same player in terms of per game value to their team(with Flacco having a slight edge), but that Flacco is clearly in the "Average" QB category and not the "Elite" or even "Very Good" category." It's your one particular way of looking at this one particular stat. I guess you can look at it as some added data. Plus, I'm not sure where you see that Aaron Brooks is higher at weighted AV than Brees or Rodgers. Looks to me like what they say is that Brooks has a 65 weighted AV, 755th overall since 1960 while Rodgers has a 124, 26th overall and Brees has a 153 7th overall. Weighted AV has them in fairly reasonable places. Could it be your new stat that has them strangely? IMHO it's a bit questionable whether AV, a stat meant to be good at making sweeping generalizations about large amounts of data, a large groups of seasons together for instance, should usefully be divided up in an attempt to pretend that it is like a knife and can make small differentiations. That's not how it was designed. He (Doug from PFR) himself points out that there's a good reason he included the "Approximate" in Approximate Value. In any case, it looks to have been an interesting project. I am impressed by your energy and interest.

While you're certainly right that the odds are high to find a guy that late, anytime you add the phrase "and later" and then look at all that big "there and later" group, you're adjusting the odds to make it look even less likely. I mean, if you say, "Look at guys who were drafted in the first round or later ... the odds are terrible." Well, yeah, but the addition of "and later" does a lot to make the stats look impossible. Here's another way to look at this that doesn't put the guy with a huge group of guys most of whom do indeed look less likely than him to have major success. In the last 20 - 25 years or so, how many franchise QBs or guys who are still possible franchise QBs, or guys who came close were not drafted in the top three or four rounds? From Brady to Siemian (as I say, I'm counting guys who still have a legit chance to get there), throwing in Tyrod, Foles, Cousins, Romo, Hasselbeck, Marc Bulger, Mark Rypien, Warren Moon, Jake Delhomme, Brad Johnson, Kurt Warner, Dave Krieg, Trent Green, Steve Beuerlein, Rich Gannon, Grbac, Flutie, Bobby Hebert ... Fitz looked like he might be one of those guys for a while, though I somehow never bought in to him. Oh, and Dak Prescott. It's not that rare for somebody who has a real shot to turn out not to have been drafted high, and a decent amount have had real success. Looked at as a percentage, sure, it's a really low-percentage deal, but how many of those guys who weren't higher picks and didn't make it have had anything like the reviews Peterman has had in terms of the Gruden quote and so on. He's not an average low pick. Does that mean I give him a high chance to make it? Nah, but it's not infinitesimal as some here are making it seem either. Some guys do it. Maybe he won't be one but maybe he will.

This, and very much so. It's hard to find a good guy anywhere.

Ah, you only got on board in 2001? Well, that makes total sense to me. I've been a fan long enough to have cheered for them to lose the last game of the year so they'd get the first pick and be able to draft O.J. Simpson. I saw the four SBs. Making the playoffs as a fodder team would do nothing to me. But if I were in your shoes, I'd probably feel the way you do. But I'm right with you that they have so often valued the short term with or even over the long term. Which is why I absolutely loved seeing them pick up that 2018 first round pick this draft. Long term thinking. I love it.