Orton's Arm

I did a little searching and found this. That KC defense you mentioned was listed as one of the best of all time. One could dispute the order of the ranks however. In fact, I'm about to do exactly that! The Ravens of 2000 got listed at #3, one rank behind the Bears of '85. And yet, the Bears allowed 11.7 points per game during the regular season, whereas the Ravens allowed only 10.3. Not only that, but the Ravens offense was bad. It went five games straight without scoring a touchdown. It became somewhat less bad as the season went on, but was never comparable to the offense the Bears had in '85. Because of the Ravens' offense high number of 3-and-outs, the Ravens' defense was under a lot more pressure than was the Bears' defense of '85. Another point to consider is that the NFL's rules were more passing-friendly and offense-friendly in 2000 than they'd been in 1985. For the Ravens defense to prevent 12% fewer points per game than the Bears' defense, despite the twin disadvantages of different rules and a significantly worse offense, says very good things about that Ravens defense indeed! Not only that, but that Ravens defense got better in the postseason! They allowed only 3 points in the wildcard round of the playoffs. They faced the Titans in the divisional round. They allowed 10 points, but scored 7, which means they allowed 3 points net. They allowed 3 points in the AFC Championship Game. Then in the Super Bowl the Ravens defense did not allow any points to the Giants, whose only touchdown came on a kickoff return. Over the course of four postseason games, that works out to a net points allowed of just 2.25 points per game!! Granted, the '85 Bears defense also did better in the postseason, allowing a miserly 3.3 points per game. (And that's gross points allowed, because I wasn't able to subtract away the points the Bears' opponents scored on special teams or defense, or take into account whichever points the Bears' defense scored.)

You are reading waaaaaay too much into my earlier post. I simply indicated the link was depressing (which it was) and that thus far Spiller has done nothing to justify his draft selection (which he hasn't). I did not comment on whether Spiller might turn into a good RB someday, nor did I suggest the Bills should cut him. As a general rule I am very loath to cut a first, second, or even third round pick after just one year. Spiller hasn't (to my knowledge) done anything drastic enough to justify a quick removal. He will (and should) get at least another year in which to learn and to prove himself. Maybe NoSaint is right, and maybe Spiller's anticipated improvement in pass protection will help his rushing stats as well. (By making the offense less predictable when he's on the field.)

Just to add to my earlier post: Bradshaw's career average was 7.2 yards per attempt; Peyton Manning's is 7.6 yards per attempt, and Trent Edwards' average is 6.5 yards per attempt. (Bear in mind that Bradshaw played during a different, less passer-friendly era.) Ryan Fitzpatrick's career average is 6.0 yards per attempt, and he averaged 6.8 yards per attempt this past season. The assertion that Fitzpatrick is at least as good as Bradshaw is not borne out by the stats.

During the regular season, Bradshaw averaged 7.2 yards per attempt. That's not too shabby, especially considering Peyton Manning's career average is 7.6 yards per attempt in a different, more passer-friendly era. It's also worth noting that Bradshaw played at a high level once he got to the postseason. From Wikipedia: In the second Super Bowl between the Cowboys and Steelers, the vaunted Steel Curtain defense had allowed Staubach and the Cowboys to score 31 points. Staubach's success against the Steel Curtain demonstrates that even a great defense can have a lot of trouble stopping an offense led by a franchise quarterback. Fortunately for the Steelers, Terry Bradshaw was able to bail them out with a franchise QB-level performance of his own; and led them to 35 points. That game demonstrated the importance of franchise QBs, even back in the '70s when DBs were allowed to murder WRs! Jeff Hostetler is another guy who played a lot better in the postseason than in the regular season. From Wikipedia: "Perhaps most impressive about Hostetler was his ability to perform very well in the post-season. In five playoff games, he completed 72 of 115 passes (62.6 percent) for 1,034 yards, seven touchdowns, no interceptions, and a 112.0 passer rating while going 4-1." To describe Hostetler as a career backup is inaccurate: he was a starter for four years with the Raiders, and was elected to the Pro Bowl for one of those seasons. Almost every Super Bowl winner has had either a franchise QB, or at very least a player like Jeff Hostetler who played at a Pro Bowl level or higher during his team's championship postseason. You can point to exceptions to that rule. The Ravens of 2000 won the Super Bowl without getting impressive play from the QB position. But those are very, very rare exceptions.

That was a depressing link I just clicked on! The first eight picks in last year's draft were mostly A's, with a few B's or B+'s mixed in. Then there was the Spiller pick (C+), after which it went right back to A's and B's. Every player taken within the first 28 picks of the 2010 draft was rated B or higher, except for Spiller. I'm more concerned with Spiller's average yards per carry (3.8) than with his lack of pass blocking. Normally pass protection is the thing rookie RBs struggle with the most. But Fred Jackson averaged 4.2 yards per carry with the same OL Spiller had. The whole point of taking a RB 9th overall was supposed to be to get an upgrade at the RB position. A reduction of 0.4 yards per carry versus the current starter doesn't seem to provide that. I couldn't care less what Spiller does or doesn't do on special teams. You don't take a RB 9th overall just to be a return man. Especially not when punt returner and kick returner are the only positions your team may actually have even more depth than it did at running back!!

If I understand your point correctly, you're arguing that a team which wants to win the Super Bowl should prioritize defense first, and franchise QB a (distant?) second. If that is indeed your argument--which I'm not sure of--it seems odd for you to cite the '49ers Super Bowl victory in 1984 over the Dolphins. Granted, the 49ers had the #1-ranked defense that year. But they also had Joe Montana. As for the 1978 Super Bowl when Pittsburgh beat Dallas--the Steelers had Terry Bradshaw. I realize Bradshaw's regular season stats aren't exactly the stuff of legends. But he played at a much higher level in the postseason than the regular season. That Super Bowl also represented a game in which both teams were getting franchise-level play from their QBs, regardless of whatever Bradshaw had or hadn't done in the regular season. One could make the argument that when both teams have franchise-level QBs, those two players will cancel each other out. That would leave the game to be decided by players at other positions. The Super Bowl between the Colts and the Saints was like that: Manning and Brees played at about the same level as each other for that game--the outcome of which was decided by the Saints defense playing better than the Colts defense. (And probably also by the Colts offensive line not playing well enough.) But the only reason that was the case is because both teams brought franchise QBs to the table. If for example one team had had Peyton Manning and the other had featured Trent Edwards, whichever of those two teams had Manning would have won. More generally, having a franchise QB gives you a very solid edge over any team that doesn't have one. If you want to beat a team with a franchise QB, and if you don't have one yourself, you have to outplay them by a lot in some other aspect of the game to make up for your lack of a franchise QB. The deeper into the playoffs you go, the harder that is to do. If it's your good defense + your Trent Dilfer of a QB against someone else's franchise QB + lousy defense, then you stand a good chance! But teams like the Steelers and the Packers have good defenses and franchise QBs. If you face a team like that, will your own defense be enough better to make up for your disadvantage at QB? Unless your defense is Ravens of 2000 good, the answer is clearly no.

Steel Curtain of the '70s and Bears of '85.

1) Edwards being the guy on day 1 was Gailey's call, not Nix's. 2) Let's suppose for the sake of argument that the rumors about the Saints are true, and they would have offered up a third round pick for Lynch. The Saints were defending Super Bowl champs, so it was reasonable to believe their pick would be toward the end of the third round. The deal Nix actually got for Lynch--a 4th and 6th rounder--isn't much worse than that. Plus, for all we know Nix held onto Lynch for a while after the draft in hopes someone would offer more than just a 3rd rounder for him. If that was the plan things obviously didn't work out that way. But Nix wouldn't exactly be the first GM to hold onto a player if he wasn't satisfied with what he was being offered for a trade. Belichick threatened to hold onto Bledsoe if no one offered a first rounder for him--which was clearly a lot more than Bledsoe was worth. Belichick looks like a genius because TD fell for his scam. But he'd look a lot less impressive if Bledsoe had remained on the Patriots' roster for years despite someone offering him a second round pick. 3) I agree Cornell Green was an embarrassing signing. But according to a football rankings site, his play while with Oakland was actually somewhat decent. Obviously that wasn't the case during his time in Buffalo. One possible reason I've seen thrown around for his dramatic drop-off in play was that he might have been playing hurt this year. The fact that he's getting on in years doesn't help anything either. Besides that, any given GM isn't necessarily going to be able to sign the best available free agent at a given position. There are 32 GMs in the league, and a lot of holes on a lot of teams. While the Green signing was a mistake, Nix recognized it quickly, moved on quickly, and didn't sacrifice much in the process.

I disagree. Tim Graham simply listed each team's payroll, and then divided that payroll by its number of wins to determine its cost per win. He did not advance any particular statistical or economic theory. He merely provided raw data for us as fans to chew on. But apparently a few fans decided to chew on him instead! I'll grant that Graham isn't exactly going to be in line for a Pulitzer Prize over this piece. It's just a niblet of potentially interesting data--no more and no less. Maybe when you clicked on the link you expected a full-fledged article, which this clearly was not. But that said, there's nothing wrong with knowing how much the Bills spent in relation to their divisional rivals, or where that payroll ranks in comparison to the other teams in the league.

Thanks for the compliment. I agree that Baltimore significantly overestimated the benefits Grbac would bring, and may well have underestimated what Trent Dilfer had brought to the table. Grbac was released after just one season with the Ravens. But not every free agent QB signing will turn out to be another Elvis Grbac. A few years ago the New Orleans Saints signed Drew Brees despite concerns that had been raised about his injuries. That decision worked out very well for them!! There are both Elvis Grbac and Drew Brees stories at every position on the field. Nearly a decade ago, the Atlanta Falcons traded away a first round pick for Peerless Price. That decision didn't work out so well for them! Jonas Jennings and Nate Clements weren't the players San Francisco had been hoping for when it signed them. Nor was Cornell Green the RT Nix had been hoping for when he acquired him. Marcellus Wiley's career went downhill after he left Buffalo, and he did not turn out so well for the Chargers or (especially) the Cowboys. The Ravens' acquisition of Grbac was not much different than those. The problem with evaluating a player at any position is that a football team is like a machine: the success or failure of one of the machine's components directly affects the productivity of other, related parts. A WR could say that his play would look a lot better if he had a better QB and if the OL blocked better. An offensive lineman could tell you that his blocking would look better if the quarterback did a better job of moving around in the pocket/being aware, and of getting rid of the ball faster. That OL might even add that it would be nice if the receivers got open faster. Similarly on defense, a DL could say that he'd have better stats if the secondary could hold its coverage longer, and if the other defensive linemen did a better job of attracting double teams. Fitz has thrown a lot of inaccurate passes. To me as a fan, those passes look like they're his fault. But it's theoretically possible that those passes would have been spot-on had the WR run exactly the route he should have. Chan and Buddy will doubtless be able to figure out which plays to blame on Fitz, and which to blame on the WRs. But my own impression here is that despite Fitz's other virtues--and they are many--he isn't a particularly accurate passer. That's something that can and should be evaluated now, as opposed to waiting until some future time when his supporting cast is exemplary. When evaluating a player--whether a potential draft pick or someone already on your roster--you should make a list of the qualities you want, and then determine the extent to which the player in question provides those qualities. Fitz gets high marks for his ability to read defenses, on-field generalship, leadership, and other important things, but low scores for accuracy. You don't need to surround Fitz with a top-10 supporting cast to determine that! Nor will improving his supporting cast make him into something he's not (an accurate QB). There are plenty of reasons why drives can stall. Holding penalties. Runs that get stuffed. Receivers dropping passes. A few bad play calls. A QB who sometimes throws very inaccurate passes creates one more thing to go wrong, and one more reason for drives to stall. A QB who's accurate 90% of the time has a 1/100 chance of throwing two inaccurate passes in a row. A QB who's accurate 75% of the time has a 1/16 chance of throwing two inaccurate passes in a row. Obviously, the second QB is going to stall a lot more drives than the first!

Over the course of his career, Kerry Collins averaged 6.6 yards per pass attempt. Trent Dilfer averaged 6.5 yards per attempt. While I'll grant that Collins was probably a slightly better QB than Dilfer, it's not like the difference between the two was Earth-shattering. I also feel that Baltimore had the better offensive line. But let's take a step back and look at the bigger picture. The Ravens played a 4-3. Each of their defensive lineman was good enough to justify getting double-teamed every single play. Obviously offenses couldn't do that, which meant the Ravens would win the battle up front with a high degree of consistency. Behind those four very good DL there were three LBs, the worst of whom played at or near a Pro Bowl level. That LB corps was led by Ray Lewis. Then in the secondary you had a pair of shutdown corners to go along with safeties such as Ed Reed. There were no weaknesses on that defense, which was one of the three best in NFL history. But most years aren't like that. As I'd indicated earlier in this thread, of the past 14 Super Bowl champions, 11 have had franchise QBs, and another two have had QBs which played at a Pro Bowl level during the postseason in which their teams won the championship. Only in one year out of 14--the year you've mentioned--did a team win a Super Bowl despite receiving less-than-Pro Bowl-level QB play in the postseason. The reason for that is simple: while finding a franchise QB is no easy task, building a Ravens-of-2000-like defense is even tougher still. The Bills would need to upgrade ten of their starting defensive positions if they wanted to equal that defense. But not any old upgrade will do: you basically need a guy at or near the Pro Bowl level at literally every starting position on defense if you're going to be like the Ravens of 2000! Even the Ravens found that difficult to maintain, as demonstrated by that team's failure to get more than just one Super Bowl ring. In contrast, Joe Montana owns four Super Bowl rings, Tom Brady owns three, John Elway owns two, etc. While a team without a franchise QB can still win a Super Bowl--as the Ravens of 2000 proved--the odds are a lot more in your favor if you have a franchise QB! The other advantage is that a franchise QB stays with your team for many, many years, giving you plenty of chances to get enough other stuff right to come away with a Super Bowl ring. The Colts took Peyton Manning in 1998, and he's still going strong today. Look at all the opportunities Indy has had over the years to win Super Bowls because of that one guy! While Indy hasn't done a great job of taking advantage of those opportunities--only two Super Bowl appearances and only one ring--at least they're a contender year after year. The 49ers of the '80s were similar to that, except that they were able to achieve four Super Bowl wins because they surrounded Montana with more talent than Manning has been surrounded with. How many defensive players taken in 1998 are still in the league, and are still playing at a high level? How many defensive players elevate the overall prospects of their teams the way Montana elevated the 49ers, Warner elevated the Rams and Cardinals, or Manning has elevated the Colts? If you don't have a franchise QB and you have the chance to get one, you take advantage of that chance. Period. Anyone who would dispute that proposition is simply wrong.

Well said!

I'll address the above list of QBs. You are right to say Brad Johnson had a career year in 2002. He was selected to the Pro Bowl that year, and his play was an important part of the reason why the Bucs won the Super Bowl. Kurt Warner was one of the best QBs to ever play the game. Over the course of his career, Peyton Manning has averaged 7.6 yards per pass attempt. Kurt Warner's career average is 7.9. Kurt Warner owns the record for the most, second-most, and third-most passing yards in a Super Bowl. He owns the record for the most passing yards in a single playoff season, and is tied with Joe Montana for the most TD passes in a single postseason. He has the highest career completion percentage of any NFL QB ever. Of the games Kurt Warner has played in, he's passed for over 300 yards 41.9% of the time. That's a higher percentage than any other NFL QB ever. He's tied with Peyton Manning and Ben Roethlisberger for most games with a perfect passer rating. He tied Dan Marino for the fastest player to reach 30,000 passing yards; accomplishing it in 114 games. He's second to Peyton Manning in average passing yards per game. Kurt Warner was not just good--he was first ballot Hall of Fame good. There were a few years when he was playing hurt. During that time, injuries clearly affected his performance. That's why the Rams let him go, and one of the reasons why the Giants benched him in favor of Eli Manning. But he came back with the Cardinals; averaging about 7.6 yards per attempt during his five years there. He played well enough to lead them to a Super Bowl appearance, despite that team's problems on the OL and on defense. As for Eli Manning: it's true that his career average is a solid but unspectacular 6.8 yards per attempt. But over the last two years, he's averaged 7.9 and 7.4 yards per attempt. That shows he can put up the type of numbers you'd expect from a franchise QB. He also played well in the 2007 playoffs. He completed 74% of his passes against the Bucs, with 2 TDs and no INTs. He completed 67% of his passes against the Cowboys, with another 2 TDs to go with no INTs. And against the Patriots in the Super Bowl, he played well enough to be elected the MVP. So let's look at that list again: 10 New Orleans - Franchise QB 09 Pittsburgh - Franchise QB 08 NY Giants - Franchise QB-level play 07 Indy - Franchise QB 06 Pittsburgh - Franchise QB 05 Patriots - Franchise QB 04 Patriots - Franchise QB 03 Tampa Bay - Pro Bowl-level QB play 02 Patriots - Franchise QB 01 Baltimore - Defense (Dilfer was mediocre) 00 Rams - Franchise QB 99 Denver - Franchise QB 98 Denver - Franchise QB 97 Green Bay - Franchise QB Totals: 11 of those Super Bowls were won by teams with franchise QBs. Two others were won by teams that received Pro Bowl level play or better from their QBs for that particular postseason, even though the QBs in question (Brad Johnson and Eli Manning) had not done enough over their careers to be considered franchise QBs. Only one out of those teams (out of 14 total) received less than Pro Bowl-level play from the QB position during the postseason in which they won the Super Bowl. That one team was the Ravens of 2000--a team that had one of the three best defenses in NFL history. Not to mention Jamal Lewis at RB, and an OL led by Hall of Fame-level LT in Ogden. The deeper into the postseason you go, the fewer weaknesses your opponents will tend to have. If you build a team that's defense only, or defense + run the ball, then sooner or later you'll run into teams like the current Steelers or Packers. Those teams also have very good defenses, so it will be hard for you to build a defense that's that much better than theirs. Plus, their defenses will probably shut down your running game, so there goes that source of offense. And, those teams will bring franchise QBs to the table. If you're not getting very good play from your own quarterback, odds are heavily against you being able to do enough other things well to offset the significant advantage a franchise QB gives those teams. Nearly every team on the above list followed the this model: a franchise QB + good offensive supporting cast + good defense. This year's Super Bowl winner will also adhere to that model.

I agree that if you asked someone in the Jets organization about Sanchez that they'd tell you good things about him. Whether those things represent an honest analysis, propaganda, or wishful thinking is not immediately clear. This past season Sanchez averaged 6.5 yards per pass attempt. Trent Edwards' career average is also 6.5 yards per attempt. The difference is that Sanchez has a great OL, a great running game, solid talent at the TE and WR positions, and an offensive coordinator who's apparently able to outthink Bill Belichick. For most of his starts with the Bills, Trent Edwards had none of those things except maybe the solid WR corps. If you're going to put the "franchise" label on Sanchez, he needs to put up franchise numbers. Which he hasn't, at least not in the regular season. I'll grant he's done well in the postseason--especially against the Patriots. In contrast, Jay Cutler's career average is 7.2 yards per attempt, and this past season he averaged 7.6 yards per attempt. This, despite the fact his offensive supporting cast/situation was worse than Sanchez's. To put that in perspective, Peyton Manning's career average is 7.6 yards per attempt. Earlier in this thread, someone mentioned Cutler as a "statistically average" quarterback. That couldn't possibly be more untrue: even Cutler's career average is closer to Manning's than to Edwards', and this season he's averaging exactly as many yards per attempt as Manning has over his career. "Statistically average" indeed!

I'll respond to the bolded text. We absolutely need a franchise QB right now, because we don't have one. That's far from being this team's only need. It's also worth noting that there are much weaker starters at other positions--such as RT--than at QB. But a franchise QB is the single most important upgrade you can make to your team and the single hardest to find. In another thread I looked at eleven years of data which shows that a new franchise QB enters the NFL on average less than once a year. An average NFL team acquires a new franchise QB on average once every 44 years. One could argue that my definition of franchise QB is more stringent than most people's. But even if a broader definition would lead one to conclude it was only once every 30 years or so, the point still remains that you absolutely have to take a franchise QB if you don't have one and there's a chance to get one. It does not matter what your other needs are, or even if the guy you have under center is or isn't respectable. Unless that guy is franchise--which Fitz clearly isn't--you absolutely have to go with the franchise QB in the draft. But like I said, franchise QBs appear less than once a year. There may not be any franchise QBs in the upcoming draft. If there are, Ponder has as good a chance as any other QB to be that franchise guy, and odds are he'll be available in the second round. There's no sense in wasting the 3rd overall pick on a QB who a) is likely going to be a bust, and b) whose presence on the roster will prevent the Bills from using some future early draft pick on a real QB should there be one available later. As for the poster who pointed out that nearly all the Super Bowl winners over the last decade or more have had franchise QBs: his point wasn't that a franchise QB, alone, will win you the Super Bowl. His point was that if you don't have a franchise QB, you will not win a Super Bowl unless you're the Ravens of 2000.

Over the course of his career, Bill Polian has had four picks in the top-5. He used those picks on Bruce Smith, Kerry Collins, Peyton Manning, and Edgerrin James. Bill Polian has also had successes later on in the first round. But none of those successes compare to Manning or Smith, and very few compare to James. There were also a fair number of busts mixed in with Polian's picks later in the first round. An early draft pick is more valuable than one later in the draft, period. Giving a good GM like Polian the choice of any player he wants is much better than putting him in a situation where the players he wanted the most have already been taken. There will be a number of talented front-7 players available when the Bills pick. But the bust rate for DTs taken in the first 16 picks of the draft is over 50% (as someone pointed out earlier). Nearly all those busts had glowing pre-draft endorsements from a host of mock draft sites, much like this year's current crop of DL have. But those players failed to come close to expectations. More generally, the league is littered with players at all positions who had all the talent in the world, but who turned into busts anyway due to a lack of work ethic, football character, and passion for the game. Think of how much Mike Williams could have achieved had he worked as hard as Kyle Williams! The Bills' front office gets paid to identify (and avoid drafting) future Mike Williams-type players. Suppose that the two defensive front-7 players with the best work ethic and passion for the game get snapped up first and second overall. That would leave the Bills in a situation where they could still take a front-7 player with plenty of athletic talent, but who's more likely to become the next Mike Williams. Should that happen, they might want to consider going in a different direction in the draft instead. On the other hand, there might well be more than two early DL picks from the 2011 draft who go on to have very good careers. As for Mallett and Newton: I'm not overly interested in drafting either player at any point in the first round. Had Luck been an option, I would have been very happy to see the Bills trade up to get him. Even if they had to pay a king's ransom to do so. Having a franchise QB is just that important. Since Luck isn't available, the Bills might want to consider using a second round pick on Ponder. While Ponder may not have the arm strength of Mallett or the physical tools of Newton, he's done far more than either of those two have to show he's an accurate QB, a guy who can quickly read defenses, and generally be a pocket passer at the NFL level. The higher your team's draft pick, the more able your GM will be to do what he wants. Giving a Matt Millen the ability to do whatever he wants won't help your team much, because he'll just waste the pick anyway. But giving a Bill Polian the ability to do exactly what he wants is a very powerful thing! If Nix is more like Polian than he is like Millen, we should want to see him have as many draft-day options as possible. That means early picks. If, however, Nix is more like Millen, we should want to see nothing but miserable seasons for however long it takes to get Nix fired. Either way, a losing record in 2010 made sense. But obviously if Nix is good, you want to see that losing record get turned around before someone gets the ax.

A good article from Walter Football about this. There have been eight safeties taken in the top-15 since 1991. The author looks at how those safeties did, and draws a conclusion about whether it's worthwhile to use an early pick on a safety.

Great post! Just to add to what you've written: Tampa Bay was a defense-first team. But they still had Brad Johnson, who had a Pro Bowl year, and put up Pro Bowl numbers the year the Bucs won the Super Bowl. Like you said, if you don't have a franchise QB--or at least a guy who's close--odds are heavily against your team winning the Super Bowl. But football is a team sport, so once you have that franchise QB you need to surround him with talent or else you're not going to win anything. More simply, having a very good QB is close to being a necessary condition of winning a Super Bowl, but is not a sufficient condition.

My favorite QB stat is yards per attempt. Fitz's career average for that stat is 6.0, but this past season he averaged 6.8. Cutler's career average is 7.2, and this past season he averaged 7.6. I haven't seen a lot of Cutler's play recently, but at least statistically he's doing better than Fitz. As for Sanchez: this past season he averaged 6.5 yards per attempt, so he's playing less well than Fitz. (Especially after you factor in the Jets' stronger supporting cast.) In reference to your response to my post, I agree that AlphaDawg's point was very obvious. Upgrading the QB's level of play is more impactful than an equal upgrade at a DL position. But I nevertheless saw that point being disputed! With Luck out of the draft, I don't see any of the available QB prospects as worthy of the 3rd overall pick. I have no objection to the Bills taking a front-7 defensive player with that pick, as long as they're reasonably comfortable that he'll live up to his draft position. I fully agree the Bills' defense needs a major talent infusion. I'm intrigued by Bowers, who I think has the size and strength to be a DE even in a 3-4. As for the Bills' W/L record, the first two games the whole team seemed to collapse. Then the offense started putting up points because of the OL playing better, going up against defenses not as good as the Packers' or the Dolphins', and because of the QB switch. But the Bills' own defense was allowing insane numbers of points per game, so the vast improvement Fitz created in the Bills' point production didn't result in many wins. Then in the second half of the season the defense started playing better and keeping opponents' scores more reasonable. But the offense's point production began dwindling down--in part because of Fitz--so there weren't too many wins at the end of the season either, except against weaker teams. All of which goes to show that football is a team sport, and that you're not going to win a lot of games unless you're strong on both sides of the ball. Since the Bills aren't particularly strong on either side of the ball, they could justify going in a number of different ways in the draft. But an elite DE is the second- or third-most valuable thing you can come away with in the draft (after an elite QB and maybe an elite LT), which is another reason for the Bills to strongly consider Bowers. There will be no QBs or LTs worthy of a top-5 pick in this draft.

Alphadawg has said that a lot of times, and he's right! You mentioned Matt Stafford. In his rookie year, he averaged 6.0 yards per attempt. That declined to 5.6 yards per attempt his second year. To put those numbers in perspective, Trent Edwards' career average is 6.5 yards per attempt. Maybe the Lions will start receiving good QB play from Stafford someday. They aren't getting good (or even decent) performance from their QB now. That lack of good QB play is one of the most important reasons why they were in a position to draft Suh; and why they will have a very early pick in this draft. Alphadawg's point was that if you dramatically improve the quality of your QB play, you'll make more of a difference to your win/loss record than if you dramatically improve the quality of play you're getting from one of your DL positions. That point is absolutely correct.

First, thanks to everyone who complimented my earlier, longer post! As for what you've written above, I probably should have created a third category of quarterbacks. Guys who are more than just solid starters, but less than true franchise QBs. Many of the people you mentioned would fall into that borderline category. Now I'll address the guys you mentioned specifically. McNabb has a big name and was taken very early in the draft. But his career average of 6.9 yards per attempt didn't impress me as being franchise QB-worthy. It's not a bad average by any means. Just not as good as what I'd expect from a franchise QB. Likewise, Michael Vick has put up franchise-QB caliber numbers this year, and did a solid job last year as well. But his career averages are significantly less impressive due to his time in Atlanta. Eli Manning is another guy in that category: these last two years he's averaged 7.9 and 7.4 yards per attempt, which puts him in the same statistical category as his brother's career average. But Eli Manning's career average is a far less impressive 6.9 yards per attempt. You could point to Eli and Vick as two guys who have evolved into franchise QBs over the last year or two, but who have played well below that level for most of their prior careers. Jay Cutler's career average is 7.2 yards per attempt, and I felt a bit guilty about lumping him in with the solid starters. This past season he averaged a gaudy 7.6 yards per attempt. He clearly belongs at very least in the borderline category I mentioned above. If he maintains his current 7.6 average over the next few years, he'll have become a franchise QB. Matt Ryan had a very, very impressive 7.9 yards per attempt his rookie year, which is why people started throwing his name around as a franchise QB. But in years two and three of his three years in the NFL, he averaged 6.5 and 6.5 yards per attempt, respectively. That's not a franchise QB average. In fact it's exactly the same as Trent Edwards' career average for yards per attempt! While I'm sure Matt Ryan is a better QB than Trent Edwards, there's no way I'm putting a guy with numbers like those in the franchise QB category! I thought about putting Flacco in the franchise QB category: his career average is 7.2. But he didn't look like a franchise QB in his playoff game against the Steelers. He probably belongs in the borderline category. Mark Sanchez averaged 6.7 yards per attempt his rookie year, and 6.5 yards per attempt his second year. That 6.5 in particular is a Trent Edwards number, not a franchise QB number. As long as we're comparing Sanchez's stats to Edwards', it's important to bear in mind that Sanchez was given a much better OL, a better running game, better TEs, a better offensive coaching staff, and an overall better situation than the one Edwards had. For Sanchez to fail to put up better sophomore year numbers than Edwards strongly suggests very unimpressive things indeed about the quality of Sanchez's play. I'll grant that Sanchez stepped up his play in the playoffs, and was an important part of the Jets' recent win against the Patriots. But a franchise QB should be able to play that way all the time, not just in the postseason. As for identifying which players will turn out to be franchise QBs in the NFL draft, teams should (and probably do) go about it in the following way. You make a list of the things you want your franchise quarterback to be able to do. Throwing the ball accurately, reading defenses well, hitting receivers in perfect stride, etc. Then you rate the quarterback from -10 to 10 on each trait, with -10 meaning that he's proven he can't do that thing at all, and 10 meaning he's conclusively proven he can do it at a very high level. Zero means that you have no idea one way or the other if he can or can't do it. The more high numbers you see on that list, the safer it is to draft that guy.

I didn't think of Matt Schaub as a franchise QB myself until I looked at his numbers. His career average is 7.8 yards per pass attempt, compared to 7.6 for Peyton Manning. His passer rating is 91.5, compared to 94.9 for Manning. Schaub has a TD/INT ratio of 1.6, compared to 2.0 for Manning. That last number is solidly in Manning's favor, but otherwise the two QBs are statistically comparable. I'm not saying Schaub is as good as Manning. But any time you see a quarterback put up numbers as good as Schaub's, you feel like you're doing him a disservice by lumping him in with the "solid starters" QBs--guys whose numbers are significantly worse than those--rather than with the franchise QBs, where Schaub's stats fit right in. If someone who watches a lot of Texans games wants to make the argument that his stats overstate the quality of his passing, I'm certainly willing to listen. As for passing up a franchise QB--I did not mean to suggest that there was such a thing as a zero risk player. But there are players generally considered low risk guys likely to have good careers. Bruce Smith and Peyton Manning come to mind for their respective positions. A quarterback in that category is particularly valuable because franchise QBs are so rare. But the fact that franchise QBs are so rare also means that there might not be a franchise guy available for the Bills to take in the 2011 draft.

My goal was to find out the pace at which franchise QBs enter the league. To achieve that, I looked at previous drafts, starting in 1998. Let me know if I missed anyone. 1998 Franchise QBs added: 2. Peyton Manning, first overall, 7.6 yards per attempt. Kurt Warner, UDFA, 7.9 yards per attempt. Solid starters added: 1. Matt Hasselbeck, 6th round (6.9 yards per attempt). 1999 Franchise QBs added: none. Solid starters added: 1. Donovan McNabb, 2nd overall. (6.9 yards per attempt.) 2000 Franchise QBs added: 1. Tom Brady, sixth round. (7.4 yards per attempt.) Solid starters added: 2. Chad Pennington, first round. (7.2 yards per attempt.) Marc Bulger, sixth round (also 7.2 yards per attempt). 2001 Franchise QBs added: 1. Drew Brees, second round. (7.3 yards per attempt.) Solid starters added: 1. Michael Vick, first overall, 6.9 yards per attempt. (Though Vick had 8.1 yards per attempt this year, and looked like a franchise QB.) 2002 Franchise QBs added: none Solid starters added: none 2003 Franchise QBs added: none Solid starters added: 1. Carson Palmer, 1st overall, 7.1 yards per attempt. (Close to being a franchise QB.) 2004 Franchise QBs added: 3. Philip Rivers, 4th overall, 8.0 yards per attempt. Ben Roethlisberger, 11th overall, 8.0 yards per attempt. Matt Schaub, 3rd round, 7.8 yards per attempt. Solid starters added: 1. Eli Manning, 1st overall, 6.8 yards per attempt. (Though Manning has averaged 7.9 and 7.4 yards per attempt the last two years.) 2005 Franchise QBs added: 1. Aaron Rodgers, 1st round, 7.9 yards per attempt. Solid starters added: 4. Jason Campbell, 1st round, 6.7 yards per attempt. Kyle Orton, 4th round, 6.5 yards per attempt. Matt Cassell, 7th round, 6.7 yards per attempt. Ryan Fitzpatrick, 7th round, 6.0 yards per attempt. (This past season he averaged 6.8 yards per attempt.) 2006 Franchise QBs added: none. Solid starters added: 1. Jay Cutler, 11th overall, 7.2 yards per attempt. (This past season he averaged 7.6 yards per attempt.) 2007 Franchise QBs added: none. Solid starters added: none. 2008 Franchise QBs added: none Solid starters added: 2. Matt Ryan, 3rd overall, 6.9 yards per attempt. Joe Flacco, 1st round, 7.2 yards per attempt. That's eight franchise QBs added to the league over an eleven year period, or less than one a year. Assuming the above is typical, it means that each NFL team will, on average, add a franchise QB once every 44 years. Because such opportunities are so rare, and because franchise QBs are so instrumental in determining the outcomes of games, a team should never pass up the chance to get a franchise QB if it doesn't already have one.

Just to add to what you've written, some quarterbacks play better in the playoffs than in the regular season. If you look at Terry Bradshaw's stats for the regular season, the idea of inducting him into the Hall of Fame would seem laughable. But his postseason stats were dramatically better. During the '70s, the Steelers' defense--not Terry Bradshaw--got that team into the playoffs in the first place. But once they were in, Bradshaw would initiate his usual postseason improvement. His very good play in the postseason was one of several reasons why the Steelers won those Super Bowls. While Sanchez can't really be compared to Bradshaw in most respects, they seem similar in one way. Both QBs are associated with much better postseason stats than regular season stats. Sanchez had a good game against the Patriots. Had he turned in a (for him) standard-issue regular season performance, his team would have lost. You could say the same about a lot of the postseason games in which Bradshaw led his team to victory.

The text I've bolded is misleading. Whoever wrote it begins looking at quarterbacks drafted in 1985, and looks at the 20 year period from 1985 - 2005 in evaluating his data. The problem there is that for many of those years, there were three Hall of Fame first round quarterbacks drafted in 1983 who were busily skewing the data. If (for example) some QB drafted in the first round in the mid to late '80s failed to lead the league in passer rating because Dan Marino led the league instead, that hardly illustrates the futility of picking a QB in the first round. If the author wanted to examine the period of 1985 - 2005, fine. Except that when a first round QB taken in any year led the league in passer rating or led his team to a championship during that span, it should have been marked down as a success for a first round QB. But even though his methodology is deeply flawed, there is still some truth to a lot of what he'd written. There were a lot of first round QB busts during that period. One contributing factor was that offensive coordinators or front office personnel got it in their heads that they wanted mobile, athletic QBs. They wanted QBs who would create a dual threat, and extend plays by fleeing from pass pressure. As a result, a lot of quarterbacks were chosen in the first round based on their athletic gifts rather than their abilities as pocket passers. Those quarterbacks had an exceptionally high bust rate. That particular fad seems to have died down a little. But not before it produced more than a few first round busts! It's also worth nothing that the period the author examined contained three non-first round quarterbacks who collectively had a significant impact on his data about championships. Joe Montana was drafted in the third round and led the 49ers to two championships during the period he examined. Kurt Warner was an undrafted free agent and led his team to a championship. Tom Brady was a sixth round pick and led his team to multiple championships during the period in question. But stories like that are rare. Tom Brady was drafted back in 2000. Since then, no quarterback acquired in the third round or lower, or as an UDFA, has come remotely close to playing at the Montana/Warner/Brady level. None. Point to any franchise quarterback who joined the NFL in 2001 or later, and I'll show you a guy who was drafted in the first or second round. The only second rounder in the mix will be Drew Brees, taken with the 32nd overall pick.