Looking back at FEI's preseason projections, we find that most teams did about what they were supposed to do -- but not in the Big Ten, where things got screwy.
17 Apr 2007
by David Lewin
This year, as usual, a number of teams at the top of the draft are looking quarterback. The apples of scouts' eyes this time around are Brady Quinn and JaMarcus Russell. Quinn is the All-American boy, a four-year starter at Notre Dame, with all the measurables that teams look for. Russell, an early entry after his junior year, is more of a dark horse. He never achieved the level of success Quinn did in college, but he soundly outplayed Quinn in their head-to-head matchup at the Sugar Bowl and has ridiculous physical talent.
Still, if we've learned anything about the NFL draft in recent years it's that there's no such thing as a can't miss prospect, especially at the quarterback position. After Ryan Leaf, Tim Couch, Joey Harrington, Akili Smith, David Carr, etc. teams are more wary than ever about taking a QB high.
Last year, in Pro Football Prospectus 2006, I introduced a system for predicting the NFL passing performance of first- and second-round quarterbacks by looking at their on college statistics. (I may try to project running ability in the future, but it is not currently included.) After gathering and analyzing extensive data, I determined that games started and completion percentage in college were strong predictors of NFL success.
In fact, for first- and second-round quarterbacks these two variables explain roughly 65 percent of the variation in NFL performance. This is a remarkably high number given how much of an impact we believe that teammates have on a quarterback's play. The system accurately predicted Philip Rivers' breakout season, and the success or lack thereof, of many other young passers. With an eye on this research I take a closer look at the pro prospects of Russell, Quinn, and the other quarterbacks of the Class of 2007.
Russell is possibly the most physically talented quarterback prospect since John Elway. He's 6-foot-6, 260 pounds with an arm that makes Brett Favre look like Danny Wuerffel. Russell is also a good runner and his arm strength allows him to hit any open receiver while he is on the move. He has huge hands, which help him avoid fumbles, and remarkably quick feet for such large man. Physically, Russell compares favorably to Daunte Culpepper, Byron Leftwich, and Ben Roethlisberger, all of whom have had some degree of success in the NFL. Assuming he can get his weight under control, Russell figures to be more mobile than Leftwich and Roethlisberger, but not as good of a runner as pre-injury Culpepper.
However, hidden behind Russell's immense physical talent are a couple of major red flags. For starters, as recently as last summer's training camp, Russell was engaged in a fierce quarterback battle with Matt Flynn and Ryan Perriloux. There were rumors that Russell was gone to the NFL at the end of the season no matter how the year went, simply, because he was unsure that he'd be able to hold onto the starting job again. Russell emerged from the controversy to have a great season, but it was the first of his career in which he was statistically the best passer on his team. While this history doesn't doom Russell (Tom Brady had similar issues), I would still be wary of spending the number one overall pick and the 25 million dollar signing bonus that comes with it on a guy who wasn't even the clear-cut starter his final year in college. Less than six months ago, very few experts considered Russell a first-round pick. FO's own Michael David Smith, for example, wrote "I don't see Russell as a first-round pick at all" in the discussion thread for Seventh Day Adventure on September 17th. Late risers usually fail to live up to the hype.
An even bigger issue with Russell is that he is an atrocious decision maker. He consistently throws the ball into double and triple coverage. Like Favre he often gets away with it by making spectacular throws. Still, field vision is the single most important quality for a quarterback. You can get away with being average in this regard if you have superior arm strength and accuracy, but Russell still has a ways to go before he can be considered average at seeing the field.
Russell also has the tendency to simply throw the ball up for grabs when he gets confused. If you watched the Sugar Bowl you saw this happen on an early in the fourth quarter interception. Russell clearly blew his underneath reads, panicked, and just lofted the ball up. It went pretty much unnoticed because he played an otherwise stellar game, but it is a recurring problem. These passes didn't always end up picks in college because Dwayne Bowe, Craig Davis and Early Doucet (all future first day picks) did a nice job of winning jump balls. That won't fly at the next level.
Another concern is that Russell ran a fairly simple offense at LSU. The Tigers eschewed complexity partly to avoid confusing Russell, and partly because complexity was not necessary for an offense as talented as LSU's. Russell can make very difficult throws when he knows where to go with the ball, so the LSU coaching staff didn't have to get fancy. This will still be true to a degree in the NFL. In fact Drew Bledsoe did pretty well in his first few years in the league in a Patriots offense using similar principles. Unfortunately coaches often seem reluctant to play to their quarterback's strengths (see Vick, Michael) and try to force unique players into the cookie-cutter mold. Russell will struggle with an NFL-style offense at first, but if he's coached well (like Vince Young was last season) then this shouldn't be a prohibitive factor in his development.
All of these issues are related. Because of Russell's knack for making boneheaded plays at the worst possible times, he had Matt Flynn and Ryan Perrilloux breathing down his neck his whole college career. Any team that drafts him must be prepared to tolerate some very questionable decisions interspersed with Russell's trademark spectacular throws. He is also known to get down on himself and get rattled when things are not going well. All in all he has a number of characteristics that you don't like to see in a young quarterback.
Russell's understanding of the game is questionable at best. He constantly infuriated LSU fans with his inability to manage situations properly. Pretty much everything that Russell is can be summed up by describing the final drive from this year's LSU-Auburn game.
LSU took over with 1:04 to go at their own 20-yard line trailing 7-3 with no timeouts remaining. On the first play from scrimmage Auburn sat back in a soft Cover-2 zone with the safeties deep. Russell didn't notice this until it was too late and threw the ball out of bounds over the head of Craig Davis who was open running a vertical route down the sideline. The LSU coaching staff called Russell over and pointed out that he could pick the zone apart by throwing the ball to Davis in the Cover-2 hole at the sideline 20 yards downfield.
LSU then went out and ran the exact same play two more times. Russell hit Davis for gains of 20 and 21 yards. Davis was immediately drilled out of bounds both times. The Auburn safeties came flying downhill because they knew what was coming, but they couldn't stop it. Russell simply got the ball there too fast.
All of a sudden LSU was driving. They had the ball on the Auburn 39-yard line with 50-odd seconds to go. Auburn, knowing they were in trouble, changed up their defense getting out of the soft Cover-2. Russell went back to pass, and seeing something different from what he was expecting, got confused. Luckily the protection was good and he was able to scan the field for four or five seconds. Still not seeing anything he liked, Russell began to roll to his right. He escaped the pressure and continued to move towards the sideline at a leisurely pace. Instead of throwing the ball away to stop the clock Russell decided to tuck it and duck upfield. He was tackled in-bounds at the line of scrimmage for no gain.
He got up without any sense of urgency and with a grin on his face. He seemed to be quite pleased with himself for avoiding a sack and maybe picking up a yard. After about a second he finally heard all the people screaming for him to spike the ball and began to hustle to the line. However by then it was too late. By the time LSU got set and spiked it there were only 25 seconds remaining. Russell had just wasted 20 crucial seconds. I have only rarely been angrier while watching a football game (not because I'm an LSU fan, I'm not, but because I like to see the game played intelligently).
Russell ended up bouncing back on the next play by hitting Dwayne Bowe for 20 yards putting LSU back in striking distance with under ten seconds remaining. However he followed that up with a stupid illegal formation penalty that pushed them back to the 24-yard line, and then hit Craig Davis for a 19-yard gain with no time left on the clock. Overall Russell's numbers for the game looked pretty good (20/35, 267 yards, no TDs or picks), but when it counted he made certain LSU would come up short. That pretty much says all there is to say about Russell. He will tantalize you with impossible throws, but there is significant evidence that mentally he doesn't have what it takes to be a great quarterback.
So, overall Russell has a pretty risky profile from a scouting perspective. He might be great, but he could be the next Jeff George (which, unless you're Jason Whitlock, is not a good thing). That could be enough to keep teams from taking him first, but probably not. Here's something that should: Russell started only 29 games at LSU. Over the past ten years, collegiate games started has been the single greatest predictor of NFL success for early first-round quarterbacks. Since 1997 seven quarterbacks who started fewer than 30 games in college have been drafted in the top ten: Ryan Leaf, Tim Couch, Akili Smith, Michael Vick, David Carr, Joey Harrington, Alex Smith. Who's the best player on that list? Michael Vick? Alex Smith? I don't even know. All I know is that list does not leave me saying, "I gotta get me some of that."
There is one positive indicator among Russell's college stats: The player to whom he is most similar is Vince Young. Young started 32 games, Russell 29. Both completed 61.8 percent of their passes. Both lost starts because they platooned with an upper classman during their freshman year. Both are tall, strong, and obviously both are African-American. But the similarity only goes so far. Young started three more games than Russell. That doesn't seem like a significant number, but it is. Young has also been only moderately successful as passer and will probably never be one of the best passers in the league. As long as he's above-average, it's not a problem, because he brings so much to the table with his running ability. Russell, on the other hand, will succeed or fail based on his ability to pass the ball. He's decently mobile, but he is nowhere near Young's caliber as a scrambler. Even if he lives up to the comparison and becomes a slightly worse passer than Vince Young, I doubt people will consider that a success.
All in all, I would be very wary investing $25 million guaranteed in a guy who barely won his starting job in college, doesn't see the field well, is known to be immature, and has an unfavorable statistical profile. At least the Raiders will be able to admire the velocity and distance of Russell's passes on their way to the arms of opposing defensive backs.
Quinn is pretty much your stereotypical stud quarterback prospect. A four-year starter out of Notre Dame, he is used to having the national spotlight focused on him. At 6-foot-4, 225 pounds he is physically reminiscent of another Charlie Weis pupil named Brady. In fact, he is definitely more mobile, and probably has a better arm than Tom. The big difference between the two is the skill for which the Patriots' Brady is best known: staying cool under pressure.
Quinn has been in the national spotlight for so long that people have started to feel the need to tear him down. If you look for faults long enough you will find them. This is a major part of the reason why games started in college is so predictive. Players with few starts have not given scouts time to find their faults.
Quinn's slide is reminiscent of another recent top quarterback prospect who chose to stick around for his senior year, Matt Leinart. People nitpicked Leinart to pieces his senior year which caused him to drop farther than he should have; teams that passed on him are starting to regret it.
The big knock on Quinn is that he buckles under pressure. Last season, against the top defenses that Notre Dame faced, those that really pressured the quarterback well, Quinn came up short. This was particularly evident in LSU's manhandling of Notre Dame in the Sugar Bowl. Now, I think it makes sense that a quarterback might struggle when his offensive line is obviously overmatched, but still Quinn's ineptitude was worrisome. This is an important area for improvement, but it is far from a crippling weakness. Peyton Manning faced similar criticisms coming out of Tennessee and he seems to have done fine.
Quinn has a lot of things going for him. In addition to his good size and mobility, Quinn has an excellent arm. He doesn't throw the deep ball as well as Russell does, but he has great zip on his intermediate throws. Quinn has no problems hitting the deep outs, and loves to throw crossing patterns over the middle. He has had plenty of experience making these NFL-type throws after spending the last two years under Weis' tutelage.
Most people consider Quinn's time under Weis one of his greatest selling points. However, a few are somewhat concerned by the huge jump in Quinn's numbers after Weis' arrival. Was Quinn's success just a product of the system? I would suspect not, especially given the continued success of Tom Brady. Weis works wonders with quarterbacks, no doubt about that, but it seems that whatever he does to them is permanent.
In my opinion, Quinn's college stats match up pretty well with his scouting profile. He completed 58.0 percent of his passes in college and started 46 games. This projects Quinn as a good pro quarterback, but not a great one. Here is the complete list of players drafted in first two rounds over the past ten years who started at least 35 games and completed at least 57 percent of their passes: Peyton Manning, Donovan McNabb, Daunte Culpepper, Chad Pennington, Drew Brees, Carson Palmer, Byron Leftwich, Eli Manning, Philip Rivers, Ben Roethlisberger, Jason Campbell, Matt Leinart and Jay Cutler.
Are there any bad players on that list? The worst according to my numbers is Eli Manning, and he's been decent (maybe not up to expectations, but decent). I would be happy to spend a top ten pick on most of those players.
Brady Quinn's closest comparables paint an even prettier picture. Carson Palmer, Donovan McNabb and Jay Cutler all started 45 games, and completed 59, 58 and 57 percent of their passes respectively. If Quinn's pro career is as similar to those players as his college stats are, then he should be good enough to justify a high pick.
Charlie Weis recently said that Quinn is "a combination of Tom Brady and Peyton Manning." I'm a bit skeptical of that; it will be tough for Quinn to find time to both impregnate B-list actresses and teach the local kids how to boost an Escalade. All joking aside, Quinn is far from flawless as a prospect, but, in spite of his occasional shortcomings on the national stage, all evidence points to Quinn becoming a very good NFL quarterback.
I've read in a couple of different places recently that the Raiders are thinking of passing on a quarterback in round one because they want to select Trent Edwards of Stanford in round two. I am unsure what to say about this decision. On the one hand, if they are really sure that Calvin Johnson is going to an outstanding player then they should take him. Drafting Russell so early would be a mistake, so avoiding that would have to be considered a positive.
On the other hand, Trent Edwards is not a good prospect either. He started 31 games at Stanford and completed 56.3 percent of his passes. He was never a particularly outstanding player, and he got hurt all four years. The projection system is not designed to handle injuries, and if he had stayed healthy Edwards would have played forty-some games, so perhaps he is better than the numbers say. Still, it's hard for me to believe that a guy who couldn't make it through a college season even once will be able to survive the NFL.
Teams looking for a QB in the second round should give Drew Stanton a long look. His play at Michigan State was uneven at best, but he has prototypical size, a strong arm and excellent mobility. Though he didn't start many games in college (29), he did complete 64.2 percent of his passes. It's tough to say too much about second-round QBs because there have not been many in recent years, but based on the limited data, completion percentage seems to gain greater importance in the second round. I wouldn't spend a first-round pick on Stanton, especially given his erratic play at Michigan State, but he has all the tools to be a quality NFL starter and could be a good value in the second round.
John Beck of BYU is another guy that may sneak into the end of the second round. He posted an excellent completion percentage at BYU and was a four-year starter. He lacks lacks elite talent, but he makes good decisions with the ball. He has the makings of a good but not great, prospect, a la Matt Schaub.
Last year at this time I wrote that, "I was almost hoping that [the system] would indicate that one of the three top prospects was going to be a bust, just so I could say that I called it. But the system projects all three to be good pros." Well, disappointed as I was with the lack of controversial predictions, I am glad that I stuck by the numbers. This year is different. The numbers are not quite black and white (Quinn's completion percentage is a little low, Russell has a number of starts a little above complete bust territory), but they are pretty clear: Brady Quinn is a much better quarterback prospect than JaMarcus Russell.
I am hardly pleased to call out Russell as a likely bust, and given the right situation and good coaching I am sure he could defy the odds and become a good pro. However, players like Russell rarely do.
Here's that under 30 starts list again: Ryan Leaf, Tim Couch, Akili Smith, Michael Vick, David Carr, Joey Harrington, Alex Smith.
Sorry, Raiders fans.
For more about the method behind the College Quarterback Projection System, and projections for young quarterbacks already in the league, see David's expanded article in Pro Football Prospectus 2007, available this summer.
214 comments, Last at 13 Nov 2012, 4:58pm by quality software