Get to Know 'Em: Sage Rosenfels

It seems only fitting that as I spent my time yesterday writing about a long shot at quarterback, Pat White, the Vikings were busy getting to the brink of a deal with Houston for Sage Rosenfels. So while I'm happy to hear that the competitor for Tarvaris Jackson isn't going to be someone that Tarvaris will likely beat (I'm looking at you Gus), it just figures that the news of a trade would break right after I hit "Post" and used up what free time I had on Pat White.

Anyway (it's rough being me, isn't it?), as you probably recall, the Vikings' chased Rosenfels before this season. As I wrote then, Rosenfels is a very accurate passer (he completed 66.7% and 64.2% of his passes the last two years) who is one of the better quarterbacks at avoiding sacks:

"Another thing Rosenfels has going for himself is that he rarely gets sacked. He was only dropped once every 41 times he went back to pass, an astoundingly low rate. Rosenfels has only been sacked 10 times in his entire career and was sacked only once every 39.8 pass attempts. To put that in context, Tarvaris had about as many pass attempts (313), but was sacked 19 times last year, or once every 16.5 dropbacks, which isn't that bad of a sack rate (David Carr's sack rate is once every 9.4 drop backs)."

Rosenfels wasn't as succesful at avoiding sacks last year, going down 9 times in 183 drop backs, a rate of once every 20.3 pass plays. His sack rate wasn't helped by an offensive line that dropped from 10th to 16th in Adjusted Sack Rate last year. Of course, the Vikings were 28th in ASR last year, so he will be facing more pressure than in Houston if the Purple don't upgrade their line.

Another reason that Rosenfels may have been sacked more is that he was more willing to hang in the pocket for a deeper route to come open. His yards per attempt jumped from 6.67 YPA to 7.50YPA. He also completed 4 passes for more than 40 yards, the same as 2007 and 18 passes for more than 20 yards, which was one more than in 2007, despite running 63 fewer pass plays. He can get the ball downfield accurately and is willing to hang in the pocket to do so.

Of course, there's a reason why Rosenfels is available to the Vikings for a fourth round pick--he throws a lot of interceptions. He's thrown 22 in the past two years, which, when you consider he only had 414 pass attempt, is horrible. That's an interception rate of 5.3%, which would have lead the league last year (Frerotte was the actual leader at 5.0%). That interception rate is a big reason why he had fewer DYAR and a lower DVOA last year than Tarvaris Jackson.

His propensity to throw interceptions is the main reason Rosenfels likely isn't the best answer to the Vikings' quaterback situation. But, as I've said before, the time to find the best answer is long past. Cassel is going to be expensive and has big question marks about his ability to play under center without Randy Moss and Wes Welker, and while Garcia is probably a better answer, Childress doesn't seem to like him and he gets hurt too frequently. Rosenfels is an improvement on the "Gus and Tarvaris Traveling Sucking Spectacular" that we saw for most of last year, and that's probably enough. He's a legitimate contender for the starting job (unlike Frerotte), which means that if Tarvaris beats Rosenfels out, he'll have earned it with strong play. And at this point, that's enough for me to be happy about the Purple's quarterback situation.

Get to Know 'Em: Pat White

For awhile now, I've been expecting (maybe even hoping) the Vikings would take quarterback Nate Davis out of Ball State in the second round if they decided to go the "sign a veteran and draft a future starter" route to improve at quarterback. That assumed that Davis would step up and claim a spot as a sure second round pick, something it seems he's not doing. But enough negativity (of which there is more than enough due to this threat from Brad Childress). Let's talk about a player that might actually be a long term answer to the Vikings' quarterback problems: Pat White.

Now, as I mentioned when I looked at the various quarterback prospects, the two most predictive statistics for college quarterbacks talented enough to go in the first or second round, are games started and completion percentage. And those are two stats in which Pat White did very well. He started 42 games in his college career, and completed 64.8% of his passes. And it seems that he just might have the arm to actually make the transition from Spread QB to NFL QB, which might raise his stock high enough to make the Lewin Forecaster applicable (it's only for QBs taken in the first and second rounds).

Of course, there are a couple of reasons why most teams don't see White as an NFL QB. He's only 6'1 and is under 200 lbs, likely making him the smallest quarterback in the league. The man's very breakable, and he had to sit out (or leave) games with injuries his sophomore, junior and senior seasons. And he was a run first quarterback in a spread offense--he threw less than 800 passes in his four years. If you're worried about Matt Cassel's ability to run the Vikings' offense from under center, you should definitely be worried about Pat White's ability to run an offense that isn't 80-90% shotgun formation (a problem almost every quarterback in the draft has, including Josh Freeman and Nate Davis thanks to the widespread use of the Spread).

So is Pat White the best answer to the Vikings' quarterback issues? Probably not, but I don't know if there is a "best answer" at this point. He obviously has some drawbacks, but if the scouts think he has the arm strength and the accuracy to play quarterback, he was more than productive enough in college to make him an interesting long term option, especially if the Purple put him on the same diet as Pat Williams.

Get to Know 'Em: Matt Cassel

The more I look into Matt Cassel, the more unsure I become about whether or not I'd like the Vikings to trade for him. Unlike last year, when an in depth look at Derek Anderson showed his inconsistency and the unlikeliness of his continued success, Cassel's stats send a lot more mixed messages. And since I've been spending the past two days writing memos, I'm going to break this post up into reasons Cassel is the answer for the Purple at quarterback and reasons the Vikings should stay away.

Why the Vikings should trade for Cassel:
He's young (26), he's accurate enough for the West Coast Offense (he completed 63.4% of his passes) and he's been well coached. He had the 10th highest QB rating in the NFL last year, the 11th highest completion percentage and he doesn't really turn the ball over, only throwing 11 interceptions. He also had some monster games (and when I say monster games, I mean actual monster games-not the 11-17 for 163 yards and 4 touchdowns type monster games we've been used to the past few years), including two games with over 400 yards passing and another with 345 passing yards. He had seven games with a QB rating over 100 and five games with 3 or more touchdowns. While it's true that he inherited a passing game that included Randy Moss and Wes Welker, he was also behind an offensive line that finished 26th overall in adjusted sack rate and gave up the fifth most sacks in the league. He showed he can successfully lead an offense that scored 42 touchdowns when given the right pieces. And it's not like the Vikings are completely devoid of talent in their receiving corps--Bernard Berrian and Bobby Wade aren't Moss and Welker (there's an understatement), but they aren't Troy Williamson either. And Sidney Rice is entering his third year, which, if he can stay healthy, is the year that many wide receivers end up breaking out. And, of course, Sammy Morris is no Adrian Peterson. (And let's all just gloss over the whole coaching comparison so we don't get too depressed).

Why the Vikings Should Stay Away:
There's a lot of evidence that Cassel isn't going to be an elite quarterback, which means that the Vikings shouldn't pay him or the Patriots like he's an elite quarterback. Here's his projection from this year's Pro Football Prospectus, which is based on quarterbacks that are similar to him (which, prior to this year did not include a lot of good quarterbacks) would do if they got to play 16 games in New England's offense against New England's schedule:
299-478, 62.5%, 3751 YDs, 27 TDs, 23 INTs, 7.5 YPA 17.0% DVOA

And here's the actual statistics he put up:
327-516, 63.4%, 3693 YDS, 21 TDs, 11 INTs, 6.2 YPA 6.4% DVOA (20th)

Pretty similar if you ask me. He threw a few more passes, completed a slightly higher percentage of them and threw a lot fewer interceptions, but his DVOA is lower than his projection, as is his yards per attempt. In fact, he had a lower DVOA than Tarvaris Jackson (though, one has to remember that while DVOA is adjusted for competition, Tarvaris Jackson didn't exactly play against a murderers row of pass defenses, and, since it's a rate state, it's easier to put a higher one in fewer games). It definitely makes it seem like Cassel was a product of the talent around him, rather than an equal part of the offense. If that was true, however, it would seem to follow that he'd do well against the bad pass defenses and struggle against the good ones. So was this a case of Cassel putting up big stats against bad defenses and struggling against good ones?

This following chart has Cassel's QB Rating per week compared with his opponent's QB Rating against.

It shows a young quarterback that had some struggles, but also had some amazing games. His best game came against Denver's awful pass defense at home, while his worst game was quite clearly, his 12th, which came against the Steelers. Not all that surprising, considering it was on the road, in the cold, against a top pass defense and the Patriots were missing Wes Welker for most of the game. He also bounced back afterward and finished strong, with a QB rating over 100 each of the last three weeks. He did only face five teams that had above average pass defenses by QB Rating against (and the Steelers were the only top five pass defense he faced). He did have some success against them, however, posting a QB Rating of 114.0 against Miami and a QB rating of 108.1 against Oakland.

Recommendation:
Even after going this in depth, I'm not really sure whether Cassel is the long term answer for the Vikings. The problem is that I'm not sure if there's any other realistic option for the Vikings' this year. I wish I trusted Childress to evaluate quarterbacking talent (of course, if I did, I doubt I'd be writing this post about Cassel because the Vikings wouldn't have an issue at quarterback), because then I'd say that it came down to what the scouts said--were Cassel's big games due to Moss and Welker and Belichick, or were they a result of finally getting a chance to show off his talent (along with having Moss, Welker and Belichick's offense). The stats seem to support the latter, but there's enough there to think that the struggles were an inexperience thing, rather than a lack of talent, and I'm not sure if anyone can tell the difference without breaking down Cassel's tape. There isn't a clear drop off in performance as the season went on like there was with Anderson. So if the Vikings' think he's the answer, I'm not sure I disagree.

I guess that means it comes down to what the Vikings would have to give up to get him. As I've said before (and will expand on later), the window of opportunity for this team to make a run at the Super Bowl is a lot smaller than one might think. Giving up a first and third round pick for the second straight year is worth it to acquire a franchise quarterback, just like it was worth it to acquire a franchise defensive end. I'm not so sure that's Cassel, however. If they give up less than that, such as only a first rounder, or even less than that, then I think they should jump on it. Yes, they'll need to pay Cassel a big chunk of change, but I trust that the cap wizards at Winter Park can make it work (and they currently have $28 million in cap room to work with). At the very least, the Purple need to inquire about trading for Cassel and find out whether they have a similar definition of Cassel's value as New England does.

Get to Know 'Em: Quarterback Prospects

[This post has been edited to reflect reality. I had Matt Stafford's, Mark Sanchez', Josh Freeman's and Nate Davis' numbers completely wrong, completely changing my conclusion. Sloppy research on my part, but in my defense, its not easy to find statistics on college players for some reason. I blame ESPN (who had some wrong numbers and don't give career stats for college players). Thanks to the Anonymous commenter who pointed it out.]

It’s quite obvious that the Vikings need to do something about the quarterback position this off season. Another year with Tarvaris (or a mediocre to bad veteran that was signed to back him up) under center isn’t going to cut it. One of the ways they might address the position is by drafting a QB (which, would, I think, probably be a lot more rational than trying to acquire Matt Cassell or wagering it all next year on an over the hill veteran like Jeff Garcia or *shudder* Brett Favre). Of course, if they’re going to draft a quarterback the need to find one that can contribute by 2010 at the latest, which likely means drafting one in the first or second round. Of course, if they’re going to draft a quarterback, they need to draft a good one. So how do you know which college quarterback is going to be good? Well, luckily for us, Dave Lewin at Football Outsiders has developed a system that’s pretty accurate at determining which quarterbacks are worth taking. In a nutshell:

Lewin found that there are only two college QB statistics that reliably, and independent of all others, predict NFL QB performance: completion percentage and games started. Completion percentage is obvious: QBs kind of have to be able to hit their target. Generally, you’re looking for a 60% completion percentage and around 35 games started in college.

Lewin’s system only applies to QBs drafted in the first two rounds because, as Lewin showed, pro scouts are good enough to separate top talent from lesser talent. Which is the more games a player started, the better (not to mention the more talented quarterbacks tend to play earlier in their career in college). The guys who go after the first 2 rounds are generally there for good reason (Tom Brady excluded). When a team ends up with a bust, it’s usually because they either reached (*cough* Tarvaris Jackson *cough*) or because the quarterback didn’t play enough for their flaws to become apparent. Or the team ignored the flaws because of other things they thought mattered more, like arm strength, which are usually vastly overrated.

So how does this year’s crop of quarterbacks look? Well, from looking at the various draft sites, I’ve come up with a list of 7 quarterbacks that might go in the first or second round. Not all of them will, but I’m trying to be comprehensive and I’ve included where most sites have them going.

Quarterback	College	Predicted Round	Games Started	Completion Percentage
Matt Stafford	Georgia	1st	32	57.1%
Mark Sanchez	USC	1st	16	64.2
Josh Freeman	Kansas St.	Late 1st	31	59.1%
Nate Davis	Ball State	2nd	37	60.3%
Pat White	West Virginia	2nd/3rd	42	64.8%
Graham Harrell	Texas Tech	2nd/3rd	41	69.8%
Rhett Bomar	Sam Houston St.	2nd/3rd	26	56.5%

So what does this tell us about this year’s quarterback class? Well, there isn't a quarterback that meet the two standards and will go in the 1^st or 2^nd round. Matt Stafford isn't close to the mark completion wise, but he hasn’t started as many games (and while 32 games is a solid amount, he platooned his first year playing). Mark Sanchez didn't start anywhere near enough games (though with USC quarterbacks, that isn't necessarily about a lack of talent) and Josh Freeman didn’t hit either mark (though his games started and completion percentage were close), which makes sense, considering he’s seen as a very raw quarterback. Nate Davis is like Matt Stafford--he started 32 games, but platooned a few more his freshman year. He was more accurate in college than Stafford was though. Pat White and Graham Harrell both hit each mark, but quarterbacks like them are why this projection system doesn’t work past the first two rounds—neither of them is seen as having the physical tools necessary to be successful in the NFL. And Rhett Bomar is an interesting case because he probably would have played 35 games for Oklahoma if he hadn’t been kicked off the team for taking money from boosters. He doesn’t seem to have the accuracy to be successful however, as he only completed 56.5% of his passes in college. Don’t be surprised to see him talked up as a big sleeper based on his arm strength and mobility, but it’s never, ever a good sign to see a quarterback completed so few of his passes in 1-AA.

So where does that leave the Vikings? Well, none of the quarterbacks made the benchmarks that predict success in the NFL. So, that makes this year's draft particularly unappetizing when it comes to quarterbacks. Which is great, because the Vikings probably need to draft one this year. Matt Stafford and Mark Sanchez aren’t likely to fall to the 22^nd pick. Josh Freeman might be there, but he’s a project and didn’t hit either of the marks that predict success in the NFL. Nate Davis is intriguing, however, and he might be available to the Purple in the 2^nd Round, allowing them to address another one of their issues in the first, but he's a little small for a quarterback and would have to adjust from the MAC to the NFL (then again, the MAC seems to turn out very good NFL quarterbacks). I’m going to get into this a lot more (look for a breakdown of the Vikings’ needs next) and hopefully take a more in depth look at the prospects the Vikings might take in the first round, but as of right now, it doesn’t look like they should be drafting a quarterback with the 22^nd pick.