Welcome, Sage Rosenfels!

It's official -- Sage Rosenfels is a Minnesota Viking, and a reasonably well-paid one as well, after agreeing to a two-year, $9 million contract with the team following a trade with Houston that sent the Vikings' fourth-round pick to the Texans.

Reaction from Vikings fans seem to be generally positive, though I still caution against thinking that Rosenfels is some savior, and that he's a significant upgrade over Tarvaris Jackson. He's definitely a different style of quarterback than Jackson, but saying he's "better" overall simply ignores his deficiencies -- most notably his relative immobility and propensity for turnovers -- while lambasting Jackson who makes fewer mistakes but looks "worse" doing it.

That said, Rosenfels probably is a better overall QB than Jackson, but this isn't like replacing Ryan Leaf with Peyton Manning. Jackson's not the worst QB in the world, no matter how he looks, and Rosenfels isn't spectacular, despite his big arm. Still, with no other major free-agent acquisitions on the horizon for the Vikings, sending a middle-round pick and spending $4.5 million a year over two years on a guy who might be a good QB (and at the least is a better backup plan than Gus Frerotte) is a decent deal. Fact of the matter is, there wasn't an elite quarterback available, and Rosenfels likely represents one of the better options.

The best-case scenario for the Vikings is that Rosenfels turns into a latter-day Rich Gannon, another QB who got only occasional chances to start the first part of his career, showed equal parts promise and poorness, and, in his mid-30s, found the right team and right place to blossom as a star player in the latter years of his career. Rosenfels turns 31 next week, while Gannon was 33 when he finally secured a starting job with the Oakland Raiders in 1999, going on to have four straight excellent seasons with the silver-and-black. If we could get four seasons of stellar -- heck, I'll settle for "steady" -- play from Rosenfels, I'd consider it a success.

Right now, nobody knows exactly what we'll get from Rosenfels, and we won't know that until the games start up in September. Now that he can rely on his running game and defense to help him win games, he could develop into a dependable QB, and he might even have the talent to sniff the Pro Bowl. All this depends on whether he can reign in the bad plays, however, avoid the crazy mistakes -- oh, and beat out Jackson in training camp, which might be harder than it appears. This one ain't decided yet, folks.

Rosenfels vs. Jackson

With Sage Rosenfels set to join the Vikings as soon as the trade can be consummated -- 12:01 Eastern, Friday morning, to be exact -- many fans are already predicting that he'll easily oust Tarvaris Jackson as the Vikings' quarterback. While that may yet happen, it shouldn't be the slam-dunk, no-questions-asked decision some think it ought to be.

Somewhat conveniently, both quarterbacks have career pass attempt totals that are about in line with a full season's worth of passes, albeit for a rather pass-happy team: 561 for Rosenfels and 524 for Jackson. This allows us to compare their stats, side-by-side, with no real manipulation needed and to consider how they'd look if they really were single-season stat lines.

Rank	Cmp	Att	Pct.	Yards	TD	Int.	Sack	Yds.	Rush	Yds	TD	Fum
Jackson	306	524	58.4	3442	20	18	41	204	95	476	4	14
Rosenfels	351	562	62.5	4156	30	29	19	127	45	98	1	11

I've italicized the better numbers for each quarterback. Yes, I realized Jackson's better rushing numbers come with more attempts, but I don't think anyone will dispute that Jackson's a better runner than Rosenfels.

In this rather simple comparison, the numbers come out in favor of Rosenfels, 6-3. He's racked up more yardage, at a better rate per attempt (7.4 to 6.6) than Jackson; even his yards per completion is slightly better (11.8 to 11.2), showing that that's not just the result of Jackson's mediocre completion percentage. While clearly not a better runner than Jackson, Rosenfels has nonetheless done a much better job of avoiding sacks, going down at about half the rate of Jackson and fumbling with less frequency.

But there are negatives to Rosenfel's superior numbers, the most glaring of which is his interception total. If this were a full season's stats, 29 interceptions would be practically unacceptable for a starting quarterback. I've not seem him play much, but it's possible that part of the reason for Rosenfels' lack of sacks comes from him throwing the ball away and into coverage, thus resulting in the very negative play (interception) instead of the somewhat negative play (sack).

In any case, for all that Vikings fans -- such as this one -- complain about Jackson's prediliction for turning the ball over, it's clear that Rosenfels is markedly worse at protecting the football. Assuming that 2/3 of each player's fumbles were recovered by the defense (the typical NFL rate), that would give Jackson about 10 lost fumbles to Rosenfels' 8 for their careers. That gives Jackson 28 turnovers and Rosenfels 37. That's a rate of 4.2% for Jackson (when dividing turnovers by attempts+sacks+rushes) and 5.9% for Rosenfels. Finally, we can look at the rate stats for each QB. Rosenfels clocks in with an 81.2 passer rating and 4.48 TYA. Jackson's a 76.5 and 4.13, so the slight edge there goes to Rosenfels.

Still, if there's anything the Vikings need at the quarterback position, it's someone who doesn't turn the ball over and gives the defense and running game a chance to win (i.e., more of a "game manager" than a "gunslinger"). Rosenfels, at least through the first part of his career, doesn't fit that mold. Perhaps, as a career backup, he's been trying to "force" plays to make himself look better and impress in his limited playing time (the "Rosencopter" again comes to mind). With Adrian Peterson to hand off to and a world-class defense backing him up, perhaps he won't feel the need to try and do so much in Minnesota. (Then again, he won't have Andre Johnson to throw to anymore.) And there's little doubt he's got a stronger arm and looks better in the pocket than Jackson.

But, as I've said before, looks aren't everything. Rosenfels is probably better than Jackson, but only marginally so, and if the turnovers come in bunches in 2009, Brad Childress will need to do whatever he can to save his job -- even if that means going back to the safer, more careful Tarvaris Jackson. And who ever thought that would be the case?

Do you have to draft a QB high?

One of the topics (or, more accurately, one of the many tangents) of the most recent Pro-Football-Reference.com podcast was about the subject of "star quarterbacks" predominantly coming from very high (i.e., primarily first-round) draft picks and how teams can best find their quarterbacks. Says JKL around the six-minute mark:

The best option, I think, is to go for the elite talent at the top of the draft...Yeah, there are busts, but the upside there is just too great.

Recent busts -- Ryan Leaf, Tim Couch, Cade McNown, et al -- are well known, as are the success stories, like Peyton Manning, Donovan McNabb, and Ben Roethlisberger. But are they absolutely necessary? Do you have to pick a QB at the top of the draft to succeed? After all, if you don't pick a QB with your top pick, you're picking another (probably very good) player. And even the most jaded QB-loving fan would probably admit that quarterbacks tend to be a touch overvalued and definitely overdrafted.

So, where do starting quarterbacks come from? I compiled a list of starting quarterbacks* in 2008, what round they were drafted in, and whether they were with their original teams -- in other words, a first-round pick playing for a team that he wasn't drafted by didn't help his original team in 2008, so, in a sense, that team's first pick was a "bust."

* Here's the rub, though...rather than try to pass off guys like Ryan Fitzpatrick and Ken Dorsey as "starting quarterbacks," I defined each team's "starting quarterback" by the following two rules. He is:

A) The guy the team would have started if there had been a week 18; and
B) The guy the team would have started if healthy.

Point A lets me not worry about subsequent free-agent moves, trades, retirements, and so on. Point B lets me take the guy who "should" be the starter for the team (like Tom Brady over Matt Cassel) rather than a guy forced into the role. Here's the list:

Team	Quarterback	Round	Orig. Team?
Baltimore Ravens	Joe Flacco	1	Y
Oakland Raiders	JaMarcus Russell	1	Y
Philadelphia Eagles	Donovan McNabb	1	Y
Atlanta Falcons	Matt Ryan	1	Y
Pittsburgh Steelers	Ben Roethlisberger	1	Y
New York Giants	Eli Manning	1	Y
Denver Broncos	Jay Cutler	1	Y
Washington Redskins	Jason Campbell	1	Y
Cleveland Browns	Brady Quinn	1	Y
San Diego Chargers	Phillip Rivers	1	Y
Cincinnati Bengals	Carson Palmer	1	Y
Green Bay Packers	Aaron Rodgers	1	Y
Indianapolis Colts	Peyton Manning	1	Y
Detroit Lions	Daunte Culpepper	1	N
Miami Dolphins	Chad Pennington	1	N
Tennessee Titans	Kerry Collins	1	N
Minnesota Vikings	Tarvaris Jackson	2	Y
New York Jets	Brett Favre	2	N
Houston Texans	Matt Schaub	2	N
New Orleans Saints	Drew Brees	2	N
Buffalo Bills	Trent Edwards	3	Y
Chicago Bears	Kyle Orton	4	Y
Jacksonville Jaguars	David Garrard	4	Y
New England Patriots	Tom Brady	6	Y
St. Louis Rams	Marc Bulger	6	N
Seattle Seahawks	Matt Hasselbeck	6	N
Kansas City Chiefs	Tyler Thigpen	7	N
Dallas Cowboys	Tony Romo	U	Y
San Francisco 49ers	Shaun Hill	U	N
Carolina Panthers	Jake Delhomme	U	N
Arizona Cardinals	Kurt Warner	U	N
Tampa Bay Buccaneers	Jeff Garcia	U	N

Of the starting quarterbacks for the 32 NFL teams:

16 were first-round draft picks
19 are with their original teams
13 are "1Y" players -- first-round picks with their original teams

So, that means that 13 of 32 teams in 2008, or about 41%, found their "starting quarterback" by drafting him in the first round. That's a solid percentage, but maybe not enough to be considered as the "only" way to do it.

What about the quality of these quarterbacks, at least as compared to the later-drafted quarterbacks? The only 1Y I see on the list who might have competition next year is JaMarcus Russell. Matt Ryan and Joe Flacco could be one-year wonders, granted, but everyone else is pretty firmly entrenched as their team's starters. The list of non-first-rounders includes a Hall-of-Famer (Brett Favre), a potential Hall-of-Famer (Tom Brady), two of the best quarterbacks of 2008 (Drew Brees and Kurt Warner), and a slew of former or current Pro Bowlers and overall above average QBs (Matt Hasselbeck, Jake Delhomme, Jeff Garcia, Marc Bulger, Tony Romo). Overall, if I had to choose who the best QBs are on the list -- the first-rounders or the non-first-rounders -- I'd probably give the first-rounders the edge, but only barely.

This ignores the fact that there are two more notable 1Y players (Matt Leinart and Vince Young) lurking around who could be their team's primary starters very soon, depending on how the former first-rounder (Kerry Collins) and undrafted free agent (Kurt Warner) ahead of them play out. I also haven't taken draft position into account -- there might be a difference between being the #1 overall pick and the #23 overall pick (Brady Quinn). And, admittedly, this is a one-year sample size, though I have conducted a similar exercise, just for fun, the last few years. The list of starting quarterbacks hasn't changed too much, so it's always been around 1/2 first-rounders. Maybe I'll glance back ten years or so in a future post.

In any case, my conclusion is that, while it's not a bad idea to take a Matthew Stafford or Mark Sanchez early in the draft if your team needs a franchise QB, I don't think it's absolutely vital either. As with any position, good -- even great -- players can be found later in the draft, and quarterbacks probably aren't an exception to that rule.

Sage Rosenfels?

For a fourth-round pick? That's not awful, I suppose.

But still...does this really solve anything? Or is it just another grab at a cheaply available, mid- to low-talent veteran QB with no real upside (like Kelly Holcomb and Gus Frerotte)?

Probably the latter.

Instead of addressing their quarterback needs with either a solid draft pick (like I advocated Brian Brohm last year; the odds against fifth-round draft picks, like John David Booty, ever developing into anything useful are low) or trading or acquiring an above-average QB (such as Jeff Garcia or Donovan McNabb), it looks like the Vikings are heading back down the path of looking for the cheapest possible option at the position and hoping he can miraculously turn into a quality starter.

That might be possible with a good head coach, offensive coordinator, or QB coach. Unfortunately, we have Brad Childress and Darrell Bevell, which are about as far from "good" as can be imagined.

Rosenfels would bring a few positives to the team. He's a significant upgrade in accuracy (65.6% completion percentage in three years with the Texans), without sacrificing the deep ball (7.5 yards per attempt over that span). The downside is that he's a little mistake-prone, with 23 interceptions (and 24 TDs) in his 453 pass attempts with the Texans and is no threat to run, though he's taken only 16 sacks in three years. (Remember when we all though the Texans' offensive line was awful and then realized it was just David Carr?)

And then, of course, there was this play, which goes down with "The Orlovsky" as the worst plays by a quarterback in 2008. Must. Not. Do. Again.

Again, a fourth-round pick isn't a terrible price to pay, even if Rosenfels does nothing. And the deal's not done yet. If it's made, though, I hope the team doesn't view Rosenfels as a "savior" at the QB position. He's OK-but-not-great -- and not nearly as good as another QB the Vikings got for a fourth-rounder, 15 years ago.

Why 3/4 of quarterbacks are good

At least that's what you'd think to hear most media members talk about them (inspired by the most recent post on the PFR blog):


And even guys in the upper left quadrant often get credit for "veteran presence" or "leading their team to victory" (see Frerotte, Gus).

DI on Advanced NFL Stats Community

Last last year, Brian Burke over at Advanced NFL Stats started up a new site, Advanced NFL Stats Community, where he invited like-minded stat geeks (you know who you are) to submit their work to be published on the site. Following much procrastination on my part, I finally summarized my thoughts on TYA and submitted it to Brian earlier this week.

The finished article appeared today on the site -- which, along with its parent site, has a lot of great content you should definitely check out. Just bring an open mind and a willingness to cast aside preconceived notions about things like "low interceptions are always good" and "the best teams don't rely on luck to win."

While you're at it, check out some of the other sites on my new and improved blogroll to the right. In fact, I heartily recommend this feature to anyone using Blogger. It's like a personal RSS feed for blogs, and it seems to work for blogs running just about any kind of framework (Blogger, Wordpress, whatever PFR uses...).

And, obligatory Vikings note: Maurice Hicks is gone. Though some would say he never should have been here.

Is it too late to get Mewelde Moore back?

Tuesday snark

Most mock drafts I'm seeing have the Vikings picking some combination of QB, OL, WR, and CB for their first two draft picks -- all solid need areas.

This one has us picking a defensive end in the first round and a safety in the second. The DE position opposite Jared Allen isn't even remotely a concern, with Ray Edwards and Brian Robison notching a combined 7.5 sacks, and the Vikings drafted a safety in the second round last year (Tyrell Johnson) to serve as Darren Sharper's eventual successor. Someone ought to give this Draft Dog the Ol' Yeller treatment.

(Even better, look at their 2010 Mock Draft, which has the Vikings picking 10th. If that's how they think the team will do, then I feel very good about next season.)

* Sure, the Vikings have the Love Boat scandal, but at least one of our players never beat up a paper-towel dispenser.

* Is it any surprise that a Chicago Tribune writer came up with this?

Favre's numbers were skewed by an awful December in which he threw only two touchdown passes against eight interceptions.

Funny, I thought they were skewed by that September game where he threw for six TDs. Take that away and his passer rating is 76.7. Oh, that's right, you only remove a game or a couple games from a sample when it helps your argument, not when it hurts it.

* Not sure how long it will last, but go to the Alex Rodriguez story here. The headline is "A-Rod says he took energy booster." Look to the right, after the related headlines. I see an ad stating "Tired of Being Tired?" and pimping an ad for a product that says it "fights fatigue."

Yeah, maybe an ad for an energy-boosting supplement wasn't the greatest choice for that page. But hey, as long as everyone's making money, steroids are OK. It's when they start costing MLB fans (and fans' dollars) that they'll actually do something about it.