With Matt Flynn signing in Seattle, Robert Griffin III headed to Washington and Peyton Manning seemingly not interested, the Dolphins for the 13th season in a row are left wondering who their next franchise QB will be. In the 12 seasons since Marino retired, the Dolphins have struggled to find a merely average QB, let alone a true cornerstone player. They’ve trotted out a seemingly endless string of veteran retreads (Chad Pennington, Gus Frerotte), has-beens (Daunte Culpepper, Trent Green) and never were’s (Joey Harrington, Cleo Lemon, Sage Rosenfels, Brian Griese). A whopping 16 different QB’s have started for Miami since their Hall of Famer departed after the 1999 season. Take a look at the numbers:
Looking at it another way, take a look at the Dolphins team QB Rating in the last 12 seasons, compared to league average:
Only Chad Pennington in 2008 was significantly above average. Matt Moore, to his credit, did a pretty decent job last year – posting a QB Rating of 84.9, a little better than the 82.5 league average. To that end, Moore was statistically better than Michael Vick, Cam Newton Joe Flacco, Matt Hasselbeck, Mark Sanchez and Jay Cutler. The issue for the Dolphins is that Moore at his best doesn’t seem like the type of guy who can take a team deep in the playoffs without a great defense behind him. Of course it’s possible that he is a late bloomer and could continue to improve, but if you’re Miami, do you really want to put your future in the hands of Matt Moore? Probably not.
That said, Moore’s success in 2011 is a good reason to avoid Matt Flynn who is no sure bet to be any better than Moore and is only 10 months younger. That Flynn’s former offensive coordinator and current Dolphins coach Joe Philbin wasn’t willing to make a big play for Flynn (and it seems that Miami’s interest in him was tepid) says a lot about the Dolphins valuation of Moore vs. Flynn. While Flynn might end up the better player, it’s tough to blame Miami for not seeing him as a guy worth $19-24M.
Where the Dolphins do deserve some criticism though is not addressing the QB spot in the draft. They’ve wasted 2nd round picks on Pat White (2009), Chad Henne (2008) and John Beck (2007) and took a late round flier on Josh Heupel (6th round, 2001) but have otherwise ignored the position since Marino left.
Looking ahead, the Dolphins will have to decide whether or not they want to invest a high draft pick in a rookie QB. Ryan Tannehill is raw but has high upside and the Dolphins could easily sit him behind the reasonably competent Moore for a year. However, will Tannehill still be around when the Dolphins are on the clock? It seems likely, but there are still more teams in need of a QB than quality veterans available so it wouldn’t be a shock if someone jumps Miami to take Tannehill or perhaps the Browns pull the trigger at #4. The other option would be Brandon Weeden, but at soon-to-be 29 years old, is he really worth a high investment for a team with a mediocre 28 year old QB at the helm? Certainly not in the 1st round but will Weeden last until Miami’s 2nd round pick? What about 2nd tier guys like Kirk Cousins, Nick Foles or Brock Osweiler? Again, it’s tough to project any of those guys as a long term answer or even a better solution than Moore.
Unfortunately for the Dolphins, unless Moore totally bombs they will not be in position for Matt Barkley next year so they might have to get aggressive with how they address the position. Either they will have to secure Tannehill (trading up if necessary) or they will have to be prepared to go hard for Barkley (or another franchise QB type) next year. Of course, if the Dolphins go with Moore + Cousins (or another 2nd/3rd tier QB) they will just be reliving the pattern of the last few years which has gotten them nowhere. At some point they have to make a splash at the position because they won’t hoist the Lombardi trophy without an upgrade from the Matt Moore’s of the world
From the mailbag:
(In reference to my examination of QB’s selected outside of the top 50 picks in the draft)
that is an apparently telling statistic regarding the 90% “failure” rate of non-top-50-pick quarterbacks. However, what is the failure rate of top-50-pick quarterbacks (didn’t start for 5 years or throw 1,000 passes)?
That answer would provide a context for the failure rate among non-top-50 picks
It’s a good question. Let’s take a quick look at these QB’s :
Between 1995-2010, there were 49 QB’s selected in the top 50.
18 have started for 5+ years
29 have thrown 1000+ passes (with Bradford and Stafford likely to get there by the end of the year, health providing)
Of the 35 QB’s drafted from 2000-2010, only 5 are out of the league (Chad Pennington, Patrick Ramsey, Joey Harrington, Jamarcus Russell, Pat White).
Here are 40 years of QB’s
As you can see, a majority of Top 50 QB’s hit 1000 pass attempts in their career, whereas only 16% of QB’s taken after pick #50 do. In that 40 year time period, 51 Top 50 QB’s were 5+ year starters compared to an aggregate 31 from the later portions of the draft.
Ben also writes:
Per your stats, a full 30% of the top 20 NFL quarterbacks [in 2011 passer rating] were not among the top-50 picks. As I see it, that’s a success story, not a failure story.
Here’s the thing though, those 5 QB’s (Brady, Romo, Schaub, Fitzpatrick, Hasselbeck) came from 4 different draft classes (1998, 2000, 2004, 2005) over a 14 year span (1998-2011). In that time period, there were 127 QB’s drafted after pick #50 and probably 300+ undrafted guys. Most years will produce one “decent” QB (short term starter, journeyman starter or long-term decent backup) and every so often a very good (or better) long-term starter emerges. That 25% of the “best” (by passer rating) QB’s were later selections/undrafted shouldn’t really be a surprise. Good QB’s tend to start for a long time, so there only needs to be a “good” QB to come out every 2-3 years to consistently have a few such players in the top 20 QB’s.
For reference, here the best of the drafted “2nd Tier+” QB’s since 1995:
2010: Colt McCoy, John Skelton
2009: Curtis Painter
2008: Chad Henne, Josh Johnson
2007: Trent Edwards, Tyler Thigpen
2006: Tarvaris Jackson, Bruce Gradkowski
2005: Kyle Orton, Matt Cassel, Derek Anderson, Ryan Fitzpatrick
2004: Matt Schaub
2003: Chris Simms
2002: David Garrard, Josh McNown, Shaun Hill
2001: Quincy Carter, Sage Rosenfels, A.J. Feeley
2000: Marc Bulger, Tom Brady
1999: Aaron Brooks
1998: Charlie Batch, Brian Griese, Matt Hasselbeck
1997: Koy Detmer, Jake Delhomme
1996: Danny Kanell, Jon Kitna, Damon Huard
1995: Kordell Stewart, Rob Johnson, Kelly Holcomb
Not many top QB’s. Even many of the “successes” were guys who had 1-2 really good years in otherwise average (or worse) careers (Delhomme, Brooks, Orton, Garrard).
Individually, Brady, Romo et al. are success stories. But the chances of finding a top QB outside of the top tier of prospects are very, very slim. It’s a good strategy to take a QB every year or 2 in the hopes of hitting on one, but a QB-needy team that passes on the elite prospect in hopes of finding a guy later on is probably going to fail to find their quarterback of the future.
(originally posted by me at Mocking The Draft)
Most people who follow football know that Tom Brady is one of the all time late round draft steals – selected in the 6th round of the 2000 draft sixteen selections after Spergon Wynn and three spots before Todd Husak. He’s outlasted 10 the 11 other QB’s taken in 2000; only Chris Redman (3rd round) is also still active. More knowledgeable fans might also know that there are a few high-quality NFL QB’s who didn’t get much hype on draft day: Tony Romo (undrafted), Matt Hasselbeck (6th round), Matt Schaub (3rd round) and Ryan Fitzpatrick (7th round). These guys are often used as examples of how great QB’s can be found anywhere in the draft. Unfortunately, that’s usually not the case. Almost all successful NFL QB’s are found in the top 50 selections of the draft. However, that won’t stop NFL personnel men and general managers from trying to unearth the next Brady or Romo.
Is there a hidden gem to be found in 2012? Well, by now, most NFL draft watchers know the top QB prospects: Andrew Luck, Matt Barkley, Landry Jones, Ryan Tannehill and Robert Griffin III. As we move into the Bowl season and pre-draft workouts, those five guys will get the bulk of the hype (assuming they declare for the draft) – and deservedly so. However, there are a few other guys that will go later on who could find success in the NFL as starters.
Before taking a look at some of the less known (or less hyped) QB prospects for 2012, it should be noted that the chances for success of any QB taken outside of the top 50 picks are remote. While Brady, Romo, Hasselbeck, Fitzpatrick and Schaub were great finds, they are the only non Top-50 QB’s in the top 20 for passer rating:
As you can see, 9 of the top QB’s by passer rating were top 10 picks with Cutler and Roethlisberger both going 11th overall. It should be noted that of the five aforementioned non-Top 50 QB’s, only Romo and Brady found success with the team that drafted them.
How unlikely is it to find a good QB once you get beyond the middle of the 2nd round? Check out these facts:
QB’s drafted after pick #50 since 1995: 148
# of those who have attempted 1000 or more passes: 15
# who have been/were full time starters for 5+ years: 9
# of undrafted QB’s since 1995 to throw 1000+ passes – 5
# of undrafted QB’s since 1995 to start for 5+ years – 4
It probably goes without saying, but if your favorite team needs a “QB Of The Future” come April, 2012, they should take one early. If they decide to try to get lucky later on, here are four guys who could beat the (daunting) odds and carve out a long-term starting spot in the pros:
Brandon Weeden (Oklahoma State) – Based on talent alone, Weeden could be a top 50 pick in the 2012 draft. Unfortunately, he’s already 28 years old (which makes him a month or so older than Aaron Rodgers). He’s the prototypical pocket passer with the size (6’4 220) that NFL scouts love. His accuracy on short and intermediate routes is excellent, although his deep ball is somewhat lacking compared to the top QB prospects. He plays in a shotgun offense in college, which can make for a tough adjustment for some QB’s. His awareness is above average and while he’s not athletic, he’s got good enough feet to escape the rush moving outside of the picket. In order for Weeden to succeed, he will need to go to a team which can quickly work him into their offense (probably a team that runs an offensive scheme similar to Oklahoma State’s). Over-aged QB prospects have not fared well in the pros (Chris Weinke, John Beck) but Weeden has the tools to overcome the odds if he lands in the right city. I expect Weeden to go somewhere in the Top 100 picks, as too many teams need a QB to let him drop too far. In the right environment, Weeden could be a 5-7 year starter.
Kirk Cousins (Michigan State) – As we run up to the draft, Cousins will get labeled with two of my least favorite (and over used) terms: “game manager” and “has intangibles”. Those types of labels are often applied to guys who get it done in college, but lack an elite attribute (e.g. arm strength, mobility, etc). When I watch Cousins I see a guy who has a well rounded, but definitely flawed, skill set: he’s got good arm strength, his accuracy at times is very good and his mobility – while not especially noteworthy – is good enough to allow him to make the occasional play. He comes from a run heavy team, which has allowed him to develop his playaction ability which is as good as any prospect in this draft. What I don’t like is that he sometimes (too often for a top prospect) seems confused by opposing defenses and has a tendency to lock onto his first read. His mechanics are sloppy and he will need a good QB coach to clean them up if he is to succeed in the pros. It’s tough to predict where Cousins will go – he needs a lot of work but he’s the type of guy which teams tend to fall in love with. I like him as a 3rd rounder but he could sneak into the middle of round, possibly even earlier.
Ryan Lindley (San Diego State) – Lindley will draw a lot of attention at the combine and in his Pro Day due to his excellent arm strength but he’s definitely a raw prospect. While he can get the ball deep, he often gets the ball too deep – overthrowing his targets far too often. His short range accuracy is a bit better although still lacking. His struggles could be partially a result of overcompensating for a bad receiving corps (even more so since Vincent Brown has graduated). In an effort to avoid interceptions, he tries to put the ball where only his receivers can make a play and he lacks the finesse to do so with regularity. His mechanics tend to fall apart completely under pressure and he will need a lot of coaching before getting into an NFL game. That said, he’s got an NFL caliber arm, has shown improvement every year in college and possesses a lot of untapped potential. His rawness will probably drop him down the board a bit and he’ll almost certainly have to wait til the 3rd or 4th round before he gets drafted.
Aaron Corp (Richmond) – A small school sleeper, Corp will probably have to wait until the 6th or 7th round to hear his name called (if he is drafted at all) but there is some reason to think Corp could find a home in the NFL. He does a pretty good job at reading opposing defenses and finding the open man. His arm strength isn’t great, but he’s got enough to succeed in the NFL (especially with the proliferation of the spread offense and short passing game). When he’s able to set his feet in the pocket, his accuracy is very good – he recently set the FCS all time record for single game completion percentage. Unfortunately for Corp, his O-Line is awful which has led to him getting knocked around and chased around a lot. As a result, his accuracy and decision making have been inconsistent at times. He’ll need a few years to develop, but he has some good tools with which to work.
If the pundits and (most) scouts are to believed, Stanford QB Andrew Luck is the next big thing. Often trumpeted as “the best QB prospect since Peyton Manning”, Luck has NFL fans and personnel men drooling. Watching him play, it’s easy to see why. He has prototype size (listed at 6’4 235 lbs), a strong arm, good accuracy and a terrific release. On top of that, he is very smart, comes from a pro-style offense and has enough mobility to extend plays (a la Ben Roethlisberger). There’s very little to dislike about Luck. Sure, his mechanics need work (so did Aaron Rodgers’) and he needs to get less air under his intermediate passes but those are problems which are fixable with good coaching. So it’s not surprising that the “Suck For Luck” mantra has caught on. The NFL is very much a QB-oriented league and true franchise QB prospects come along very rarely. That said, QB’s who are selected 1st overall rarely live up to the hype.
Since the AFL/NFL merger in 1970, there have been 18 QB’s selected 1st overall:
If you want evidence that that NFL is a QB-centric league, look no further than the fact that a QB has gone 1st overall in 8 of the last 10 drafts. Anyone who follows football can probably tell you that the list of #1 pick QB’s is more hit than miss. While Luck may be thought (now) to be a better prospect than Tim Couch, David Carr and Carson Palmer (etc), those guys also came out of school with much fanfare. Most were thought to be “can’t miss” and “franchise QB” worthy (a somewhat obvious point, since they probably wouldn’t have gone #1 overall were they not so highly regarded). Sure, there were the “Peyton Manning or Ryan Leaf?” debates in 1998 and “Drew Bledose vs. Rick Mirer” in 1993. But, for the most part, these guys were regarded as high draft picks throughout their last season in school and in the run-up to the draft (Russell is the most obvious exception).
Earlier, I discussed the concept of Rate+ (Passer Rating compared to league average) as a way to measure a QB’s performance from year to year. Here’s a list of the 1st overall QB’s, the number of seasons where they had 150+ passing attempts, their average Rate+ over those seasons, the number of seasons with a Rate+ of 100 or better and the percentage of 150+ attempt seasons which were 100+ Rate+ (in other words, the percentage of seasons where they were better than league average):
And here is the data in graph form:
The X-axis is the QB’s season # with 150+ attempts and the Y-axis is the average Rate+. So, 1st overall QB’s average a Rate+ of 85 in their first 150+ attempt season. They see a pretty consistent increase, peaking at Rate+ of 117 in year 7 (if they survive that long in the league) and then decline. For their careers, 1st overall QB’s average a Rate+ is 103 – or 3% better than league average (in qualifying seasons). “Elite” QB’s (Peyton Manning, Tom Brady, Aaron Rodgers, Drew Brees) typically end up around 115+.
This isn’t to suggest that Luck will be below average for his first few years, or that he will bust, but he will need to be as exceptional as the hype makes him out to be in order to be a great QB from Day 1.
(Originally posted by me at Mocking The Draft)
“Suck For Luck!” has become the rallying cry emanating from Miami, Indianapolis and even St. Louis (where apparently Sam Bradford’s honeymoon is about over). Many fans of bad teams or teams with bad QB’s are hoping that their team will end up atop the 2012 draft, in position to draft uber-prospect Andrew Luck (or trade the “Luck pick” for a king’s ransom). With all the hype surrounding Luck, one can’t help but think back to other QB’s who have gone 1st overall in the draft. One of whom is currently playing for Luck’s former coach – 2005 1st overall pick Alex Smith
Coming into the 2011 season, Smith had put up numbers ranging from terrible to mediocre, but he seems to be blossoming this year under Jim Harbaugh and former Stanford assistant (now 49er’s Offensive Coordinator) Greg Roman:
The column all the way to the right is the one of the most interest: Rate+. If you’re unfamiliar with the “[stat]+” type metrics, it’s a way to compare a player’s stats (in this case, Passer Rating) to the league average for that particular year. It’s calculated as follows*:
In other words, a 110 Rate+ means a QB’s rating is 10% better than the league average. A Rate+ of 90 would be 10% worse than average. In 2007, Alex Smith’s Passer Rating was 57.2 and the league average was 80.9. So his Rate+ is [100*(57.2/80.9)]=71, or 29% worse than league average. This type of statistic is especially useful for things like Passer Rating which (like ERA in baseball) varies significantly from one era to the next. Look at the average QB Passer Rating in 10 year intervals:
So, Alex Smith’s career best 82.1 passer rating in 2010 was about league average. But had he played in 1980 and put up a 82.1 rating, he’d be considered an upper-echelon QB.
If Smith can maintain his pace, he will become the 124th QB since 1970 to post a 110+ Rate+ in a season where he threw at least 150 pass attempts. While that’s not particularly notable, the fact that he could do it for the first time in his 6th 150+ attempt season is unusual. Only 7 of the 123 QB’s who have already gotten to 110+ have taken longer to hit the milestone. Of the 123, 90 of them accomplished the feat before their 4th season (47 of them did it in their first 150 attempt season). Here are the other 33:
The middle column is the number of 150+ pass attempt seasons the QB had before his 110+ Rate+ season. The column to the right is the year in which he finally got to 110.
Notable in that list is former 1st overall pick Eli Manning and Smith’s current head coach, Harbaugh, both of whom also got to 110+ in his 6th 150+ pass attempt season. Playing for a guy who understands late-blooming QB’s can only help Smith’s career. He may never shed the “bust” label entirely, and will almost certainly never be the best QB to come out of the 2005 draft (Aaron Rodgers), but Alex Smith seems to be on the road to being a decent starting QB. At only 27 years old, he could be a solid starter in the league for the next 5-7 seasons.
*Note: Pro-Football-Reference.com (and perhaps other places) use a slightly more complex formula for coming up with [stat]+ for QB stats. My method is the “traditional” method first popularized a decade or so ago in baseball.
(Originally posted by me at Mocking The Draft)
Just over three months from now, Super Bowl XLVI will be in the books and a new NFL champion decided. Almost immediately thereafter, draftniks, NFL pundits, analysts and the like will start dissecting the next wave of rookies to enter the league. Leading the way will be the Class of 2012 QB’s, guys like Andrew Luck, Landry Jones, Matt Barkley, Nick Foles, Robert Griffin III and a few others. College gametape will be broken down endlessly, results from the scouting combine will be analyzed and re-analyzed and college stats will be thrown around as fodder for discussion (and perhaps used at the the basis of some lazy scouting reports). What, if anything, can we learn about these young quarterbacks’ chances of NFL success from looking at their college stats?
Before we take a closer look at stats, it should be noted that football is a very difficult game to quantify. There are a number of variables which aren’t represented in stats that only game tape (or live scouting) can tell you. Quality of competition, quality of a QB’s receivers (bad receivers could drop more passes, leading to more incompletions, etc), weather conditions, offensive scheme and a lot of other things all should be considered. For that reason, stats aren’t particularly useful to make specific predictions. Stats are much more useful in football as a basis for broad generalities. So, with that caveat, let’s take a look at some numbers.
Below (and you’ll have to click the picture to make it bigger), I’ve listed a group of 40 QB’s, along with both their college (on the left) and pro (on the right) stats. All 40 fit the following criteria:
1. They were active in 2010 and at least somewhat relevant in 2011
2. I could find their college stats (Sorry Jon Kitna)
3. They have attempted at least 500 NFL passes
Additionally, I’ve calculated each QB’s college passer rating using the NFL formula (which differs from the NCAA formula). The college stats are updated through last weekend, the pro stats are current through October 23rd.
The QB’s are then sorted by NFL passer rating.
A number of things stand out:
1. Only one of the top 10 active QB’s in terms of passer rating had a college completion % less than 61.5% (Carson Palmer).
2. Nine of the top 15 had at least 1,000 pass attempts in college, whereas 6 of the bottom 15 did
3. College QB’s with completion percentages less than 60% are bunched up in the middle. In other words, high completion percentage in college isn’t predictive of NFL success but is almost always a trait top end NFL QB’s have. I have come to think of this as the “Gaudy Stats Principle” – college QB’s with especially good college stats tend to either flop (e.g. Matt Leinart, David Carr) or hit it big (e.g. Aaron Rodgers, Tony Romo, Ben Roethlisberger). There aren’t many “middle tier” QB’s in the NFL who had especially impressive college numbers.
Breaking down the QB’s college stats into 4 tiers we can see this a bit more clearly:
What’s really interesting here is that the second tier is worse than the first tier across the board, but the third tier is slightly better than the 2nd and the bottom group is better than either middle group.
How do the 2012 prospects look? Surprisingly good, minus one:
Russell Wilson is the only guy who has a completion % below 60. Weeden, Griffin and Foles all are completing >66% of their passes. Of the 40 NFL QB’s listed, only Sam Bradford, Matt Schaub and Alex Smith had a college completion % that high. One thing to consider here is that completion % has become inflated over time, so if guys like Tom Brady and Peyton Manning played in today’s NCAA, their numbers might be even better than they appear. That said, take a look at the Class of 2011:
In general, the class of 2011 pales statistically to the 2012 group – except for Cam Newton whose small sample size should be taken into account, although the early signs look very good for him in the NFL.
In the end, almost all the top QB prospects in 2012 have stats which indicate they could have elite NFL upside based on the broad observations we can make from the currently active NFL QB’s. So, come April, when some talking head on TV is telling you that QB Prospect X will be better than QB Prospect Y because of his college stats, remember that all of these guys (except Wilson) have college stats similar to – or better than – Drew Brees, Tom Brady, Aaron Rodgers, Peyton Manning et al. That they all seem to fit in the “Gaudy Stats Principle” suggests that the class of 2012 will have a few future superstars and a few huge busts.
Hopefully this exercise has shown that while stats can be fun to play with and provide us more things to (over)analyze in the otherwise slow months leading up to the draft, they are no substitute for real scouting and analysis.
Yesterday I discussed LSU CB Tyrann Mathieu’s size (or lack of) and how it could affect his draft stock. The conclusion was that there’s a good chance he could overcome his small stature and become a productive NFL player. However, size matters at some positions more than others. 2012 QB prospect Russell Wilson is a good example of a successful college player who is almost certainly too small to succeed in the NFL.
Wilson, who is currently sporting a 74.8 Cmp% and a gaudy 12.53 YPA, is listed at 5’11 191. Not exactly the prototype size for an NFL QB. But is he too small? Recent history tells us that he most likely is. Since 1980, only 4 different QB’s under 6’0 have attempted more than 10 passes in a season:
Not exactly a great track record. Seneca Wallace performed well for half a season in 2008 and Doug Flutie had a solid 1998 campaign. But otherwise, short QB’s have not fared well in the NFL. The most successful short QB is Drew Brees, who is often used to counter the “too small for the NFL” argument. Unfortunately, Brees is a rare exception. Were we to include 6’0 QB’s like Brees in the discussion we’d see that there haven’t been too many more success stories:
Michael Vick and Joe Theismann stand out as other success stories, but other than 2010 Vick has not had much success as a passer and Theismann might be one of the most overrated QB’s of all time (check out that TD to INT ratio). Take out Drew Brees and 2010 Michael Vick and 6’0 QB’s have thrown a combined 294 TD and 286 INT with a 56 Cmp%. Not too great. The one thing that Wilson has going for him is that his college numbers aren’t significantly worse than Brees’:
While Wilson’s performance has been pretty similar thus far, it should be noted that it’s rare for a top NFL QB to ever post a <60 Cmp% in college. That he hasn’t hit that benchmark until this season could spell trouble for him if he even gets a shot in the pros.