QB Draftees and the Completion Percentage Correlation.
by iverson2169 on Mar 30, 2009 1:19 AM PDT
50 comments
Hypothesis: Completion percentage is the single best statistical indicator of NFL QB success.
There hasn't been a more discussed draft topic lately than the annual debate about what types of college quarterbacks translate into successful NFL quarterbacks. Opinions seem to differ as to whether the quintessential "gunslinger", or the more cerebral "accurate" passer is a safer 1st round pick.
Specifically, as it relates to the Seattle Seahawks, the debate over whether or not the Seahawks should draft Stafford (or Sanchez for that matter) is complex to say the least. Conventional wisdom would lead us to believe that any single statistic cannot predict success in a draftee. I thought the theory deserved to be tested and found that, as odd as it seems, there may be a statistical category that can predict NFL QB success better than others.
After some research, there does seem to be a very interesting correlation between QB completion percentage and subsequent success in the NFL. The premise is that collegiate QB's that complete more than 60% of their passes in their final season, have a better chance of NFL success, while QB's that complete LESS than 60% are substantially more likely to fail.
The following group of QB's drafted in the first round (1997-2004 ... stopped in 2004 to allow for a development window) completed LESS THAN 60% of their passes during their final collegiate season. This list is complete .
1997: Jim Druckenmiller, Virginia Tech (Completion Percentage: 54 percent)
1998: Ryan Leaf, Washington State (Completion Percentage: 55 percent). The (1) QB taken in Round 1 with better than 60% (Manning).
1999: Akili Smith, Oregon (Completion Percentage: 58 percent); Cade McNown, UCLA (Completion Percentage: 58 percent). The three QB's taken in round 1 with better than 60% (Couch, McNabb, Culpepper)
2000 Only 1 QB selected in Round 1. Pennington had higher than 60% completion percentage.
2001: Michael Vick, Virginia Tech (Completion Percentage: 54 percent)
2002: Joey Harrington, Oregon (Completion Percentage: 59 percent); Patrick Ramsey, Tulane (Completion Percentage: 57 percent. The (1) QB taken in Round 1 with better than 60% (Carr).
2003: Kyle Boller, California (Completion Percentage: 53 percent); Rex Grossman, Florida (Completion Percentage: 57 percent). The two QB's taken in round 1 with better than 60% (Palmer and Leftwich)
2004 J.P. Losman, Tulane (Completion Percentage: 59 percent). The three QB's taken in round 1 with better than 60% (Manning, Rivers, Rothlisberger).
10 out of the 21 QB's completed less than 60% of their final collegiate season:
Legends: None
Pro Bowl: Michael Vick
Stars: None
Some Success: Rex Grossman
Limited to No Success: Kyle Boller, Patrick Ramsey, Cade McNown, Joey Harrington, JP Losman
Failure: Ryan Leaf, Jim Druckenmiller, Akili Smith
11 Completed more than 60% of their passes in their final collegiate year.
Legends: Peyton Manning
Pro Bowlers Carson Palmer, Donovan McNabb, Ben Rothlisberger, Eli Manning, Chad Pennington, Dante Culpepper
Stars: Philip Rivers
Some Success: Byron Leftwich
Limited to No Success: David Carr, Tim Couch
Thats a staggering 80% of the "Less Than 60%" group that have experienced little or no NFL success (based on a subjective common sense filter)
. . . and an equally staggering 73% of the "More Than 60%" group that are stars or better (Obviously there is some room for debate as to which categories I have ranked them in, but they should be fairly representative of the collective consensus).
With this is mind:
What do this years 1st Round Draft Picks bring to the table?
Mark Sanchez: 65.8%
Matt Stafford: 61.4%
Josh Freeman: 58.6%
Time will tell if this Statistical Model holds up for these 3.
A place to bury strangers.
1 recs |
50 comments
Email
Print
Comments
how does Staffords fresman year impact his numbers?
i’m at work so I can’t be digging through stat-sheets, but I assume he was pretty bad his freshman year and those numbers are pulling down his career stats. Starting as a freshman should HELP him, but using this metric it hurts him.
just a thought.
by cro-mag! on Mar 30, 2009 9:37 AM PDT reply actions 0 recs
It doesn't affect it.
Since it says 60% or higher in their final collegiate season.
by LantermanC on Mar 30, 2009 9:57 AM PDT up reply actions 0 recs
oh, my bad. didn't catch that.
thanks!
by cro-mag! on Mar 30, 2009 11:49 AM PDT up reply actions 0 recs
Awesome find.
A couple of qualms, which have been discussed before. If a player throws 59.5% as opposed to 60.5% I don’t think there is much difference. A throw here or there. It doesn’t tell me if I’m throwing in a spread or a pro-type offense. It doesn’t say if I’m throwing deep or doing a bunch of short passes to my RBs. It doesn’t tell me if I throw 30 times per game, or am a Pat White-type who throws as little as 10 times in a game. It doesn’t tell me how good my o-line and supporting skill position players are. It doesn’t say how good the opposing competition is. If a player throws 58% in his junior year and then 62% his senior year, does that now make him a success, whereas he wouldn’t have been before? It could be that he just had a Jeremy Reed-type thing going where several years at AAA for a (possibly) AAAA player makes it so that he can shine at that level with enough repetition, but at the major league level, he can’t make the jump.
But as far as a cursory study based on one number goes, this study seems pretty damn solid.
by LantermanC on Mar 30, 2009 9:56 AM PDT reply actions 0 recs
This idea is ripped off from David Lewin
And rather shamelessly. It’s also pretty shaky. Completion percentage matters, if only because it is a stat with a large enough sample to be meaningful, but with a sample size of quarterbacks as small as the above, there’s bound to be a stat that somehow bifurcates them. It doesn’t mean it’s meaningful. It certainly wasn’t meaningful for Matt Ryan: 59.3%.
by John Morgan on Mar 30, 2009 9:56 AM PDT reply actions 0 recs
I knew I had seen this info somewhere before...
by Fear on Mar 30, 2009 10:29 AM PDT up reply actions 0 recs
In fact, now that I think of it
I think you already talked about this when you first started dissecting the QBs, bringing it in as part of the overall dissection as a tool to help evaluate, but certainly not one that gives any answers by itself.
by Fear on Mar 30, 2009 10:32 AM PDT up reply actions 0 recs
One thought I have is
that looking at completion percentage doesn’t account for ‘surrounding talent’. I know we’ve discussed this, but Ryan’s surrounding talent wasn’t good (and some would say worse than that). His receivers, running backs, etc. could have had a negative effect on his comp. On the other side of that, if this is valid criterion, would be that Sanchez, for example, probably had his comp benefit due to his surrounding cast. Another example of this might be Oakland’s Russell while at LSU. It’s been said that Ryan did more with less, therefore his success may have translated better. I imagine the offense run may be yet another factor disregarded when looking strictly at comp%.
Just throwing that out there. I’m not sure how relevant that stuff is. I realize that team and what the coaching staff’s did with their respective NFL QBs are huge in why each did have done what they’ve done, as well.
I’ve read the Lewin forecast, and you’re right about the um, similarity.
by Misfit74 on Mar 30, 2009 11:35 AM PDT up reply actions 0 recs
What? I didn't rip this idea off...I cited my source of information with a link.
Completion percentage matters, if only because it is a stat with a large enough sample to be meaningful
Every QB drafted in the first round from 97 to 04 ISN’T meaningful? Weird.
It certainly wasn’t meaningful for Matt Ryan: 59.3%.
Every statistics set has outliers (see Vick above who made the probowl).
there’s bound to be a stat that somehow bifurcates them. It doesn’t mean it’s meaningful.
I guess I should stop creating meaningless fanposts?
by iverson2169 on Mar 30, 2009 6:12 PM PDT up reply actions 0 recs
I liked the post,
I thought people were being a bit rude myself :-/
by germpod on Mar 30, 2009 6:41 PM PDT up reply actions 0 recs
Just for the sake of Sh__t's and Giggles
I’ll looked even farther back (maybe even bordering on too far as game trends (schemes such as spread offenses) can change)
1996: No 1st Round QB’s
1995: QB’s with over 60% completion (Steve McNair, Kerry Collins); Without: None
1994: QB’s with over 60% completion (Trent Dilfer and Heath Shuler)
1993: QB’s with over 60% completion (Bledsoe)… Under 60% completion (Mirer)
That’s another 4 years (Total of 1993-2004) and all but Shuler conform.
OVER:
McNair
Bledsoe
Collins
Dilfer
Shuler (obvious bust)
UNDER:
Mirer (bust)
by iverson2169 on Mar 30, 2009 7:08 PM PDT up reply actions 0 recs
Shoots?
Sheets?
Schwots?
…Shits?
"Part, fools!
Put up your swords. You know not what you do."
by Fearless Frog on Mar 30, 2009 7:17 PM PDT up reply actions 0 recs
Collins was a horrendous bust for the Panthers.
by John Morgan on Mar 31, 2009 11:31 AM PDT up reply actions 0 recs
Collins made the pro-bowl for the Panthers!
Also, I thought we were discussing NFL success.
Collins:
37393 yards passing and 187 TD’s is a decent career I’d say.
by iverson2169 on Mar 31, 2009 9:50 PM PDT up reply actions 0 recs
No point in discussing NFL success
when we’re talking about our draft pick. Tony Mandarich was a decent offensive lineman with the Colts, but I don’t think that made Green Bay any happier about drafting him.
by SeaTownBlueDevil on Apr 1, 2009 10:12 AM PDT up reply actions 0 recs
People aren't saying that you ripped an idea off.
They’re saying the guy you cited, Wes Bunting, ripped off another guy. Apparently this study has already been done by a guy named David Lewin.
by LantermanC on Mar 30, 2009 8:12 PM PDT up reply actions 0 recs
I never read any of the other articles
Lewin or Bunting, but I found this pretty interesting.
by SeaTownBlueDevil on Mar 30, 2009 8:36 PM PDT up reply actions 0 recs
You didn't rip it off
the person you cited ripped it off. Even the 60% rule has been done.
And no, finding a stat that roughly conforms to the performance of 21 quarterbacks is not very meaningful. A good stat must be predicative, not just able to conform to pre-existing results. Nothing magic happens when a quarterback reaches a 60% completion. Losman, for instance, was 59.5% his final season. His team compiled a 5-7 record. Losman could have missed 60% because he spiked the ball to stop the clock—twice.
It’s just a very unscientific use of numbers.
by John Morgan on Mar 31, 2009 11:31 AM PDT up reply actions 1 recs
I don't get how you can say this John...
A good stat must be predicative
A statistic does NOT need to be predicative. A stat is simply numerical datum… unbiased and unfiltered. A good THEORY should be predicative.
not just able to conform to pre-existing results
This really shocks me. Throughout the history of scientific method, the ONLY way one can go about creating theorems is to sort through data and find pre-existing correlations. PERIOD! When a theory is formulated and the pre-existing data doesn’t support the premises, it is scrapped. In this case, ALL 27 QB’s over an 11 year time period were looked at, and 21 out of the 27 conform to the statistical model.
And no, finding a stat that roughly conforms to the performance of 21 quarterbacks is not very meaningful.
A. I went back farther and included 6 more QB’s.
B. That’s 27 QB’s over 11 years!!! How big of a sample size do you want? I don’t remember anyone saying it was a universal law (like gravity), but when 21 out of 27 QB’s in the set conform I just don’t see how one can say it is not meaningful.
by iverson2169 on Mar 31, 2009 8:03 PM PDT up reply actions 0 recs
60% is an arbitrary number to make those groupings look good
What would change if it was 59, 58, or 57? There is nothing about a quarterback who’s able to complete 60 of his passes that makes him a star. The number is a symptom, not a cause. The Losman and Ryan examples show that.
by Nate Dogg on Mar 31, 2009 8:21 PM PDT up reply actions 0 recs
Correct Nate
It is arbitrary, and it does make the groupings look good. What’s the problem with that?
If the number were changed to 59% for example it may look more like 50/50 and have no relevance. The whole point of the exercise is attempting to find a threshold that works.
For 1st round draft picks from 1993 to 2004 (05-07 accounts for a reasonable learning curve before making a judgment) 60% completions is a relatively accurate indicator (21 of 27).
Does that mean that any QB drafted in round 1 with more than 60% completions is a guaranteed NFL success? OF COURSE NOT, but the data suggests that there is a likelihood that they could be.
by iverson2169 on Mar 31, 2009 9:56 PM PDT up reply actions 0 recs
Can I also add this....?
A good stat must be predicative, not just able to conform to pre-existing results.
John is that not what you did here? Find a stat category to support a point that conformed to pre-existing results? Doesn’t a stat category, by definition, have to conform to results that pre-existed?
Aggregate Completion Percentage: 58.8%
Kyle Boller 2005-2007: 58.8%
Adjusted Net Yards an Attempt: 4.8
Jon Kitna 2005-2007: 4.7
Air Yards
2006 AY/A: 4.4 (4)
2007 AY/A: 4.0 (11)
2008 AY/A: 3.7 (23)
DYAR
2006: 22
2007: 937
2008: -334
Three Year Average: 208
Jon Kitna’s Average 2005-2007: 208
and you later posted this reply:
I didn’t pick or exclude any numbers. It’s simply the past three season. My point was only that with all the talk about 2008 being an exceptionally bad season for Matt Hasselbeck, how does he look when we expand the sample to its greatest meaningful size.
I don’t see this study being much different in execution. No 1st round QB’s were left out during those years, and the results look interesting (60% at least appears to be a meaningful threshold for the first rounders). ALso remember that to be a 1st rounder, you MUST posses some sort of intangible qualities as well.
I believe this may be where you are having problems with the argument (that its a paint by numbers for selecting QB’s) I dont believe this. I think that it’s a stat that can aid in sorting QB’s worthy of 1st round consideration (remembering of course that to be a 1st round caliber QB, other litmus tests had to have already been passed.
I’ll agree however; If the analysis were expanded to all rounds, it wouldn’t hold water.
by iverson2169 on Mar 31, 2009 11:42 PM PDT up reply actions 0 recs
Settle down.
No one was trying to slight you.
by BrianL on Apr 1, 2009 7:14 AM PDT up reply actions 0 recs
I wasn't addressing you...
…and the issue has already been cleared up. John’s original comment was that “this idea has been ripped off”… without mention of whom “this” belonged to. I then asked for clarification to which he replied that the link was the reference, not my fan-post. case closed.
by iverson2169 on Apr 1, 2009 8:36 PM PDT up reply actions 0 recs
I assumed the meaning was implied in the context, but I guess not
A stat that claims to be able to point to quality quarterbacks and busts must be predicative and independent of other variables. It’s an apples and oranges comparison between what I wrote about Hasselbeck and the “study” done above. I simply showed that Hasselbeck has been bad two of the last three seasons, I did not say it was a projection system for his future success. The 60% stat claims prognosticative value. It doesn’t control for important factors like that the players with a high completion percentage were selected with, on average, the ~5th overall pick while the low completion percentage group was selected with the ~14th pick.
This is a case of multiple end points. It’s a statistical trick. Brian Burke does a good job of describing it here.
Statistical Trickery
Why did Football Outsiders pick 370 as the cutoff? I’ll show you why in a moment, but for now I’m going to illustrate a common statistical trick sometimes known as multiple endpoints by proving a statistically significant relationship between two completely unrelated things. I picked an NFL stat as obscure and random as I could think of—% of punts out of bounds (%OOB).
Let’s say I want to show how alphabetical order is directly related to this stat. I’ll call my theory the “Curse of A through C” because punters whose first names start with an A, B, or C tend to kick the ball out of bounds far more often than other punters. In 2007 the A – C punters averaged 15% of their kicks out of bounds compared to only 10% for D – Z punters. In fact, the relationship is statistically significant (at p=0.02) despite the small sample size. So alphabetical order is clearly related to punting out of bounds!
Actually, what I did was sort the list of punters in alphabetical order, and then scanned down the column of %OOB. I picked the spot on the list that was most favorable to my argument, then divided the sample there. This trick is called multiple endpoints because there are any number of places where I could draw the dividing line (endpoints), but chose the most favorable one after looking at the data. Football Outsiders used this very same trick, and I’ll show exactly how and why.
Further, an outlier is not a data point that defies a “rule”. It’s not the so called exception that proves* the rule. It is a data point that for clear reasons does not conform to what is otherwise being measured. In other words, the only real outlier of this bunch is Michael Vick, and he’s not an outlier because his success defies the 60% rule, but because his success was not passing, but rushing, something the 60% rule does not hope to measure, and something that truly puts him apart from the others in the sample.
Matt Ryan is not an outlier, he’s a clear and obvious example that 60% rule is nothing more than statistical chicanery and that any team that puts any stock in it, is making a grievous error.
*Proves, in this case, being anachronistic. “Prove” meaning test, as in “proving grounds”.
by John Morgan on Apr 1, 2009 10:41 AM PDT up reply actions 0 recs
I think we are closer on this than you know...
The Hasselbeck reference was in regards to this comment:
if only because it is a stat with a large enough sample to be meaningful, but with a sample size of quarterbacks as small as the above, there’s bound to be a stat that somehow bifurcates them.
The Hasselbeck stats are an even smaller subsample by which to draw any conclusions. My point was: What’s the difference between the Hasselbeck data set, and the set form this study, in terms of meaningful sample size, or the ability to draw any meaningful conclusions from it?
The Hasselbeck data shows that ,in terms of the statistics being compared, he is on par with Jon Kitna and Kyle Boller.
This study’s data set (as limited in size, or as misleading as it may be), does show an interesting correlation (strike that)….maybe your more comfortable with the word “irony”?…. between 1st round picks and their completion percentage.
You’ll note that the majority of the posts content is paraphrasing from the source. The only editorial I have added were these comments:
… there seems to be an interesting correlation….. and Only time will tell if this statistical model will hold up….
I thought the post would provide interesting “food for thought”, but I never in any way championed this as anything more than a statistical curiosity.
by iverson2169 on Apr 1, 2009 8:56 PM PDT up reply actions 0 recs
Coincidence
and, as I said before, it’s not the completion percentage that provides the small sample, it’s the set of 21 quarterbacks.
by John Morgan on Apr 2, 2009 3:00 AM PDT up reply actions 0 recs
Can we agree to call it a trend?
trend (trnd)
n.
2. A general tendency or inclination. See Synonyms at tendency.
21 out of 27 would seem to be a general tendancy.
and, as I said before, it’s not the completion percentage that provides the small sample, it’s the set of 21 quarterbacks.
It’s 27 (not 21…that was amended to include 1992-1996) QB’s over 11 years, and I never said it was the Comp % that was the small sample size.
by iverson2169 on Apr 2, 2009 7:47 PM PDT up reply actions 0 recs
Interesting stuff
The problem is that there’s a big difference in “expected” NFL success between a QB drafted in the Top 5 and one drafted later in the first round. Comparing the draft positions of the players in your two lists (draft position in parentheses):
Low completion %:
Michael Vick (1)
Rex Grossman (22)
Kyle Boller (19)
Patrick Ramsey (32)
Cade McNown (12)
Joey Harrington (3)
JP Losman (22)
Ryan Leaf (1)
Jim Druckenmiller (26)
Akili Smith (3)
High completion %:
Peyton Manning (1)
Carson Palmer (1)
Donovan McNabb (2)
Ben Rothlisberger (11)
Eli Manning (1)
Chad Pennington (18)
Dante Culpepper (11)
Philip Rivers (4)
Byron Leftwich (7)
David Carr (1)
Tim Couch (1)
Even a cursory glance suggests that, even without knowing anything about completion percentage, the latter group was much more highly-regarded (and probably highly-skilled) overall.
by cyberwulf on Mar 30, 2009 10:02 AM PDT reply actions 0 recs
This suggests, once again,
that you need to adjust for draft position when performing such an analysis. I threw something together quickly, taking the final-year college stats of QBs drafted in 2005-2007 (all rounds). I looked at NFL games started as the outcome (which has its own problems, but is at least a “hard” outcome not subject to interpretation), and adjusted for pick # and year drafted.
The result:
No statistically significant association between the following statistics and NFL games started: Yards, Yards/Att, TD, Int, Rating, Attempts/game
I’d like to do this looking back over a longer time period. Anyone know where to find a compilation of individual college QB stats from the past, say, 10-15 years?
by cyberwulf on Mar 30, 2009 10:10 AM PDT up reply actions 0 recs
As flawed as it may be
It’s more promising than it could be deflating.
by Nick Andron on Mar 30, 2009 1:00 PM PDT reply actions 0 recs
It may be flawed,
but it is not worthless. Look at this list, the list of under 60% percent has way more washouts then the list of over 60%.
Small sample size, compared to baseball maybe, but it is spanning 8 years of drafts. In football, every sample size is going to be small when you look at the big picture.
I like it, though I also liked the ESPN system. When you can take a stat, or series of stats to make a formula, and the players who are seperated by the stat system end up in good player and bad player catigories that are clearly defined, it seems like it is pointing to something and can not just be written off.
by germpod on Mar 30, 2009 1:42 PM PDT reply actions 0 recs
The ESPN system was a statistic for the sake of having a statistic.
It offered absolutely nothing substantial in terms of real analysis. It’s the WHIP of football stats.
by BrianL on Mar 30, 2009 3:28 PM PDT up reply actions 0 recs
WHIP isn't totally useless.
If someone has a WHIP of 3.00 after 200 IP, I can safely say that he is not a very good pitcher. Same for a WHIP under .5 after 200 IP and good pitchers.
by LantermanC on Mar 30, 2009 3:44 PM PDT up reply actions 0 recs
If a system predicts,
even a little bit then it is usefull. Do I think Ruskell sits back, say’s “Sanchez has the highest completion percentage, draft him”, of course not. But if there is a difference in success between QBs who complete over 60% vs under 60% then that should be part of the equation in predicting future success. If the two colums of players did not have a difference in quality of players then it would be a useless gauge, but there is a big difference.
WHIP is usefull, good pitchers have a low WHIP, bad ones have a high one. If a pitcher has a WHIP of .5 and another has a WHIP of 3.0 then you can make a safe assumption on which will be the better pitcher next season.
Due to only playing 16 games, and being such a team sport, real analysis of any individual is difficult. In baseball a person batting a ball only has him and the pitcher involved, what his teamates in the dugout are doing does not affect his at bat. On the other hand Jenning’s success is directly related to the safeties behind him, and the defensive line creating pressure. If you were to judge a baseball player on 16 games, you would be crazy, yet we do that every year to every player in football. It is our only option. By the time you get a big enough sample size the player has gotten older and now age throws in another variable.
by germpod on Mar 30, 2009 4:03 PM PDT up reply actions 0 recs
Something important to note:
What do this years 1st Round Draft Picks bring to the table?
Mark Sanchez: 65.8%
Matt Stafford: 61.4%
Josh Freeman: 58.6%
I don’t think this study is saying that higher the completion percentage is the higher the chance of success in the NFL, but rather that the 60% threshold is an on and off switch, or a dummy variable. Either you have a good shot at success or you don’t. Otherwise, it seems like it would imply the chances of Colt McCoy failing are as close to 0% as possible.
by LantermanC on Mar 30, 2009 3:46 PM PDT reply actions 0 recs
I like Stats
unfortunately, as was pointed out above, you can get into trouble when the sample size is small or there are other variables that affect the results. I haven’t checked, but I suspect Graham Harrell’s comp percent was 70. OK, I just checked and it was 70.6 (and 71.8% in 2007). Damn I’m good. The point is, that doesn’t guarantee his NFL success. The problem isn’t the conference or level of competition, it’s the system that took advantage of his skill set. A system that will not be, cannot be, duplicated at the NFL level. However, if you have a guy with the arm strength and skills and good competition, and his completion rate is under 60%, it is cause to ask why.
by diehard82 on Mar 30, 2009 8:11 PM PDT reply actions 0 recs
Good point with the Graham Harrel reference
but I suspect he won’t be drafted in the first round. That being said, that’s another flaw in the study. It’s almost as if it were saying IF Graham Harrell were drafted in the first round, he’d have a high chance of success, but if not, then the study can’t predict his success.
by LantermanC on Mar 30, 2009 8:14 PM PDT up reply actions 0 recs
No actually Lanterman....
What it says is that being drafted in the first round is a “filter” of sorts for some of those intangibles we all talk about.
by iverson2169 on Mar 30, 2009 10:07 PM PDT up reply actions 0 recs
Good point,
I guess intangibles (not necessarily character, but arm strength, decision making, and other things that can’t be measured accurately) can be lumped into round 1 level QBs. A bit rough, but I can’t think of a better way to do it.
by LantermanC on Mar 31, 2009 8:33 AM PDT up reply actions 0 recs
AND that makes me question Stafford
Surounded by talent, so why wasn’t his completion rate a little higher. BTW, I believe the study by David Lewin that John mentions above looked at the final 2 years of the QB’s college career, and yes, three and four year starters typically saw improvement so that the final 2 years were better than their collegiate career in total. I think Stafford improved every year. Still, why under 60%???
by diehard82 on Mar 30, 2009 8:17 PM PDT up reply actions 0 recs
Bad o-line, perhaps.
Maybe not, just throwing that out there.
"Part, fools!
Put up your swords. You know not what you do."
by Fearless Frog on Mar 30, 2009 9:29 PM PDT up reply actions 0 recs
Stafford did lose a couple of O-linemen at the beginning of the season
If I recall correctly.
Weez the juice!!
by Carl Shinyama on Mar 30, 2009 9:38 PM PDT up reply actions 0 recs
That could explain it.
"Part, fools!
Put up your swords. You know not what you do."
by Fearless Frog on Mar 30, 2009 10:25 PM PDT up reply actions 0 recs
He completed 61.4% of his passes last year
I think something that doesn’t get looked at much is how many passes he threw each year. He went from 256 to 348 to 383, a signigicant climb from year to year. Each year his numbers improved while his load increased. His adjusted yards / attempt jumps each year from 4.8 to 6.5 to 8.5, his completion percentage improves each year and his sacks/attempt holds steady at ~4.4 his last two years (which he did despite rampant o-line injuries). It’s fair to say that if he stuck around for a senior year he would put up some great numbers.
by Nate Dogg on Mar 31, 2009 4:50 PM PDT up reply actions 0 recs
Yeah, the o-line injuries
and the amount of confidence his coaches had with him for audibles allays a lot of my concern on Stafford. From earlier stories, I was under the impression that he made poor decisions and was not very bright, but the opposite seems to be true. It’s hard to make smart decisions all the time when your line is very young and inexperienced, and his audibles and wonderlic test score indicates a smart and capable QB.
by LantermanC on Mar 31, 2009 5:03 PM PDT up reply actions 0 recs
It would be relevant to bring up Graham Harrell
If he met the criteria that he was expected to be drafted in the first round, IMO. But I do see the point that you’re trying to make.
Weez the juice!!
by Carl Shinyama on Mar 30, 2009 9:01 PM PDT up reply actions 0 recs
Comments For This Post Are Closed
Secondary Sidebar
FanPosts
Community blog posts and discussion.
Certain photos copyright © 2009 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
