*This is an on-going series of putting college QBs in our mathematical analysis. See this link for details on the College QB rating system --  Predicting the Unpredictable…Projecting a College QB to the NFL with a Mathematical Formula

We have beaten the importance of the QB Wonderlic testing scores and their correlation to NFL success, to death recently. As a re-refresh, the Wonderlic is a timed test in which you are given "X" time to finish "X" questions. For the NFL players at the NFL Combine they are given 12 minutes to complete 50 questions. The questions range from easy to a little complex on math and logic questions. It's not so much that the questions are hard, but it tests your ability to think quickly under pressure -- and many of the questions are noticing patterns in a hurry. It really is an interesting mini-trial on paper of what it is like to be a QB: recognizing defenses, patterns, and making split second anticipations. That may be overstating it some, but I think it is a great indicator for future success (or more to the point a red-flag of future NFL struggles).

Greg McElroy scored one of the all-time high scores for a QB (or for anyone), with a score of 43. Giving him national headlines for a few days, during a very slow football news cycle.

The reason the Wonderlic score is a great indicator is not because I think so. I know so...statistically. The recent history of certain levels of Wonderlic scores for QBs has been a great predictor of failure, not so much for success per say. Scores below a 24 on the test have been increasingly accurate at predicting NFL failure more recently. No college QB drafted in the past 10 years (2001+) has appeared in a Super Bowl in the past 10 years if they scored below a 25 on the Wonderlic. Donovan McNabb made it to the Super Bowl within the last decade, but took the test 12 years ago...outside of last decade. I have a theory that a decade plus ago, players could have cared less about that test...now it's everything. Players go to Wonderlic "coaches" and take the test several times now. A prospective NFL QB who scores low on the Wonderlic these days has a real red-flag going forward. 

**You may enjoy an article we did about the Carolina Panthers QB situation, that is really more about an extended analysis/debate of the Wonderlic test and it's relevance/correlation to NFL results/success. FF Draft Picks 2011 - Carolina Clausen - Fantasy Football 2011

Low Wonderlic scores have been amazingly predictive the last 5-10 years, but what about higher Wonderlic scores? I've heard a few people opine that a high Wonderlic score indicated that a QB may be "too smart", and that it might be tough for coaches to work with an "ultra-smart" QB, or it could hurt QB performance with a "paralysis of analysis", or simply over-thinking things too much. A Wonderlic score of 40+ is really world class smart. There is not enough data on the 40+ QB Wonderlic scorers to say whether "big smarts" is bad or good or detrimental. 40+ Wonderlic scoring QBs that we do know of (that actually played at all in the NFL, or now ones entering):

• 48 = Ryan Fitzpatrick

• 43 = Greg McElroy

• 42 = Blaine Gabbert (now a reported 31)

• 42 = Drew Henson

• 41 = Hugh Millen

• 40 = Alex Smith

OK, great...a whole section about Wonderlic scores and no closer to any idea of an analysis on whether Greg McElroy can translate to the NFL as a QB. Let's get back to football....

 

Good decision maker/highly accurate or playing it too "safe"?

There is the famous moment in the movie "Tin Cup", where Kevin Costner's character is a journeyman/no-name golfer who suddenly has a critical choice to make during the final day of the US Open. The Costner's character is in a 3-way tie for the tournament lead -- and faced with the option of...should he lay up on the 18th hole and play it safe and keep the tie likely, or should he not play it safe and "go for it" -- and go after the outright win? Keeping in mind the 3 rounds/days prior he has "gone for it" on the 18th and put it in the water every time. (The movie is a sports classic and 15 years old so you should know how it ends, if not - fast-forward!). Costner's character "goes for it", hitting it into the water multiple times, ultimately taking a 12 and not winning the tournament.

We know a QB who "goes for it", trying to defy odds and stick the ball into tight spaces/multiple coverage -- is not likely going to be consistently good in today's NFL. Nobody likes a high turnover QB. However, there may be an anti-"Tin Cup" mentality that "fakes" us into thinking we have a possible good/great QB on our hands. QBs that have a seemingly high efficiency, but do so on a much lower/limited passing attempts per game basis...they may be somewhat "tricking" us into thinking statistically that they are great/highly-efficient. We fans/NFL execs are all looking for the next Tom Brady. Tom Brady escaped everyone's radar screen way back when...and now we are (to me) somewhat over compensating looking for any signs of the new "messiah", a college QB who is a highly accurate Brady-like QB waiting to be discovered. There are a couple 2011 QBs that fit into this questionable position for me, and Greg McElroy is one of them.

On the surface, it all looks great for McElroy: a 70.9% Completion Percentage and a 20-TD to 5-INT ratio. That looks incredible. Before we think 70%+ Completion Percentage has rarely/never been done before and put it on too lofty a perch, we have to be careful not to overlook the ever increasing population of high Completion Percentage QBs coming from college in recent years. The 4-to-1 TD/INT ratio is great, but I have a bunch of QBs researched who have reached there and have not made it as a success in the NFL. I'm not saying McElroy somehow did a "bad thing" with these nice stats; just cautioning not to over-value it to far (relatively speaking).

(My pure opinion) There is such an emphasis on no turnovers and west-coast passing system variations for a QB -- being in range of 70% Completion Percentage used to be almost unheard 5-10+ years ago, now multiple QBs are at or approaching it yearly. The much "safer" college passing game is upon is. How can you find the balance of accurate/efficient QB, and yet gunslinger QB when necessary/appropriate?

Is there something to be said for comparing a QB that has a 70% Completion Percentage on just 20 throws a game, versus another who has 65% Completion Percentage on 40 throws a game? Are some QBs just "laying up" to pad stats,  or maybe "checking down" more to play it safe and not draw the wrath of the coaches, crowds, or media?

Which of the two hypothetical QBs (above) is the more accurate? The pure percentage view point would say the one with 70% completions, but if that same QB was relied upon to throw 35-40 times a game all the time -- could we rely upon the 70% pace to uphold? Does the 40 throw-a-game QB come from a team/system that defenses gear up for the pass (as obviously they pass a lot), thus a more difficult pass defensive strategy for a QB to face? Does the 20 throw-a-game QB give off a "tell" that the coach doesn't really "trust them" fully? Does the 20 throw-a-game QB come from a high/great run-game system, where defenses have to gear up more for the run -- and allows the QB (in the limited times that he throws) to face a more wide open/low DB-count pass defenses that are overcompensating for the run? Does all this call the great pass metrics of low pass attempt QBs into question?

Let's look at QBs we have studied who had very high Completion Percentage metrics in our system (analyzed for stronger opponents) in their final college season, and how many pass attempts per game they had against this stronger competition (again, not full season attempts per game -- just key game/stronger opponent pass attempts per game):

1. 41.3 pass attempts per game = Matt Schaub, Virginia

2. 39.4 = Chase Daniel, Missouri

3. 37.0 = Philip Rivers, NC State

4. 36.3 = Ben Roethlisberger, Miami, Ohio

5. 32.0 = Andrew Luck, Stanford

6. 30.0 = Alex Smith, Utah (was 26.4 for his overall season)

7. 26.6 = Vince Young, Texas (yes, Vince Young was pretty accurate in college...check out his stats if you don't believe me)

8. 25.9 = Greg McElroy, Alabama

9. 24.5 = Jason Campbell, Auburn

10. 23.0 = Tim Tebow, Florida

11. 21.7 = Dennis Dixon, Oregon

Most all of the future elite QBs we looked at pushed 30+ Pass Attempts per game their final season of college in our look at just stronger opponent matchups. There may be something to question on highly accurate + low pass attempts. We're not saying high pass attempts alone are awesome, if so Graham Harrell and Colt Brennan would be NFL elite. It's just that high pass attempts and high accuracy/Completion Percentage may be a better "tell" than just looking high Completion Percentage alone...and thus, it is also better than high Completion Percentage with lower (historically) pass attempts.

I don't penalize McElroy in our system for the following statements, just something I am pointing out -- McElroy had 7 games with 20+ passing attempts (6 games with 20 or less pass attempts), and his Completion Percentage in those 7 games dipped slightly to 68% which was off of his season overall of 71%; and down from his 75% in games where he had 20 or less pass attempts. In the 4 games that McElroy threw the ball 30+ times in a game he kept a 70% Completion percentage, but lost 3 of the 4 games. It's not that he became dreadful with more throws, it's just a point to note that he comes down just a little bit off the gaudy 70%+ Completion Percentage that is turning everyone's heads right now. It's an odd psychology, 70%+ Completion Percentage catches our attention for a season total, but a season total of 67-68% wouldn't really turn our head any more. That 70% mark overall number is possibly building up in our minds to mean maybe just a bit too much. I can't quantify that, it's just a theory.

I feel like I just spent a lot of time tearing down a great season and great metrics put up by Greg McElroy, as I read it back I feel bad. My intention is not to tear McElroy down, as much as raise the point that his lofty stats have been obtained by others -- and the QBs who did hit these highly accurate/efficient levels on lower pass attempts, typically didn't translate to the NFL as well as QBs who were accurate/efficient and had many more throws per game over the course of a season.

 

Physical Red-Flags...

If you do not buy fully into the potential issues with the low pass-attempts, you do have to acknowledge the glaring physical issues ahead for McElroy. He has measured under 6'2, with smaller hands and arms than most current day successful NFL QBs. Those limitations are not an absolute for future doom for a QB headed to the NFL, but they are not good odds in their favor historically either. QBs who match physically to McElroy are Troy Smith, Max Hall, Colt McCoy, Matt Flynn and Tony Romo -- so there is some hope seeing Romo in there. Drew Brees is smaller in height than McElroy, but has much larger hands. Again, this is not a total predictor...just historical odds not in his favor.

 

More than held his own in the SEC

What attracted me initially to Greg McElroy was looking at his performance in games against winning record SEC teams; and comparing that to other QBs from high win SEC teams Cam Newton, Ryan Mallett, and John Brantley. You will see that McElroy performed just as well as the elite SEC QBs on a "per" basis. I threw John Brantley in there to show that just being on a very good SEC team (Florida) and being the QB of said team, doesn't automatically mean you are a good QB/will have great stats.

You can take these numbers (below) a few different ways:

• Greg McElroy holds his own on performance with more highly thought of Newton and Mallett

• Newton and Mallett are not necessarily so much better of a passing QB than McElroy is, even though they are portrayed as better

• John Brantley has a lot of catching up to do...

 

Greg McElroy	Comp	Att	Yards	TD	INT	Cmp Pct		Cam Newton	Comp	Att	Yards	TD	INT	Cmp Pct
at Ark	18	26	194	1	2	69.2%		at Miss St	11	19	136	2	1	57.9%
Fla *injured 4th Q	11	17	84	0	0	64.7%		So Carl	16	21	158	2	0	76.2%
at So Carl	27	34	315	2	0	79.4%		Ark	10	14	140	1	0	71.4%
at LSU	21	34	223	2	1	61.8%		LSU	10	16	86	0	0	62.5%
Miss St	12	18	227	2	1	66.7%		at Ala	13	20	216	3	0	65.0%
Aub	27	37	377	2	0	73.0%		So Carl	17	28	335	4	0	60.7%
TOTAL	116	166	1420	9	4	69.9%		TOTAL	77	118	1071	12	1	65.3%
		27.7	12.2	18.4	41.5					19.7	13.9	9.8	118.0
		Pass Att per Game	Yds per Cmp	Pass Att per TD	Pass Att per INT					Pass Att per Game	Yds per Cmp	Pass Att per TD	Pass Att per INT
Ryan Mallett	Comp	Att	Yards	TD	INT	Cmp Pct
Ala	25	38	357	1	3	65.8%		John Brantley	Comp	Att	Yards	TD	INT	Cmp Pct
at Aub *injured 1/2 game	10	15	96	1	0	66.7%		at Ala	16	31	202	0	2	51.6%
at So Carl	21	30	303	1	1	70.0%		LSU	16	24	154	0	1	66.7%
at Miss St	17	26	305	3	1	65.4%		Miss St	24	39	210	0	1	61.5%
LSU	13	23	320	3	2	56.5%		So Carl	16	31	130	0	1	51.6%
TOTAL	86	132	1381	9	7	65.2%		TOTAL	72	125	696	0	5	57.6%
		29.3	16.1	14.7	18.9					31.3	9.7	#DIV/0!	25.0
		Pass Att per Game	Yds per Cmp	Pass Att per TD	Pass Att per INT					Pass Att per Game	Yds per Cmp	Pass Att per TD	Pass Att per INT

There is nothing like facing SEC defenses for multiple seasons. Better talent, higher paid coordinators, higher profile coaches, bigger crowds and scrutiny. In our analysis there are a few SEC QBs who our metrics pegged as weaker than the public may have first realized -- JaMarcus Russell, Brody Croyle, Matt Flynn to name a few. Greg McElroy doesn't appear to fall in that box, as the competition got tougher -- he held his own with the big name QBs of the SEC.

 

The current NFL QB Greg McElroy most compares to...

There was not as good of a "match" in our system as there usually is (covering statistical and physical). In our system, I would say a hybrid of John Beck and Alex Smith are the most likely candidates. John Beck shows more of a physical/performance match for McElroy -- Beck just wasn't as highly accurate. Alex Smith is more a match on accuracy and pass attempts per game, but Smith is much taller/bigger physically.

A clip/glimpse at just a few of our system's comparison data on these 3 QBs:

• "Adj" means just key games/better competition -- weighted for strength of opponent (our own proprietary work, weighted for strength of opponent)

• "per 35 att" numbers are the key games, weighted for strength of opponent and then translated into an average as if every QB had an equal 35 Pass Attempts per game all the time, and thus what would each QB produce (in college) if they had 35 passes per game based on the key games their final college season. In an attempt to somewhat equalize the college performance and show you what our systems sees.

QB	Yr	College	H	W	adj Comp Pct	Adj Yds per Comp	adj Pass per TD	adj Pass Per INT	x	Yds per game 35 Att	TDs per game 35 Att	INTs per game 35 Att
McElroy, Greg	2010	Alabama	73.8	220	69.0%	12.7	17.4	48.7		305.6	2.0	0.7
Beck, John	2006	BYU	74.1	215	63.8%	12.9	16.1	72.6		287.3	2.2	0.5
Smith, Alex	2004	Utah	76.0	217	66.9%	13.1	11.5	81.4		307.1	3.0	0.4

 

Greg McElroy Overall Score = 0.760  *See original work and scoring tables from the following link -- Predicting the Unpredictable…Projecting a College QB to the NFL with a Mathematical Formula

McElroy is a sleeper for many, again I think that is somewhat pushed by an overvaluation of his high Completion Percentage numbers. I am a fan of McElroy from what I have watched of him in 2009-2010. However, I call into question the effect of the low pass attempts per game, the Alabama run game that teams had to over-compensate for, and his physical size issues -- in trying to project McElroy as a hidden Tom Brady/Chad Pennington like QB for the NFL. McElroy's score in our system analysis (just based on what we can measure), sets him just outside of a potential good/elite NFL QB -- and has him more in the mediocre/good camp with some "bust" possibilities.

I think McElroy will be drafted in the 3rd-4th Round of the 2011 NFL Draft. I would be a buyer on McElroy in the 4-5th Round on a team with an established QB, so I could have a cheap, no-risk look at McElroy...but I would not plan to take McElroy early on the basis that he might be my potential franchise QB. However, I would want McElroy ahead of Cam Newton, Blaine Gabbert or Jake Locker for sure; and probably ahead of Ryan Mallett if I got to interview both of them personally. I still would take Ricky Stanzi, Andy Dalton and possibly Christian Ponder (and possibly Pat Devlin) ahead of him....those QBs all rate higher than McElroy in our system analysis (and they also scored well on their Wonderlic tests too...).

I'm intrigued by many of the things I see on McElroy, but I don't want to pay a lot/use a higher draft pick for the look-see.

 

 

Select a position from the tabs below to see stats and scouting information for that respective position.

• 
• 
• 
• 
• 
• 
• 
•