By the Numbers: Comparing Elite NFL Quarterbacks to 2012 Draft Prospects

Using statistics to project athletes from college to the pros would make a scout's job much easier. Over the years, writers like John Hollinger have come up with formulas to take college basketball players and predict what kind of NBA players they might make.

It's an inexact science to be sure and there is no silver bullet, or scouts would already be using it.

Trying to the do the same with NFL players proves much more difficult because the college and NFL systems a player is in can make a significantly greater difference in football than it does in basketball.

I broke down the top 10 quarterbacks in the NFL this season by different statistical categories. Often, writers will want to use historical context, but given the constant flux of the pro game, using the most up to date information we can provides us the most relevant information.

Given the inherent flaws of gross statistics and even some efficiency statistics, I used QB Rating to pick a "Top 10" for comparison purposes.The following are the Top 10 QB's career college numbers.

Comp Att Yards TD INT Comp % TD:Int YPA Aaron Rodgers  424 656 5,469 43 13 64.70% 3.3:1 8.33 Drew Brees 1026 1628 11,792 90 44 63.00% 2.0:1 7.24 Tom Brady 443 711 5,351 35 19 62.31% 1.8:1 7.52 Tony Romo 560 892 7,816 82 34 62.78% 2.4:1 8.76 Matthew Stafford 564 987 7,731 51 33 57.14% 1.54:1 7.83 Matt Schaub 716 1069 7,502 56 26 66.97% 1.69:1 7.02 Eli Manning 829 1363 10,119 81 34 60.82% 2.38:1 7.42 Matt Ryan 807 1347 9,313 56 37 59.91% 1.51:1 6.91 Alex Smith 389 587 5,203 47 8 66.27% 5.86:1 8.86 Ben Roethlisberger 854 1304 10,829 84 34 65.49% 2.47:1 8.30 Averages 661.2 1054 8,112.50 62.5 28.2 62.68% 2.21:1 7.69

A brief look at the numbers and you'll see Alex Smith is actually the most impressive statistical player on this list. Tom Brady is particularly underwhelming and Ben Roethlisberger is perhaps the most accomplished overall. One of the important lessons of Alex Smith appears to be that without a sufficient sample size, we tend to lose the reliability of these numbers. Smith was wonderfully efficient against mid-major talent and threw by far the fewest passes on this list as the quarterback at Utah. Smith's success this year, however, illustrates my first point about the necessity to put a player in the proper system to succeed. The statistical proficiency of the quarterbacks in college systems vary, but completion percentage and yards per attempt seem like translatable talents as does the ability to score versus the propensity to turn the ball over (TD:INT). In other words, quarterbacks who can't complete high percentages in college, won't change much in the pros. Furthermore, dink and dunk college quarterbacks will be the same game managers in the NFL. Comp Att Yards TD INT Comp % TD:Int YPA Andrew Luck 686 1033 9083 80 21 66.41% 3.81:1 8.79 Robert Griffin III 776 1159 10,071 77 17 66.95% 4.53:1 8.69 Landry Jones 1005 1603 12,218 92 40 62.79% 2.3:1 7.62 Ryan Tannehill 452 725 5,053 41 20 62.34% 2.05:1 6.97 Nick Foles 938 1403 10,068 67 33 66.86% 2.03:1 7.17 Brandon Weeden 737 1060 8,861 72 26 69.53% 2.77:1 8.36 Kirk Cousins 696 1078 8,831 65 27 64.56% 2.41:1 8.19 Ryan Lindley 933 1683 12,277 87 47 55.44% 1.85:1 7.29 Kellen Moore 1131 1624 14,374 140 26 69.64% 5.38:1 8.85 Russell Wilson 888 1464 11,424 107 20 60.66% 5.35:1 7.8

These are the top 10 prospects entering the 2012 NFL Draft at this point. Landry Jones says he hasn't made a decision, but we'll include him for perspective purposes. Right away Andrew Luck and Robbert Griffin III jump off the page. Their efficiency and overall production are well above our composite average from above.Does that mean they'll necessarily be top-level NFL quarterbacks? Of course not. They need the right coach, right system and right players around them. On the other hand, the reliable indicators like completion percentage and YPA appear to be part Luck and RG3's already impressive resume.

Using statistics to reenforce what we see with our eyes is the best way to use statistics when it comes to their predictive powers. If we thought Andrew Luck and Robert Griffin would be excellent NFL quarterbacks and what we found was startlingly below average, it might give us pause.

Looking at the list above, Ryan Tannehill and Nick Foles both seem overrated based on their rankings. The two are big, strong quarterbacks who play on mediocre teams. Neither has a pedigree as big-time winners, nor have they had marquee games to prove their abilities.

That being said, both are well below average when it comes to production, particularly in TD:Int ratio and YPA.

Tannehill is only a two-year starter while Foles has more experience (making his lack of success even more frustrating).

Basically, they are what they are at this point.

Based on the numbers above, he doesn't belong above Brandon Weeden or even Kirk Cousins when it comes to translatable NFL skills like accuracy and driving the ball down the field.

Speaking of Weeden, the Oklahoma State prospect does appear underranked based on these numbers. An interesting doppelganger emerges when you look at these numbers side by side.

	Comp	Att	Yards	TD	INT	Comp %	TD:Int	YPA
Ben Roethlisberger	854	1304	10,829	84	34	65.49%	2.47:1	8.3
Brandon Weeden	737	1060	8,861	72	26	69.53%	2.77:1	8.36

If you take into account the talent-level difference between the Big 12 and the MAC, the two are very similar.

Both are tall, lanky quarterbacks who slide step and move in the pocket to create plays. Weeden, at 28 years-old, doesn't have a ton of upside, but has the experience and maturity that could follow a similar path to Big Ben's with Pittsburgh as he lead them to the playoffs as a rookie.

If I needed a quarterback for the next 10 years, perhaps Tannehill makes more sense because of age and upside, but if I'm an 8-8 team like the Seahawks or a 7-9 team like the Chiefs who may just be a quarterback away from the playoffs next year, I would like Weeden's chances to play right away.

When it comes to interesting similarities though, there is a somewhat shocking pair.

	Comp	Att	Yards	TD	INT	Comp %	TD:Int	YPA
Drew Brees	1026	1628	11,792	90	44	63.00%	2.0:1	7.24
Landry Jones	1005	1603	12,218	92	40	62.79%	2.3:1	7.62

Jones, of course, has put up these numbers as a standing junior. Brees was a senior when he left for the NFL with these Big-10 leading stats.

Oklahoma's signal-caller is more physically gifted, with a bigger frame and stronger arm than Brees. Jones isn't as pinpoint accurate as Brees, but even Brees in San Diego wasn't as accurate or efficient as Brees in New Orleans.

Oklahoma's recent list of NFL quarterbacks isn't exactly a who's who of All-Pro's, but Jones was basically unstoppable with Ryan Broyles before Broyles tore up his knee. Jones is certainly better as a prospect than Blaine Gabbert and there was some discussion he'd the #1 pick last year.

Jones, as a thrower, has as much natural ability as any quarterback in this class and while Griffin put up most of his numbers for middling teams (until this year), Jones has been playing in big games all three years in Norman.

Some uneven play this season has dropped Jones' stock, but he has considered a top 10 pick before the season began.

If we re-ordred the top 10 list based on these numbers it would look more like this:

1.) Andrew Luck
2.) Robert Griffin III
3.) Landry Jones
4.) Brandon Weeden
5.) Kirk Cousins
6.) Nick Foles
7.) Ryan Tannehill
8.) Kellen Moore
9.) Russell Wilson
10.) Ryan Lindley

Lindley couldn't break 60% completing passes in the Mountain West Conference, dropping him down the board and both Moore and Wilson just aren't big enough to rank any higher.

It is interesting to note though, both would be a few spots higher if they were a few inches taller, particularly Moore who has bar far the most impressive stats on the list.

I'm not a fan of expecting dramatic improvement out of players (e.g. Tannehill has to actually show me he can do it before I'll buy that he can), although, we do see guys like Tom Brady and Matthew Stafford outperform their college careers.

College numbers don't tell the whole story, but they do expose important flaws to potential teams.