Geno Smith: A Tale in Accuracy
From the FanPosts -Joel
I’m not going to sit here and say that I can predict the future, because I can’t. But I’m pretty sure what the talk on AP will be about for roughly the next three months. There will be an endless discussion (which will end in 3 months) on what the Chiefs are going to do with the number 1 overall pick in the 2013 NFL draft. While perusing through the limitless comments on AP, one voice prevails through the masses. That voice says, "Draft Geno Smith."
The numbers are impressive. In his career, Mr. Smith threw for 11,662 yards, tallied up 98 touchdowns to only 21 interceptions, and completed 67% of the passes that left his hand. But, many believe those impressive stats are a result of the system he played in. Short screen passes and two explosive receivers that warrant draft talk of their own undoubtedly could inflate any QBs success.
After going through the AP scouting process (YouTube), many can form their own opinion of what they like and don’t like. Depending on the observer, things like arm strength, footwork, poise, accuracy, release, and pocket awareness can all be discerned from prospect to prospect. Opinions are formed and debates ensue on exactly what each prospect can offer.
It was my attempt, in this post, to stray from this type of analysis. I wanted something more objective. You can show 10 different people tape of Geno throwing the ball and you will get 10 different opinions. Some may be similar, but all will vary in one degree or another.
Specifically, I wanted to tackle the debate on Geno’s accuracy. As stated earlier in the post, Geno has a very impressive career completion percentage, one that he was able to increase every year as a starter. Also stated earlier, this percentage is said to be due, in large part, to the short and "easy" passes that he was asked to make.
To explore this idea, I observed 8 games from Geno’s 2012 campaign. Those games were Texas, Texas Tech, Maryland, Baylor, Kansas State, Oklahoma, Iowa State, and Kansas. I used the YouTube videos of Geno Smith vs. insert opponent here in my research. The goal of this research was relatively simple. One comment in particular sparked the idea when a user stated that, "90% of his passes were screen passes." Well AP, were they?
I present the following chart:
As I watched the games, I charted where each completion and incompletion were in regard to the line of scrimmage. It didn’t matter where the receiver ended up, all that mattered is where he caught it or attempted to catch it. After completing 8 games and 300 throws, I felt I had an adequate and representative sample size.
The double lines running down the middle of the chart can pretty much be thought of as the hash marks on a football field. With the exception, for example, that sometimes the ball can be located on the right hash pre-snap and a pass will be completed with the receiver going over the middle. The receiver might catch it outside of the hash marks, but for all intents and purposes it was a pass over the middle. I thought this better represented the types of throws that he was either missing or hitting.
The “NLOS” (Near Line of Scrimmage) is any pass that is generally within 3 yards of the LOS (i.e. screens, swing passes, and maybe a few shallow drag routes). The “4 to 9” section refers to passes that were completed/attempted within 4 to 9 yards of the line of scrimmage (usually outs, ins, and slants). The farther up the chart you go, the farther the passes were from the LOS. From there, I think most get the idea.
To give an example on how to read the chart, the bottom left section describes that Geno completed 41 of 46 passes (89.1%) that were directed to his left and were near the line of scrimmage. The far right column gives totals for each “level.”
Feel free to take all of the screens away from Geno’s game. He completed 131 out of 213 passes that were 4+ yards past the LOS for a completion percentage of 61.5%.
28% of his pass attempts were NLOS, 33% were from the 4-9 range, and 24% were from 10-20 yards past the LOS.
36% of his pass completions were NLOS (not quite 90%), 32% were from the 4-9 range, and 23% were from 10-20 yards past the LOS.
42% of his passes were directed to the left side of the field.
Geno completed 18 out of 43 passes (42%) that were 21+ yards past the LOS.
Geno is deadly accurate over the middle, check out that middle column. He had 6 of 9 passes that were 31+ yards down field completed, impressive. Geno’s completion percentages in the left, middle, and right were 70%, 72%, and 66% respectively.
Cause for concern? Completing 10/21 for a 48% completion percentage on passes that were 10 – 20 yards downfield and on the right side of the field.
Reason for optimism? Completing 19/28 for a 68% completion percentage on passes that were 10 – 20 yards downfield and on the left side of the field.
These last two notes were of particular importance to myself when doing this evaluation. NFL quarterbacks are separated by what they can do 10 to 20 yards downfield and outside the numbers. I’ll give you a math problem. How far does a ball travel from a quarterback that drops back 5 yards from the LOS, throws it halfway across a football field that is 53 yards wide, and the ball travels 15 yards past the LOS? My buddy, Pythagoras, told me that 15 yard completion is actually about a 33 yard pass. Is he right?
The overall results of the research seem promising. Initially, I defined 15 parts of the field. I had 5 levels of the defense separated by 3 sections. Through my evaluation, Geno was able to effectively attack each of these 15 parts. Not all equally, but he made the defense account for the whole field. Do the screens inflate his completion percentage? Yes, but those passes are hardly a given… cough*Matt Cassel to Eric Weddle*cough. The percentage of those he completes is beyond impressive by itself (88%). And those passes can travel around 20 yards in the air, at times. The only way those plays work are to get the ball there very quickly and very accurately, which Geno does nearly robotically.
I was a Geno Smith fan before I put this together, and this did nothing but confirm my thoughts on him. I have nothing to compare these numbers to, so I’m thinking about doing a similar evaluation on a couple of the other quarterback prospects. In the end, I hope people find this interesting and can find a way to use this information in their own evaluations. Go Chiefs.