Alex Smith getting some statistical love

A couple days ago, hudd07 posted a FanShot to an article written by K.C. Joyner over at ESPN.com discussing Alex Smith. Unfortunately it's Insider-protected, so most folks can't read it. I can't copy and paste the whole article (copyright issues), but I thought I'd pull out some parts of it because it goes a lot farther in one direction than I think any article I've read recently.

The general gist of the article is that Alex Smith might already be better than a lot of people think, and also might be close to making that jump to becoming a very solid quarterback. Sensationalism is obviously good for page views, so one has to take the closing line of the article with a grain of salt for now:

"The numbers say it would behoove the 49ers to do all they can to get him to that mark as soon as possible. If they do, they might end up with one of the best quarterbacks in the NFL."

I think Smith can become a solid QB in this league, but we'll see if what he can do beyond that. The number that is mentioned in that first sentence is 10,000. The article discusses some of the concepts Malcolm Gladwell has discussed, including the idea that true mastery of a field does not occur until one puts in 10,000 hours of work within it. He certainly hasn't put in 10,000 hours in the current offense, but I think this concept goes more towards the idea that it just takes a lot of time doing something to improve.

But, after the jump, we'll take a look at some of the stats Joyner points to as evidence of Smith's improved performance.

Before going into Joyner's stats, let's be clear that sample size issues do come into play. So, while this first group of stats is impressive, always remember the idea of sample size.

QB Rating has plenty of detractors, so we'll take this first stat with a big grain of salt. Last year, Smith was able to post a rating of at least 88 in 6 of his 11 games last season. According to Joyner, only 16 of the 32 qualifying quarterbacks (14 passes per game is the minimum qualifier) were better. And of those QBs ahead of him, all played in 14 or more games.

Joyner also found strong performance in Smith's deep passes (20-29 yards downfield). His yards per attempt was 13.4 on those passes, and ranked 11th in the NFL, which put him just behind Donovan McNabb and just ahead of Peyton Manning. In looking at passes 30 or more yards down the field, Smith's 25.3 yards per attempt was tops in the league. Of course, as mentioned, this was based on only 11 such passes. Given that it's yards per attempt, I would assume this means passes that were in the air for 30+ yards, but I'm not completely sure.

One of the biggest weaknesses for Smith was on short passes (I'm guessing under 10 yards), where his 5.4 yards per attempt was 31st in the league. Apparently the median is 6.4 yards so, clearly it's a tight bunch in that area. Here's where it gets interesting and provides Joyner with optimism:

This is notable because it wasn't a lack of accuracy or even completions that stunted Smith's short pass YPA. His 75.7 percent short pass success rate (a metric that divides completions and defensive penalty pass plays into short pass attempts) ranked 16th in the league, or right in the aforementioned middle.

Joyner thinks Crabtree's struggles on short passes (4.3 ypa on such plays) is a major reason for Smith's own struggles, and argues that we could reasonably expect a big jump in this category in year 2 for Crabtree.

I wonder how much an improved offensive line would help improve that short category (if not the longer ones as well)? How many times has Smith had to roll out and throw the ball away because of the pass rush coming at him? He struggled to make his reads, let alone check down underneath. Combine that with an improved season from Michael Crabtree and maybe we'll see improvement in those numbers.

As always, certain statistics can be used to prove virtually any argument, so we do have to take this with a grain of salt. But for fans of Smith, it validates some of the excitement, and for non-fans, it at least provides something to consider.