Football statistics do a horrible job of telling the whole story of a player's productiveness. We can measure the end production such as yard gained and scores produced, which in the end is what it all comes down to, but each player on the field is so connected to each other that one player may be playing excellent yet his stats still suffer, or vice versa.
No where is this problem more apparent than at quarterback. And the unique problem with judging QB that unless you have access to game film there is no way at judging a QBs overall performance on any given play. Say a QB completes a ten yard pass on a beautiful throw into coverage. Watching TV it is going to look like that QB did a tremendous job, but if there is a more open receiver further down the field then in reality he failed on that play. Scenarios similar to this are infinite for QBs, and it starts before the ball is even snapped. Did the QB make the proper pre-snap read? Jump over for more....
All the things that a QB can do that will never have a direct result, and fans would have no way of thoroughly evaluating, coaches look at all the time, and the Raiders just got a new system of evaluation. In this article by Seth Wickersham on ESPN (it's an insider piece) we get a look at Al Saunders grading system. He calls it SHAPE:
• Site: Did the QB recognize the pre-snap defense?
• Hot: Did he direct the "hot" receiver where to go?
• Alert: Did he correctly alter the play at the line of scrimmage?
• Progression: Did he correctly read the defense and locate his options?
• Execution: How were the mechanics on his drop and throw? What was the result?
The article goes on to give us some insight into how Saunders grades his QBs on this scale:
Generally, Saunders expects his starting QB to be close to 100 percent, assignment-wise, and to receive A's on efficiency. The genius of the Coryell offense, from which Saunders, Joe Gibbs, Mike Martz and others descend, is that, in theory, it has an open receiver on every play. Many of today's offenses -- those of the Colts and Pats, especially -- place a lot of emphasis on "conversion routes," patterns that are adjusted on the fly depending on the coverage. You need a great QB to direct those schemes.
He then explains how he uses it to raise the level of play of his QBs:
...he groups the stats according to different facets of the game. For instance, if he wants to judge his QB's effectiveness on third down against seven-man pressures, he can pull every grade from every play under those circumstances. Once there, he can view stats such as completion percentage and yards per attempt in the proprietary context of the play's intent. If his QB's SHAPE stats are subpar -- 80 percent, C -- Saunders will spend more practice time on those situations.
"That's how you use statistics to help QBs play better," he said. While it may sound complicated, it's just another way to stay ahead of the competition in an increasingly stat-driven—and pass-happy—league.
It is reassuring to me that such a proven offensive mind will be helping Jason Campbell improve in these areas. Campbell made great strides as last year went on in these areas, but that is not saying much. As much as you could tell by the TV broadcast he was an absolute disaster to start the year. I am sure it is what drove Hue Jackson and Tom Cable to start Gradkowski over Campbell. He was just not running the offense the way it was designed to be run.
Now Campbell will be moving into his second year with the same play-caller for the first time in his professional career, and he will have one of the most experienced and successful offensive minds helping him learn how to run that offense efficiently. These are all good things—let's get this lockout lifted and this season started.