Matt Cassel Throw by Throw
By William T. Cloake IV
November 23, 2009
WILDBILLSCHIEFS.COM
In spite of the Chiefs exciting win over the Steelers this week and a two-game winning streak, Matt Cassel continues to have his detractors, those who wonder if the Chiefs did the right thing by sending a 2nd round draft pick to New England for a quarterback with one good season under his belt, while ultimately sending away the popular Tyler Thigpen.
On of the biggest criticisms of Cassel and rightfully so has been his accuracy. But how fair a criticism is this? Too often we only remember the one or two awful throws or the one or two great ones. Psychology (specifically cognitive behavioral theory) tells us this is natural. People have beliefs and generally, we unconsciously seek information that upholds these beliefs. Hence, if you think that Cassel is an accurate quarterback you remember the good throws. If you think he is inaccurate then you remember the bad ones. But what about the "neutral" throws? This is where the lens of our beliefs really comes into play. The tendency is that we will construe those neutral throws to be "nothing good" if we think Cassel is inaccurate, and will see them as "solid throws" if we think he is accurate. So how do we break these beliefs? (known, incidentally as schemata) There are two ways. First option is that the "evidence" becomes so overwhelmingly against our belief that we have to change. Often times this never occurs. The second way is to scrutinize the "evidence" in such a way that we cannot blindly continue our pattern of thinking. This process will either confirm we were right, wrong or often somewhere in between.
To this end, I decided to take a look an objective look at Cassel's accuracy in the Steelers game, I broke down the tape pass play by pass play. On each play, I gave Cassel a rating on his throw including perfect, good, fair, poor and terrible.
Now, as you can imagine such a rating could be a subjective thing, so here are the criteria that I used in order to make the determination. My emphasis was on creating a system that would be objectifiable, so that it might be replicated and hence remove personal bias from the equation. Below, you can see my criteria and point value for each type of throw.
Perfect (4 points): I determined a perfect pass to be one that hits the receiver exactly where the receiver was expecting it, does not break the receivers stride and/or is thrown around the defense. Sometimes a perfect pass may not mean an easy catch (Ben Roethlisberger's game winning throw to Santonio Holmes in the last Superbowl, for example). In looking at a perfect throw, I often asked myself; "is there a better place this ball could have been thrown?"
Good (3 points): I determined a good throw to be one that hit the receiver in his "box" meaning between his shoulders, no more than a foot above the helmet or a below the knees, without breaking his stride substantially (in a situation where the defense does not make a throw outside these parameters the best one).
Fair (2 points): A catchable ball that falls outside of the receivers box but is caught or should have been caught. This is where the announcers will say things like, "A touch catch but one you expect a player in the NFL to make".
Poor (1 point): A ball that is well outside of the box that the receiver can make a play on that leads to either a spectacular catch or a miss. In particular, I note here that the defenders to factor into the equation. A pass that might be perfect against "air" (no defense) can be poor or even terrible if there is a defender right there. To some extent, this criteria ties accuracy to decision making, but I saw no reasonable way to parse these two factors completely out.
Terrible (0 points): A pass that is nowhere near the receiver that he was not going to have any chance to catch.
To be fair, passes that were deflected at the line of scrimmage, passes where Cassel was hit while throwing (that greatly impacted the throw) or passes that were clearly thrown away were not scored. Additionally, I also scored throws on plays cancelled by penalty, since most penalty's don't effect the throw and those that do would be thrown out by the rules above, anyway.
Here are Cassel's scores pass by pass. On each throw, I provide the intended receiver, type of pass and my score, with any relevant comments and points in parenthesis. I conclude by giving final averages and a brief interpretation of scores.
1) Slip screen to Mike Cox for 19 yards. Perfect. Gain was largely a result of Cox not needing to break stride (4 points).
2) Flat pass to Jamaal Charles, incomplete. Good. Hit Charles on his left side and he dropped it because he was looking upfield before bringing the ball in (3).
3) Out pattern to Bobby Wade, incomplete. Terrible. In Cassel's defense, there was a small window and I am not sure he wasn't just throwing it away, but it looked like a try to me (0).
4) Curl pass to Wade for 9 yards. Good. Pass right at the top of Wade's helmet (3).
5) Swing to Charles for (-2) yards. This was between good and fair, but I scored Cassel as fair since his throw slowed Charles down. Also, technically this play was a run but I scored it as pass since Cassel threw the ball overhand to Charles (2).
6) Batted ball at LOS no score.
7) Screen to Charles for 6 yards. Perfect (4).
8) Cassel hit while throwing and ball fluttered out no score.
9) Wide receiver screen to Bobby Wade for one yard. Perfect throw defense jumped the "automatic throw" if the ball had not been perfect, would have been a pick-6 (4).
10) Cassel under pressure, ball thrown away, no score.
11) Curl pattern to Bobby Wade. Poor throw at his ankles (1).
12) Curl pattern to Mark Bradley. Terrible overthrow. May have gotten tipped at the line, but I couldn't tell for sure so I scored it (0).
13) Out Pattern to Chris Chambers. Good (3).
14) Post pattern to Mark Bradley. Good throw, but a case where a perfect throw leads to more yards. Ball hit Bradley in the gut, but could have been out in front so he would have run to it and gotten some YAC (4).
15) Flat pass to Jamaal Charles. Poor throw. Cassel may have been throwing this one away, but it wasn't obvious, so I scored it (1).
16) Corner pattern to Leonard Pope for 21 yards and a touchdown. Good throw. A perfect throw would have been about a foot to the left (4).
17) Post pattern to Chris Chambers. Poor. The throw itself would have been good against "air", however, the defender who made the play on the ball made it impossible for Chambers to do more than get a hand on the ball (1).
18) Curl pattern to Bobby Wade for 7 yards. Perfect throw if you watch you can see Wade's hand coming up out of the cut to exactly where the ball arrives (4).
19) Short curl to Pope, incomplete. fair throw, this was a check down route and Pope had little separation from James Farrior and the defender got a hand in, even though it was in Pope's box. A good throw would have been to high to Pope's left, away from the defender (2).
20) Flat pass to Mike Cox, incomplete. Even thought this was incomplete it was a good throw. Cox had little separation and Cassel put the ball in a place where only Cox could make a play, anything further left gets intercepted (3).
21) Wide receiver screen to Chambers for 1 yard. Good throw (3).
22) Seam route to Lance Long for 30 yards. Perfect pass, which drops in over the defense and doesn't break Long's stride at all (4).
23) Post and Go to Chambers for 47 yards. This is a good throw. However, a perfect throw leads to a touchdown. This ball covers nearly 60 yards in the air (including the angle) and you wonder if Cassel was coming to the end of his range (3).
24) An out pattern to Mark Bradley in the endzone, incomplete. Terrible. Another one where I think Cassel may have been throwing it away, but it wasn't clear so I scored it (0).
25) Slip screen to Charles for 2 yards and a touchdown. Perfect (4).
26) Sideline route to Charles, incomplete. Perfect. Although Charles went out of bounds and came back in, this is a tremendous throw, just outside of the defender but into Charles' chest (4).
27) Out pass to Long, complete for 7 yards. Perfect (4).
28) Out pass to Mark Bradley, incomplete. Good throw dropped by Bradley (who has been dropping a lot of balls lately) (3).
29) Middle screen to Charles for 13 yards. Perfect throw (play called back on penalty) (4).
30) Curl to Pope for 5 yards. Perfect (4).
31) Post pattern to Chambers, incomplete. Poor throw. Chambers mis-times his jump or he maybe makes a play on this, however, the throw was poor, either way because it should have been much further to the inside (1).
32) Crossing pattern to Chambers for 61 yards. Perfect. On 2nd glance, this is a brilliant throw. The ball comes out in front of Chambers so as not to break his stride but is seemingly a little high. However, the height actually is to get the ball past the outstretch arm of Lawerence Timmons, who was recovering on the route. A lower pass is deflected (4)
When all is said and done, Cassel had 29 score-able throws good for a total of 81 points for an average of score of 2.8 per throw, which is close to a "good" rating of 3.0. In grade terms, I would have to give Cassel a grade for accuracy in somewhere between the C+ to B- grade. This seems about right for the way he played on Sunday.
In addition, Cassel had three "terrible" throws, good for 10.3%. This is the biggest problem in my opinion as this seems way to high. However, on the upside, Cassel did have 20 throws in the good or perfect categories good for 69%, which seems like a solid number. Overall, it seems that Cassel's accuracy was at least adequate in the Steelers game, although there is still plenty of room for improvement. Frankly, given that this is the first time I have used this system, it is a little difficult to completely evaluate Cassel without any referential data. However, my plan is to use this system again in the future so it can be an insightful tool in evaluating the Chiefs quarterback play.
Digg It
del.icio.us
Facebook
StumbleUpon
MySpace
Reddit
Twitter
Yahoo! Buzz