What's wrong with Robertson?

After getting rocked his last time out, I decided to download the Pitch f/x data for two Nate Robertson starts. The first being his gem against the Mariners on July 6, while the second was his abortion Wednesday night. I really wanted to download some 2006 data to look at successful Robertson, but that's not available. Pitch identification wasn't done until this year, so looking at some data from last year are to no avail. Of the pitches tracked, here is some of the noticeable data I grabbed.

# of times Thrown, July 6:
Fastballs: 53 (53.5%)
Changeups: 20 (20.2%)
Sliders: 17 (17.1%)
Curveballs: 2 (2.0%)

# of time Thrown, July 30:
Fastballs: 38 (56.7%)
Changeups: 9 (13.4%)
Sliders: 6 (9.0%)
Curveballs: 14 (20.9%)

If we assume the pitch classification is correct, the first thing I notice is that Robertson threw just two curves in his start on July 6, while throwing the pitch ten times as frequently on Wednesday. Not sure what to make of that, but shall we take a look at some of the data from his breaking balls?

His average slider on July 6 moved1.50 inches horizontally, while on July 30 it moved just 0.6 inches. Meanwhile, his two curveballs moved an average of 5.5 inches horizontally compared to just 1.36 inches on Wednesday. I have a hunch that some of his sliders were being counted as curveballs Wednesday.

Regardless, each of his pitches showed significantly less horizontal and vertical movement in his start on Sunday.

I will admit to having very little expertise when it comes to reading pitchfx data, and it's a nice day outside today. I'll try to come to some conclusions later. If anybody wants my downloaded data, comment in this tread.