Nah, Coop, you're not a jerk.
Redsland brought up essentially the same point.
Basically, you're right that we'd need more variables to answer the question better, and we'd most definitely need a larger sample size than just one season of stolen base data (preferably many many seasons). It's also possible that last season could have just been a fluke.
Here's the SB data in a bit more detail for 2005:
HCPA SB CS SB% PA/SB Attempt
64893 1118 507 68.8% 39.93
PCPA SB CS SB% PA/SB Attempt
57738 735 280 72.4% 56.88
Obviously there's quite a bit more SB attempts in hitting counts than there are in pitching counts, and the success rate drops in hitting counts. My guess as to why this happens is primarily A) more fastballs thrown in hitting counts, as you and RL have stated, B) prolific basestealers likely accounting for a greater percentage of attempts in pitching counts than they do in hitting counts (mainly because non-prolific basestealers will rarely run in a pitching count) and C) more failed hit-and-run scenarios in hitting counts.