This morning I was thinking more about the relative weather issue on these two widely-spaced dyno runs, when I remembered that there's online almanac data. I looked up the pressure and humidity in Houston for the first dyno date and yesterday, combined that with the temperature data from the dyno's logs, and plugged the full set of values into the SAE calculator
HERE, and it came up with a dyno correction factor of 0.995 for my old run and 1.050 for the new run.
Assuming the almanac data is decent information source for these purposes, those correction factors put the numbers at SAE-corrected 298.7 and 294, respectively, rather than 300.2 and 280. So really I'm betting most of the difference was the weather. 4.7hp could be anything, likely the dirty filters
