Based on a quick online search, it looks like the meter (not sure if the video I saw is version 1 or 2, it had a small label for the chart of good cap values) alternates applying a positive and negative current on the meter. It determines the ESR by looking at the voltage step as it switches polarity. The test signal is on the order of many kilohertz. However, the frequency is variable, and inversely proportional to the capacitance.
My intuition is that if the measurement period is much shorter than the RC time constant of your capacitor in parallel with the bleeder resistor, the ESR will be accurate. I'd suggest playing with a spice simulation to confirm.