Thanks, Aaron - good data. The tubes seem to be getting 4.84-4.86 volts which is 3% low, well within the usual +/-10% and the more strict +/-5% that premium tubes often specify.
EML specifies +/-4%. For a quick estimate, let's assume your Mills resistors are exact and that your meter is also exact. Then averaging the two tubes, your actual current would be 5.00v/4 ohms, or 1.25 amps. If you increased to 1.3 amps, the resistor drop would be 5.2 volts, leaving 9.85-5.2=4.65 volts which is 7% low - below the EML spec.
As you can calculate, a 56 ohm resistor in parallel with the two 8-ohm resistors would, in this case, give 5.00 volts on the 300B filament. I am not recommending this, at least not yet, and I won't be making any recommendations until we have a greater variety of data on the variability from one build to another. But you can see where I'm headed. I expect this will prove to be better than adjusting the regulated voltage, which is a delicate balance between the available voltage and the heat wasted in the regulator.