The 8.2k/2W resistor is not across the cap, it's in series with the cap.
It's the second half of the RCRC filter and voltage dropping network. The first half is contained on the power amp board and since it has higher current through it (feeding node C and downstream), the Pd is 15W. The second half is part of the preamp, it's the same value but rated at 2 watts feeding node B.
The 8.2k/2W resistor has a drop of about 180V across it (DEPENDING ON THE SCHEMATIC), therefore, the Pd in circuit may be about 4W, BUT the voltages may not be accurate since experience tells us that this part lasts a long time in most amps. What is the actual voltage drop across the resistor on a normally functioning amp steady state? Wasnex's example shows different voltages and thus different Pd. Many older wire wound power resistors have a rating at a higher ambient temperature than they are used at, therefore at a lower Tamb with some derating allowance 2.1W may be ok (in this case).
It's critical to understand how the voltages are specified, especially since they feed cathode followers whose current is very much proportional to AC signal. It's easy for there to be more to the story than what we are seeing in a static environment.
Right.
Acutally, the 79 SVT has original 8.2k / 5W (!) resistor there - checked now, so I think they upgraded it over time. I just ordered 8.2k/15W ,military spec-ones and will put them in the R47/R49 places in both amps. I'd take measurements then and will be back here. Thank you all for the input!