Load factor in various classes

[Peter47]

I'm not sure that this is the right forum, but I can't think of anywhere better. An legacy carrier will generally offer various classes - up to four on long haul services. Getting the split of seats across classes must be difficult and results can be surprising, As an example, from Nov - Feb BA offered 620 premium seats a week to Seattle (140 F / 480 C). The full summer offering is 1,134 (196 F / 938 C) Does demand actually go up this much between seasons? Its interesting that BA 's 4-class 777s have 40 WT+ (W) seats against 30 or 36 in larger 744s. 789s have a lower proportion of premium seats. Presumably aircraft configuration and the the allocation of fleet types to routes can be refined to get close to the ideal mix of fleets. 4 class a/c substituting for 3 class a/c is on odd days is probably not optimal!
My question is:
We know that airlines can achieve very high load factors - typically 85% over the year in long haul but I have seen load factors as high as 95% on a the busiest direction during the peak month. How does this vary by class, route, time of year, day of the week, etc. I realise that there is no simple answer with award redemptions, upgrades, seat sales, etc but I would be interested if anyone in the know could give me some very rough examples.