I think there is an error in Bkool Pro which over calculates the power and speed. Or is speed which is slightly wrong and power is calculated from that- I think its the latter. I really don't know, just that everyone who moved away from the Bkool Pro and has used either a calibrated trainer with a built in PM or a separate PM has come down to earth with a bump, so to speak
Aside, I did a little number crunch with that calculator for hills. A 75kg rider needs to 240W to climb a 6% hill where I need 95Kg, 300W to climb at the same speed.
Firstly, absolutely agree that the Bkool over-states both watts AND speed (as I said earlier), and its made clear from those rides I did. Whilst I wouldn't dream of arguing with the physics of speed vs power, I think applying the watts reported by the Bkool to this kind of analysis is a bit pointless really as it would have to assume the watts were in some way accurate in the first place which is clearly not the case. Based on the inaccuracies we already know about, I don't think it's that much of a stretch to suggest that the relationship between watts and speed on the Bkool is also inaccurate, and I think that's what the data from my two rides suggests. A close look at the Zwift Power results still makes me definitely think the watts are even more over-stated than the speed. I won't bore you with a load more analysis (unless you really want me to!) but looking at these two sets of results I think it's clear that if I'd done 4.01w/kg on the Neo I'd have gone a fair bit faster than I did on the Bkool - https://www.zwiftpower.com/race.php?id=13370, https://www.zwiftpower.com/race.php?id=13326
As for the reason why the Bkool figures are wrong, I think the speed increase comes partly from inaccurate or slow application of resistance compared to the Neo. The Neo resistance is powerful, accurate, and fast to apply, and the Bkool is much less so. This means on the bkool you sometimes can maintain speed or accelerate quicker when the Neo's resistance would be holding you back.
As for the excessively high watts compared to speed, I do think this makes some sense when you think about it - and I think one of the reasons is the fact the Bkool reports power based on the speed of the rear wheel rather than how much actual power you're producing. This means it will record power quite high and for quite a while (until the wheel stops spinning) even when you back off and decelerate, whilst the Neo or other PM knows you've backed off and instantly reduces reported power. So on both trainers your speed will be decreasing (or increasing on a steep descent) but the Bkool is reporting a load of watts and the Neo zero watts. These are extra watts for no real speed gain and will build up over a 50 minute race.
The amount all of this affects things is obviously going to vary depending on loads of factors like the terrain, the weight of the rider, what kind of mood the Bkool trainer is in that day etc etc which is why it's impossible to put a specific figure on it that applies in all circumstances.