I just addressed this concern. Any analysis of damage per attack has to establish chance to hit, and as this is highly varied from game to game, any such analysis is inherently limited or flawed. When you attack over 70 times in a row with GWM and never miss due to advantage, AC of enemies, ally bonuses, etc..., that is a free 10 damage per strike, essentially. When you miss by 5 or less on 25% of your attacks, there is a huge impact on your expected average damage.
And being honest, I did one of the most meaningful studies on GWM and found that it is SIGNIFICANTLY overrated, even for powerful PCs. Why? It really only matters if using the feat reduces the amount of resources a party uses up in a battle. Sometimes getting the benefit of the feat will not reduce the number of attacks required to fell an enemy. Even if it does delay it, the enemy might still fall before the PCs lose more hps or use more abilities that are limited. If the PCs use at will abilities, the fact that the monster may get an extra turn may not matter if the monster misses on an attack or a PC makes a save. Even when the DPR is increased by 6 per strike, that may not be a meaningful change.