Theoretical before rake and after rake win rate

So this question is not exclusively an online question, but I figure more people here would be familiar with how to think about this question in theory.

I am a live pro and was having this conversation with another live pro. We were trying to figure out the profitability of a highly raked game, in this case 10/25 with 10% rake up the $75. In theory the game would ideally play deeper and always have the $50 straddle on. I believe there is no rake preflop.

My friend's contention is that to figure out your win rate in this game, one would estimate what the average rake per hour (could also be per 100 hands) at this table and divide that by the number of players to determine the average rake played. Then they would subtract this from the win rate before rake to figure out what your average win rate would be. Note, the actual alternative games are time raked, although for the most part if these private games are running, no similar stakes are going to be available in a time raked setting.

While I think that is sort of on the right track, I felt like there are some problems with that thinking. For one, we will probably participate in less hands than the average for the table since the average of the table driven up by the higher VPIP recs.

Basically what he was saying is he thinks that his win rate unraked would probably be around $200/hr for that stake and the average rake for the table would be $110/hr, so his after rake win rate would be only $90/hr, while his win rate on games that are on average a little smaller would be more like $115/hr. Without questioning the $110/hr average rake paid per player, is my friend right? Or would the affect of rake paid by him be smaller than the average rake paid?

Can anyone point to some empirical evidence they have on before rake win rates compared to after rake win rates?