Just because we don't know how some edge cases are handled doesn't mean we don't understand the general idea of the number. When you see a player at 100%, as was the example above, you know that it means this player either wins all their 2 player games, or wins half their 4 player games, or some other combination of those that averages to 100%.
It gives you an idea of how frequently they win while adjusting for player count. Sure, it doesn't you tell the whole story. It contains less information than if you had all the separate win rates for each player count. But it gives you infinitely more information than if you just saw a raw win percentage and had zero idea of how many players this player typically plays against.
Because like I explained, the wins/games doesn't tell you anything, since you don't know how many players were involved in those games.
It may have a meaning for you when you look at your personal profile, because you know by experience how populated the games you play in tend to be. But when you check the profile of someone playing a game that can be played at 2,3,4,5 players, you have no idea what their play history is like. So the number doesn't tell you anything and you need the adjusted number to get an actual rough idea.