With SK or any game I can see someone playing many hours trying to love it and just not doing so. I'm ok with that.
But to go onto the Steam review page and give it a down vote? Sorry, I think that's an incredible dick move. I don't care what your justification is, if you've played a game 50+ hours you got your money's worth. It would be like me watching a movie 5 times and then going onto IMDB and giving it a 1.
Well, one problem with steam is that it's binary. Yes or no. Nominally, 4/10 is a No and a 6/10 is a Yes, even though that's closer than 6 and 10. Furthermore, the analogy breaks down. In a lot of strategy games it may take a very long time to find the flaws (or to try to find the good) for some people. I can't think of a strategy game that I didn't like/dislike relatively quickly, but Skyrim is a game I would personally not recommend even though I spent a lot of time on it, primarily because I spent a lot of time TRYING to find the fun in that game. It'd be more analogous to compare it to not recommending a show or movie because its flaws are not evident until the ending. Off the top of my head, AI (the movie with that Sixth Sense kid) and Yuuki Yuuna is a Hero are things I would personally not recommend because the ending simply ruins them, even if everything before that point was excellent. AI is particularly egregious because they literally could've stopped the movie at a point and had a much better movie, but they decided to continue for no good reason.
Heck, if anything spending a bunch of time trying to find the fun in a game is going to make me more angry because I wasted my time. For a very long time I actually recommended GalCiv 2 to friends even though I personally did not like it because of how money worked in the game. It took only a couple of relatively fast sessions for me to figure out I really didn't like that (in combination with some other things as well). If there's two games of otherwise similar quality I'll view the one that says "no, this game is not for you" more highly than one that says "hey come over here, we might have some fun for you maybe.... nah, never mind".
This also isn't controlling for various user expectations. For example, strategy gamers may expect much MORE than 80 hours into a game. EU4 has an average of 145 hours according to steamspy. Civ 5 has an average of 150. And both those are waaaaayyyyyy above the median (30-40 hours). If you think about it that way, the typical user puts in 30 hours into the game. The hardcore user has to put in a LOT more than that to make the average 150. Beyond Earth has an average of 32 hours and a 50% userscore.
Yet other genres where the amount of gameplay is easier to understand or with NO gameplay, less time is expected. Long Live the Queen, a life sim, has a 95% userscore for a $10 game with only 5 hours on average. On the kinetic novel front (absolutely zero gameplay pretty much), fault milestone one has a 95% steam userscore with a 2 hour long average game time. For a $15 game.
Basing a rating on length or hell, even quality of game mechanics oddly enough, is often insufficient. Magical Diary, another life sim by the same person as LLtQ has only an 89% despite having an extra hour of playtime on average (20% more!) and pretty much indisputably better gameplay. Why? Well, the theme is generally considered not as good as LLtQ. Low tension fem Harry Potter is less interesting than being a princess with a bunch of people out to kill you. I've often said that Master of Orion 2 is the best 4X game, but Alpha Centauri is the best TBS. On the surface that seems ludicrous because both those games are indisputably 4X TBSes. But while MOO2 has better 4X mechanics, AC has better everything else going for it. It does a much better job of making the aliens feel "alien", it does a great job making the planet feel alive and making it a unique setting, it has way better lore. A lot of stuff goes into making a good game.
This also doesn't account for human weirdness in ranking things with "potential". I believe Maro said that people will consistently rate a vanilla creature higher than one with identical stats and an overcosted activated ability like say... 7 mana: regenerate this creature. Even though the latter is strictly better. I mean, it's almost certain they're functionally identical, which would suggest a nearly identical rating, but no, it was actually consistently lower.