In 2018, 9,300 games were added to Steam. It's probably over 10k in 2019. Whenever you have such a large number of data points, rules of statistics apply. So I am pretty certain that if you listed all of these games by let's say Metacritic score, you would get very few excellent games with scores over 90, very few really bad games with scores under 50, and a huge number of mediocre games with scores around 70.

Now one of the strange problems of the market for games is that there is very little correlation between a game's quality and its price. Many of the mediocre games cost the same as those excellent games, and some good games are indies that cost less than some mediocre or bad games. So normally, if you buy a game and it turns out to be mediocre, you are cursing because you could have gotten something better for the same money. Which is why some people really, really hate mediocre games.

However, mediocre often means that the game is flawed in some aspects, but good in others. Playing a mediocre game for a few hours and then moving on can be fun enough. Which makes me think that a game subscription service could actually change the way we play games. I mean, if you can get a game without any additional cost because you are subscribed to some game subscription service, does every game you download have to be excellent? If you download something, play it for a few hours, then think that the replayability isn't great, and you uninstall it again, you haven't lost anything and you still had some hours of entertainment. At least I feel a lot less negative about playing a mediocre game on a subscription service.