Nope, 4 is the best one. Me too can't understand common love for Crimson Butterfly.

Well, that was during the time Survival Horror were Survival Horror. Now they are supposed to be TPS with zombies and jumpscare.

50 is supposed to mean average. Whether a reviewer uses the full range of the scoring system or not depends on the reviewer. I'm not saying that that's always the case, because it's not. I'm saying that's how it's supposed to be.
And there are reviewers who use the full scale of the numerical scoring system. But instead we have the vast majority of people just automatically assuming that no one does, without understanding the review or reviewer in question. And that's the problem.

If people care about their money then they should read the reviews rather than just look at scores that mean nothing unless you understand the reviewer in question who wrote it. And if you're going to make assumptions about what the reviewer thinks, then stick as close to the norm as possible, instead of the opposite. Read the review, and if you find that the reviewer is not using the full scale, people can complain, and they can change and do what they're supposed to do. I've seen magazines and sites change their review scoring system for this very reason. And recently more sites are abolishing the scoring system completely, because readers tend to abuse and missunderstand them. Primarily through metacritic. But that site has become so important that many sites are afraid of becoming irrelevant if they leave.
There are even have people who think that something below 90 isn't good. And not from a specific reviewer, but from Metacritic, which makes it even more ridiculous.

And I'm not saying that you might enjoy a game more than the review suggests. (Although that's possible too) I'm saying the actual review may suggest something much more positive than your pre-concieved conception of what the score means to that particular reviewer, if you actually read it.
There are those who give a game a 6/10, and put a "Good" stamp on it. But many people will just assume that it means bad, and not read the review.
But yeah, even if you understand that particular reviewers scoring system, you may find that he/she gave it 4/10 because he/she doesn't like long cutscenes, but you do. Or a fighting game doesn't have a story mode, but while you find that to be a nice bonus, the most important thing is the online vs mode. Etc.
You can't understand what the score means unless you understand the reviewer. You can guess. But if we're going to guess based on conjecture, then we should guess closer to the norm. Not the opposite, and just assume that every reviewer is the same, which isn't the case.

Only 2 out of 30 reviewers gave it a below average score.
50 is average. Not bad. But sadly it seems that a lot of people don't use the full scale of the scoring system.
Instead people look at it like this.

1-50: Unplayable.
50-60: Horrible
60-65: Bad.
65-75: Average.
75-85: Good.
85-90: Great.
90-91: Very good.
91-92: Amazing.
92-93: Increadible. GOTY contender.
93-95: Greatest game of all time contender.
95-100: Can't breathe! I need an ambulance!

We have a full range of numbers from 0-100 for a reason.
Use them.
But of course always read the reviews of the games you are interested in. It can get a 40 and still sound like something you'd enjoy. Or 90 and sound like a snorefest.

I'm not expecting people who have no interest in the game to read reviews. I'm expecting those people to not jump to the conclusion that every reviewer, or the vast majority of them, ignores to use the full scoring scale.
If they are interested in the game, and care about their investment in it, then yes they should read some reviews or watch review videos.
However, the reality of today is that many people just don't read reviews any more. Even when it comes to games that they are interested in, there are people who are just going to look at the final Metacritic score, look at a couple of trailers, and nothing else. And they all have different ideas of what that score means.

There are alternatives to numerical scores that are much more clear though. A number can be missinterprited without context. While with something like: "Is it worth playing?: Yes/No." you can't missunderstand the reviewers intention regarding if he/she recommends you to play the game.

Metacritic didn't change much. I agree with that. But there are a few things. When you jumped from site to site to check the final score, you could also some times see what a particular score meant according to that site, because they would specify it. And you'd realize some differences between the way some sites score things.
This concept gets lost without that information, and instead people interprit the scores based on their own preconceptions of what they represent. And I've seen someone who thinks that the difference between 89 and the "at least 92" on Metacritic (not even a specific review) that he expected meant canceling his preorder. Three points. And three points at the very end of the upper scale of the spectrum, for that matter.
Another thing is also that, even if you only came there to look at the final score, there's a greater chance of something in the article catching your eye, and making you actually read the review even though you initially didn't plan to, if you go to the review rather than just look at the final score somewhere else.