Marvel's Wolverine: the score arrived before the game, and so did the disappointment

Marvel's Wolverine arrives on PlayStation 5 on 15 September 2026. As I write this, the game is not out. The press reviews are, and the internet has already decided it is a disappointment. Notice what is strange about that sentence: there is a collective verdict on the quality of an experience almost nobody has had.

On OpenCritic, at the time of writing, there were 75 critic reviews and fewer than 20 player ratings, so few that the site will not display a player average at all. On the other side, thousands of furious comments. This case is worth looking at calmly, because it is one of the cleanest portraits yet of something psychoanalysis has been describing for a century.

First, what the reviews actually say

The numbers, before the interpretation. The game sits around 78 on Metacritic across more than a hundred reviews, and at 79 on OpenCritic top critic average, where 71% of critics recommend it. For comparison, Marvel's Spider-Man landed at 87 and Marvel's Spider-Man 2 at 90, both from the same Insomniac.

More interesting than the average is the spread. IGN, GameSpot and GamesRadar all gave it 6 out of 10. Giant Bomb gave 2.5 out of 5. At the other end, Hardcore Gamer gave 5 out of 5, TheGamer 4 out of 5 and DualShockers 8.5. The same object was scored anywhere from 50 to 100 by professionals who played the same thing.

And there is consensus on the content, which tends to go unnoticed. Almost every review praises the same things: the characterisation of Logan, the performances, the weight and brutality of the combat. And almost every review criticises the same things: linearity, repetitive encounters, uneven pacing and limited player agency. Critics are not divided about what the game is. They are divided about how much that matters.

78 is not a failure, it is a distance

Here is the first knot. A game at 78 on Metacritic is, by any reasonable measure, a good game. The word the internet is using, however, is failure. And it is not measuring the game, it is measuring the gap between 78 and the number people expected.

The arithmetic is this: 90 was expected, 78 arrived. The twelve-point deficit is not one of quality, it is one of expectation. Had the very same game been announced eight months ago by an unknown studio, the identical score would read as a pleasant surprise. The object would not change in the slightest. What would change is how long it had been anticipated.

Bion: five years of expectation without an encounter

Wilfred Bion, in Learning from Experience (1962), proposed a model that describes this case almost literally. He starts from the idea of a preconception: an expectation that is still empty, a form awaiting content. When the preconception meets a realisation that approximates it, the outcome is a conception, and that is what Bion calls learning from experience.

But there is the other path. When the preconception meets the absence of realisation, what is produced is frustration. And Bion draws a distinction that matters a great deal here: what happens next depends on the capacity to tolerate that frustration. If it can be borne, a thought is born. If it cannot, it is expelled, evacuated, turned into something to be discharged outwards.

Marvel's Wolverine was announced in September 2021. That is five years of preconception with no realisation whatsoever. Five years in which each person filled the empty form with their own content: the Wolverine they wanted, the story they imagined, the game that would do for Logan what Spider-Man did for Spider-Man. A real object would need enormous luck to coincide with that.

The 2023 leak did something worse than reveal the game

There is a detail in this story that makes it singular, and which almost nobody is factoring in. In December 2023 Insomniac was hit by a ransomware attack. The Rhysida group leaked around 1.3 million files, among them development material for Wolverine: maps, gameplay footage, story elements. An unfinished build, two and a half years before release.

From the standpoint of fantasy, that is extremely high-octane fuel. It was not a controlled trailer showing a finished product. It was raw material full of gaps, and gaps are exactly what imagination fills best. Everyone completed the unfinished thing with their own version. When an official trailer came out later, outlets noted the game looked considerably better than the leaked pre-alpha build, which proves the point in reverse: people had been judging, for worse and for better, an object that did not yet exist.

The score became the object

And here we reach the newer phenomenon, where this case parts ways with the old story of hype. This week's mass disappointment is not with the game. It is with a number that summarises the game.

Something remarkable is happening there. A weighted average of other people opinions has taken the place of one own experience, and taken it in advance. The person does not merely say the game is bad, they say it is a 78, and cancel the purchase on that basis. The aggregator stopped informing and started deciding, ahead of time, whether someone has permission to enjoy a thing.

This has a clear defensive advantage, which is why it works. Whoever anticipates the disappointment cannot be caught off guard by it. It is more comfortable to arrive already disappointed than to risk hoping and take the hit. The cost is high and quiet: in protecting yourself from frustration, you also give up the chance of being surprised.

Freud: disappointment is what installs reality

In 1911, in a short paper called Formulations on the Two Principles of Mental Functioning, Freud describes how the psychic apparatus handles desire before it learns to handle the world. At first, whatever is wished for is simply presented in a hallucinatory manner, just as still happens in dreams. The wish satisfies itself, with no object at all.

And then comes the sentence that matters. Freud writes that it was only the non-occurrence of the expected satisfaction, the disappointment experienced, that led to the abandonment of this attempt at satisfaction by means of hallucination. It was disappointment, and nothing else, that introduced the reality principle, in which what counts stops being what is agreeable and becomes what is real, even if it happens to be disagreeable.

That repositions the whole thing. Disappointment is not the accident that spoils the process. It is the mechanism by which someone leaves hallucination behind and meets an object that actually exists, with edges, limits and flaws. A game that matches exactly what you imagined has taught you nothing, because it was you. The game that disappoints is the first one with an existence of its own.

What to do with the game arriving tomorrow

I am not saying criticism is worthless or that scores are useless. They are useful, and the criticisms of this game look consistent: linearity and repetition are real problems, flagged by those who liked it and those who did not. Anyone after an open world and freedom of approach has enough information not to buy. That is using a review as a review, which is what it exists for.

What is worth separating out is something else: deciding that a game disappointed you before touching it is not an opinion, it is a defence. And there is a simple question that puts the problem back where it belongs, useful well beyond this release. The thing you are about to dismiss, have you actually experienced it, or are you dismissing the distance between it and the version you built alone over five years?

The Wolverine that shipped will not be anyone's Wolverine. That is the bad news and it is also the only way it gets to be a game rather than a mirror. Bion would say that frustration, if borne, becomes thought. Take it as practical advice: play first, be disappointed afterwards, and notice what is left. There is usually more left than the score suggested.