I believe the ideas of fantasy and wish fulfilment in game design have been greatly misunderstood and over applied.

Ideally, each game is a new experience, although certainly there are clones and reskins. Given this, the player does not necessarily have an idea going into a new game about what they want from it. Even within a genre, games should be striving to better the art and science involved, so in a way, if you’re concerned with fulfilling backward-looking fantasies about horse-riding, this “logic” would lead you to not invent the automobile.

Where did this idea come from? In Level Up! The Guide to Great Video Game Design, Scott Rogers says:

This is, frankly, nonsense.

Using the example of chess, as is commonly done in thought experiments about games, what can players possibly be feeling that is different from the real world? Being smart? It’s pretty hard to win chess without being smart in the real world; same with sneaky. Powerful then? Both players have exactly the same pieces; you only become more powerful by eliminating their pieces from the board, which gets back to smart and sneaky. Successful? You feel successful if you win, which is a possibility in every game, but again has requirements of its own.

I’d actually argue that a player cannot be made to feel something they have not in the real world, and it’s arrogant to pretend that a game could do that. If my real world experience is that I’m stupid, can you really make me feel smart? Probably not. This is the problem I, and, I venture many others had when we studied Romeo and Juliet as high school Freshmen: It’s generally agreed to be one of the masterpieces of English Literature, but by and large, we had no experiences to connect with it—instead, we just wondered what these two were blathering about (and generally gravitated instead to Mercutio’s wit and bravado). The player brings their own set of experiences into the game; those are what you have to work with.

Even in the specific context to which wish fulfilment seems to belong, role-playing games, I’d argue that someone playing an assassin in a D&D game has to already have that side to their personality in order to enjoy it. Therefore, I’d say instead that games are heightened experiences. Chess is an arena in which someone already smart and sneaky can harness those traits and master the strategy and tactics of the game in order to become powerful and ultimately successful within that context.

Also note that there is zero authorial intent in chess to foster any of those feelings or traits. Although we can’t ask the game designer what they intended, it seems a clear abstraction of warfare, with movement limitations and therefore strategic value placed on the various pieces reflecting the subject matter. The gameplay arises from this ruleset's application to controlling territory, defending units, attacking your opponent, and so forth.

Let's dig deeper into where wish fulfilment came from. Shafer took up the thread again in an interview with Game Studies:

This is a terrible way of looking at designing characters. To begin with, actors and gameplayers are fundamentally different roles, where the first is part of the work and the second is the audience of the work. Ideally a screenwriter is thinking about conveying meaning to their audience rather than appealing to their co-creators (obvious Hollywood realities aside). Plain and simple this is pandering. There have been many attacks on the gaming medium’s puerility, and this certainly smacks of that. But more importantly it places artificial limits on the kinds of games that can be made, keeping creators too, in a state of arrested adolescence.

Papers Please provides an excellent counterexample. The role of immigration officer is one few players would aspire to, but it is vital to the experience of the moral choices and emotions that accompany them that this game deals in. And the game’s no exception in this regard: This War of Mine, Abe’s Oddysee, and Life is Strange are just a few excellent games that are far from wish fulfilment.

I think the error here, as has been said of comics, is that a medium has been mistaken for a genre. Every game does not need to be the equivalent of an action movie. Certainly designers need to know what they are choosing and why, but our medium is a rich one, and I think it's capable of representing a much broader spectrum of experiences than this “rule” would dictate.

Another possible source for the fantasy idea is Raph Koster, who says in his blog post, Game Grammar:

At the risk of putting words in his mouth, I believe he is referring to fantasy in a very different way than how I often see it interpreted in game development: his verb roughly means an action a game allows a player to perform, and the game designer here is being asked to consider the things allowing that action implies.

So goals of course—games revolve around the achievement of goals, and it’s important that the player understand how a verb can help them achieve those goals. On the fantasy side, this means thinking about player expectations of what a verb might let them do. And to the designer this also might imply some of the limitations they need to place on a verb: allowing players to battle NPCs definitely creates the fantasy that they can kill Lord British.

Even in the relatively narrow context of a game that involves the player taking on the role of a character, the idea of this character being a wish fulfilment is one I’d approach fairly cautiously—which of these provides a more interesting character arc: a character who starts out awesome and keeps being awesome, or one that changes over the course of the work?