Freedom to Fail

Why must players beat every challenge laid before them to progress within a game? This structure suits many gamer hobbyists, but what about mass market players? Why should they jump through the same hoops as the hobbyists? In short: should we give players the freedom to fail?

The games industry, as we have discussed before, has built itself into its current state by making games which are effective at supplying the emotion of fiero – triumph over adversity. But, as we discussed before in Emotions of Play Revisited, not all players are suited to games of this kind. In fact, I hypothesise that a significant but unknown proportion of players feel disappointment (sadness) when they fail, rather than the anger that motivates the fiero-seeker to throw themselves against challenges until they achieve victory, and the endorphin hit they crave.

A recent study by Dr. Jesper Juul (not currently published – I’ll provide a link when it is) on “hardcore” gamers and failure demonstrates that these players seem to require failure as part of their enjoyment. This result lies well without our expectations – if the player wins without failure, there is little or no sense that a challenge has been overcome, and thus little or no fiero is provided. The report does not repeat the experiment with “casual” gamers and thus it remains an open question whether or not this finding applies across the entire audience for games. My firm suspicion is that it is not the case, but at the current time I cannot prove this contention.

For several years now, my company and I have been experimenting with game structures that provide new ways to balance the needs of the fiero-seeking hobbyist against the less demanding requirements of a mass market audience. The principle structure we have been investigating is the fail-continue structure, which I talked about before in the context of Air Conflicts, and its semi-sequel Attack on Pearl Harbor. The idea here is that the game is organised in such a way that the player does not need to complete a sequence of challenges to progress, they just need to reach a minimal degree of success in order to proceed.

In many respects, the fail-continue structure used in these aforementioned games seems to be a success – no reviewer has marked the games down in this regard, and the needs of obsessive-compulsive players to repeat a challenge until they are victorious is met by a “retry?” option. However, unlike the vast majority of games on the market the player is not forced to retry until they succeed. They can, if they prefer, proceed with the game anyway, the cost of failure being represented in these games by the loss of a plane. (Players who crash constantly run out of planes, and have to retry the campaign, but they are never required to repeat a specific mission until they master it).

However, publishers have resisted this application of fail-continue structure, in one case even claiming that a game using such a structure “wouldn’t even be a game” (despite expressly intending the game in question for a casual audience). We are even getting some resistance from our development partners. There is no real comprehension among people in the games industry as to why such a structure might be beneficial to a wider audience, yet I remain confident that there is a sound commercial benefit to approaches of this kind.

Platform games since Mario 64 have almost uniformly used a structure that requires less than perfect success to continue – following a “70% rule”, checkpoints in game progress require that the player collects some but not all of the tokens available in the game space in order to continue (usually about 7 in 10, hence the “70% rule”). This structure was a huge step forward from earlier platform games which generally required that challenges be completed in strict sequence, and I believe that a fail-continue structure has the potential to take this even further.

For instance, in Air Conflicts, I believe one need only complete about a third of the missions in order to reach the end of each campaign – and ‘completion’ need not be success; the player is also allowed to retreat from a battle in preference to being killed. Gamer hobbyists facing this structure play the game like they would any other: they want to succeed, they want to get fiero, so they push themselves against the challenge in the hope to win. However, mass market players have an option in this structure they would never normally have – facing a challenge they cannot beat, they can just ignore it and carry on. A narrative structure (in this case, the story of World War II) links together the progression; it feels natural to most players that the war goes on, whether or not you succeeded in your assignment.

The only viable argument against fail-continue I have heard thus far is that it would allow a person to complete the game too quickly by just killing themselves. True enough – but such a player would have none of the experience of the play of the game, they would gain nothing from doing so. Furthermore it is possible to set gateway targets for progression between sections of play such that the player does need to reach some degree of performance to continue (as happens in Air Conflicts) or, perhaps even better, that the next section unlocks either from success or from passage of time (when you have played the previous section for more than two hours, you automatically unlock the next section).

Videogames in their current form train the player to perform certain tasks through repetition and rewards for success, in a manner similar to the way one trains an animal to perform. The player is expected to complete the task set, to learn the skills the game is trying to teach. The idea that the game might be organised in such a way as to allow the player to get out of this enforced training – to avoid being made to jump through hoops – is alien to most gamer hobbyists, trapped as they are in the idea of videogames being entirely about the powerful drug-like hit of victory in the face of struggle.

Yet the mass market isn’t about fiero and hoop jumping. If we look at, for instance, Bejewelled, it is an error to say this is simply a remake of Tetris Attack – this older game was build very much like a conventional gamer hobbyist experience, while Bejewelled has been simplified to the point of triviality in the eyes of most hobbyists. It remains compulsive and fun to the mass market players because what this audience is looking for is very different indeed. Consider also the way the Brain Age games (which, let's not forget, have sold more than 8 million units each) unlocks new content based on the number of days the player has been playing, irrespective of their degree of performance.

If games for the hobbyists are designed as training programs for beating specific challenge sequences, games for the mass market need a different paradigm, quite possibly many different paradigms. On the one hand, we have the simple compulsiveness of PopCap’s titles, including Bejewelled, and also the more traditional puzzle designs of something like Brain Age. But I believe there is another paradigm that can be gainfully applied: that of the amusement park.

When someone goes to an amusement park, they expect to be entertained; to be delivered fun experiences with zero challenge beyond the patience to wait in line, and the willingness to face ones fears in the case of certain rollercoaster rides. The global amusement park industry is worth about $25 billion, making it quite similar in size to the videogames industry (worth around $32 billion and still growing). I believe that if the games industry wishes to reach out deeper into the mass market than it currently does (the casual games industry is currently worth only about $2 billion), it needs to look at the amusement park industry for guidance. Combining amusement park sensibilities with compelling virtual worlds (and a little narrative glue to hold it all together) could uncover a whole new market for videogames.

Freedom to fail means allowing players the opportunity to continue in a game whether or not they can beat all the challenges. This can happen in many different ways – through further use of the “70% rule”, or the reduction of the margin to 50% or less; in the use of fail-continue structures requiring only a minimal level of performance in each stage to progress; in the use of game structures which unlock both in response to the completion of challenges, and also in return for the player’s investment of time (a model already deployed on a small scale in Tetsuya Mizuguchi’s Rez).

Perhaps freedom to fail doesn’t take it far enough – perhaps there are players out there who want freedom from failure, who like the patrons of an amusement park are just looking for some fun and excitement in a way that they cannot possibly fail, and which the only challenge is to experience the maximum amount of enjoyment. Without further studies into the mass market players who lie beyond our current awareness we can’t know these players are out there, but there are surely signs that there are players beyond our usual reach, players who perhaps avoid videogames because they see the anger-contorted faces of the fiero-seeker and know that that experience is not for them.

Gamer hobbyists will always find plenty of developers and publishers willing to deliver what they want to play, but it’s about time the games industry looked beyond its habitual obsession with fiero and explored other models for play and entertainment. The money is there to be discovered, as the billion dollar franchise The Sims proves. When are we going to start seriously pursuing it? 

The opening image is Frustration by Laura Walker Scott, which I found here. As ever, no copyright infringement is intended and I will take the image down if asked.

Comments

[Bezman]

Mario 64 only needed 70/120 stars - a '58%' rule!

I think that even retro games have had a 'freedom for failure' in some small sense, with the margin for error simply becoming larger and covering more of the game.

My mind flicked to Sonic 2 (Megadrive version) and its 3D tunnel-like bonus stages, in which - at the first few checkpoints at least - only a certain percentage of rings were required to be gathered. Of course, in the later checkpoints in a given stage, the % collected needed to be far higher, meaning even those bonus rounds would only appeal to those who enjoy repetition for the sake of improvement.

Another point that came to mind was any 2D scrolling shoot 'em up, wherein only a small number of targets need be shot, the remainder only destroyed to get better chances for survival or a better score.

Last point - any game where dying results in continuing from that point, with no penalty other than the life loss. Streets of Rage and various shoot 'em ups come to mind.

Of course, the 'accepted level of failure' is far lower here but my basic point is that in any game where a less-than-perfect performance is required there is some tacit allowance for some failure (nearly any game other than certain puzzle games, such as the notably mass-market Sudoku or various other puzzlers seems to fit this criteria).

What you're talking about isn't a new approach - rather simply expanding the leeway that's always been given.

As you may have guessed, I'm all in favour of this and imagine that as long as a score is also supplied, both parties would be catered for. Maybe there could be a selectable option to turn on the highscore board, so casual gamers need not even see the stat.

On a different tack, I note that in Super Mario 64 and the other games that followed the model, there were still sections (e.g. the Bowser battles) that didn't allow freedom to fail - requiring eventual success in order to end the game. Would the answer again be 'different degrees' of winning, giving assistance if problems are had? Would such a game not be robbed of its 'climax'?

Are true climaxes possible or even desirable for the audience you talk about?

[Chris]

Bezman: in the case of Mario 64, the question is not necessarily the proportion required to finish (although I accept, as you say here, that this was closer to 60%!) but rather what proportion was needed to reach the next area. I don't have the answer to what this was at my finger tips, but "the 70% rule" isn't intended to be taken literally anyway. ;)

"Last point - any game where dying results in continuing from that point, with no penalty other than the life loss."

Of course, one still has to overcome the next challenge to continue in such games...

"What you're talking about isn't a new approach - rather simply expanding the leeway that's always been given."

I completely agree with this statement! That's why I was keen to reference existing games, such as platformers, which already afford a degree of "freedom to fail".

"Are true climaxes possible or even desirable for the audience you talk about?"

Great question!

I don't think the climax of a game need be 'fiero earned in blood' - there are many ways to get a narratively satisfying climax without forcing repetition until victory.

We had a game in predev a few years back called "Beneath a Waning Moon" which used a sorter question to determine if the player was challenge-oriented (i.e. a fiero-seeker) or not. For the boss battles, if the player was identified as "casual", a narrative endings to the boss fights were provided - i.e. you don't have to win, because someone you've met in the game comes and bails you out if/when you get into trouble.

I think approaches like this are completely viable. After all, film goers accept the victory of their heroes in purely narrative terms - why shouldn't mass market videogame players?

Thanks for a great comment!

[Michael]

I have to agree wholeheartedly with this. There are a bunch of games that follow this theme, although in variations. The old Wing Commander series did a fantastic job, letting you fail miserably and continue (although do that often enough and you ended up in a dead end campaign against impossible odds and a failure ending). Many newer games sneak it in in various ways. Splinter Cell, as I recall, did a good job of giving you an obvious idea of perfection, but letting you err in small ways while still completing the mission. I honestly find flexible, open games with loose objectives and lenient policies fascinating, replayable, and enjoyable, and wish more developers and publishers would adopt the model.

[Chris]

Michael: I keep forgetting that Wing Commander had something of a fail-continue going for it - thanks for reminding me!

"I honestly find flexible, open games with loose objectives and lenient policies fascinating, replayable, and enjoyable, and wish more developers and publishers would adopt the model."

Trouble is, such games can be very expensive to make... Of course, now that all the upper market games are grossly expensive to make, what's the excuse? ;)

Best wishes!