Games and Toys and Defeat Horizons

I’ve continued playing Wesnoth and trying to figure out what it is that annoys me in the game. I think I’ve got at least a part of the answer here: the issue is that what we call “games” actually consists of two different types of interactive objects, and mistaking them for each other is a recipe for disaster. Just an idle thought, let’s see if it goes anywhere.


A game is (for our purposes here) a ritual space with simplified rules and objectives – simplified in comparison with reality. A good game has goals and performance, the players derive enjoyment from performing the skills necessary for the game. We can talk about good play and bad play, and about strategy and technique of play. Chess and tennis are games.


Some things called games are actually virtual experiences of various sorts of multimedia art. Playing them is not a matter of performance, but of experiencing content. Much of what passes for video “game” design is actually about designing toys. The issue is blurred by the fact that many of these software toys provide a default game embedded inside; it’s just that often this game is not very fun or functional, or it might be arbitrary, or it might simply get ignored by the player. Console rpgs are almost always foremostly toys in this sense of the word.


OK, so that didn’t turn out as sublime as I thought it would – a pretty routine analysis. Still, back to Wesnoth: the reason for why I wouldn’t want to resolve the annoying issue of dying units by simply making unit death a non-issue in campaign play is that it’d turn the game more toy-like. It would make improving units into a grind – a routine process with psychological values that is very typical of toys, but rare in games, and often reviled.

The fundamental problem of computer war game campaigns (and tabletop as well, in fact) is very important – it would actually deserve to have a name… let’s call it the Defeat Horizon. That deserves its own heading.

Defeat Horizon

Defeat horizon is the limit that prevents us from introducing genuine loss into certain types of games. It is caused by conflicting assumptions:

  • A game needs to remain feasible as a challenge proposition through its entire length.
  • A toy needs to remain fun and rewarding for the duration of the psychological process (f.ex. following a plot) that is engaged when you fiddle with it.
  • Negative feedback cycles in games make players give up when they find the position untenable. In toys they give up when the toy refuses new content to them.

The above is not an issue at all for games that have graceful exit strategies from the play situation. Consider Chess: whenever you find your position untenable because of a stupid mistake, you ragequit (or just quit, if you’re an adult). This does not break the game, as it’s intended to end, and it would have ended even if you had won. You can start a new game and hope to do better. Compare with Battle for Wesnoth, in which the defeat horizon gets crossed every time I lose too many units for my liking and decide to reload; instead of the game ending conclusively like Chess, I just go through the arduous punishment of waiting for the game to reload and then giving my units almost the same orders as before, trying to hone my strategy.

The defeat horizon is a real issue in several types of game and toy design. For example, a bog-standard console rpg like Final Fantasy has a very strict defeat horizon, because it is very toy-like (you have to actively speedrun the game or something of the sort to find an actual challenge in these), with the plot and character advancement being the main axes of development. Consequently the defeat horizon resides on those axes: you the designer can’t feasibly punish the player by backtracking the plot or taking away character development, as those are the only things that are actually keeping the player there and playing. As we all know, this has led to a situation where these games actually have really wimpy and consequence-free gameplay: because there are no meaningful ways of punishing players without crossing the defeat horizon and making them quit, you don’t punish them ever. The closest these games come to punishment is really low-tech psych stuff, like having to sit through the same stupid monologue six times as you try to win the boss fight. I certainly hope this stuff is just design oversight and not intentional punishment for failing in the fight, anyway.

Conclusion the second

Wesnoth has a too strict defeat horizon because it is too toy-like, is my conclusion (which is based on what I like in games). If I played the game in multiplayer mode with other real people, then the single scenario would be the battlefield, esteem would be on the line and the defeat horizon would be far away out back; I’d fight until I couldn’t win the scenario anymore, instead of quitting when my first unit gets a scratch. Instead I’m playing the campaign mode, which is toy-like and really just about me micromanaging my ever-growing army towards higher experience levels. This activity has a strict defeat horizon, I find my position untenable after the slightest set-back and act quickly to rectify.

(I don’t claim that the multiplayer mode would be any good here, incidentally. I haven’t tried Wesnoth, but I’ve tried other games like it, and I personally don’t care much for arbitrary single scenario play in this sort of war game. The game-like experience is much more robust, but choosing your army without the constraints of campaign play is a pain in the ass in many games of this sort, as they haven’t really been designed with that in mind. Could be good if I really got into it at some point, don’t know.)

To fix this sort of thing would require making the campaign mode less toy-like by introducing real skill elements that actually would allow me a sense of accomplishment, and more importantly, relaxing the defeat horizon by making sure that while making mistakes made my overall position worse (to keep it game-like) and defeat more likely, they would also heighten the sense of urgency and hope, and a will to fight. When I play table tennis and am one serve from a loss, I don’t quit the game, because the game’s defeat horizon is correctly adjusted: as long as the game continues I have an obligation of sportsmanship towards my opponent and a slim chance of victory, which both make me play the game until its conclusion – and I might even start a new game right afterwards, it’s that well-designed. This is not true of the Wesnoth campaign mode, which presents me with an uncertain future towards which my only safeguard is manic army improvement in the hopes of growing a strong enough army to keep up with the rising scenario difficulty. In these conditions I will of course find my position untenable after most minor losses, as I don’t know how good I should be doing in comparison to the next scenario in the campaign (or the last one, for that matter).

Looking at it from this viewpoint, the situation is actually pretty simple: the whole concept of a string-of-pearls wargame campaign with incrementally improved army sucks, because the player can’t know (in most implementations) how well he is doing, and is therefore reduced to either obsessive optimization or actually playing for 40 hours only to find out in the last scenario that he’s hopelessly outmatched by the enemy (this one actually happened to me in FF7, I think). Both are pretty sucky outcomes.

The above issue is not that difficult to fix when we understand it, of course. We could introduce some real information about the future for the player, for example. I’d find it pretty cool if this sort of campaign set-up gave me an army list for the last scenario’s enemy forces right off the bat at the start, for example, allowing me to run my own comparisons, or even allowing the game itself to tell me how I’m doing in relation to the expected level of performance. The classic strategy game Colonization did this, allowing the player to track the development of the king’s continental army in real time. (Of course if we’re copying Colonization, we might as well go full-hog and allow the player some say in when the final battle is going to take place. I’d find that sort of thing pretty cool in a game like Wesnoth – you run through this gauntlet of preliminary battles not because the plot dictates them, but because you’re trying to set the parameters of the decisive confrontation at the end to be to your advantage.)

4 thoughts on “Games and Toys and Defeat Horizons”

  1. That is so true. While I was playing “Heir to Throne” for the first time, I got stuck at the end with lousy units and no chance of winning the fight. That far I thought the game would safe guard me against that. Only later I realized it was because I had lousy units.

    The only survival strategy I could think of was moving my units into a narrow space between a lake and the top of the screen, as that would reveal as little attack surface as possible, and then cheating the enemy units with a mermen. It always resulted in defeat by turns ending, but I took a nice picture of it as a souvenir. See

  2. Good insight on Defeat Horizon. This is a concept that must be thoroughly understood by anyone hoping to make a moderately successful game. I work at a game design and dev firm and this is one of the first things we ask our prospective employees if they know about.

  3. There is a mismatch between the scale of your performance (campaign) and the scale it is actually gauged against (level). You know how good you are inside your level, but not if you are doing well in the campaign. Still, your performance of being good or bad really is relevant on the scale of the campaign. Thus, you are running blind. Which I hate, really.

    Another mismatch is then between the stated goal of the game (win campaign) and the inherent focus of the game (level up units). I think the urge to make things better and keep what you have got is rooted deep in our psychology. Also, it is the only thing of measurable success in the game (except for how many gold coins you obtain). It, thus, moves into our focus. This mismatch however leads to frustration creeping in, as it becomes more and more apparent that the game is not about what it is about (winning / leveling). That is the toy effect you are mentioning: developing your own goals (toy) instead of accepting the stated ones (game).

    That said, while reloading is a flaw, it is also explicitly discouraged almost everywhere Wesnoth. I don’t think there is a satisfying solution to this one, as removing save/load would just make the game nigh unplayable to me.

    One final remark: With what I said, I am agreeing with you and just stating things differently so they become more clear to me. 🙂

Comments are closed.