Jon, Dirk and David discuss the role of failure in games and how it ties in with how people learn how to play. What does punishment add, and where can it go wrong? How do you effectively teach players strategy games? What do PLAYERS get out of failure? And how the heck do you learn a tabletop game without tooltips?
This is a very important topic, and one I feel isn’t given due attention by most developers. A point I made during the show that bears repeating here is that even people who do this for a living feel helpless the first time they play a new game.
In this age of F2P, demos and good ol’ word-of-mouth it’s absolutely crucial that we developers do whatever we can to ease the learning curve in our games. Writing tooltips and help systems and top-notch manuals isn’t fun, but the learnability of your game is really just about as important as whether or not it’s fun.
Some people are willing to grit their teeth and fight their way to the good stuff – but many aren’t. Is the cost of spending another 5% of the effort you’ve put into the rest of the game worth the reward of possibly doubling your audience? The answer seems obvious to me!
– Jon
Jon Shafer on Design 
I wonder if something worth doing is to have a post-game analysis engine. That is, after a game ends or the player resigns, have an algorithm look at the player’s game and make some suggestions based on it. “Looks like you didn’t build enough workers here…” “Oh man that one battle really screwed you!” “That enemy is just going to get stronger and stronger, why didn’t you attack?” And so on.
Then again that might be really hard to code…
A game wrapup screen could be a good supplement your main tutorial features, but any logic you put into that could also be utilized (and would be much more effective) during a game. Let’s use Civ as an example. How often do people actually finish a full playthrough all the way to the end screens? If someone is confused and quits 30 minutes in then all of that hard work you put into a recap-and-teach screen is wasted. Your #1 goal needs to be making players comfortable at all times, and that means prioritizing just-in-time help over all else.
– Jon
I guess the idea here is that it might be more effective if the player sees the consequences of the errors. Like if the player finds himself without resources later in the game, and the game explains to him that he’s in this situation because of not enough workers in turn 2, say, that might make more sense and be more meaningful to the player than if the game just told him to build more workers.
The other point is that the endgame recap might be helpful for people who *win* the game, but don’t realise that they played (very) non-optimally. This is especially the case for people who play on an easy difficulty, and then find the game inexplicably impossible on a higher difficulty. And I guess personally I just find post-game stats and game logs to be very cool things to look at and share anyway!
Great points, and I certainly agree with all of them.
However, the percent of people who start even fairly basic strategy games like Civ and put it down before finishing a single game is really quite alarming. My focus is on bringing people into the game and getting them excited about sticking with it. Sure, just telling players to build some workers on turn 2 isn’t going to teach them what’s going on, but you first have to develop a feel for a game before you can appreciate its nuances and actively pursue different strategies.
People who are eager to learn more about a game and get better at it already have several resources at their disposal, even if it means having to watch YouTube videos. Supplementing that with effective in-game tools is certainly a worthy cause, but when time and money are running out it’s more important that you just get players to that point.
– Jon
A few points- beware wall of text. 🙂
1) I’m not sure that I agree with the departure point for the ‘failure’ discussion. I do not play primarily to win (ok, I’ll admit it – I don’t play to lose either…). Games are for having fun and I often find that making ‘winning the game’ the objective makes the actual playing just a grinding task. Sure, you get a number of people, particularly aggressive ones on forums and the like who cannot concieve any other ‘playing’ but really I suspect more people quit playing because they are not having fun rather than because they are ‘losing’. (Note that for learning this is particularly apt as it is not ‘fun’ to have your game change markedly because of a previously unknown mechanism or consequence.) I play a lot more strategy games than I finish because with many of them at some point it stops being fun. For me the fun is in playing the game and this often means slowly building up a civilisation, or creating a special city, or character, etc. It becomes more an artistic process of crafting (the map or whatever) into something beautiful rather than a Machiavellian process of win at all costs and damn the means or consequences. But really a good game would cater for all experiences.
Corollary: One of the games I had most fun with over the years was Civ 2. The quickest and highest scoring wins I had were bloodlust stompfests – I didn’t enjoy them much. The reason I enjoyed it so much were all the ‘little wins’ along the way. Building a road, yay, little win. Finishing a temple, yay, little win. Gaining a new tech, yay, little win. Completing a diplomatic deal, yay little win. Founding a new city, yay, little win. Winning a combat, well you know. Building a wonder, double yay, littleish win. Building wonders was rarely strategically efficient but it was fun (and doubly so if it was a step on the road of some preplanned supercity…). I would agree with you that the Imperialism games were much more strategically sound (especially II with the resource management side) and I played them a fair bit, but as much ‘fun’, I’m not sure…
2) As far as learning a game goes one idea applicable to faction based empire builders might be to use particular factions to teach certain aspects of the game. Make ‘scenarios’ which are basically sand box games with certain slight modifications and set starting conditions with some suggested or somewhat scripted objectives. Use the ‘capitalist’ faction to teach the economic aspects of the game; their scenario would include starting next to a gold mine resource (maybe a prescripted region which was dropped into an otherwise random map) and having 100 turns to achive economic victory and would include prompts/advice along the way of mechanisms to use and actions to carry out to achieve this goal. This way the player is playing the game from the start and learning a viable strategy to use. Another scenario might be to lead the ‘atlantians’ off their island and on a conquest of the mainland and would teach the water based aspects of the game. A strategic designer could with a small number of scenarios teach the important/basic mechanisms of the game as well as introducing a few basic strategies and get the player familiar playing with different factions all in one go. The guide sounds like a good idea for AtG though.
3) I think the first few moves of strategy games are a worthy topic in their own right. One of the reasons they are so special, both exciting and confusing often in equal measure, is that this is when the richness of strategic possibilities of the game are present. Good strategy games sometimes preserve something of the richness of options and possibilities (‘difficult choices’ on the strategic rather than tactical or mundane level) into the mid-game. Unfortunately the end game generally becomes a convergent grind where the previous strategic choices get played out in a limited number of ways.