Players Shouldn’t Miss…

It’s been another two years since I posted anything, and apparently I had made the blog private, so nobody has been reading it anyway. Never mind. Today, a game design post, in tabletop roleplaying games but with underlying principles equally valuable (if not more) in video games.

I’m going to talk about a couple of related tweaks that some GMs might benefit from making to their general design principles (and hence their game system, in some cases). They’re vaguely linked by the principle of reducing frustration.

It’s Your Game

I’m going to give some game design advice. You don’t have to take it. Game design isn’t an exact science and I can’t prove my approach is better than yours. It may not be, for your game.

Some players don’t mind being frustrated; they might argue that it’s what makes the game interesting, or ‘difficult’ enough. Certainly the tension of not knowing to what extent the protagonists will succeed is important: if success is guaranteed you have no conflict, and hence no drama.

(I’d argue that those people should try it, that ultimately they can feel challenged without being frustrated, and that the frustration of undermining player agency and character effectiveness is no good to anyone. But just as surely as players don’t often predict which things will be most enjoyable, sometimes you have to let them have what they ask for anyway.)

Read, and make your mind up.

The Psychology

When I first got into game design, logical little me tried to insist the logical or mathematical outcomes were all that mattered; that two solutions with the same numerical outcome (by whatever metric) had to be equivalent. It was a hard sell, even to myself, and it took me a long while to accept that this was because it’s nonsense.

The presentation of your game mechanics matters a great deal. Even those who fancy themselves as super-rational players of the odds will feel differently and have a different game experience when you manipulate the presentation properly, and as a game designer (or a GM) you’re in the business of delivering that experience.

In this particular dimension, picture the design task as being split into two layers:

• The raw results: how long, what probability, etc.
• Player perception of those results: what pace, how close, etc.

Players Shouldn’t Miss Critical Clues

Let’s start with the less contentious one. I’d go so far as ‘bloody obvious’ in a ‘why didn’t I think of this’ sense.

The Gumshoe system includes the directive that players don’t need to make skill checks to find critical-path clues that their characters have the skills to find. I gather there’s something similar in the latest Call of Cthulhu, although I haven’t read that yet.

It doesn’t say that you can’t ask for the skill check anyway, and have a better clue or some other benefit for those who pass. It just says that you shouldn’t put critical path plot progression behind failable checks. It doesn’t say you have to complete the puzzle for them, just that you shouldn’t let luck withhold any of the pieces.

It doesn’t mean that you have to give your computer-illiterate party the clue that’s stored in the villain’s mainframe. However, this is a sign that the adventure doesn’t suit the party; as GM you should adapt the adventure or give the party more guidance on what abilities you’d like them to bring.

Some of you (especially those running mystery games) may have had a principle like this for some time, but it was an eye-opener for me; one of those things that I’d always needed but never quite thought hard enough about to find. It’s one of the best things about Gumshoe, but there’s no reason not to incorporate it into whatever else you’re doing.

Players Shouldn’t Miss Opponents

This one is more contentious. MMOs in general have been getting ‘friendlier’ for some time now, and it’s telling that World of Warcraft is one of the most successful and also one of the most ‘dumbed down’ (meanwhile EVE Online continues to soldier on, proving that you can still run a hardcore game provided it can survive on a smaller market).

The principle is this: the outcome set for a typical RPG attack is {miss, hit} or {miss, hit, critical hit}. Tweak the rules a little to change it to {hit, strong hit, critical hit}.

A Redesign?

It needn’t be very complicated, and you don’t need to redesign the whole game (although there are probably some neat opportunities if you wanted to take this as the basis for a new system). For a very quick fix that will only mess with combat duration slightly:

• Double hitpoints for everyone and everything
• Keep all the same attack rolls and hit chances
• Add a normal attack’s worth of damage to all attacks
  • A former miss is now a ‘standard hit’, with regular damage
  • A former hit is now a ‘strong hit’ with double damage
  • A critical hit does an additional regular damage; depending on system (and sometimes even weapon, etc.) it might now be triple damage or thereabouts
• Give similar treatment to spells, etc.:
  • Enemies can save for standard damage, or fail for double damage (although more about ‘enemies fail’ later)
  • Effects that never had saves now do flat double damage

This will keep combat to approximately the same length. It may not be exact, depending on how frequent critical hits are and how much extra damage they do. If you’re mathematically minded you can probably analyse a given system and work out exactly what to change, but since the duration of combat varies with dice rolls anyway this is almost certainly close enough.

Window Dressing

The main benefit is the psychology: players don’t miss through bad luck and waste their turn any more. This is particularly important when the player is spending more than just the time: spending a spell slot for no benefit because of a saving throw is especially annoying, particularly at lower levels (when you have fewer to spend).

Some players may try to nit-pick and claim that this is just window-dressing. My response is two-fold:

• It is, but most players will be taken in by it
• My experience as a game designer suggests that the feel-good psychology of making ‘less progress’ rather than ‘no progress’ still works even when rationally you know that you’ve still made minimum progress, and that even whiners may appreciate this, even if they’d never admit it

Of course, your mileage may vary and I can’t promise that it works on everyone. Some players will take great offence at the ‘dumbing down’ of the game. You know your party.

An Objective Difference

It’s not just appearances, though: there is a potentially powerful difference in the mathematics. On average combat will last approximately as long; you also haven’t changed the minimum possible time by much (actually we’ve made it longer, because in the case where everyone hits critically all the time we’ve increased enemy HP by 100% but damage by approximately 50%, thereby increasing combat duration by 33%).

What we have done is set a maximum combat length, because even if everyone ‘misses’ every time (and we’ve all had those days), combat will still end in only twice as many turns as if they always managed ‘old normal’ hits. In the old system that combat could have lasted indefinitely.

Sure, on average combats will end in reasonable time, and in practice no combat can last literally forever (although I’m sure I’m not the only GM to have fudged NPC health to bring a more dramatic end to a fight that was starting to drag). But suppose a game is balanced for hitting half the time (as D&D has been in some editions), then during a three-round combat there’s approaching half a chance that a party of four contains someone who missed every attack.

Note that crucially we haven’t changed the chance of the party winning the combat. Although there’s now no chance of the enemy lasting forever, we haven’t significantly changed the chance that the party dies first.

Some players will object to this change , because there’s always someone to object to everything. I maintain that missing isn’t fun and that anyone who tells you they enjoy missing can probably get used to the adjusted system (and masochistically savour those ‘suboptimal hits’). I can’t really see how removing the possibility of combat lasting basically forever is to anyone’s detriment. Know your party.

The ‘Average’ Player

It’s another great fallacy of the maths of game design (along with ‘mathematical equivalence is sufficient for equivalent experience’): that there is always an ‘average’ player/party/combat.

Let’s be more precise here: our normal average is the ‘arithmetic mean’, the sum of the values divided by the number of outcomes. The mean value is not guaranteed to be a valid data point: we can see this from dice, most of which have a decimal mean that the die can’t roll (3.5 on a d6).

You can model an ‘average’ (mean) player, but you need to remember that in practice you might not have a real player who actually fits the profile. You can consider a ‘typical’ player, either using the median (the one in the middle if you put them in order) or the mode (the kind of player that appears most frequently). You also need to model the outliers and check what your system does for the worst/unluckiest player and the best/luckiest one.

The change above does its best to maintain the mean experience while improving the minimum experience. If you think in these terms (and can manage tricks like this) you can make big improvements to a game design that most players will never notice. In a table-top RPG there’s a valid question of whether you should add complexity to the system (which is a cost in most cases) for a benefit that many players won’t see (and some reviewers won’t understand). In a video game I propose that this complexity is a cost of doing business, along with explaining it to colleagues who don’t see the benefit.

Psychology and Presentation

One last note about presentation. You’ll notice that I’ve labelled the outcomes as {hit, strong hit, critical hit} rather than {poor hit, hit, critical hit} or similar. This is an important part of the illusion: you want a situation where the minimum result is OK and anything more is a bonus. Players know that ‘hit’ is the worst they can do, and some will beat themselves up about not doing better: they don’t need your help.

It’s a challenge to maintain this kind of framing, but worth it. You don’t need to worry about underplaying the value of better results, because that’s already built into the game mechanics. We’ve made a big change so that ‘no progress’ has largely been replaced by ‘some progress’: you just need to find all the tricks you can to make that minimum progress an acceptable default.

This ties in with a persistent argument about ‘participation trophies’. Perhaps I’ll talk about them another day.

Grounds for Missing

Some enemies might be evasive, or have some kind of stealth technology. In this case you can introduce a chance to be dodged; not to miss, but to have the enemy successfully downgrade the player’s hit. That’s fine provided it’s an unusual characteristic of some enemies; if you make it a common feature you’ve undone the changes we made. Note that it also softens the upper bound on combat length, reintroducing a probability that combat can last longer than desired; keep that chance manageable.

Sometimes players might choose to do daft things, things that probably shouldn’t work. Feel free to give these a chance to fail, perhaps the old ‘failure’ chance. It’s up to you whether anything can be salvaged from one of these failures: players should understand that they are risky.

Both of these points come down to a question of who is responsible for failures. I will talk more about that another time.

Honorable Mention

Another brilliant mechanic you should be aware of (which I have less to say about today and can’t quite shoe-horn into the theme of the article), is that of Apocalypse Engine’s move resolution, where the outcome space is {failure, success with complications, outright success}.

The failure has a little less emphasis than some games (42% for a neutral character, 28% with even one point of bias toward that move); it hasn’t been removed completely but many of the moves are narrative tools for which base success like I’ve been describing doesn’t make sense. The stroke of genius is that many results (almost half of them, depending on the value being rolled) include some player progress with added plot-driving complications.

Agency and Responsibility for Failure

Underlying these ideas are some thoughts about managing who is responsible for failure. I’ll post them next time.