Today's post is about an unusual topic: Luck. Luck may be an intangible resource, but it has become a major component of any game built around risk: From rogue-likes to strategy games and anything with abstracted systems. Luck can be a cruel mistress; a fact that anyone who missed a 95% shot in XCOM can relate to, but its use in game design presents an interesting debate around the player's lack of control.
Luck is probably one of the toughest things to talk about, as technically it's not a tangible element, but it is something that we all feel when dealing with anything that has to do with chance. For video games, luck's impact varies based on the design (which is a point we'll talk about in a minute,) but we see it the most in games built around abstracted mechanics like strategy and RPG titles.
The millisecond you introduce probability and chance in your game is when luck becomes an intended or unintended part of your design. The allure of luck-based elements is that they add tension to a title and take away control from the player. In turn, it allows a title to still challenge expert players when things don't go their way.
- Games built around probability and chance take control away from the player and force the player to rely on the luck of the dice
In games built around rogue-like mechanics, luck helps to keep each play varied and has been a major part of the success of titles like The Binding of Isaac, FTL and Spelunky.
It's important to note that expert-level players can use their knowledge to mitigate poor luck to some extent; presenting the very tricky challenge of creating a game that has luck elements, but still letting skill be a factor.
At its extremes, luck can be the saving grace or the knife in the back for the player, and takes us to the main point of the debate.
As mentioned, luck's impact varies based on the design of the game, and this can be a big deal for trying to play certain titles. Regardless of percentages and to-hit chances, the player only sees one thing: The result. It doesn't matter if they had a 5 or 95% chance, when all they care about is whether or not they succeeded.
- mechanics based on chance and probability can have a huge impact on the player's enjoyment, depending on how vital they are to succeed
And that chance of success can be anything from a minor victory to your entire playthrough relying on that one roll of the dice. When things don't work out, it can make the game frustrating to the point where players quit.
I've spoken about my love of XCOM Enemy Unknown in the past, but it is not a game that reciprocates.
I've had times where I missed 90% and up shots and lost full health squaddies in full cover to a critical hit from an enemy turning around the corner from the fog of war. Another case would be the Darkest Dungeon, where I went from a fully rested party to dead party members after an ambush and critical hits.
In both cases, there was nothing more I could have done skill-wise to avoid the situation, no other steps to mitigate it; this was simply the game telling me it was time to die. And this presents the debate gamers have when talking about luck in games: How much should it affect the experience?
This is why you see experiences in games built around luck being so varied: One person may never be able to beat the game and someone else will beat it on their first try without much effort. With XCOM, I'm starting to think that the Let's Players I watch on YouTube are cheating somehow, or have figured out a way to siphon off my luck for their encounters.
When designing and balancing your game, you need to be really careful when you start implementing luck into your design. With both XCOM and Darkest Dungeon, percents do matter, and the law of averages dictate that a really lucky break can be as common as a game ending one.
One other balancing challenge is when you adjust the probability in the AI's favor. Both XCOM and Darkest Dungeon tilt the probability in the AI's direction as you go up in difficulty; increasing their chance to be successful and lowering yours.
Playing XCOM on classic and up, the first two months are such a linchpin to the success of your campaign, that if you squad wipe or aren't able to get the bare necessities, you might as well just restart. That's when luck can go too far in my opinion: Where all the choice in your game are replaced with only a few (or one) outcome that will be the winning one; see also the boss designs of FTL and Convoy as other examples. I've lost count of the number of times I've had to cancel a campaign in XCOM for the as mentioned reason.
In Darkest Dungeon, all it takes is one poor fight during a run to cause the whole thing to come apart. While luck is an intangible concept, what you can do to influence it isn't.
Being able to mitigate poor luck in abstracted games is a very hard challenge for a game designer; there are two popular options that are often used. The first is to add skills or mechanics that explicitly ignore luck and probability: Skills that always hit, dodge certain attacks, etc.
- When luck becomes too much of the focus, it can reduce choice down to only a few options that will let the player the win
Normally, abilities of this power are limited either in use or functionality. In XCOM, abilities that allow you to free aim limit their use, and the lightning reflexes skill that negates overwatch was only possible once per turn.
The second is to give the player the means of adjusting probability in their favor, which can be very hard to balance. Some examples would be a skill that raises to-hit chance, improves dodge and so on. The difference between the two is that the former completely negates chance, while the latter simply influences it.
This presents another problem: If you're altering probability, then how do you know that what you're doing is working to improve your situation? Again, the player only cares about the outcome, and they will never know if that extra 10% buff actually mattered.
Unfortunately, this is one of those times where I don't have an answer, because as we've talked about, luck is something you can't quantify. The only thing you can do is playtest the hell out of your title and draw a line in the sand of how much luck plays a role in your design.
The best games manage to find a happy medium between luck and skill; keeping each play random, but still letting the player influence things to some extent. We're all not high rollers, and it's important to give us at least some shot at beating the house.
(For posts and podcasts on game design and the industry, check out Game-Wisdom, along with the new Youtube channel for daily videos. And my Patreon campaign to secure some much needed monthly funding.)