The Failure of Skill Challenges

I don’t like D&D 4e. The current narrative its apologists are building is that people only disliked 4e because it was sold under the D&D brand. The old narrative was that it got really good like eighteen books in if you just use MM3 stats, and I didn’t really care about that narrative, because no one reasonable is going to question my decision to stop giving 4e second chances after the first seventeen books. This new narrative, the idea that 4e is any good as a tactics game, is more troublesome, because if people believe it, they might actually think that 4e is a good model of how to design tactical gameplay. And dear God is that not the case.

I don’t really want to get into a whole rant about it myself because that would require digging up my 4e books to talk about it, and it’s been the better part of a decade and two or three moves since I last cracked one of them open. Plus, I can’t open those books without getting a little depressed, because they were part of the 3-4 years or so of my life when I lost the ability to get hyped about anything, and settled into that position so many modern consumers are in, where our ability to anticipate things caps out at cautious optimism, and that rarely. The book is made of broken promises and lies. I don’t really play Guild Wars 2 much at all for the same reason, even though it totally is one of the best MMOs on the market right now.

Anyways, since I can’t find my 4e books and don’t really want to bother anyway, I am instead reprinting the words of some other guy who examined why 4e skill challenges were a disaster. This might not seem like it’s much of an issue for the new “4e is a tactics game” narrative, but demonstrating that the team working on 4e were bad at accomplishing their goals absolutely helps build the plausibility of their tactical combat being just as bad. Talking about 4e’s skill challenges doesn’t provide direct evidence that 4e combat is bad, but the team being unable to get one system functioning properly should make you skeptical of claims that they did such an amazing job with another. The obvious follow-up here would be to actually examine combat, but I’m pretty sure the only people who care about 4e at this point are deluded fanboys, so I probably won’t bother unless someone else’s words fall into my lap and I can use it to avoid writing a blog post for a day.

Comments from our unwitting guest poster have all been italicized, but only because I couldn’t find a way to put them into Da Vinci Forward Regular font.

So as we know there are many different design goals you can have in making a rule or a subsystem. And as such it can be difficult to determine if a rule is functioning correctly. When a halfling slinger throws a rock at an ogre’s head, is the rule functioning right when the ogre goes down or when the ogre stays up? That depends on what your goals are. And yet, we know that 4e Skill challenges are a failure. Not just subjectively, but objectively. How do we know that?

Well it goes back to design goals. And for that, we’re going to take relentless Skill Challenge booster and designers Bill Slavicsek and Mike Mearls’ actual word for it. see, skill challenges are something that really excite him, and considering how infectious that excitement is, it seems that his stated design goals probably have a fair amount of resonance.

From the first discussions about D&D 4th Edition, we knew that we wanted a mechanical subsystem as robust as combat that could handle the other things PCs do in an adventure—namely, social encounters and challenge encounters. We didn’t want a system that reduced all the intricacies of a situation to a single die roll; we also didn’t want a system that failed to add to the fun of an adventure. What we did want, for the situations that called for it, was a system full of tension, drama, and risk… the very essence of any D&D encounter.

-Bill Slavicsek


Get everyone involved!
The first goal of the Skill Challenges is to keep people from feeling that their characters have nothing to contribute. That is, to get everyone trying to do something every round of the challenge rather than just sitting back and eating Doritos while the Diplomancer talks. Again.

A worthy goal. But wait a minute, Skill Challenges don’t do that, do they? Indeed, since any failure on the team counts against the team’s failure numbers, anyone who isn’t the half elf diplomancer or bullysaurus who so much as opens their mouth during a social encounter to let words out instead of filling it with Doritos is actively hurting the team’s chances. Each roll has a chance to add to the failure quote, so if you don’t have the bet roll the entire team is better off with you not rolling at all. That’s bad, but it’s actually worse than that, because in addition to relegating the rest of the team to Doritos munching, they take longer to resolve than the old system. So not only has the core objective of pulling the excluded players into the game not been achieved, the excluded characters are actually excluded for longer in real time. 

What to do instead: One of the key components to getting people to try to contribute is to make their contributions be positive, or at least neutral. That means not using up party resources to act. The party could be limited by the number of total challenge rounds, or individual characters could be knocked out of the challenge after they individually rack up enough failures. Either way, a character who was ill suited for a challenge could still pull a success out of their rounds and the team would be richer for that assistance (however minor).

Be Dynamic!
The second goal of the skill challenge was to get people to throw around different techniques round after round. “Each skill check in a challenge should grant the players a tangible repercussion for the check’s success or failure, one that influences their subsequent decisions.” In short, people shouldn’t just spam their best skill, they should be responding to the tests tactically, making different choices each round and over the course of the challenge the results of their actions should “Introduce a new option that the PCs can pursue, a path to success they didn’t know existed.”

Cool concept, right? Doing all kinds of different stuff on a round by round basis. Why doesn’t it work out? Well, he reason that never happens is because the difference in a Bullysaurus’ Intimidate bonus and his Heal check is generally more than +/-10. That means that even if next round you find out that another skill is two steps easier than your focus skill (and remember kids: there are only three difficulty steps), you’re still better off just using your focus skill again. It’s not even a question. If your focus skill could work at all, you just use it next round without fail. 

What to do instead: This is more complicated, because you could attack it from several directions. The first is the skill bonuses themselves. If you tightened up the bonuses a lot you could just tantalize people with a shot at an easier skill check and have them jump ship willingly to a secondary or tertiary skill. Or you could go after it on the resource management end. If individual skills couldn’t be used every round, you would obviously end up using different skills now and again. If skills had some kind of skill fatigue where using the same skill over and over again was increasingly difficult you would eventually want to switch over to another technique voluntarily no matter how far apart your skill bonuses were.

End Binary!
The third goal is to keep things from being a boring and static binary choice of success or failure. No longer are things just a die roll to see whether you succeed or not, there’s… stuff.

Another worthy goal. But um… it totally is binary. As things stand, it’s even more binary than rolling a d20 because you can’t do degrees of success. The challenge ends the moment you get sufficient successes, so there’s really no possible way to get more than the minimum success. Really, for all the stuff where you go round by round and make all kinds of rolls, you still only get 2 end results: success or failure. And there is nothing in there to allow you to get a better success or a worse failure. 

What to do instead: There’s no real excuse to have a dozen die rolls be incapable of generating more than 2 end results if that’s your goal. Obvious methods include setting the task to a finite number of rounds and having a minimum number of successes to count as an overall success with additional successes raising the level of awesome – or having a terminating number of failures for each participant with characters allowed to just keep adding cherries on top until they are forced to stop. In either case you could cut it short when player were just trying to get across a chasm or something essentially binary while still allowing dice to keep getting rolled during a tense negotiation to see if you could get an extra plate of shrimp out of the deal.

Other Difficulties

The Difficulty level has been discussed Extensively. With charts. A key portion of any mechanic would be to make it so that the results weren’t mathematically untenable.

It is highly problematic to call success on an individual die roll “success” while success on the overall challenge is also called “success.” The fact that “failure” has the exact same confusing double meaning is equally bad. The part and the whole need to have distinct terminology so that we can talk about them. The individual die rolls could create “steps and setbacks” I don’t even care. It just has to have a different name from the result that comes from tallying all the rolls together.

And finally, for goodness sake, whatever your system is, actually use it. When Mike Mearls describes using skill challenges, he says stuff like this:

As the characters travel through town, it is important that they all make an effort to keep a low profile. When the PCs take one of the actions above, each PC in the group must make a separate skill check. The group, as a whole, must have more successes than failures in order to succeed overall. Otherwise, the group fails (including on a tie).

The PCs can each use a different skill, provided each skill is allowable for that action. Each PC can also aid one other PC. One PC can receive aid from more than one ally.

-Mike Mearls

I man seriously, what the heck is that? It’s not recognizable as a skill challenge out of the book. Which basically tells us what we’ve always known: that the designers just did random stuff and never even paid lip service to the skill challenge rules they were actually writing down. Don’t do that. If you come up with something that seems to work better than the original methods, you should write that one down. You should not publish something that has little or no relation to the rules you actually use in your game that seem to be working.

Adding on to what was written above: The weird thing is that what Mike Mearls is describing sounds like a reasonably serviceable skill system. It still has the problem where the party is going to wish they’d left the Fighter home as soon as they need a Stealth check, but since it requires the Fighter to participate, that’s pretty much okay. I’d prefer a system where the worst thing the Fighter can do is nothing, but there’s perfectly good reasons to go with the alternative. So why didn’t they actually print this version of the rules? Maybe it was in one of the umpteen skill challenge revisions they posted, but that really just gets us into another problem, where the 4e team treated D&D like it was an MMORPG where you can push an update to make minor balance tweaks and your audience will mostly not notice. Every rules tweak is more workload for both the GM and any halfway competent players. You know that guy who doesn’t know any of the rules for playing his own character? No one likes that guy. If what those rules are is quicksand under your feet, though, it’s hard to avoid being that guy.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s