On 24th May, the future Edition of D&D embarks on the next stage of its journey toward completion. This is when Wizards of The Coast begins the first Open Playtest of D&D Next. With that in mind, it’s worth looking at what playtesting involves. After all, if you are going to be playing a part in the future direction of D&D, perhaps it’s worth knowing what you are getting into ahead of time.
Technopedia defines playtesting as:
Playtesting is a method of quality control that takes place at many points during the video game design process. A selected group of users play unfinished versions of a game to work out flaws in gameplay, level design and other basic elements, as well as to discover and resolve bugs and glitches. In addition, the process mainly involves clarifying the vague points, adding fun elements or reducing boredom, balancing the victory situations, and so on.
When D&D Next launches, I expect blogs, forums and Twitter to erupt with people complaining that “this doesn’t work”, “it’s all broken”, “bring back 3rd Edition”, “5e sucks”, etc. Every single one of these people are entirely missing the point of playtesting. The objective is to find out what does and doesn’t work, not to give something for the interwebs to gripe about. Don’t be That Guy.
Playtesting is for the benefit of the designers
It’s not about you, dude. It’s about the game. The goal is to improve the game by getting as many eyeballs (see Linus’ Law) as possible on the rules so that the vast majority of issues are fixed ahead of time. This means actually playing the game rather than merely reading it (though that can be quite helpful too). All too often RPG rules can read just fine, but when it comes to playing the game there’s too much room for misinterpretation or confusion. If something isn’t clear or you think may be improved by wording it differently, say so through the correct channels.
One common goal of playtesting is to fine-tune the user experience. In a FPS computer game that may mean refining the heads-up display so that important information is clearly visible and in your eye-line at all times. In pencil-and-paper RPGs it’s more nebulous, but no less important. Is the character sheet easy to understand? Do the rules permit you to participate in every session or round of combat? Are you having fun or feeling frustrated, and if so, why?
Feeding back your positive user-experience is (I’m sure) very rewarding for the game designers to hear, because it means they’re doing something right – but it’s the negative user-experience they will want to hear about, even if it’ll give them headaches. These are the elements which need sorting, and that’s why you’re playtesting in the first place.
You help them by giving them more headaches. Trust me. They will thank you for it in the end.
Playtesting usually has a fixed purpose
Most (but not all) playtests come with a set of instructions or a request of which features the designers want to see if you can break. I suspect the Open Playtest for D&D Next will be no different (though I have no specific knowledge, and could be wrong).
It could be as simple as “run through this adventure and tell us what problems you had”, or as specific as “kill this roomful of goblins 20 times. Did combat speed up as you became familiar with the rules and what was the average time your last five combats lasted?”. Again, I dunno, but that’s how I would do it (disclaimer: I’m a business analyst. Testing things is what I do. I know this stuff).
Sometimes it is to the benefit of the designers not to state their purpose. This eliminates the problem where actually stating the goal causes the problem itself. For example, if the designers asked “Did the fight with the Ogre take too long?” they are more likely to get “Yes” replies than if they say nothing, but listen out for comments about the length of the Ogre fight. Asking the question sets a degree of anticipation in the players’ mind, which then increases the likelihood of the perceived issue occurring. Let’s call it Quantum Playtesting, or something.
The initial Open Playtest will contain pre-generated characters, and for good reason. This means every single playtesting group will be on exactly the same page, both literally and figuratively. Your Rogue will be the same as everyone else’s so if there’s a problem with a certain ability or feature, it should be widely reported. A single group having an issue may well just be a blip on the radar or a misreading of the rules, but if 500 groups find the same thing is a problem, that’s a whole different kettle of orcs.
As the playtest continues, I expect to see playtests of higher levels, character generation and all of the various subsystems which make up D&D Next. Fingers crossed, Open Playtesting will carry on after launch for future books and modules. That depends on you, willing playtester, and the success (or failure) of the Open Playtest itself.
Playtesting only works if you provide feedback through the established channels
The most important aspect of Playtesting is feedback, and this has to be through the communication channels stated by Wizards of The Coast. Putting your opinions and thoughts about the game on your blog, Twitter and forums is one thing, but before you do that PLEASE feed it back to the Coastal Wizards in the format they ask.
I doubt the designers have time to spend their days surfing the net looking for comments about D&D Next wherever they may be. If you are going to play by the rules, play by the rules. That will make the designers’ job so much easier, and guarantee that your feedback will be heard. Putting it up on your blog won’t.
Playtesting will make the game better, and that is to everyone’s benefit
If you care about D&D, regardless of Edition, play style or preference, please take part in the D&D Next Open Playtest and help to make it a game to be proud of.
This is your chance to shape the future of the hobby as a whole. Why not participate, and be a part of gaming history?
Thanks for listening.
I would also expect that they would send out different variations of the rules to different people. That will let them see what works and what doesn’t in a faster (parallel) way than just iteratively changing it and re-testing. They should have plenty of testers to do that kind of thing. I would not expect that the playtest I’m doing would be exactly the same as yours. (this would have the added bonus of emphasizing to people that they’re not just going to do things whatever way they thought and playtesting is just marketing… )
I would be surprised if any differentiated parallel testing takes place because of the delivery mechanism. I think they will keep things simple and offer the same downloads to all – but they might ask if different groups can focus on one area or another under the same rules set. That said, anything is possible :)
Excellent post. I playtest for companies and I also run playtests of adventures very often. There is a real art to both asking for and providing feedback. It is very easy for playtest feedback to be of very limited value. When I find a great playtest group I go to them time and time again, because it is so much better. This includes some of what that Borderland article mentions: it is less important to hear why something is wrong or a possible fix and far more important to get details around what took place. An adventure can seem way too hard for one group, way too easy for another, and perfect for the third (seen this many times). Any of those groups could react with pleasure or anger and miss the point. When they provide enough information I can find the issues (the party that was overwhelmed had 4 players and the domination effect from the statue was too strong; the table that had too easy a time took out the evil wizard in round 1 and without the wizard the combat becomes easy). Once the issues are identified, solutions can be found that don’t just please one group but will please most groups.
Also, I cite my early 4E experience. When I first previewed it at D&DXP, I didn’t like it much at all. I had to play it a couple of more times to see its potential and then it slowly became really fantastic as I learned to work with it. Playtesting often requires staying level-headed despite initial reactions. Maybe the rules turn you off when reading them, but are great in practice. Maybe the reverse is true. If you want to contribute to a better game, that is best done via patience, good analysis, and reporting what you found in detail, without bias, and without trying to be overly prescriptive.
Thanks for this article. For those of us who’ve never participated in a videogame playtest, it’s good, useful information on what to expect. Let’s hope playtesters can keep the criticism constructive and funneled through the proper channels.
You forgot “Bring back 4E.”
Good points, all. I’ve only hung on the peripheries of a couple of producer-sponsored playtests in the past and discovered that quite a bit of the “this sucks, bring back X” commentary floated around the groups I was with. Something I can’t say being very helpful.
Now, that being said, my specific game group does a lot of playtesting of games just trying to decide if we want to run with a game or dump it. We found with our initial playtests of 4e years ago that running that with pregen characters just didn’t work for us: we hated the experience all around. Later, when we had the chance to actually design our own characters, we understood the design mechanics and 4e has since become our game of choice (though we have been working with a few comic book hero RPGs lately trying to decide which one we want to use as our go-to game for superhero action).
Thanks for the post and a very clear set of guidelines for feedback.
It’s not good if the instructions are _too_ specific, though. For example, if there were a version (let’s call it ErouF four now) where playesters are given a combat scenario to run and then asked about that miniatures battle, but not asked about, oh, let’s say roleplaying, you could get very positive feedback but still end up with a game that is unsatifying to most (potential) players even if there is a minority who really like it.
A company I worked for previously for had a habit of unit testing features by people who knew exactly how things were supposed to work, and the testing showed that the features worked perfectly – but those features often failed miserably when they were deployed and normal users had to use them.
Games Workshop was even more notorious for this kind of thing – the designers bragged about “playtesting” things themselves, often not even using the books because they knew in their head how they planned for things to work, including playing specific armies instead of seeing what would happen with different lists. Literally within hours of a new book being released the players would find dozens of ways in which the new rules were broken, to which the reply was usually “well, that’s not how we played it at the studio” – not exactly useful to the thousands of players who were stuck playing by how the rules were printed.
For beta testing it’s often better to make it an open-ended test – give the users the tool, and let them use it however they want. That lets you see how it works in many different circumstances, and how well the features work together. The users as a group can always come up with more things to try than the developers can. With a roleplaying game, you could probably get better feedback by giving the full game to five groups (completely unrelated to the development effort!) and letting them run a series of adventures (at least some of them created by the DM instead of prepackaged modules) than you would by giving a hundred groups pieces of the game to try in isolation. And the “completely unrelated to the development effort” is essential – if they already know how things are “supposed” to work, that is going to influence the way they play and weaken the test.