... now with 35% more arrogance!

Thursday, October 3, 2013

Rant: I Hate Play Testing

Here's another rant to go along with the rant about game balance.

I hate play testing. It's possibly caused more damage to RPGs than even the damnable quest for game balance.

I'm sure some people reading this are choking on their own frothy rage-spittle already. But it's true.

It's not the actual act of playing a game before it's finished that is the problem. I've done that myself, playing in a couple games before their publication, and play testing a couple games I wrote and discovering that they just didn't work. There's never anything wrong with playing a game and then deciding it's not fun, either abandoning it or making changes.

No, what I hate is the nefarious Cult of Play Testing that seems to have emerged.

The Cult of Play Testing doesn't advocate play testing for any sensible, like making sure it's fun and that there are no major mistakes. The Cult of Play Testing is batshit insane. The Cult believes in Objectively Good Design. There are certain numbers you have to hit to make a good game, in the eyes of the Cult. And as knowledge of Objectively Good Design advances, old games become obsolete. Dice expire, you know, and must be replaced with dice that have more sides.

Let's take the numbers racket first. Stuff like "average damage output". Sometimes, this is linked to "game balance", for example the damage output of a particular spell may be higher than the damage done by a fighter of similar level. But even in other situations, Cultists like to pick some magic number as the goal and will add or drop features of the game based on whether they hit the expected number. They don't realize that "5" isn't any better or worse than "7", depending on the circumstances.

The absurdity of this, of course, is that the numbers chosen are completely arbitrary, as are the means of testing, in many cases. Consider  the classic example of comparing Fighter damage to that of a Magic-User with a Fireball spell. A 5th level M-U -- lowest level where Fireball is available -- does 5d6 damage, or half that on a save, to multiple targets. A 5th level Fighter, though, only does 1d6 in pre-Greyhawk D&D, maybe double or triple that with a spear set for a charge. That's nowhere near the same damage output, and the *Fighter* is supposed to be the one who deals damage!

Except, of course, that the Fireball is cast only once, while the Fighter can fight all day, and in some cases multiple opponents. A Fireball can potentially kill 5 orcs with 1 HD each in one round... but so can a 5th level Fighter, getting 1 attack per level against 1 HD creatures. And then again in the second round, and the third... and later that day, during another battle. By choosing damage per second instead of damage per expedition, you artificially inflate the damage potential of the 5th level M-U.

The Cult also buys the idea that everything should be play tested. Especially an old game, because maybe it stopped working when you adapted it to post-apocalyptic dance-offs. Better be safe and design the game from scratch, instead of starting with a tried-and-true base and adding modifications until you get what you like.

What this means is that taking, say, D&D, limiting the importance of levels, changing the classes, and tacking on broad backgrounds to adapt it to modern occult horror is something the Cult considers "bad design". Not, oddly, because they play tested such a design and found it lacking, but because everyone knows D&D is a class-and-level game designed for primitive dungeon-crawl game play, and therefore you can't adapt it to mystery/investigation or any other kind of game play.

"Modern" game design is better because RPGs are like computer software or automobiles, employing new technical improvements every year. Which means you can play test an old set of rules to set some base numbers, then play test new iterations of the rules to see if you can "beat" those numbers. Old games had +1 hit die every couple months? Well, this new game speeds it up to every couple weeks! Old ability score range was 3 to 18? Version 2.0 will fill in the missing numbers to get 1 to 25, and Version 3.0 will have scores all the way to 100, with scores increasing every couple levels! And none of this limiting ability score bonuses to +1 or +3... we'll give you a +1 every 2 points, indefinitely!

And all those numbers and components have to fit together tightly. You need synergy. Everything should interact with everything else, so that creating a character is an art. And it's all about numbers. Beautiful, soulless numbers.

What a load of bullshit. Half the time, maybe even *more* that half time, even new computer software or new automobiles aren't any better than the old ones. All the good solutions have pretty much been invented already. Maybe one new mechanics idea -- just an idea, mind you, not a complete system -- will come along every 5 to 10 years.

"Game Design" is really a matter of picking stuff off the shelf and assembling it, then changing a few numbers to fit a campaign concept. Play testing, such as it is, is just to see if the numbers you picked are viable. The numbers and mechanics in RPGs are just a convenience for making up new stuff for an ongoing game; it's the non-numerical part that's the most interesting.


  1. Replies
    1. I was going to say "I wonder what that would be?" But I think I have an idea.

      Also, I need to make a "Random" icon, for stuff like the previous post (which was actually supposed to be today's post, but whatever.)

  2. Does this D&D apated for horror and mystery actually exist?

    1. Don't know, but it seems like a trivial task. I will probably include a chapter on investigation in Alternative V, and could always do an occult mystery spinoff, if I feel like it.

  3. I thought crap like your Fireball example came more from faulty theory than any actual playtesting. There's your real culprit: the armchair theorists

    What really bugs me about all these "new" ideas is that many aren't even new at all. The best (and not-so-best) ideas I've heard from 4E and Next all appeared in 3.0, and were probably around before then. Yet they're somehow touted as being "revolutionary"

    1. It probably *did* come from faulty theory, originally. But the Cult cites stuff like that as a reason to play test stuff that's already been tested -- old games -- to find a solution to the "problem".

      And yeah, very few truly new ideas in gaming. Mostly it's a matter of changing a few numbers, or rearranging steps.

    2. It's actually pretty amazing to see how old some of these ideas really are. So many things I thought were new in the 90s or later, like spell points, were around way back when the game was being published! Not that being old makes spell points any better, 'cause I still hate 'em, but I definitely see them in a different light and it's kinda cool to see how experimental even the earliest players were. It's a bit disheartening to see passages like the following in the lbb's, though:

      "Intelligence will also affect referees' decisions as to whether or not certain action[s] would be taken"

      Maybe that was meant to apply strictly to NPCs, but I wonder if that started the trend of using mental abilities as a sort of override ("your character isn't smart enough to come up with that", "your character wouldn't know that", "your character should..."). Gygax certainly didn't help matters with his alignment rules in AD&D. People went from players to actors, simulating a character who most explicitly was NOT them. This is what leads to dozens of needless skills and desires for greater levels of customization

  4. The whole concept that somehow 'getting the numbers right' will lead to a good game seems to ignore that people/humans/players play the game differently... that even if your game is the perfect darling of Numberwang fans everywhere some guy can come along and play it different... and render a lot of that time with the slide-rule moot.
    Beyond basic functionality of the rules it's all in the hands of the folks at the table.