Thursday, August 24, 2017

40K - The Playtesting Fallacy

One of the things that the "community" has a lot to be thankful for is the playtesting that went into the creation of the new edition of Warhammer 40k.

That is the myth that is being perpetrated by a section of commentators out there, be it on podcasts or blogs. Strangely - if you are a moron, that is - is that this is most commonly advanced by those that were "invited" into the process.

Roger and Bruce - Warhammer 40K's Top Analytical Minds At Work

I'm sorry to disappoint you guys but this is becoming a bit of a circle jerk. If you are going to say you were a playtester then you need to own responsibility for the errors that are emerging within days of release - who would have thought that Flyers with a high rate of fire were good; the ability to smite enemies off the board with 2 point Brimstone Horrors, what about RG sitting in the middle of a bubble of lascannon conveyors?

Either the playtesters involved weren't very good or they were only involved in the most cursory manner (in their role as Internet Celebrities). So which was it guys? Did they miss what were pretty apparent glitches - spam + stats = Good - or were they only really shown the finished product. I'm going to go with the latter, because if they were there for any significant part of the ride they didn't do the job.

Perhaps I'm being uncharitable? Perhaps they did find these holes and GW ignored them. Perhaps they've now been found right. I guess I'm an old fashioned cynic.

My guess is that they were flattered when GW came and asked them to "playtest". What they were actually offered was early access and the opportunity to "cheerlead" at the same time growing their own individual brand. Is that a bad thing? Does that make them bad people? Of course not. But let's not dress it up as playtesting.

Playtesting is an intense process designed to break the system to identify faults that can be corrected. I don't think that that is what happened here. Playtesting in the new Warhammer 40k context was just another form of marketing.

Well played GW, well played.


  1. Couldn't agree more, there doesn't seem to be much playtesting or even editing going on at the moment at GW. I for one believe the "new" GW to be pretty much the same old one but with a better marketing and customer relations department. I still don't think they have it in them to put out much in the way of a robust ruleset...

  2. This is one thing I don't miss from WHFB at all. It's another reason I'm not interested in AOS; I have zero faith in GW to write a balanced, low-error ruleset.

    Jeff's right, the new GW is just the old one with better outreach. Which isn't a bad thing, but there are so many low-hanging fruit they could pick, why they don't do it mystifies me. How hard is it for the 'Premier Games Company IN THE WORLD' to write a tight set of rules.....?

  3. I think the rules themselves are very good. Fast and fun to play.

    I just don't think they were "externally" playtested to a rigorous degree.

    As far as editing goes...well I expect more than what I got from the Codex:CSM.

    However the rules have got me playing 40k for the first time in 9 years and that's a good thing in my eyes.

    And I do love the fluff and boy have I missed it.

  4. I recall, waaay back in the day, there was quite a bit of in-house playtesting. The problem was that some of the senior playtesters cough*jervisjohnson*cough were actually not very good players. This lead to a number of units in each codex being mis-pointed. God knows what they do now.

  5. I believe as long as Kirby has any say in the matter, this issue will remain to some degree. Based on what was shared, some of the big names centered around our hobby were allowed to play in the new ruleset- to what degree was not disclosed. I know that Reece and Frankie did provide input to GW for their experience, with the game they were allowed to participate in.

    It was an excellent PR move, and, brave first step for GW- the white citadel closed off from the community. To even allow personalities famous in our hobby to participate prior to release was jaw dropping. Many, actual competitors to Games Workshop in one form or another. Moving forward, I would like to see Games Workshop double down on this, and, allow players (competitive, and, narrative) the opportunity to provide input into making the game the best it can be.

    Alix is right- JJ is notorious for knocking competitive 40K, and, echoes Kirby's philosophy regarding Games Workshop games. I support the direction that Roundtree has set for the Company, and, the communication with players. I think we are witnessing those brave first steps.

    40K 8th has a net positive resonance when compared to previous editions 6th, 7th. I am actually going to play- again (I stopped playing in 6th because of the Company's lack of support for competitive play). There is still room for improvement, and at the very least Games Workshop is listening to player feed back- also unheard of in the last 10 years.

    My suggestion is to forward your constructive criticism to Games Workshop, along with suggestions on improvement. The Company has a very talented team of designers, and a very passionate fanbase. It wouldn't hurt for passionate players to volunteer to play test moving forward, as, I believe now more than ever Games Workshop would actually welcome it.

    1. Thanks for your comments.

      We had a local group here (Wellington) back in 2003/2004 that were playtesters for GW....Late 3rd Ed, early 4th....Al Borthwick coordinated the feedback from their testing. Their names are in the various codexes they worked on. They went out of their way to bust things and provided extensive feedback on points costs and combos. As I remember it all died around 2005 due the NDA leaks.

      This was proper playtesting in that it was try this list and these set of points values versus Orks, Tyranids etc. The GW contact was Pete Haines (blessed is his name). He left mid 2005 as I met him at the first round of the UKGT that year and he was just going out the door.

      The problem with external playtesting is leaks and that's why GW killed it.

      Hopefully they will engage more fully with the external contacts they've made in this process and they'll be given an opportunity to engage in intensive testing. However I'm still firmly of the view that what we've seen recently has been an extension of marketing rather than design.

  6. An interesting side effect of their lack of playtesting/game design is that they're actually leaving a lot of money on the table. A great example of this is with Dark Eldar who have received a massive increase in playability thanks to 8th edition. The downside is you can't actually buy most of their range now (and especially not the 3-4 units that really pump the army) as GW simply hasn't made those units in sufficient quantities. If they'd actually playtested the new lists they could have forecast this.


    1. They have got some more streetwise internal resources recently. Ben Johnson for instance was the English 2014 ETC Captain. Although he comes from a Fantasy background you'd think he'd be able to give rules/codex a study and say "Dark Eldar? We're gonna need a bigger boat"

  7. You need to treat playtesting like software testing. each of the conditions needs to be exercised.

    I helped Al in a few games back when he was involved in play testing and we specifically designed scenarios to test particular conditions and demonstrate the interactions of rules and units.

    To appreciate how effective or not a unit is, you need to see it in context of all the other units and scenarios it may encounter.

    The problem is there are thousands of test cases that need to be run as part of a test phase of the complexity of 40K. Thats probably thousands of hours of test games.

    That doesn't excuse sloppy proof reading and editing errors.

    1. There are ways to reduce that load.
      Unit Vs Unit duels are largely a matter of probability, a suitable computer pogram can crunch through zillions of these in an hour testing all potential match-ups in every type of terrain.

      It certainly gets more complicated when assessing very powerful and very weak units (since they are most likely to end up in one against many match-ups.

      Computer simulation works really well where numbers are involved.
      It's less effective where vaguely defined "special rules" are in play.
      GW had used special rules heavily in the past, so perhaps this explains their lack of / poor testing.

  8. I'm hoping that the largely positive reaction they've had so far will lead to an expansion of this approach, and some genuine exhaustive 3rd party playtesting. We'll see if it ends up going that way, or sliding back into Tom Kirby style.