Friday, December 2, 2011

M13 Is Lair Good Enough?

While discussing how best to use lair in our core set, some fundamental concerns about the mechanic have been raised. That's great and I'll tell you why. First, it means people are thinking critically which is the very thing I asked them to do and the act of such thinking implies learning. We are realizing not only features (good and bad) of lair, but also of core sets, Magic and game design in general. Second, being aware of potential flaws gives us the opportunity to address them in design or—should they prove logistically insurmountable—back up and try something else entirely.

I don't want to keep you in suspense, so I'll let you know right now that I am not abandoning lair... yet.

HavelockV wrote "I'm afraid I now believe that Lair is not a core set common mechanic" and explained that lair will create overly complex board states and cause confusion among players. His concern closely echoes Greg Marques' original concern with the mechanic. That's exactly the concern that we playtested against in September before deciding to proceed with lair.

What my personal and concrete experience with lair shows is that remembering whether lair is on or off is trivial—exactly as simple as remembering whether landwalkers are unblockable or not—even for newer players.

That testing all but eliminated my concerns about tracking which lair effects are on or not. Is it possible that the playtesting that happened was insufficient or biased? Of course. If anyone still believes that on-board complexity is a problem for lair, you have been given the tools to playtest it (instantly and free) and I urge you to do so. If you find results that differ from what we've seen so far, I very much want to see that.

One big caveat is that my test only had 10 unique lair cards and 5 unique lair abilities. Even if they came up more often since the decks were small and I was running double of the lair cards, adding 5 uncommons and 5 rares to the mix, for example, all with unique abilities may well contribute—in occasional extreme scenarios—to an overwhelming board state. It seems comparable to the use of transform in Innistrad, but it's something to watch out for.

The metalcraft analogy is useful. Dan is right that we can't take it as 1:1 because metalcraft was designed for an expert set where the goal was to deviate from the norm rather than establish it, but there's still relevant data to mine. Of the 12 commons in Scars four have the same +2/+2 ability, 3 were ETB effects, 2 were spells, there was 1 static ability, 1 tap ability and 1 combat trigger. Actually, that's more variety than I expected. Uncommon adds three more unique abilities while rare and mythic add seven more unique abilities.

That's 19 unique abilities among 22 cards. I'm legitimately surprised to see that much variety. Apart from the core/expert difference, the other big difference is that there is only one condition for metalcraft, where we are proposing five different (albeit closely related) conditions for lair. Even so, knowing whether a player has a forest and therefore whether his Kird Ape is big or not is easier than knowing whether his Carapace Forger is big or not (because it's a boolean test rather than a count). Lair also changes much less frequently, so once you've noted that Crested Dodo can fly, you never have to think about it again unless you play land destruction, in which case that change is probably foremost in your mind. What I can't say for certain is how easy it is to remember what your Mastodon Calf, Champion of Purity and Avatar of Sylvan Angels from Rivendell (or something) all do at once. Based on the variety in metalcraft, it's probably not a problem, but then again it is a core set.

I'm more concerned whether lair is flavorful enough, interesting enough and exciting enough (as pointed out by @Kirblar024. While we knew when we chose lair that it was medium on the flavor scale, we've recently identified some disconnects that might push it toward light. Still waiting for feedback on Seastorm Moccassin: Is that any better than Vaportrail Imp?

The existence of lair in the set will affect the way people draft and build their decks, and that's interesting, but it won't really affect how they play the game. Not like bloodthirst, anyhow. Is that enough to give 2013 a unique feel? Maybe. I'm not sure and that concerns me. The fact that it's threshold-1 makes it a fairly casual mechanic: It's always nice to have, but doesn't often make or break a game; It's usually pretty easy to achieve; The fact that it's counting land, something your deck needs regardless, means you don't have to make many hard choices like cutting spells or running otherwise dead cards. I guess I kind of see it as a cherry-on-top. If the rest of set is rotten, lair won't redeem it, but if the set is tasty, lair will be that last little sweetness to top it off.

While it's not visibly a cost-reduction mechanic, one could look at lair in that way. Rather than pay 4W for a Siege Mastodon, we can pay 3W for Mastodon Calf, a conditional Siege Mastodon. We could pay 4W for Castle Raptors or we can pay 3W for Crested Dodo, a conditional 3/3 flier. If it's enough to increase the bonus for lair to make it more exciting (like metalcraft giving +2/+2 instead of +1/+1) we can do that, but if the mechanic just feels like a cost-reduction mechanic, then it's never going to appeal to Timmy or Johnny. It's worded as an all-upside mechanic, and so far that's how I've seen players react to it, so we should be okay here, but it's something to watch out for.

MaRo's article on threshold mechanics relating to metalcraft is pretty relevant too. Our condition is pretty easy to reach. Not sure whether it's more or less easy than requiring 1 artifact (since those require fewer color commitments rather than more) but it's plainly easier than requiring 3 artifacts. If the hoop is too large, the prize can't be that big and the whole process feels futile. Our testing showed that a WU deck with white island lair and blue plains lair might not get lair right out of the gate, but would get it eventually, while a WuB deck with white island lair and black island lair was less likely to achieve lair and, in some ways, a little more fun.

After all our progress, there's no way I'm going to dump lair without being completely convinced that it's bad. I've seen too many sweet lair cards and played too many fun games to believe it's bad now, but I will be doing a lot more testing and keeping a keen eye out for the presence of these possible deficencies. If lair really isn't good enough, the sooner we figure that out the more time we have to start over. Fortunately, this is where playtesting shines and that's why I want to finalize what lair looks like in 2013: So we can playtest in real sealed and draft games to get a more complete vision into how it plays and is received.

Finally, I want to point out that while lair is a big part of what will make 2013 different, it's still just a very small part of the set. Scry made 2011 special because it introduced the concept or returning mechanics in core sets, but it was on very very few cards and didn't affect the environment appreciably. Bloodthirst was twice as numerous and had a large impact on the nature of the set. It seems entirely legitimate that lair appear in relatively large numbers like bloodthirst but have a limited influence on actual play like scry. As long as it gives the set a unique identity and doesn't reduce the overall quality, it's done its job.


  1. About testing:
    I am looking forward to draft and sealed tests!

    Since up till now contributors have been playtesting only with their own play group and reporting about it, it would be great if the contributors can playtest with each other and discuss how the cards felt and why it felt that way.

    Will we be testing online with + Magic Workstation? That may allow many people to play with the set in progress and give feedback (such as Limited discussion forum crowds). There are many cool Limited players on Twitter that I would love to have feedback from.

  2. Is that a thing? I don't know, does it let you import a custom set? (I've also only used MWS enough to know how to export its filetype, having preferred Octgn, but I'm sure it's easy to pick up.) I didn't think it would be possible to test *together* but if so, that would be AMAZING.

  3. Yes, it allows importing custom sets. Or rather, typing in cards one by one. There are a few custom sets in there that you can draft.

    I just remembered, I think the free version of MWS allows only a few custom cards. If Octgn allows custom cards, maybe we should use that.

  4. MSE has an OCTGN export template. For that matter, it looks like it has an MWS template too.

  5. Metalcraft feels easier to notice: 1) whether my opponent has 3 artifacts in play. 2) which of his permanents have metalcraft.

    Lair, on the other hand, feels more complicated: E.g. if my opponent has an island, swamp, Godless Shrine, and several permanents that possibly have island/swamp/plains takes some effort.

    Turning off metalcraft with an artifact removal is easy to imagine. But figuring out whether to target the island, swamp or Godless Shrine with Land-D to turn off some of his lair can be @__@.