Wednesday, August 14, 2013

Lessons from Tesla Playtest #1

We test-drafted our Tesla set on Tuesday, August 6.

We used a mockup set consisting of core set-like cards as well as Infect cards to provide a counterpoint to the Mecha faction.

Click here to see what Mechas are.

Here's the list of cards we used.

» Click to open «

The Objective
The main purpose of the test was to determine the right number Mechas per pack to make them draft well. We can't actually nail down the exact numbers yet since the mechanic is still changing, but hopefully we can gain a rough sense of the ratios needed — enough to inform the world-building process in which we determine how Mechas fit into the world and what they represent.

We also wanted to see whether Mechas should be colored. If they need to be colored to draft well, that would definitely affect how we concept the Mecha's identity.

Finally, we wanted to see how Phyrexians worked as a counterpoint to Mechas. Would they provide an alternate draft-around strategy for players so that the entire draft table doesn't have to compete for Mechas? Do they produce interesting battles when they tangle with gigantic Mechas? That was another thing we wanted to find out.

The Draft
4 players were able to join this online draft. We used packs with 2 guaranteed Mecha slots per pack. It was an all-Common draft.

A variety of deck types were drafted. James Bartolotti and I both drafted a GB Mecha deck. Jules drafted a WR beatdown deck, while Hubatish drafted a UR Infect deck.

This is how the drafting went for my seat:

I start by first picking a Goliath Body, a 6/3 trample Body for 5G that turns into a 8/5 trample when combined with a Head. The front side isn't quite first-pick material, but the mechanic is so intriguing for me that I can't help picking it.

I follow it up by picking almost every Mecha that I see in my colors for the next several picks. I set my course for a Black-Green deck because those were the colors of Mechas that I saw. It seems that an unusually high number of Black and Green Mecha cards were opened in this pod.

As I pick Mecha after Mecha, I'm worried that this might be one of those all-in strategies that require blind-forcing from an early pick, as I suspected. Once I forced that first pick, the mechanic definitely made me want to secure a number of Mechas early on.

But later on in the draft there were many decision points as well, especially after I started to approach critical mass. I found myself picking cards that are good for staying alive long enough to draw and cast Mechas, rather than taking every random Mecha I see.

I pass a few late Thopter Bodies (a 2U 1/3 flyer that turns into a 3/5 flyer when combined) because they're not in my colors, even though I have the capability to splash it. Hubatish exclaims, "What? Nobody wants a Thopter Body?" It seems that the coloredness of Mechas is having a good effect on drafting. Not every Mecha drafter is going for the same cards. It also helps that their front side is weak. Mechas aren't an autopick even for the Mecha drafter, which is great draft dynamics.

In the end, my deck has 3 Deathshot Heads and 1 Heatray Head to combine with 2 Roadblock Bodies and 3 Goliath Bodies. I also have 2 Gift of the Gargantuans to help draw the Mecha parts, and 1 Morbid Plunder to recover destroyed parts.

The Games
I played a match against hubatish's UR Infect deck:

On turn 2, I lay down a 1/2 Deathouch Head while Hubatish plays a Thrummingbird.

Then on turn 4, I am able to assemble a Deathshot Roadblock. It felt pretty impressive for me as a 4-drop, but it was also balanced due to Hubatish casting an Infect creature around the same time.

Hubatish blocks my 4/4 Mecha with a Molten Necropede, then proliferates the counters with a Thrummingbird to quickly whittle the Mecha down to a 1/1 size. He chooses not to kill it the next turn though, to keep me from recovering one half of it.

He then summons a Brine Shrike and starts attacking me with it alongside the Thrummingbird, causing serious Infect damage.

I have a Heatray Head (a 0/4 with "{2}: +1/-1 until end of turn"), a Dawntreader Elk, and a shrunken 1/1 Mecha on my side, but their racing speed is pitiful compared to the flying infect/proliferate tag team. But when I get to 6 lands I manage to cast a Goliath Body (6/3 trample) and combine it with the Heatray Head.

Hubatish casts a Redcleave Goblin and attacks with everything.

I chump-block the Goblin with the poisoned 1/1 Mecha I had from earlier, taking damage from the flyers and going up to 9 poison. Then I strike back with the Elk and the Heatray Goliath (becoming a 11/2 trampler after some intensive pumping) to eke out a victory.

In the second game, I also manage to create an early 4/4 Mecha. Hubatish attacks into it with a Redcleave Goblin (2/2 haste, infect). I block, and he casts a Rally the Forces. The first strike synergizes with Infect to neuter my Mecha and save his Goblin.

My Mecha becomes a 1/1, and then dies. (I forgot what caused the death. Proliferate?) I get half of my Mecha back, but it's Hubatish's turn so he gets to choose which side is salvaged. He returns a Part that's unimportant in this board state — maybe it was a Heatray Head.

I have trouble stabilizing and keep taking Infect damage. Hubatish drops a Chained Throatseeker, but I cast a Goliath Body myself, hoping to block it. Then Hubatish casts Neurok Invisimancer and Vulshok Heartstoker within the same turn to make the Throatseeker into an unblockable 7/5 infect for the winning attack.

On the other half of the virtual game table, Jules was playing an aggressive RW beatdown deck against James Bartolotti.

James is playing defense with a Wall of Tangecord and some small creatures including a Head, trying to buy time to assemble a Mecha, but at one point Jules has three flyers on the board (it could have been three Charging Griffins), overwhelming the Wall of Tanglecord.

I probably should have reduced the number of effective flyers in the set to slow down the environment. The ground creatures weren't slouches either. In one game, Jules is beating down with creatures such as Riot Devils and Tangle Ogre. I had put those creatures in the set hoping they could create some temporary stalls due to their high toughness, but Jules' deck demonstrated that they could create a brutal offense if you curve out with them.

At another point (maybe in the second game) I see him sacrifice a Fanatical Ritualist to power out a Flameborn Hellion.

James manages to assemble a Deadshot Roadblock (4/4 Deathtouch, {1}{B}: Regenerate) over the course of the match but it isn't enough to turn the tides.

Interpreting the Draft
Now that I've recounted the draft experience, I would like to try to interpret what this means.

First of all, I need to acknowledge that this was just one draft. I drafted a Mecha deck and it worked, but that doesn't tell us anything conclusive about whether the number of Mechas in packs were adequate.

For example, we happened to have a 2-2 split of Mecha vs non-Mecha players, but I'm sure drafting Mechas would have felt very different if this was an 8-man table with a 5-3 or 6-2 split. Also, the packs we opened contained an unlikely concentration of Mechas in two colors, Green and Black.

While one draft is affected by too much variance to use as conclusive evidence, what I can do is try to glean observations about what aspects of game play were fun when they came up, what aspects were not, and what factors contribute to those aspects.

Observations on Mechas and Infect:

The Mechas were powerful, but could be handled.
The Mecha sizes felt powerful. With stats such as a 4-drop 4/4 with Deathtouch and Regenerate, or a 6-drop 8/5 with Trample and +1/-1 pumping, they were above the level of what would be acceptable for Uncommons, although not quite bomb Rares.

Despite their impressive sizes, they weren't overly hard to handle, mostly because of the presence of Infect in the set.

 Infect was a good opponent for Mechas.
As mentioned above, Infect was the main reason Mechas could have such unconventional stats without being broken. I believe Common cards should be answerable with other Commons, and not just with an elite group of Commons like removal. Infect provided those Common answers.

We should still try other villains (suggestions are welcome) but Infect is definitely a strong candidate.

 Over-Aggressive flyers are not a neat opponent for Mechas.
In the test, flyers gave the Mecha deck less time to assemble Mechas. The Charging Griffins should have been something like Makindi Griffins. Also, I probably shouldn't have placed multiple 2-power Infect flyers in the set.

We do want some amount of evasion and steady clocking. In Rise of the Eldrazi, the evasive Levelers created a good, tense race with the Eldrazi. We just don't want the damage output to snowball every turn.

It wasn't just about the speed, though. An Evasion vs. Mecha battle provided less opportunities for Mechas to crash into opposing creatures on the ground and show off their abilities.

I would have liked to see if a ground battle would evolve into a slugfest where Mechas die and get rebuilt very often. Unfortunately, both of the matches in the draft involved an Evasion vs. Mecha battle. We definitely should have tried a Mecha vs. Mecha matchup.

Of course, if more Mechas have flying, it could be a different story. I'm not sure whether they should or not, but there's no reason there has to be only one flying Body.

 Colorful Infect synergies were cool.
I loved how Hubatish's UR Infect deck performed. Red and Blue effects formed interesting synergies with Infect cards. I would like to see each color pair get its own style of Infect decks. (I've already attempted that to some degree in the draft set list.)

Observations on Builder Satisfaction:

 It feels bad when you have to trade the Mechas before they combine. 
You don't want to get into combat with small Mecha parts because you want to save them and combine them. To fight this problem, the testers discussed some tweaks where you join Mechas from the grave, which I will discuss in another post.

Some of the Mecha stats could have been adjusted so that they don't need to trade and die in order to do a lot of work. The 2-drop 1/2 Deathtouch Head could be a 3-drop 1/3 Deathtouch Head so that it can hold off multiple creatures without trading.

 It feels bad when you're forced to combine Mechas in a combination you didn't want.
In one game, James had the goal of combining a Deathtouch-granting Head to a 6-drop 8/5 Trample Body, but he didn't have the time and circumstances forced him to combine it with a 4-drop 4/4 regenerating Body instead, which was frustrating.

Maybe there could be an activated ability to un-combine the Mechas so that you don't mind temporarily joining Mechas in a combination that's not optimal.

However, this is also what the ability that salvages half of a destroyed Mecha is for, and I would like to further examine how that plays as well.

 You don't get to choose what combination you get each game.
I like how each Head-Body combination has different strengths and is good in a different situation. Unfortunately, that aspect is not as cool as it can be when you're not in control of what combination arises over the game. In most games you just assemble what you happened to draw because drawing one Head and one Body is difficult enough. You don't get a choice between multiple parts to use.

This might not be a problem. It's good that different things happen in different games. But it would be cool if you could customize your robot to meet the situation at hand by tutoring for a silver bullet Head or Body, at least some of the time. I hope to try out some tutoring methods in future tests. Transmute as has been suggested, and maybe there can be an artifact version of that.

 There weren't many Eldrazi's-style Mechas.
The feedback contained mixed opinions: Some players thought Mechas were about the right size, a mini-bomb that can still be handled. But some wanted them to be more powerful, like Eldrazi, Titans, or the kind of towering robots you see in pop culture, with sizes starting from 6/6.

The closest that we have to Eldrazi or Titans at Common are a 8/5 trampler and a 6/6 that must attack each turn if able. The others have 3 or 4 power.

The main appeal I see in Mechas so far is that you get to be inventive and splice two creatures together. To me it's similar to the appeal of Soulbond, or what I imagine Contraptions to be like.

But another way of seeing Mechas is that they are a risky hoop to jump through, so that when you do get the two pieces, you are allowed to get a much huger creature than usual.

The type of game I believe would be best are the ones where Mechas get built, destroyed, and rebuilt multiple times, forming a different combination with salvaged parts each time. That seems to me like a recipe for a satisfying and intriguing game because of the choice-making and back-and-forth play involved. This kind of back-and-forth play is harder to achieve when the Mechas are big enough to kill in just a few hits. I'm glad to have Air Elemental and Juggernaut sizes for most Mechas rather than Shivan Dragon or Eldrazi sizes.

Also, if the Mecha mechanic is going to be on 40+ cards, that mechanic had better be doing work early enough to shape games, rather than appearing only in the late game to close them out.

The high-risk, high-reward version currently sounds to me like it would make games into a crapshoot. But if anyone has a vision of how it could work, I would be interested to hear it. Also, this could be the kind of thing we should try out and see what happens anyways.

I currently don't think most Mechas should be Eldrazi-style finishers. But even if we don't go that path, we should still have some Mechas become huge when combined, to cater to players who want something really big to build towards. We can achieve a balance between the Eldrazi-style late-game Mecha and Soulbond-style mid-game Mecha by mixing different transformation costs based on how huge it becomes.

High Transformation cost
Low Transformation Cost

 Too many Mechas had medium power, high durability?
Related to the question above, maybe the current test batch of Mechas has too much toughness and too little power. There can be a few more Mechas that have high power but aren't too hard to kill. It's not exactly the same as a board-dominating Eldrazi, but they are impactful and demand a response from the opponent (such as leaving behind blockers and double-blocking) and might satisfy the same type of players who want Eldrazi. They would also create more battles where Mechas proactively attack, get destroyed, and get rebuilt.

Observations on the Drafting Process:

 To support Mechas, I had to take Mechas every pick for the first few picks. 
If you decide to go into a heavy Mecha route, the first few picks are going to be about establishing yourself as the Mecha player at the table. You would probably have to pick up a Mecha whenever possible. There won't be many decision points. I'm not sure how this portion can be made more interesting.

By the way, this is one of the things that make me want Mechas to be colored. If the drafter upstream of me slams a first-pick Mecha to assert that s/he's going into that archetype, and I also want to first-pick a Mecha, at least I can choose to go in a different color.

 I didn't have to keep taking every Mecha I saw.
The Mechas weren't auto-picks throughout the draft, and there were many points where I passed up some random Mechas (especially off-color ones) and took other useful cards.

On the other hand, if I was being cut from Mechas in my colors, I would have taken the off-color parts and splashed them. My deck might have been a little clunky, or maybe I would have succeeded in changing colors. Either way, I wouldn't have been stranded with a do-nothing deck.

So the color scheme for Mechas seemed to be working.

 We kept seeing the same Mecha combinations.
Havelock was concerned that if Mechas are colored, people will not be able to mix and match many different Parts, causing the same combinations to form each game. That seemed to be happening here. In a 2-color deck, you only have two types of Heads and two types of Bodies as Common options.

One possibility is to make the Heads colored, and the Bodies colorless. The colorless Bodies can go in any deck, so the variety of combinations that arise would increase. If the Heads are concepted as some kind of power core system housing a magical crystal while the Bodies are just lifeless mechanical chassis, the color configuration can make flavor sense as well. I still think it would be bad for drafting since it makes multiple players at the table compete for the same cards.

However, simply increasing the number of Common Heads and Bodies can also work as well.

 Mechas can work as a draft strategy.
Overall, this draft gave me the feeling that Mechas don't need to be a wonky drafting strategy. 2 Mechas per pack may or may not be enough, but it didn't seem far off.

The conditions of this draft favored me and I picked up 9 Mechas, 2 Fetchers, and 1 Recursion spell, which ran smoothly. But I don't expect decks to be able to consistently get that many pieces in an average draft.

Nothing about this test is conclusive, but if I were to take a guesstimate, I would increase the number of Mechas to 3 per pack. There could be 2 Common Mecha slots and 1 Uncommon Mecha slot. The Rare/Mythic Mechas can appear in the Rare slot at a certain ratio.

This mix also means there can be a wider variety of Mecha parts at Common. If the Common Mechas don't have to share a slot with higher rarity Mechas, we can probably have 11 Common Heads and 11 Common Bodies. With each color getting two Heads and two Bodies at Common (plus an Artifact one), we will get to see more varied combinations occur during games.

It would also allow us to make some of the Common Mechas good and some of them bad, which is good for drafting dynamics. We don't have that luxury when there's only 1 per color.

Thank you for reading this. I would appreciate any comments on how we should test, and what you thought of the results.

We will be doing another test soon, and it's open to anyone who is interested in card design.


  1. I think colorless bodies + colored heads is definitely worth trying. Someone (I forget who) thought we could flavor the heads as pilots, which I like.

    1. This mechanic hogs all the resources of the set, but is very insular and doesn't add depth to the rest of the set. Mechanically, it shouldn't exist.

      The only reason it could be justified (if at all) is because of the joining art, which can get people excited like nothing else. I think it could be done in a way that's worth it.

      But Pilots & Vehicles aren't going to benefit much from joining art. If we're going to have Vehicles, it should be done in a more open-ended way like Licids or Living Weapon. They would have to play differently enough from Equipment, but maybe we can find something.

      If a set features Vehicles (including Mecha-style vehicles), I could see there being some Pilot cards that grant bonuses to Vehicles, but it should never be mandatory. We don't want a mechanic where you need to draw a Vehicle card and a Pilot card and put them together. That has all the problems of using Mechas as a set mechanic without any of the payoff of joining art.

    2. If a mechanic shouldn't exist, we should simply cut it, no matter how cool the art is.

    3. This comment has been removed by the author.

    4. This comment has been removed by the author.

    5. Then we should cut DFCs from Innistrad, and Levelers from Rise of the Eldrazi.

      It's one thing to read a ton of wordy text explaining to you that something has changed into something else, and another thing to actually see it change. Flip cards were a failure; DFCs were not. The Art can tip the balance between being worth it or not.

      That said, DFCs and Levelers contributed to the overall gameplay of the set, while the Mechas don't do that unless we engineer an environment where everyone is playing Mechas, and it won't contribute as much to inter-block strategies. It may not be the same situation as DFCs.

  2. You hit on something that I've been thinking about a lot recently - whether the Mechas in this set are the Mirrans, or the Eldrazi.

    The gameplay of Mechas being built, damaged, and rebuilt with different parts throughout the game is interesting for sure, and is one way of building a set. It probably requires the level of commitment in packs that you've been talking about. But on the other side, there's definitely a draw to "building a towering robot", and that's at least what I would expect as a player from this sort of set.

    My guess is that we could make either version work, and it comes down to whoever is leading the design for which avenue we pursue.

    I can see how Infect would be a good natural counter to the Mechas sizes, and helps set a distinct non-mecha path with a clear reward. My worry with that is... that it's been done before? I mean, Infect hasn't been completely cleaned out of design space; but it's tied neatly up with the Phyrexians, and something I at least would like to see is Tesla as something other than Mirrodin III. This is why I'm curious about whether the robots are part of the threat (which makes for interesting man vs. mecha, mecha vs. nature, or mecha vs. magic) or robots as part of the resistance (presumably inevitably something like the Mirrans even if they're on a different plane).

    Big, colorless mecha (with certain colors better tuned to support) against a variety of colored resisters does sound very similar to Eldrazi, but I believe the way of putting the Mecha into play or operating them would be sufficiently different from the ROE experience.

    Mostly, I think combining robot-parts together is really interesting, but not 40-cards-worth interesting.

    1. You're right that this set suffers from the problem of being similar to Mirrodin.

      I've been thinking that once we've hammered out the game play logistics of Mechas, and roughly figured out the right style for them that makes them the most fun, we have some major work to do to figure out what this set is actually about.

      One possibility is to make this a prequel to the Mirrodin series. It could be about the Thran Empire. The empire was technologically advanced, and it's also where the Phyrexians were born. They successfully fought off the Phyrexians. Urza studied the Thran for clues on how to beat the Phyrexians, and Urza also built those mech-suits for the Phyrexian incursion, so maybe the Thran had mechas too.

  3. Separate from all that though, what I mean to say is, could we give the players who want to draft Mechas the opportunity to do so without making it such a heavy part of the set?

    Would the more elegant design involve different artifacts in general that combine together, some subset of which are robot-parts?

    What if we had pieces similar to the Kaldra or "of Empires" sets, but with all of them at common? You could even have "arms" "legs" and a "head" for your robot, with incremental bonuses as you construct it on the battlefield. This gives the "build-a-mech" gameplay experience on a regular basis in Limited using a small number of cards, and when you build it feels like an accomplishment itself.

    I think there are ways to do this, that either combine cards together, use DFCs, or both, that will have a lighter-weight impact on the set. I imagine that as interesting as all of the different head-body combinations are, it's too parasitic, and alternatives will be more fun overall.

    1. That is definitely a possibility.
      It could be a sideways draft strategy that supports only one drafter, like a Storm deck or a Dampen Thought deck, rather than a mainstream playstyle of this set.

      I've been thinking that if this is a postapocalyptic world where you're scavenging and salvaging old technology, there could be a whole theme about dumping cards from your library into your graveyard and then picking something up from it. Mechas could be just one of the substrategies that make use of that.

  4. Well done. Thanks for all the detail.

    Another option for opposing creature mechanic: I’d suggest wither rather than infect. It still has the type of interaction you want against Mechs without overly speeding up the environment. Maro has already said that the creation of Infect didn’t mean Wither was now obsolete. Using wither again, would just depend on what Design needs. We might need it here.

    There is one big flaw currently in both infect and wither. When a Head or Body combines, it brings its -1/-1 counters with it and the new Mech is weakened right off the bat. -1/-1 counters are always going to create this feel bad. You’ve jumped through hoops and traded two creatures for one weakened creature. The last thing we want is players thinking “why bother” about the set’s signature mechanic. Maybe Aura, Equipment and counters SHOULDN’T remain after combination.

    Instead of an un-combining ability, just include “Sacrifice an Artifact” activation costs in the Mech aligned colors. If you use them during your turn, you can salvage the part you want to keep for re-combination, and get a cool effect. This type of activation cost also helps convey the feeling you want in the Mech decks (i.e. Build, destroy and rebuild).

    1. We do want the -1/-1 counters to be an alternate strategy to pursue, so the linear-ness of Infect is actually something that we want. However, it's so hard to balance the racing speed of Infect and Wither might be easier to handle.

      I do think we can give Infect decks a kind of late-game card advantage like Innistrad's Zombies and make it a style that wears opponents out with an unending horde, while being slow in the beginning. But it's good to have Wither to fall back on.

      So far, the pre-shrinking of Mechas hasn't felt so much like a problem in itself. For the most part, the problem has been that when the Heads block, they die anyways. When the Bodies hit play, they transform because you have a Head by that time.

      Someone suggested that when Mechas join, they refresh themselves by flickering. Maybe we should try that if pre-shrinking starts to feel like a problem.

      By the way, I'm not trying to "convey a feeling" of Build, Destroy, and Rebuild. I want it to play that way for interactivity, back-and-forth play, and choice-making. If you build a Mecha, and you just ride it to victory, it isn't interactive. If your opponent interacts by destroying your Parts or your assembled Mechas, that isn't satisfying. You have to be able to actually assemble a Mecha and fight with it in most games, but they shouldn't automatically give you victory; they should die (preferably in combat). But the Mecha dying shouldn't be the end of the fun. It shouldn't be, "oh no, I'll never get to assemble another Mecha this game. Now I have to draw some flyers or Lava Axe to win." That naturally leads to a build, destroy, and rebuild type of game play. I'm not particularly trying to convey it as a kind thematic flavor of Mecha battle or introduce effects for the sake of conveying that flavor.

      That said, I felt that the set could use a sacrifice effect like Altar's Reap for creatures that have shrunken with Infect. However, if it turns out the Mechas really need a disbanding mechanism to not feel bad, we should have it on the back side of Mechas (probably as a mirror to the front side's ability, to not tax people's memory) rather than relying on outside cards to provide that utility, because they would be unreliable.

  5. I agree on the Wither front. Infect brings Tesla too close to Scars (in my opinion), but more importantly it makes the limited format linear to the point where most of the players are just picking whatever mechs/infectors are in their colors and there's little room to mix and match.

    That all said, while the shrinking mechs gameplay is good, it's not the only route. We could have a mechanic like this:

    Disabler (This creature deals combat damage to creatures in the form of disabling counters. If a creature with a disabling counter on it would untap, remove a disabling counter from it instead.)

    1. I think that if one strategy has a lot of linear synergy, the counterpoint has to have a lot of linear synergy as well. Modern Masters is balanced like that.

      It's also possible to go another way and not have any counterpoint strategy. You can make it so that you can either pick goodstuff that's good in a vacuum, or you can pick Mechas that are good if you have the synergy. Zendikar was like that with the Allies. It felt like every Ally had a non-linear counterpart in the same color (often in the same cost) that was better than it in a vacuum.

      But even with the first route, the linearity doesn't have to make it one-dimensional to draft. In the test, I didn't feel like I had to keep taking Mechas. I picked up some Blight Widows for surviving into the late game rather than take random Mecha parts. I felt that it was possible to balance it in such a way that there's room to mix and match.

      We can get players to mix Infect and normal damage while still focusing on either Poison or non-Poison damage as the main victory condition. We can do that by putting the Infect on useful creatures like Gravedigger (making Corpse Cur work with any creature, not just Infect creatures) or Giant Spider,

      When Infect and normal damage sources get played together, it produces neat games, because the game can go back and forth along multiple axes.

      In that test game we had after the main test was over, we had a game where my Mecha dealt huge chunks of damage, then you destroyed it with removal, then my infect creatures started dealing damage. Then you got rid of the infect creatures, but I started dealing normal damage again.

      So the choices are:
      Goodstuff vs. Mechas
      Mecha emphasis decks (with useful Infect thrown in) vs. Infect emphasis decks (with strong-frontside Mecha thrown in)
      A non-linear faction vs. Mechas (not sure how I can make this work)