Ferocious vs. Fiverocious

I love it when Magic revises old mechanics. 

Most of the mistakes we see out of R&D are development errors: Thragtusk should have cost {3}{G}{G}, Pack Rat reduced variance unacceptably, the combination of Sphinx's Revelation and Supreme Verdict meant two years of control decks being only Azorius or Esper, etc. However, a reimagined mechanic is a broader admission that the original fell short. More importantly, it demonstrates how to improve a mechanic that didn't fly the first time. 

These examples are exceedingly valuable. The transition from Chroma to Devotion reinforced three important lessons:

• Flavor is important.
• Ability words should mean the same thing all the time.
• Mechanics must be supported at low rarities.

And now, we have Naya's "5 power matters" theme returning as the Temur ability word, Ferocious. Let's examine the evolution and try to discover what R&D learned along the way.

4 > 5

The most visible change is that Ferocious only requires a creature with 4 power, whereas Naya's mechanic needed 5. This makes Ferocious significantly easier to attain. It's hard for a limited deck to play more than a handful of creatures with 5+ power. Also, such creatures are difficult to cast; hitting one's fifth, sixth, or seventh land drop on time is increasingly unlikely. Ferocious will turn on sooner and more reliably than Fiverocious. 

And it is indeed problematic for a threshold mechanic to be too difficult to hit. If that's the case, large carrots are needed to compensate the player for their efforts. But for Fiverocious, big rewards wouldn't have been feasible: any deck that could stick a creature with 5+ power was likely to be winning anyway. Thus, the Fiverocious cards couldn't provide attractive bonuses without causing massive swings in the game state. 

Development responded by making Naya's best common fatties fragile: both Mosstodon and Rakeclaw Gargantuan ended up with 3 toughness. That's the critical number in Shards limited; it made them vulnerable to Resounding Thunder, Agony Warp, and Branching Bolt. However, this wasn't a full solution. The common Fiverocious support cards (Godtoucher, Bloodthorn Taunter, and Gustrider Exuberant) provided meager compensation for the difficulty of casting a fatty.

This problem was exacerbated by the fact that most Fiverocious cards provided repeatable benefits. This required these benefits to be even smaller, since they could be attained every turn. (Drumhunter is the sole exception; if anything, it was pushed so hard that it was the only reason to play a dedicated Fiverocious deck.) Every Ferocious card yields a one-time reward for having a big guy around, which allows for better calibration of benefits.

The moral of this story is to set your thresholds carefully. For example, Battalion at three was excellent. At four, it would have been awkward and swingy. (If time allows, I'll write a follow-up post explaining the curious discrepancy between Battalion and Raid.)

Cards Should Work Anyway

A linear mechanic is one that says, "Play cards of type A, and you'll get some bonus for having lots." Examples of linear mechanics include Allies, Infect, and Constellation.

Here's a new bit of terminology: a bilinear mechanic says, "Play cards of type A, and also cards of type B, and you'll get some bonus from having some of each." Examples include Metalcraft, Cipher, Heroic, and Ferocious.

Bilinear mechanics need a bit more finesse than linear mechanics. They don't just need good cards. They either need a critical mass of cards of type A that are still good without cards of type B, or vice versa. 

In Scars of Mirrodin, you'll happily play any number of Myr whether or not your deck contains Carapace Forger. Dimir's Cipher spells are pretty junky if you lack evasive creatures, but you'll run Deathcult Rogue and Balustrade Spy even without Shadow Slice. Similarly, Gods Willing and Leafcrown Dryad are strong in a deck with zero Heroic creatures.

But Fiverocious failed on this front. Cavern Thoctar hovers at the edge of playability, whereas Yoked Plowbeast and Incurable Ogre are outright bad. And on the enablers side, Bloodthorn Taunter, Godtoucher, and Gustrider Exuberant are all terrible without a fatty to bolster. At common, only Mosstodon and Rakeclaw Gargantuan are significantly better, but cards that fall in the intersection of A and B are a poor way to show off bilinear synergy; in most games, Mosstodon might as well read "{1}: Mosstodon gains trample until end of turn."

Ferocious avoids this problem by creating cards that don't suck if you're missing a big creature. Force Away is an Unsummon for {1}{U}; Feed the Clan is on curve as Whitesun's Passage (remember, life gain is much more attractive to LSPs!); Savage Punch is respectable as a {1}{G} Prey Upon.

What's more, Temur's fatties are quite playable by themselves. Alpine Grizzly, Glacial Stalker, Hooting Mandrills, Snowhorn Rider, and Woolly Loxodon are all strong commons I'd be happy to have in my sealed deck, and many others are decent filler.

Center of Attention

Perhaps the biggest design difference between Ferocious and Fiverocious is the focus of the bonus. Nearly every single Fiverocious card granted an advantage to the fatty itself: it gained trample, first strike, indestructible, haste, flying, +1/+1 counters, etc. The problem with this is that it makes the "dies to removal" problem even worse! Casting Soul's Grace (or Soul's Fire, or Soul's Might) on your Rakeclaw Gargantuan was just begging to get blown out by a removal or bounce spell.

By contrast, notice that Feed the Clan doesn't target a creature, and therefore doesn't invite the same kind of blowout. In fact, none of the Ferocious cards require a fatty as a target; even Savage Punch is easy to cast on one of your smaller creatures. This is a completely different approach to design! You can collect your bonuses as soon as the creature hits the table; no more waiting around for it to attack. If you have multiple big guys, you can retain the bonus in the face of removal. And you're vastly less likely to get two-for-oned.

Conclusion

Here's what we've learned from looking at the differences between Ferocious and its predecessor:

• Set threshold numbers as low as they need to be.
• Make at least one half of a bilinear mechanic independently attractive.
• Don't build massive risks into your mechanic. (Devour, Cipher- I'm looking at you guys here.)

I look forward to seeing Ferocious in action. Temur will play completely differently from Naya, and that's a good thing.