Craw Wurm Theory

A quick experiment: let's say that there are two players who refuse to mulligan. One has six forests and a Craw Wurm in his or her opening hand; the other has six islands and an Air Elemental in his or hers. For the rest of the game, both players -- whose poor mulligan judgment is only matched by their poor luck -- draw nothing but basic lands. Who wins?

Well, whoever goes first wins. Why is this a problem, you ask? Simple: Craw Wurm costs an extra mana. It's a 6-drop, as opposed to a 5-drop. For two offensive creatures with no special abilities, that's not acceptable. It's fine that one creature is better than the other -- in most real-world situations, a 4/4 flyer is better than a 6/4 -- but for a 5-drop nonutility beater to be the same abstract power level as a 6-drop nonutility beater is problematic. Here's a quick rundown of when an unopposed drop wins the game, on the assumption that it has no abilities and its converted casting cost equals its power:

1: turn 21     5: turn 9      9: turn 12

2: turn 12     6: turn 10    10: turn 12

3: turn 10     7: turn 10    11: turn 13

4: turn 9      8: turn 11    12: turn 14

Notice anything weird about that? If you answered, "Paying retail for big creatures is horrible," you're right. More specifically, power doesn't scale well -- its diminishing returns are very large. This next chart shows how many turns it takes a lone creature to chew through 20 life:

1: 20     5: 4     9: 3       ... 

2: 10     6: 4     10: 2     19: 2

3: 7      7: 3     11: 2     20: 1

4: 5      8: 3     12: 2          

This is fairly obvious, but it shows something interesting: as far as killing your opponent goes, 10 power is the same as 19 power. In fact, there are a handful of dog numbers for killing players: 6, 8, 9 and 11-19 are all "bad" powers. In other words, while you pay a cost to get that extra power, it doesn't make you kill your opponent any faster. (EDH changes this dynamic on generals -- something designers must be cognizant of on legendary creatures -- but the principle is the same. In EDH, since the magic number is 21, the "bad" numbers for legendary creatures are 8-10 and 12-20. Moreover, while the diminishing returns on power are still steep, they're not as bad in EDH due to the 40-life totals that start the game.) Poison effectively doubles any creature's power (or, alternatively, halves the "time-to-kill" number, rounded up).

In real-world situations, trample changes the ball game because, when accounting for blocks, it actually mediates the poor scaling of high power. A creature with a large, "bad" power is likely to deal significant damage even if it's blocked, meaning that it can crash into opposing duders with less of an impact on its "turns to kill" metric. Trample also causes toughness to have an effect on power -- a trample creature with high toughness is more likely to continue causing incremental damage to an opponent.

One more chart -- I apologize for so many damned charts -- but it's to illustrate one last point. Here, we're going to show how many turns it takes to kill an opponent if you resolve the maximum number of dudes every turn, limited by cards in hand, and they're unopposed. The dudes will be x guys, where x is their power and converted mana cost, like in the first chart. We'll skip the first draw for these figures, assuming that the (obviously very dedicated) aggro player is going first.

1/1s: 7 (unable to capitalize on resources in turns 4, 5, 6)
     2/2s: 6 (unable to capitalize on resources in turns 1, 3, 5, 6)
     3/3s: 7 (unable to capitalize on resources in turns 1, 2, 4, 5, 7)
     4/4s: 7 (unable to capitalize on resources in turns 1, 2, 3, 5, 6, 7)
     5/5s: 8 (unable to capitalize on resources in turns 1, 2, 3, 4, 6, 7, 8
     Perfect curve; 1/1 on 1, 2/2 on 2, etc: 5

What this shows is this -- when it comes to purely aggro duders, you want a granular a base of creatures. Many small creatures are mathematically better than a handful of large creatures, assuming they all have the same power-cost efficiency, for a number of reasons.

1. They come down earlier and interact sooner. For the most part, it's better to be causing a cascade of damage rather than a handful of explosive blows. This is a classic principle of aggro decks -- they force a lot of damage into opposing decks before their opponents can get their pricier, better creatures online.

2. It's easier to leverage small creatures than large creatures. Having two 1/1s attack in tandem halves the length of the game, from 20 more turns to 10 more turns, for 1 additional mana. Having two 5/5s attack in tandem also halves the length of the game, from 4 more turns to 2 more turns, but it costs 5 additional mana. You're buying 10 turns for 1 mana, as opposed to 2 turns for 5 mana. Moreover, because larger creatures attack fewer times, they're actually doing less damage per mana spent. Essentially, the more power a player has on the board, the less valuable each additional point becomes.

3. Small creatures pay for their efficiency in time to close a game -- but large creatures don't shave any time off of a game in real terms. Without mana ramp, a five-cost 5/5 isn't going to end the game on turn 4; it's going to end the game on turn 9. Two five-mana 5/5s will end a game on turn 8. In both situations, the 1-mana 1/1s closed out the match earlier.

As an aside, large creatures have their place. That place, however, isn't in an aggro deck. A five-mana beater is about the top of the curve for a purely aggro creature, with four mana being more realistic; anything beyond that needs to offer a truly irresistible suite of abilities. Even huge returns on power wouldn't cut it; the small creatures are doing enough of a job to cut out 20 life without some huge creature at six mana. After all, if you're planning on using Lightning Bolt and Shock to finish off an opponent on turn six, why would you care about a 20/20 for 4GG? You wouldn't, because it'll never be relevant.

So, what does this have to do with Craw Wurm? Besides Craw Wurm offering nothing to aggro, control or combo decks, obviously? Well, this is why:

This card was a star of Ravnica draft. It was generally a high pick and considered a very strong beater. There's plenty of reasons for this, among them that there weren't many high-power creatures at common or uncommon in RAV-GPT-DIS (and of the 5-drops, Streetbreaker Wurm is the heftiest); additionally, red was very aggressive in both Ravnica and Guildpact, meaning that Streetbreaker Wurm was a great capstone strategy that was hard to trump after after an attrition war. But a critical reason was that 1-mana discount. By dropping its cost by one mana, it means that it both played a traditional role of large beater -- hefty power, big-enough ass to survive combat -- and it also legitimately topped out aggro strategies.

(Another note: I'm fairly certain that Streetbreaker Wurm would have been just as good as a 5/5. There's only one other nonrare 5-power creature for 5 mana in RAV-GPT-DIS, and it's also very good -- Golgari Rotwurm -- but red/green was generally considered better than black/green in that draft. More to the point, 5/5 is actually better, generally speaking, than 6/4 -- it kills the opponent just as quickly but survives many, many more combats. [There are many more 4-power dudes than 5-power dudes, especially in draft.])

So, what does all that imply for designers? I give to you, dear patient reader, the (grand, unified) Craw Wurm Theory:

1. The point where power first sees diminishing returns is 6.
2. The high points, in terms of real turns to kill, of a 1:1 power:mana cost ratio are 4 and 5.

3. Due to the better scaling of small creatures and the needs of control decks, vanilla creatures are only good below 6 mana.

Therefore, high-power vanilla creatures -- aka Craw Wurms -- should top out at 6/x and 5 mana casting cost. Any vanilla creature larger than 6/x or costing more than 5 mana is a gimmick!

And why is this important? Because Craw Wurm is the baseline big dude. He's the only creature with power 6 or greater at common in most core sets, meaning he's the first large creature most new players will get to play with. He needs to be honest about what high-power creatures are capable of, since new players are likely to play him simply because he's huge. There's absolutely no reason not to revise him down to 3GG for his stats, or to do that and make him a 5/5.

And what's the connection to Legends? Simple -- the set's full of Craw Wurm variants, and they're almost all worse than Craw Wurm.