Wednesday, September 25, 2013

Designing for Low-Skill Players: Get a Life

Many instructions for games begin with that magical phrase: "Object of the Game". Players need to learn the goal before the rest of the rules because that gives them a framework for understanding the available moves. In Magic, the goal is to reduce the opponent's life total to zero. 

Because the goal is to get in twenty points of damage before the opponent, it is convenient to think of the life totals as a "score" for the game. If you're up 20 to 10, you're winning by a lot. If you're down 12 to 14, you're losing, but not by much. This way of thinking is applicable to virtually every game with a numerical scoring system. (Blazers leading the Knicks 90 - 75? Good day in Portland.) 

It is also completely and utterly incorrect when it comes to Magic.

In Magic, it is common for a slower deck to lose 10 or more life while inflicting no damage, stabilize the board with a fatty or two-for-one, and go on to win. Life totals are only a part of the game state. Cards in hand, permanents on the battlefield, and contents of the libraries are much more important for determining who is ahead in any given game. However, evaluating these factors takes foresight and experience. LSPs usually give life totals undue weight in deciding who is winning.

What does this mean for designers? Quite a lot!

Believe it or not, there is a reason
this keeps getting reprinted.
Life gain is not dumb. Cards like Angel's Mercy and Bountiful Harvest serve an important role for LSPs by allowing them to directly change the score of the game. Tournament veterans may deride them as "chaff", but LSPs need these cards.

Going to the dome is awesome. To a skilled player, Lave Axe is a fringe card for aggressive decks. But for LSPs, it's a card that always brings you a step closer to victory.

Draining life is the best of both worlds. Bloodhunter Bat and mechanics like Extort play double duty with the above desires.

One-drops make you win. Well, they make you win temporarily. A draft deck running multiple copies of Merfolk Spy and Suntail Hawk will often be ahead on life in the early game, which feels like winning from the LSP's perspective.

Losing life is terrible. Cards that damage their controller, or require life payments, feel much worse to LSPs. Use them sparingly, especially at common.

I'll conclude with an anecdote: in game three at one memorable prerelease, I was sitting on a full grip of cards that I hadn't had time and mana to play. My opponent got me down to a low life total with his last creature before it fell to removal. Amusingly, we were both disappointed when time was called: he, because he was "winning" 20 - 3, and I, because my hand was full and his was empty. For me, that moment struck home the fact that players of different skill levels have shockingly different notions of what it means to be ahead.


  1. Good points. My question is, what does all this mean in terms of how you design Angel's Mercies and Lava Axes? Obviously you have to design some of them, or the new player will be disappointed. But what is the logic behind making them "weaker" in comparison to other effects, as Wizards currently does? If anything I'd expect Wizards to be pushing their power level, so that LSPs don't have that feel-bad moment when a more experienced player tells them they're playing 'bad cards'.

    I suppose one explanation is that Wizards wants life gain / life loss cards to be a learning experience, not a trap: beginners will gravitate towards them, realize that they're not so good, and move on. But then nobody will like these cards, even the beginners. Right?

    1. Sometimes Wizards does print life gain cards that are playable to keep HSPs on their toes. Rest for the Weary did nothing in some decks but was too efficient a life swing to ignore in others. They also commonly staple life gain to cards that are playable anyhow. Most lifelink card fall into that category.

      But most of the time, yeah, the point is to make objectively weak cards so that players have a path of lessons to learn, from the painfully obvious, to the inobvious, all the way up to the questions with no definitive answers at the highest levels of play.

    2. Cards that gain life incidentally are often quite strong. (Loxodon Hierarch, Baneslayer Angel, etc.) It's usually cards that do nothing else that suck. How much life would an Angel's Mercy have to give you before it was playable in standard? 10? 13? But if you make pure life gain strong enough to be viable, you're pushing tournament games away from ending. Nobody wants to watch a Pro Tour finals between two decks running four copies each of Angel's Super-Mercy.

      Lava Axe has the opposite problem. If you print Lava Super-Axe, either you have to nerf the rest of red, or accept that games against aggro/burn decks will often be decided by the opening draw. I was around when Goblin Grenade and Fireblast were standard-legal; those were dark times indeed.

      The broader question about the wisdom of printing "newbie traps" is a very good one. In fact, I think I'll write an entire article in this series on it!

  2. Are you going to ask for designs during these or are they solely instructive/theory articles?

    1. The latter, although I hope they do influence people's design sensibilities.

  3. My fondest memories in Magic by a longshot are of playing the 7th Edition preconstructed decks against each other. Each color had a very distinctive playstyle, and they were all balanced against one another so that they *seemed* like perfect matches (though, years later, I imagine some of them are easier to exploit in the hands of HSP, but I haven't tested this to be sure). In that incredibly limited environment, one-drops really did seem good - Raging Goblin was the best of them, of course, because you could get the first hit in! If you played first, you were winning (19 to 20) before the opponent could play ANYTHING.

    I don't know what it would take from an environment to make Eager Cadets or Fugitive Wizards into playable cards. WOTC has explicitly given up - Woodland Druid seemed so cool as a 1/2 for one, and now we simply have things as clumsy as Deathrite Shaman. Maybe in the modern day it's impossible to make a fun, balanced set that accentuates one-drops without being being Zendikar-focused on exclusively aggro. Maybe I don't properly understand power creep and should be grateful for more Delvers of Secrets and Soldiers of the Pantheon. Maybe in a Gatsby-esque way I just need to learn that you can't repeat the past.

    1. I must not be understanding your point. There are more good 1-drops now than there have been at almost any other time in the game's history: Stromkirk Noble, Rakdos Cackler, Gladecover Scout, Doomed Traveler, Champion of the Parish, Arbor Elf, Elvish Mystic, Diregraf Ghoul, Gravecrawler, Avacyn's Pilgrim...

      Or are you saying that Raging Goblin doesn't measure up anymore? That's certainly true.

    2. I like Doomed Traveler, Rakdos Cackler, and Arbor Elf (and Reckless Waif!). Champion and Stromkirk seem to be toeing the line, but Gravecrawler just seems ridiculous! A 2/1 for one you can cast indefinitely, in black?

      I don't know that I have a cohesive point, except that one-drops seem to be the most concrete data point to say that there is an appreciable power creep, and that while it's tolerable at this point, it sets a scary precedent for what Magic might be like in a few years. That's not the point of this article, though, so instead, the relevant thing to take away is that I really strongly empathize with what you say in the "one-drops make you win" section, and wish I knew a better way to resolve it.

  4. I agree. After reading about 'stealth tutorials' in Sam Stoddard's Latest Development article today, I think Core Sets need to have limited environments that focus on themes that new players presume are good, such as auras, life gain, thin mana bases, and lots of spells (I'm continually amazed how new players never, ever, draft enough creatures).