What Makes a Good Mechanic

Posted in Latest Developments on March 17, 2017

By Sam Stoddard

Sam Stoddard came to Wizards of the Coast as an intern in May 2012. He is currently a game designer working on final design and development for Magic: The Gathering.

I started off this week trying to write an article about Standard, but Erik Lauer's note on the banned and restricted list announcement sums up most of what I was going to say. The lack of additional cards to the banned list now isn't a reflection that we believe things are perfect with Standard, but instead that we want more information before we decide whether to make any other changes.

So, instead of talking about Standard, I'm going to talk about what makes a good mechanic, from the development side, and what we are looking for when we start working on a set handed over from design.


This is easily the most important thing for any mechanic. While it may seem like a strange thing to list, there are plenty of mechanics early in design that aren't fun. They are usually trying to do something like tie in well to the creative side, create an overarching sense of the world, or tie some mechanical loose ends together. The most common failure for mechanics early in the process is they are balanced on the axis of "if it doesn't work, I lose; if it does, I win." What development is looking for is a mechanic that is enjoyable to play with and enjoyable to play against, both the first time and the 20th time. We want mechanics that add new and interesting choices to the game and allow for some deck building choice to maximize them.

The most important thing to figure out in early playtesting is what the designers found fun about a mechanic and making sure that it is something that we can replicate in real-world environments. Often, we find that the kind of fun the designers enjoyed about a mechanic only works within a biodome they created, and it's on development to figure out if we can fix that via number tweaks or some light reworking. For example, one of the first things development did with energy in Kaladesh was to double the amount that cards gave you. Designers enjoyed the modular feeling of being able to choose which effect to use, but they were not looking at most of the abilities critically enough and were happier making suboptimal plays than the real world would be. By doubling the amount of energy we gave, we could create more granular energy payments, like allowing for cards to enter the battlefield with three energy and spend two, or even just spend one energy for a small effect. This helped to keep the fun modular decisions the designers liked, but for more of them to be "correct" in actual gameplay.


One of the things that really helps to sell mechanics is how understandable they are from a flavor and gameplay perspective. This is kind of a hidden thing, mostly because we do a pretty good job of filtering out the mechanics that fail on this one. We also do a good job of working with creative to figure out how to name and flavor mechanics that need some help. Emerge, for example, works as a mechanic mostly because it is in the "Alien meets gothic horror world" and the idea of chestburster aliens is something that can be explained in that context, while it would have felt very out of place in almost every other setting. It was a very complex mechanic, but one that got a boost from having an understandable story related.

Flying is the best example of a mechanic that works this way. It is flavorful, and even without reminder text, people can pretty much get what it is trying to do. On the other end of the spectrum, you have a mechanic like prowl that is almost entirely mechanical and doesn't really have any flavor at all. You can kind of get at how the name is connected to rogues, but nothing about the mechanic really makes a lot of sense.

Impact vs. Word Count

The holy grail for a really great mechanic is something that is both simple and impactful in games. Flying, as I mentioned above, also fits in very well here. Menace, prowess, affinity, raid, and convoke are other examples of mechanics that do this well. The goal here is that a mechanic can lead to interesting deck building and gameplay decisions, but still leave a lot of room on the card for text. The fewer words a mechanic takes up, the greater our ability to both make simple commons and create interesting effects on the cards. Obviously, some amount of this is tied into how the mechanic works. Affinity is very nice in this regard because you can put it on about anything, but the mechanic doesn't change the cards, while raid as a whole necessitates a lot more words on the card, but most of those words are the enters-the-battlefield ability, and raid only provides a hoop to jump through.

Examples of mechanics that do this poorly include kinship and fateful hour. While these mechanics can come up in gameplay, they require a lot of text on the cards and room in the set for abilities that will show up in a very small percentage of games. In the kinship example, it's a pretty random mechanic (since you can't usually control the top card of your library very well), you need to have enough creatures in your deck that share a creature type, and the effects are often just not strong enough to really make a big difference. Miracle had many of the same problems with kinship in terms of randomness, but the size and effects of the miracle ability made much more sense—also, you could just cast a Bonfire of the Damned without miracle to get a good amount of power out of the card. Leaf-Crowned Elder, on the other hand, is going to need to hit at least once to really make up for its cost. Fateful hour, similarly, requires you to be at a low life total, and most of the bonuses aren't strong enough for you to create a deck to attempt to get to that life total. Instead, you must hope that your opponent gets you to that level, and that at that point your cards have enough of an impact to be worth the words on them, which they usually aren't.


One of the things we talk about in development, in related to mechanics, is knobs. We want to have things on a card we can tweak as we are working on it. When design hands off mechanics, we try to figure out what our strategies are for balancing the cards to make sure we can do so in interesting ways, especially late in the process. That means the ability to change things granularly enough on a mechanic that we can balance the card. While this may seem simple, not all mechanics have this quality. Examples of mechanics that are very hard, or impossible, to develop are the free-spell mechanic in Urza's Saga (storm) and cipher. In the free-spell example, adding mana doesn't do enough to weaken them, and if you have lands in the format that produce more than one mana, then anything that is cheap ends up becoming a ramp spell. You can make only reactive cards, but then you will have a hard time making enough cards to justify the mechanic. Storm and cipher have the problem that you must deal with the card getting cast a lot of times, so you need effects that scale, but not too far. As a result, storm was too strong and cipher we intentionally made too weak because of all of the problems with repeatedly casting the same spell over and over again.

An ideal mechanic that is developable is kicker because we have all the knobs in the world. We can decide what the upgrade cost is and what the upgrade effect is. We can easily balance cards at razor-thin margins. Much better than a mechanic like conspire, where you are always getting a second cast of the spell. We can balance conspire based on how likely you are to get the effect or how useful the effect is twice, but there isn't much granularity there. If we are trying to get the common removal spell in the right spot, we will often just be forced to err on the side of "too weak" because if the first cast is worth it, the conspired version is more like an uncommon power level. Burn Trail's base case is pretty weak, but it was one of the strongest cards in the set because of the upgrade, and I don't think making 3R deal 2 damage to a creature is going to make people happy, even if it is strong enough.

Mechanics like delve are developable but difficult, and I think there is a reason that whenever we take them on, we end up making some cards stronger than we intended, especially in older formats. We can test delve a lot in Limited and Standard (and did), and I think made the cards at the right level for those formats, but when provided with faster ways to fill up your graveyard, they end up too strong.

Constructed Cards

Not all mechanics need to be powerhouses in Constructed. Plenty of fun and popular mechanics that played huge roles in Limited and casual formats never really showed up Standard. Devour, for example, probably just has too much risk involved to really be strong in Constructed. In an ideal world, though, each mechanic has some way of making cards that would be fun in Constructed. Note that I say fun—we can make cards for mechanic like devour that are strong enough to take the risk on, but they may not be fun in Limited. For example, I don't think that a more efficient version of Mycoloth would've been fun when so much of the game would come down to "Did you kill this the turn it entered the battlefield? If not, I am up too far to come back from." We also want mechanics that tend to fit into normal Constructed gameplay patterns. Dash was a good mechanic for Constructed because it naturally allows aggressive decks to play around Wraths and sorcery-speed removal; we don't need to push the mechanic too hard for it to find room in Standard. We can put it on a reasonably efficient creature, and the utility that the mechanic adds makes it into something that will show up. We kept trying to create outlast creatures that would see play in Constructed, but we found that when we got them strong enough that you would rather use their tap ability than attack some amount of the time. The cards were not very fun, so we pulled back on the mechanic and let it mostly exist in casual and Limited environments.

Escalate from Eldritch Moon is an example of a mechanic that, when I created it, I joked that it was basically a Constructed mechanic grown in a vat. It has all the hallmarks derived from how we make Constructed cards. With multiple abilities per card, we could throw a sideboard-level effect to allow for metagame diversity, plus two effects that are priced individually just below Constructed power, to end up with a strong mechanic. You can see this more in Modern than Standard, where Collective Brutality and Blessed Alliance show up pretty regularly. We want some number of mechanics like this (that are highly functional and allow for easy Constructed shots). Revolt is in a similar spot; we knew that in older formats, fetch lands would create very easy abilities to make these cards work, and we could easily make cards like Fatal Push that could show up in both Standard and Modern. We don't want every mechanic to work that way—I think it is generally better when the mechanic is something that happens to work well for Constructed rather than us creating one from the blue that will work and fitting the flavor and set importance around it.

That's it for this week. Next week, I'll be back to talk about more stories from Magic's development process.

Until next time,

Sam (@samstod)

Latest Latest Developments Articles


Long Hard Road to Print by, Sam Stoddard

The entire process of designing and developing a set, plus the time it spends going through editing and the Future Future League (FFL), is about two years. In that time, the design team c...

Learn More


Development Questions with Amonkhet by, Sam Stoddard

Hello and welcome to another edition of Latest Developments! Before you read my column this week, you should probably read over the full Amonkhet set, which can be found here. Not because...

Learn More



Latest Developments Archive

Consult the archives for more articles!

See All

We use cookies on this site to enhance your user experience. By clicking any link on this page or by clicking Yes, you are giving your consent for us to set cookies. (Learn more about cookies)

No, I want to find out more