As we recently put out a pretty large balance patch, now seems like the perfect time to talk about how we approach card changes. This month's update takes a look at how we think about balance changes in general, using examples from previous balance patches, including our recently released Update 69.
Which decks should you change?
When we’re considering balance changes, we look at a number of different factors. Every week, we check in on metagame data that includes a ton of stats on cards and decks. These stats can even be stratified by level of play. While we look at a number of things, we first examine frequency and experience of playing against a deck, as opposed to simply studying the raw win rate.
For example, before Heroes of Skyrim was released, Orcs and Goblins were consistently topping the charts in terms of win rate. That said, they were both under 5% of the metagame and we believe they played out pretty well. Since people were not playing them much, the average player is not going to feel the experience of getting beats by these decks on regular basis, and therefore, it was not a priority to take action on these cards.
On the other side, Ramp Scout is viewed as a very powerful deck despite stats that tell a different story – much more so than the actual win rate. This may seem strange, but players sometimes judge decks more by how they feel than raw data. Ramp Scout tends to win long, drawn-out games, whereas Orcs and Goblins tend to get things over with pretty quick. This means that Ramp Scout may spend more time in winning positions even if it has a lower win rate. Players are likely to feel like Ramp is winning more than it actually is and thus likely to play it more.
As we make balance changes, we also want to encourage players to try out decks and cards that we feel are fun to play with and against. A deck like Tokens has had a pretty good win rate and has been reasonably popular at times, but we have been reluctant to make too many changes to it as it tends to play out quite well. It can win long and short games, interacts with the opponent and can be interacted with.
We’ve also made changes to decks without huge win rate numbers that had concerning play patterns. The Praetorian Commander and Echo of Akatosh prophecy-heavy decks were winning about half of their games, but were doing so in a way that was very demoralizing to play against, which justified action.
Which cards should you change?
Once you figure out which decks are concerning, you still have to figure out which cards to change. We try to target cards that will improve the play pattern of the deck in question in addition to reducing its win rate.
One thing we’ve thought a lot about is how to make aggressive decks more fun to play with and against. In general, we’ve found that aggressive decks that are less about attacking the opponent all the time and play more interactive games are more fun. With that in mind, we made a large change to Northwind Outpost, which tends to encourage a more “face” strategy.
What change should you make?
Once you’ve determined what cards you want to change, you still need to determine how to change them. If a card is changed for mostly power level concerns, we are likely to make a relatively small change so that it hits a playability sweet spot. If the concerns are more about play pattern, we are likely to make a bigger change and may even try to address the play pattern concerns with our change.
In general, we try to make relatively simple changes. First, it’s nice when cards still feel like the same card after a balance patch. If we wanted to make something radically different, we always have the ability to produce a new card. Second, simple changes allow us to act faster. They take less time for engineering to script, and less time for both design and quality assurance to test. When we think a change is warranted, we want to get it out to you as fast as possible.
Changing cards can also cause confusion, especially for less experienced players. For example, if we changed Lightning Bolt from dealing 4 damage to dealing 3 damage, some players would try playing Lightning Bolt on a 4 health creature and become frustrated by the result. Simple stat and cost changes are less likely to cause this frustration, because you can see those changes without having to read the text box.
How do we gain confidence on changes?
Once we have a sense of what changes we want to try, we do a few things to gain confidence in those changes. First, we try to consider the net effect of the changes on each major deck type. We look at matchup data and think about the importance of the changed cards in each matchup. It’s sometimes possible for a deck to improve from a patch in which one of its cards is changed due to its bad matchups decreasing in popularity. Our predictions are never perfect, but by estimating these effects we can get a pretty good sense of what the metagame will look like after a patch. We then playtest with each changed card on our local design server to get a sense of how each card feels in its new form.
What about Buffs?
Buffs can shine a light on cards that may have been underappreciated in the past. That said, buffs are trickier to implement compared to nerfing a card. Cards that we are considering buffing have generally not seen much play, so we don’t have as much context. If we do too small a buff the card may continue to be underplayed, and if we do too large a buff the card may dominate.
It’s also okay to have some weaker cards. Sometimes a card is more fun in arena than in constructed, so we target it for a middling power level. Sometimes a card is fun if you play with or against it once in a while, but isn’t a fun regular experience. There are also some players who enjoy winning with underappreciated cards.
A lot of the things that buffs offer can also be provided with monthly cards. We can push a variety of classes and themes with a sweet new card. When Warrior was unpopular, we came out with Sower of Revenge as the next monthly and it made a huge difference.
There’s your inside look at how we approach balance changes. We’ll be back next month with another look inside the development of The Elder Scrolls: Legends.
- The Dire Wolf Digital Team