Welcome to our January dev update for Legends. Today, we’re exploring an important and often overlooked aspect of the game: the start of the match. We’ll dig into how we arrived at our current start-of-game ruleset, and how they are working for us.
When we were creating our initial rules set we knew we wanted to come up with something that was simple. Legends features a deep gameplay experience, so we didn’t want to pile on with complicated “who goes first” rules. To that end, we knew we wanted to make our rules to limit the differences between the player going first and second. The more differences between the player who goes first and the player who goes second, the more difficult it is for players to understand what’s happening.
For example, we decided each player would begin the game with three cards. We started with each player drawing a card on turn one as the simplest implementation. Throughout testing, we tried things like giving the player going second an extra card, and having the first player skip their draw, but neither felt great. Receiving an extra card while having less access to Magicka often led to the second player getting beat down, which also presented the risk of having an overstocked hand because of the rune system. Starting with three cards and not drawing on turn one led to too many awkward starts for player one, so we ended up reverting back to the original setup.
We also knew that if Magicka also grew by one each turn without any compensation for the player going second, the first player would have too large an advantage. To start, we gave the second player a two-use activated support called the Elixir of Magicka to begin the game. Players would have to learn about activated supports anyway, so this didn’t add too much complexity.
With a rule set in place that we wanted to try, we began the part of the job that designers love most: playtesting! After testing a ton of games with the two-use Elixir, we found that it simply wasn’t enough to overcome the consistent extra magicka the player going first gets. We experimented – trying ideas such as giving player two an extra three health on top of the two-use Elixir, but this helped control decks much more than aggressive decks.
Thus, the three-use Elixir was born. It seems like a lot, but the three-use Elixir both felt good to play with and put up nearly perfect numbers in testing: win rates of the player going first were less than one percent over 50%. Three uses also lined up with the number of uses our other Elixirs had. We decided to go ahead with the three-use Elixir for the open beta.
From Elixir to Ring
One defining characteristic of the Elixir of Magicka was that it counted as an activated support card. This meant that it could be destroyed by support destruction effects, and combined with cards like Tower Alchemist to never run out of charges. We found that having your elixir destroyed by cards like Shadowfen Priest was a frustrating experience. With that in mind, we considered exploring a version of the Elixir that was indestructible. That way, it could still do the cool combos with Tower Alchemist, and would hopefully be a little easier to understand since it has a normal card type.
Eventually, we decided that the Elixir should just be a separate game piece – a ring -- because at the end of the day, its function is unlike other cards in many ways. Even though the Alchemist combos are cool, it is a little strange for Alchemist to be way better when you’re going second. Indestructible is also not a concept we use very often, and it was nice to not have to use it for the Ring.
Overall, the numbers on the first/second rules have consistently been right around 50%. In the past month in ranked, the player who goes first wins 52.2% of the time. This gap is reduced for our highest rated players, where the player who goes first wins 50.3% of the time. This difference may be due to more experienced players being more comfortable using the Ring of Magicka, or metagame differences.
More aggressive classes tend to have relatively higher win rates with the Ring of Magicka. Their games are generally shorter, so they are ahead on magicka for a larger percentage of turns with the ring. Classes like Crusader and Warrior that tend to be aggressive have similar win rates with or without the ring, whereas slower classes like Mage and Scout prefer going first by just under three percent. With such a small difference, we’re comfortable with where things are at.
In most card games, the player who goes first gets off to a faster start, whereas the player who goes second has access to more cards, either with a larger starting hand, or the first player skipping their turn-1 draw. In Legends, it’s almost the opposite. The player who goes first is not down a card, but the player who goes second gets to pick three crucial turns to be ahead on magicka. This means that, in general, the player who goes second is more likely to win a shorter game, whereas the player who goes first is more likely to win a long game. This subtle difference contributes to a lot of the confusion about the Ring’s power level.
Risks and Costs
While we are happy with the simplicity and balance of our current rules, there are risks with any rules set, and the Ring is no exception. With a perfect curve, the player going second can spend three extra magicka over the first three turns. This can create some games where one player takes over the game very early on.
We considered having the ring “cool down” for a turn to mitigate this, but ultimately decided that the additional complexity was a big cost, and that Prophecy already mitigates the risks of one player coming out to a super-fast start. In addition, the win rates currently lean slightly towards the player going first, so making the Ring even weaker would be dangerous.
There’s your look at how we came to the starting rules you are playing with now. We’ll continue to monitor both data and feedback on the setup, but for now, we are quite happy with where things are at. Until next month, good luck on the fields of battle, whether you go first or second.