Balancing Cards in Clash Royale

[Music] okay so my name is Stefan englum I come from Finland uh it's a bit cold there right now so it's kind of nice to be here in very warm San Francisco uh I work at supercell as a game designer and I've now worked in The Clash Royale team for roughly one year uh I joined the team really soon after the Global launch but I I had the privilege to follow the development like very closely as I was working at supercell in other teams and I I played very actively in all the internal playables and we are a small team there's a photo of the team Uh current developer team so we're 16 and I'm the only one with the title game designer there but we're really like all everybody is like very much involved in all kind of different uh aspects of design that so like everybody is kind of

a designer and we're like we are like super independent just like other teams at supercell and basically like all the decisions regarding Clash Royale like the game itself is done by this group so it's cool uh my main areas that I work on uh is card balancing the game's economy and system design so that's like gives me a pretty holistic view on the like and holistic take on the like the whole game uh but I of course run all Decisions by the team and we discuss everything like very openly but yeah in this talk I will focus on card balancing which is perhaps the part that I'm most passionate about and I feel that it's also the one that I've learned most about during this past year so let's get going so I'm going to talk about a few topics here so first of all I'm going to talk a little bit about Clash Ro what

is Clash Royal introduce the game a little bit and talk about some things like how we uh do the upgrade levels and how we handle the balancing there that it's kind of like a background information for the the rest of the talk then I'm going to talk about our goals for balancing uh why do we do it and like what what are our thoughts and Define some terms here then I'll talk about some of the key principles we have when we think about balancing what kind of things do we think about there and then uh how do we decide which cards we then rebalance what are the things we we think about and take into account there so let's get started Clash Royale first of all how many here have played Clash Royale okay quite a few nice how many have not played Clash Royale okay there's a few it's it's it's a free game it's like you should try it's pretty nice so all right I

so well anyways Clash Royale so it's a real-time PVP game and it has roots in in like collectible card games and and mobas uh but it's it's like a bit of a mix of those genres and then tailored for for mobile and in Clash Royale you collect cards upgrade them you build a deck of eight cards and then you go to battle against random players around the world and it's real-time PVP you generate this resource called Elixir uh during the course of the game and then you use it to cast cards on the Playfield and to rotate your deck at the same time and you have like 3 minutes plus one minute of over time to try and destroy all the enemy towers and usually like pretty much 30 seconds into the game there's already like total Mayhem and Chaos like on the field and the units they move and act by themselves when you deploy them and they still like

plenty of decision making to do with things like timing the placement and managing The Elixir optimizing the deck cycle uh defending counter pushing Etc uh something I want to talk about is the stats we have in the card or the the cards basically they actually have quite a lot of stats like this the Knight is like a very vanilla very basic card in the game and even that has quite a lot of attributes so I'd like to think of these like compare the cards in in a way a little bit more to like let's say Le of Legend Heroes than uh like let's say Hearthstone cards uh there there's way more like parameters that you can tweak and that affect it and so even like a vanilla card has that many like things not to talk about then like a more complex card like Electro wizard that shoots multiple targets stuns them and when you deploy it it does a zap effect and so forth uh so from balancing perspective it's of course nice that there's lots of these parameters that you can tweak you can do like a really small thing like adjust a little bit of speed here or or movement there and like

it can it can like tweak it just a little bit the card but one thing that's like uh tricky there is that like it actually makes it a bit more challenging to like mathematically uh balance the cars to like figure out like this is like a good sport for compared to let's say magic the Gathering where you have like Mana cost attack and defense and that's basically it for a vanilla card uh so it's a bit trickier and of course also to not make it too easy uh we have then the upgrade levels and the different different uh Rarities so I want to like use a few minutes to explain how we do that like how we how we handle that thing so I'll start with the commons so Commons there's 13 levels of Commons right and for each level when you upgrade it you we give exactly 10% damage and 10% hit hit points more but

we don't adjust any other stats we don't adjust heat speed we don't adjust movement speed or radius or anything like that because this when we touch only these values we ensure that the interactions between the cars actually remain the same um from when they are like at the same level so as an example archers and one Archer would twoot one Goblin uh at level one and at level 13 the interaction is exactly the same but the damage and hit points are higher so this helps keep me sane when when like it's so how do we handle then the different Rarities it's actually like kind of simple we start at a higher level but then we upgrade them like in the same method like 10% each so basically rares a level one rare has the value of a like level three common uh and an epic start at level six

and a legendary a level one legendary is basically like a level uh n common and then the place and what we have there is this thing we call the tournament cap that's a card cap level that we use in in tournaments and also in Friendly matches in the game we actually like not only cap cards there but we even level up cards to that cap and that is the spot where we balance the cards so when when I make like a values for a new legendary card I I balance it kind of compare it to like a level the level nine commment and see uh see from there so this ensures that the interactions remain the same at all equivalent levels uh and then like it basically ensures that like at max level we if everything is at max level then we have the same uh balance that we have at the tournament level and then like we think of the uh level differences there then like more of a bit of a spice to the

game and something you can like specialize in things and so forth so that it it's like a kind of a spice thing but the main thing is to balance it there and something to note also that I want to mention is that between Rarities it's not meant to be like more powerful as such it's uh the idea is that like the more rare the card is uh the the more exotic mechanics we introduce and the more advanced mechanics so as I had like earlier the night very vanilla card it's a common and then we have Electro wizard with like all kinds of crazy abilities and it's a legendary so that's the idea and the same applies to like when we go like upwards in arenas like things get the abilities get a bit more complicated it's also cool for like kind of introducing the cards gradually and introducing mechanics gradually for the players all right so that was the basics so let's jump to like the actual topics so let's talk about the goals of balancing so yeah wait I need some

more okay it's usually quite good to figure out like why you're doing something and and think about the goals of what you're doing so then you kind of have have something to aim for uh so for balancing I would say it's fun to have fun and interesting gameplay also includes of like have depth and and so forth and then have variety so there's a variety of different play Styles and different deck archetypes that are viable and then game feels fresh both with like new cards making the game feel fresh but also that like in between games you have like surprising moments and and stuff like that so the game feels fresh and if I would have to like sum this in one um one sentence it would be that you want to have like a healthy meta game so that's like the ultimate goal in the Quest for the healthy meta game right uh so let's define what this healthy meta game really now means so let's start with metag game

so first of all metagame is not exactly a dictionary term and I I didn't find anywhere like a proper definition for it and it can mean many things uh even like game developers seem to talk about it like with different meanings but uh players of uh like PVP games and developers PV games when they talk about balancing they usually uh the the like meta game is kind of an established term and for the purpose of this talk now metagame means the cards combinations and strategies that the community at large have found most effective so the metag game is not something that the developers create as such uh and unfortunately very rarely even predict perfectly but balancing is then the tool that we can use uh to to kind of steer it in in directions that we we want uh but like yeah as a player and a developer I actually find this concept of metag game like quite quite

intriguing and interesting uh it's I feel it's kind of like the idea that like you know we are the toy makers and we create to but we don't really know what the kids will figure out to do with them and and usually it's something like totally different that we indeed but that's also like exciting so that's metag game all right so what's like then healthy metag game and an all healthy meta game so you could think of it as uh like a of the being a state of fun variety and freshness and to kind of cover the total extremes let's first look at what would be like a broken metag game uh it's not like a realistic state but like a broken meta game would be where you have like nothing it's like yeah it's broken so an example would be like this type of card you know like you just instantly win the game by playing it it it doesn't make any sense but like I mean there's no strategy there's no interaction there's no variety there's like it's just stupid

but this is really not like a realistic state if you think about it like I mean already common sense like says that you you can't make a card like that so what's then like a perfect metag game which is kind of like as unrealistic so that would be like everything is in perfect balance everything is in seamless Harmony everybody loves everything in the game and like everything is just like absolutely fantastic like it's it's not really like it's kind of Utopia it's nice to maybe strive there and think about it and dream about it but like we need to be kind of honest with ourselves that like people have different tastes and and like it's that like so yeah we can't really reach it so instead we have something in between so we have like unhealthy and healthy meta games so they might be a bit tilted like the unhealthy State might be a bit tilted you know you maybe have like only two viable State play Styles there's not so much variety maybe it's like kind of fun uh for a while and the healthy one is like there just

several several things that are going on like but like you know it might be like maybe less varied but it's like really fun or something but and one way to look at this is kind of like think of it kind of as a diet so it's kind of like you know it's nice for a week in GDC but you know you you kind of like yeah you you you know that it's not good for you in the long run and you will kind of feel bad afterwards or then the healthy one is like that's where you want want to be and you know that like it it feels good like in the long run to be there but yeah so all right this slide kind of looks like we're talking about something completely different here all right so that's the definition of unhealthy and healthy metag games so then let's talk about some of the key principles we have for for balancing so first this is how we do it that's basically it in questions just kidding so make everything

overpowered or well make everything feel overpowered to be to be more precise so let's like let me illustrate this a little bit so let let's take like chess for example I mean it's an archetype of a very balanced game right and and a game that would only have chess soldiers duking it out that would be like extremely balanced like it couldn't be more balanced but to be honest like I it doesn't sound to me that much fun and exciting uh but overpowered like that's kind of like the Marvel Avengers so certainly they're like all overpowered and and we could argue about who's more overpowered than than the other but they all have their unique traits their unique moments when they are like the the best and kind of balanced in a way between each other so the concept of making everything feel overpowered is about making everything fun and exciting and this has of course a lot to do also with character design and and art style but that's not the

topic now so how do you with game balancing make something everything feel like overpowered so how would you make these superheroes or in Clash royale's case how would you make all the cards feel overpowered so an alternative to way way to say this is like each card needs its moment uh so there needs to be some purpose for the card uh or more technically there needs to be enough situations where the card can bring huge positive value right and you could then switch out here like the card to Heroes spells characters whatever the game has so an example from Clash Royale of a card that's like feels overpowered is speca uh she's this like giant robotic Warrior thingy that deals insane amount of damage to a single Target and can take like a lot of punches so like she can really feel overpowered in those

moments when she's like smashing through some really powerful unit and then still going for like a big counter push and if she reaches the tower she can like easily with a few few swings uh take it down so peka is super overpowered everything is good but unfortunately everything needs a counter too bad for Pekka and for Pekka there are actually plenty uh so in fact she's considered by top players to be one of the weaker cards in the game because she's so easy to distract and counter with with actually very low value cards but that's actually like the thing that that's what keeps her in check and that's what makes game play interesting and allows the meta game to evolve uh so so you have these counters to cards but you also have them counters to the counters etc etc so the meta game evolves and and this is basically what makes everything that feels overpowered powered to not actually be

overpowered and another thing that's good with having multiple counters is that it access a safety net and yeah there's a this pretty recent card Elite barbarians like if your players who have played it so you might know what happened so it's two very fast and strong barbarians and they are like extremely devastating when left alone but when it was released it was actually like very very weak like kind of the weakest card in the game and it was met with a lot of disappointment that's that's from the forums uh so uh yeah my bad for the value sorry really um so a few weeks later we we buffed it uh but it was a bit too much and um my bad again so sorry sorry also for players if they watch the video Whatever uh so I definitely definitely learned some lessons there and it's it was a really

very powerful indeed and for a few weeks we had a pretty long action-packed gam playay when they were like smashing through all the towers but but the good things is that the tools to counter it existed in the game and players did use them and players did figure out them so having counters for everything acts as a safety net uh so it's not an excuse to make bad balancing but I do think that it's like it's inevitable that mistakes you know happen so so having a safety net they're having multiple counters can turn this like disaster into perhaps a little tilted and unhealthy meta game all right so make everything feel overpowered but everything also needs a counter and something I have to say here though is that these two principles were heavily heavily inspired by blizzard uh so blizzard approach to multiplayer g game design uh as presented back in GDC 2008 by robo uh was really inspirational

talk and in general like blizzards games have really been and can continue to be a great inspiration at supercell uh so yeah just kind of wanted to say that don't thank my word for it like for the stuff that I mentioned like so yeah you can find actually this talk online still at least the audio of it so it's worth checking out all right so make everything feel op but everything needs a counter so let's move on so a little bit about new cards and old cards so something that's important to us is to make new cards meaningful uh because if the goals of balancing are to offer fun variety and freshness then it going to make sense that you have to make the new card meaning if the new card is not used it doesn't meet any of these criterias uh so you're going to have to make it meaningful enough so it's used and something that I want to like bring up here is that we often get these comments and speculations about how new car are overpowered so just we

just monetize some then uh but like the thing is that that's not true there are of course places where we think about monetization obviously but card balancing is just like it's not one of them uh and I honestly think that it's hurtful both in the shortterm but especially in the long run if you have monetization as like an actual goal for balancing because it's just like is contradicts with the other uh other goals with the actual goals so yeah just wanted to clear that out so making new cards is Meaningful uh making new cards meaningful is important so they actually have an impact on the game so you could say we aim to make new cards balanced but we going of round up a little bit hopefully not like Elite barbarians up a little bit but like you know but an important part in making new cards meaningful is to uh try to fill a game playay gap so maybe there's some spell or units that's missing or or something needs like like Alternatives or counters

uh so just to make sure that the new card clearly has its own spot and isn't just like a strictly better version of some old card in the game so as an example I could mention the Ice Spirit uh it's a single one Elixir unit that sacrifices itself uh deals a bit of damage and then freezes the building and units that it hits uh and the idea kind of came from having an alternative to the one Elixir skeleton which was the the only one Elixir card at that time and zap was and still is like a very versatile card in the game so uh that stunts the en so this was kind of also meant as an alternative to zap and then we had prior to this update we had released the fire spirits and it was kind of a cool mechanic where they sacrificed themselves and explode kind of like as a fireball so we thought that we could utilize that same mechanic and do like a more chill version of the fire Spirits so having a purpose for each card is important and it also ties to

our strategy of keeping the total card count uh reasonable and it's also important for us because uh to to it's also important for us to keep old cards relevant though so like the iconic baby dragon and mini Pekka so we do rebalancing of old cards to keep them relevant and and in the changing meta game and whenever we do that we aim to uh stay true to the soul of the card so essentially that means that we try to avoid these complete reworks of the of the cards and instead we do smaller changes like to the basic stats uh this also makes it a bit easier to predict because a complete rework is essentially like making a new card and but another thing that is important is that like when there are a lot of players playing the card so uh it this way it kind of plays like it played before uh so while we communicate about the changes quite a lot like not every reach it so it's nice if the players can

still play it the same way they can deploy it on the same spot and like play it very similarly uh so it feels familiar so make new cars meaningful but keep old cars relevant and then something we decided very early on is that we we want to keep on rebalancing we want to do regular uh semi-regular like balance changes we've been doing like balance changes roughly once per month and there's no exact science to this gate it just felt like that was the when the meta game felt like it was kind of getting getting a bit stale so we and and there was like this feel of like that we could have like a little bit of change so we we did the we chose the one month Cadence and at the beginning the reactions were uh bit negative maybe towards the balance changes but over time as we just kept on doing them because we felt that it was important for the game uh players started to like actually appreciate them and they started to like look forward to them as

well uh so so it was a good thing uh but there's also like the meta game really needs also time to evolve by itself so you you kind of need to find the sweet spot for your game like what is how do you want to approach this uh and an extreme of like not giving it enough time is that players might start to take it as a crutch like that they that instead of like figuring out like counters to the current metad decks and so forth they actually like just roll their thums and wait you to do like a balance change so it becomes like a balancing act of like being a crutch or and then like managing the frustration of the players when you're not addressing imbalances and something that we noticed uh when doing this different kind of changes is that like bu Buffs are we learn that Buffs are nicer than nerves and it's it's quite natural it makes sense because Nerfs tend to hit cards that are used a lot and that are very

powerful so there are a lot of players who get a bit disappointed and so it's not something that we want to do light-heartedly but we we of course do it do it when it's needed and and uh and it makes more sense to like Nerf one card than to buff like 69 so and Buffs on the other hand they they hit cards that not that many use so those who use them are of course super happy and those who don't like they might be happy that there's a new card coming in Into The Meta Maybe maybe uh or then they just feel indifferent because they don't have to deal with that cardly usually so balance changes are always complex complex things so uh communication is very important so it's not just about like informing players that hey this is the change we did but it's it's really helps with managing expectations because the Buffs are never we never intend Buffs to like make the card the most used card in the game um uh on likewise with nerves we never mean to kill the card it's still supposed to be relevant it's still supposed to be used but just like kind of be slightly weaker and be

in a better place so when communicating it's a good spot to give like reasons reasons background and some context to why why the balance changes were made uh we always make this in-game post about the changes and then we have in our blog uh a bit more like context and bit bigger Longer descriptions uh and then also like what helps a lot is that there's a very active community of of YouTube B streamers uh and so forth that are doing like more even more like in-depth views of the changes and all the ramifications all the indirect changes Etc so keep on balancing regularly knowing that Buffs Buffs are nicer than Nerfs so that's basically like the these key principles so next next up is to talk about how we

decide what which CS to rebalance all right so there's 70 cards in the game right now uh but it doesn't really make sense to like tamper with all of them at once obviously it's it's too big of a change and and and the ramifications become like it becomes impossible to to predict so how how do we decide then uh I think it like boils down to like three Parts it's we listen to the players and then we utilize data but at the end of the day it it's really boils down they these two can't really make the decision for you so in the end it boils down to intuition and just like knowing knowing your game all right so let's talk about listening to players a little bit so feedback and Analysis from players plays plays a very big role we we're very very fortunate that we have a big and very passionate uh and act community and then many of the best

players from all over the world are actively writing strategy articles analyzing the cards analyzing the metag game so the players the community and the players can't while they can't like make that decision for you it's it's like an invaluable tool for like uh that you you should listen to and by following the community you really can get like this pulse on the players feelings uh you have to look sometimes a little bit between the lines like what what's the actually the root of the problem there because complaints about a certain card the balance for a certain card might might not actually be like that that specific card is imbalanced but it might be that there's just like lack of variation for example rather than than you would need to like Nerf that specific card and one thing that's worth realizing is that like especially uh after a game has matured the developers

are really not the world's best players like definitely not and the the players can really really teach you about your game and so like Embrace this swallow your pride stay humble there's a lot to learn from the players and and I have an example of of this where this like kind of For the First Time really really struck on us so there's this there's this uh YouTuber I think this was the video that got make made him famous though uh yarn from orange juice and back in April last year the game was like 1 month old he made this video about how to use the one Elixir skeletons and the the team watched this video and and we were we were a bit surprised at what all things you can do with that only one Elixir and with that skeleton and and yeah like at at that moment we we really learned quite a lot like not just about about the skeletons lot of players learned about the

skeletons at that point we had to Nerf it after a couple of weeks but we also learned a lot about uh the game uh and about the fact that how like that how much depth there actually is in this game and and the fact that like players really can and will teach you about your game so that was like a valuable lesson ner there we still have these moments every now and then like with some videos that we watch it oh this is how you play this game um and something that's also related to listening to the players uh is that the top players are the trend setters in the game so I mean they have they have access to all the cards they tend to experiment they craft the new decks to out with their opponents and gain the edge and and quite often the trends then slowly trickle downwards uh and a another example where where this concretely like dawned on us was when we had our first live tournament uh so that was also in April 2016 so

that was a nice month for learnings uh we had this Invitational Tournament in in Helsinki and there was this young Finnish guy called Jason who ended up beating getting to the finals and then he beat everybody on stage 3-0 including some like of the that were like top players in the game and he was using this deck that like no one else has had ever like tried and it immediately got dubbed to the Jason deck uh it was a pretty basic deck really and and only contained like common and rare cards uh he he played it really well and and after the tournament we we really got to see for first time the effects of like Esports and and the this concept of top players being Trend sets when we saw like the how the deck deck use rate went up so yeah people saw the tournament and people definitely like tried out the deck and it helped that it was like only using like very basic cars that was quite cool cool to

see so listen to The Players but to get a second opinion you can also like check out data and you can use it to verify stuff so for us using data mostly means mostly looking at use rates and win rates and a bit definition time again so for I I suppose us rate can mean different things for different companies and different games but what we track is percent of battles where the player owned the card and used it and win rate is uh percent of battles where the player used this card and one so we look at these metrics both at individual cards uh level and also for for full decks uh so what do players use and what do they win with basically so what we read out of these metrics then first thing to mention is that use rat is the use rate is the main metric we use uh and and the thing is that that it really tells directly about like what is

the state of the meta game right and if the goal is to have a healthy metag game then it makes sense use rate is the one that uh shows you how you're doing with that so when looking at use rates at different Arenas you're going to get a feel of what the players see in the game and you can then back this up like with what the community is talking about and maybe some of the comments are actually that you see in the forums might be actually coming from just a certain trophy range where like a card might be overused or so forth and so forth so use rate is a pretty good indicator and it's also something that uh I I think it's a good indicator of something being like overpowered or underpowered because like very strong cards also tend to be used quite a lot but win rates they they still stories too so it can for example be a good indicator of sleeper cards or high skill cards like this Ty cards that not that many use but th those who do those

who specialize in them can be winning like consistently with them so it might not be a good idea to not like buff that kind of card like too much depends and win rates uh though as well as use right they they also heavily are affected by the current meta game so obviously like if if some uh good card is used a lot the counters might get used a lot and and so forth and the counters might be good for winning games and so forth so and I want to mention here now that like these are not necessarily the best metrics for some other game like you you really need to like figure out understand your own game and and figure out what you want to achieve using the metrics so I think for example a game like Hearthstone I think it would be very useful to look at things like uh win rates across different classes to get the different C classes uh balanced between each other and in a one one we one fighting game like street fight I'm sure they track uh the win rates

between different matchups with different Heroes so that or characters so that's something that you have to figure out for your own game so when we look at these metrics we we look at like different Arenas different trophy ranges and from the from the top players and also from the challenges which is the in-game tournament system which is played like with tournament cap cards uh but and when it comes to like evaluating uh the balance and like for individual cards we really mostly look at the top players and and the top challenge players because they have access to all the cards they know how to use them and you know they are they are kind of the trend setters as I mentioned before so I think that's a good place to start but we look at also like lower Arenas if like what do players see there uh like do we do we that we don't mess up like their game too much so what is then a good use rate we have in the game game currently 70 cards and it's eight cards per deck

so that makes roughly 9.4 billion different permutations which is nice not we we don't have quite yet enough players to like fill that but we'll try we'll try to get there uh so if everything would be used perfectly evenly uh it would make for an average use rate of 11.4 but that doesn't make much sense does it like we we can we can't expect all cards and all combinations to be used evenly and it can't be the goal leader like a deck with only the most expensive cards in the games it won't win you many games I see gentleman here is changing his deck right now so right so how do you figure out then like what's a good use right uh that really then boils down to like knowing your game so when we look at the use race then I Ponder like what why bonder about them we might think like okay why is a card

popular so why is it why is it used so much so like maybe maybe it is like very powerful uh probably is uh but it could also be that it's just a very very good counter in that current metag game or maybe it's just like a extreme versatile card like the zap cheap versatile card that fits almost every every deck so it makes sense that it has a higher use rate and another way to look at this think like of kind of Target use rates so for example you might look at different main Attack cards uh they should be viable alternatives to each other right like cheaper ones fit more decks for sure but ideally they are all viable and they they are kind of like you can look at them and compare them to each other or direct damage spells maybe like an even easier example that they have their own twists and situations but they should be really like viable alternatives to each other so these are all quite objective

ways to look at the use rates and think about what what is a good use rate and how does look uh but then there's also like a more subjective way to look at it which is which I I called spiciness so they might be these cards that are good spice to the game but but we don't want them to be used too much we don't want them like they're just there for variety so if we first look at like kind of bread and butter cards we would have the iconic barbarians it's an iconic common vanilla staple card that can be considered for most decks and it's fine if it's used used like quite a lot especially like if the meta is is such that that it's a good counter there and then there's like spicy cards and the xbow is an example of that so it's this building that rapidly fires arrows at a huge range and can reach the enemy Tower from the other side of the of the river so it's it's very powerful

when it's played well but we deliberately actually don't want it to be among the most used cards in the game and this is something we learned during the beta uh because there it was quite powerful and it was used by almost everybody and uh the community and in the end also we felt that it fit more as a spice to the game rather than this main strategy used by most players but is important spicy cards are important they add a lot of variety keep the game fun keep the fame fresh so I think you should have these type of cards as well and I think like metrics wise or datawise a good way to recognize a healthy spicy card is uh that it has like a low use rate but a pretty high win rate so that means that there are these highly skilled and specialized players that are using it and win with it but it's not really seen in every match next is for example a card that like it's it has a super low

use rate at lower levels it's quite hard to use well but actually top players uh every now and then like it it tends to like become quite quite us used quite a lot actually and in some tournaments in in some Esports events we can see like might see that like quite many of the players on stage are actually using Expo so I think it's in a fair okay place at the moment so you need to understand these things and you need to be able to look at the the cards and the game from very different points of view and it's it's also actually important to like understand advanced concepts of the game so especially if you're the person actually doing the balancing like you I think you really need to understand how the game works sure you I mean you will be learning all the time from players and and by playing yourself but you need to understand what the uh top players for example are talking about so in Clash Royale like there's there's plenty of these things like you could have you have like The

Elixir Advantage the concept of Defending and counter pushing multi-lane pushing a bit more advanced stuff like knowing all the card interactions or what different deck archetype exist and how they work what kind of opening moves there are identifying the deck cycle breaking the deck cycle uh understanding what spell baiting and building decks around spelled baiting guing troops uh the pros and cons of having the flying attribute in a card because it's not always good uh the pros and cons of having like slow units fast unit very fast units you need to understand these kind of things the point is not to like now go through and explain all these the point is that you need to know your game right so play your game a lot both during development and release it's it's really important and this was something like like during the development of clash Royale it was played a lot uh and as we saw with the orange

jues video thing we didn't like figure out all the things but it it was played a lot in the company and even there like kind of players outside the people outside the team were like coming up with good strategies teaching like kind of teaching teaching kind of the team about like all kinds of stuff so it's it's really important play a lot both during development and then after release so to figure out which car to rebalance we listen to our players we learn from them we utilize data but in the end you're the one who has to make the decision so it boils down to like understanding the game and and intuition so that's basically the topics covered but the Quest for the healthy metag game it continues so we have we have healthier we have unhealthier metag games but with with a persistent game that's meant to last for years it's it's it's really a never-ending process I mean new cards are added new game modes

are added even and uh the player base matures the players learn more and we learn more and more so the Eternal Eternal Quest just continues so I feel that there was like a bunch of things now in this presentation but I still feel like that I only like really scratch the surface of things regarding gameplay balancing so here's some uh little selection of of talks and articles and videos that I personally found very useful and inspirational uh some expand on these Concepts that I talked about and and some are about like other interesting Concepts about game around like game balancing things like power creep counterplay uh so I I really recommend checking this out if if you're interested in this and if you're doing this stuff so yeah you can you can take a photo I'll have a little pause before I go to the next [Music] slide all right so just to recap the things from this

presentation so the goals of balancing are to offer fun variety and freshness which feeds to a healthy meta game and some of the key principles we use are making everything feel overpowered but having counters for everything and making new cards meaningful but keeping old cards relevant and keeping on rebalancing knowing that Buffs are nicer than nerves and to decide what to balance listen to your players utilize data but at the end of the day at the end of the day just play play play play the game and know your game all right I'll leave this here and we can have a little [Music] Q&A thank you very much [Music]

any questions yeah hi uh fantastic talk thank you so much everything about it uh I would came in a little late I wasn't sure if you had mentioned this but how big is your balancing team is it just you is it like six people it's me really so yeah the the uh Clash Royale team is in total 16 persons so that's like all the developers uh and the QA guy the community manager data analy of for so yeah I'm the only one but all the card ideas come for the whole team yeah yeah and also balancing we we Bounce Around the ideas with the whole team and and like cool awesome thank you hi hey thank for the talk thank you for flying away from the cold to come visit us down here um when you were doing the internal balancing and I know you joined the team as it you know after it launched um was there you know were you testing out all the different levels or were we using kind of the tournament level as a um as a you know as your your model that you would then build off of

afterwards um both so uh there was there were plenty of tests where we started from scratch and I started from like zero and also at the same time tested the progress and the the economy and the like the first few days first seven days we also had like a 30-day like company playable where everybody was duking it out so there you go to kind of the different levels uh but then we also had these tests where we everybody had like this same like equivalent level cards and like we're playing this kind of like more like orty things and the goals of those different tests were to either test the metag game which is why you kind of you know tested the or excuse me the progression and that's why you were testing the the you know the progression Pro progress and then the goal of the other the the balance testes was more just figuring out if you actually had a competitive game yeah you could you could say that I I think there was though like also for the balancing between the two also like with the kind of the meta progress and

stuff like that so it was also to test test those like how do the early Guard Cards work and how does it work when you like early on get an epic card that's actually like kind of a level six card in a way compared to the commons that might be like a level four and so forth that all works out thank you very much thank you hello will it make sense to show realtime metrics to publicly to the users maybe uh I think one one tricky thing might be that like it it might tilt the like meta game even more so if there's like cards that are used a lot then they get used even more and so forth but but it could there could be some stuff to show I don't know yeah thanks nice idea so piggybacking on to that a little bit how transparent are you with the players about the balancing that you're doing and how do you is kind of a community question how do you manage

their expectations and keep them from there's a little bit of reluctance to mess with the right the stats too much because people get upset how do you balance all that do you use communication to do that I'd like to think that we are quite transparent about it uh but yeah I I do feel also that we could do a better job at like explaining and like of explaining this kind of stuff like what are our goals with the balancing and what what do we think about when we do the balancing so it might be sometimes feel like kind of a uh very shallow like our explanations maybe do you explicitly um expose the stat changes on a website or anything or just people find out about it by playing the game no no I mean we yeah we post like the we put always before we do the stat CH we post them like in the Inbox and we post them in our blog with explanations why we did it cool and uh and then also like we we do some cooperation also with some influencers that they they get kind of access to the

information and they can do videos about it perhaps and and some of course do like uh this proactively so they they go more in depth in in things like they might talk about like their ramifications and their own opinions about it and so forth wonderful thank you thank you so in recent months uh you guys have increased your release Cadence of new cards has that affected your process in any way um what are some of the challenges there yeah yeah it's become like infinitely more stressful so we we initial like ear in the beginning we when we released our first cards we released them always like six cards at a time four cards at the time and so forth but it was like it was in November that we decided to go that we okay let's release like one card every two weeks and these are many many good attributes like people got to focus on that one car uh at the time it wasn't like a huge amount of cards and also like it g kept

the game fresh but it was actually quite challenging for me because I had to or we had to like kind of think about the balance of like new cars that we were then developing before even see seeing what the old cards were doing to the game and we had to do like balance changes in between when there was like only one week or two weeks to the next card being released and we don't know what it will do to the game so it has definitely like added challenges and we are actually now changing we announced it already that we are we are changing that Cadence like partly partly because of this actually like the to to like be like three to four weeks awesome thank you thank you so this will kind of piggy back off that question then I have a small followup afterwards but um so you mentioned that while conceptualizing conceptualizing cards you made sure that every car had a specific scenario it provided utility for so um every card has its moment right so do you think that there would be a point that every like Niche you utility would be filled or do you think that the release of new cards it creates like new utility like

new niches for there to be utility for and do you think that like you could pretty much have an endless amount of cards or do you think there's a point where you know every every scenario has been has has its uh Niche card I guess yeah so so um yeah it it is like challenging to always like more and more challenging to come up with like new uses for the cars and and and to to again fill like a new Gap and find a new utility we still have a bunch of ideas of course uh and something we discussed like regarding the like uh kind of how many cards fit in the game how many cards does make sense that's something we we talk about a lot but we don't have it's not yet in any way a problem and we don't have like we're kind of not there but it's something we talk about and then something else um so with with Clash of Clans like supercell is also one of their amazingly successful games um

there has been just like a huge kind of subculture of the industry that just focus on like cloning that game basically um do what do you think of like the current sub industry I guess of uh Clash Royale Clans and the success of those and what do you how do you see that industry like growing in the uh re like uh close future I guess so so you mean what do I think about kind of clones and like copying I guess yeah just uh well I mean like totally blaten clones let's say that you like seriously like just copy paste Co copy like the whole code base and stuff like that and that that I think is is a bit like disrespectful and but but uh but like I don't seriously I think like most games are not really clones they are they really I think most games like try to try to find their own thing like they they are inspired heavily inspired might be like by right by a game and they they take it and I think it's it's more almost flattering and I I I like this

genre and I'm happy to try out these new games and like it's cool that you kind of guys kind of pioneered a whole new genre of game but um thanks thank you hi um so when a card is either overpowered or underpowered sometimes that's because the numbers just need to go up or down a little bit but other times it's because it's qual like what it qualitatively does is either not healthy or too weak or something what kind of factors do you use to determine whether you should be adjusting the numbers on a card versus adjusting what it does so so what's the question like kind of how do I decide like which stats to kind of change in a card right right was that the question um it it really depends on the on the card and the and the situation I I don't think there's like any any one one rule like looking like how for example just trying to like s so that it's staying true to the soul of the car that I mentioned like that that I don't like

make a fast unit slow or make a like Fast hitting unit too too slow and so forth but like kind of looking at what what makes sense where which is the moment or the place where this card is like too weak or too powerful where that needs kind of tweaking is it that it doesn't do much damage is it had does it have like a hurtful interaction with some card like that it it one shots something that maybe it shouldn't one shot and and so forth so it really depends on the situation I I guess the thing that I was specifically referring to is cases where um if you think back a while ago when alixir collector was steadily the HP was dropped the HP was dropped the HP was dropped and then eventually you decided actually we need to change the uh the cost and The Elixir generation rate at some point and then the C in heal your place after that what prompted that kind of change of approach it's there not really like I don't know I we just discuss kind of like different

different things and like what would make sense and maybe also like sometimes there might be good suggestions also in in in the community about like that this could be like tweaked in this direction I I don't know I there's not really like any process for that it's thank you sorry I can't answer that better hi hello um here um you have a Mana system that is based on on time in your game game I'm very curious how you decide if a card needs to cost five six or seven can you answer that please yeah so when we when we kind of do new cards one one thing that then like quite early comes up is like at which Elixir cost should we try to kind of get this card what what is kind of Elixir cost that maybe we we want to get like a new card there or or it it would be cool to make this card into like a seven Elixir card and then we try to kind of balance the stats accordingly so

there again like not really like any strict process to this like it's just like flying sometimes it's like we okay we start with a card like this could be cool as a 7even Elixir card but then like things maybe like the artist create like a cool character and then it's like okay this doesn't anymore make sense as some something tanky so maybe we need to like change this into something else and so I don't know doesn't really like my question is um regarding to if if if I play a card that cost seven I have to wait another X seconds to cast another card do you takeing consideration this fact yeah okay yeah okay so yeah that that ties kind of into the um like how the yeah how the like The Elixir works and the deck cycle so like it's kind of a trying to think like how to go there shortly like you know you like let's say like in Magic the Gathering or Hearthstone you have this or magic there's this concept called the

Jedi curve where you have like kind of like that a one Elixir card or a one Mana card should be one one a two Mana card should be 2 two this is the thing you're talking about right like so how do we Define like what is the stats of kind of a tree Elixir cards and that that's the thing that's like it's kind of mathematically a bit like trickier uh to Define because there are all these parameters like affecting it so we kind of compare to the existing cars that okay there other three Elixir cards or the other four Elixir cards have these kind of stats so maybe it should be like in this place maybe it's like a bit weaker with damage than compared to mini pek for maybe it has a bit more hit points and so forth okay thank you man yeah thank you hi uh when you Lun a new card uh you have this fantastic pie with data from players feedback and actual data from them so what happens uh before that how when and how do you decide uh to launch a new cart uh it boils all all boils down to intuition alone or what

sort of data do you gather before because you don't have the player data yet so it's all uh team data internal or you use external play testing for a new card so just so what's the question now that like what's the process you follow to decide that the card is ready and in a well shaped ready to launch in the game so when a new card is ready we want to release it well I mean we kind of have a have a set a deadline for the for the release for the next version so we try to get the card ready by that time I I guess but but that that's not really what you're asking I suppose but I'm I'm like trying to wrap my head around like the actual actual question maybe question maybe how you arrange what how do you get the data to to to tell that a card is ready to be launched what's the trigger point we don't really like get any data for a new card when we're making a card if that's what you mean so

just intuition based yeah so we put it out and then we see how players use it like this we just test it test it internally and and and try to figure it out I think we're running out of time uh so have to wrap this up uh there's actually like there's this wrap up room I see there's like you at least have some questions maybe others have like I'm going to go to the uh this wrap-up room is right across I think it's Overlook 2022 am I right I don't know uh so I'm going to go there if people want to come there and ask more questions they welcome to do so okay thank you very much