/dev Teamfight Tactics: Galaxies Learnings
Intro
Here we are again, at the end of another set. That means it’s time to reflect on how we made use of what we learned from the previous set, as well as discuss the lessons from Galaxies that we’ll be taking forward. We’re always trying to make TFT into a better game, so we want to share with you how we’re going to do that.
TL:DR
These articles are always deep-dives, so if you're not that into reading feel free to cut out after this not-that-short-itself summary:
Looking Back
- Spells were much more impactful and exciting this set and we want to continue that trend going forward.
- We improved the clarity of all the action happening on your arena, but we think we can do even better and will continue to work on this in Fates.
- We have a better grasp on what carries look like at various costs and we'll continue to focus on making satisfying carries for each tier.
- Overall we think most traits were viable, and most patches had many possible endgame compositions, but a few traits struggled to consistently find an identity.
- We were happy with Spatula consistency and power, so you can expect more of the same going forward… except for you Demo Spat.
Looking Forward
- We really enjoyed how distinct the Galaxies theme felt compared to previous sets, especially with the added flair of the Galaxies-specific carousel. Look for us to continue to play up the themes of future sets.
- The number of champions in the "pool" is particularly important in a game like TFT. Changing that number has cascading impacts. The next set will launch with 58 champions and will stay at that number throughout.
- The Galaxies mechanic was really exciting, even if not every Galaxy hit the mark. We're currently exploring the idea of adding a similar mechanic as a permanent part of TFT.
- There will no longer be full item carousels, and overall much less wacky variations of carousels (except the rare FoN).
- We want items to be more flexible for various compositions and vice versa. Certain items should be strong for certain compositions but we would like them to be less necessary.
- MOAR quality of life and UX updates! MOAR we say!
- We're going to be more communicative and specific about the goals of each gameplay update (patch).
Looking Back
First, let’s look back on the things we said we’d improve from the end of Rise of the Elements. (You can read that article here: https://euw.leagueoflegends.com/en-gb/news/dev/dev-tft-rise-of-the-elements-learnings/)
Spell Impact and Excitement
In Rise of the Elements we struggled to deliver some really high-impact and exciting spells, and promised to improve this to create a more exciting viewer experience.
In Galaxies, we think we made good on that promise. There were a lot of high-spectacle champions at various cost levels that created some hype moments. Darius dunks and Ziggs bombs in the early game, Vayne tumbles and Shaco outplays in the mid game, Vel’Koz lasers and Kayle barrages in the late game—and that doesn’t even include our legendaries like Miss Fortune, Gangplank, Ekko,Urgot, and of course the Super Mech! Overall we think we made great strides here, and you can expect to see us continue that trend with plenty of exciting moments and a great viewer experience in future sets.
Combat Clarity
We also promised to improve combat clarity from Rise of the Elements, which suffered from overfilled boards and a lot of invisible power. It was often quite hard to track what was going on.
For Galaxies, we made some pretty significant progress here too. Lots of spells were slowed down, given travel time, and made clearer. We also made sure that keychamps stood out so players could easily track them, including Caitlyn, Ziggs, Zoe, Rakan, and more.
That’s not to say we were perfect; there were still some instances of spells with overly subtle effects or targeting. For example, it was hard to pick out which enemies were blinded by Graves’ smoke grenades, Yasuo’s quick movements made him difficult to follow, and Ezreal’s spell faded out when it traveled too far before impact. While we’re happy with the progress we’ve made, we’ll continue to make sure combat is as clear as possible moving forward.
Champion Balance & Design
We promised that we’d improve champ and spell designs in Galaxies with more satisfying carries across the cost spectrum. We delivered on this, but not consistently. In the first half of Galaxies we added some powerful 1-costs (Xayah/Kha’Zix/Ziggs), 2-costs (Darius/Xin Zhao/Kai’Sa), 4-costs (Kayle/Vel’Koz/Jinx/Jhin), and 5-costs (Miss Fortune/Gangplank/Ekko) but failed to really deliver in the 3-cost space. At the time, Master Yi and Syndra were underperforming and most of the other 3-costs were utility champions.
In the second half of Galaxies, we added and adjusted a lot of 3-cost carries (Ashe/Jayce/Cassiopeia/Master Yi/Rumble/Shaco/Syndra/Vayne), and improved the 5-cost space as well (Ekko/Aurelion Sol/Urgot). However, we also struggled a bit in the 1-cost and 2-cost range, with little change in the 1-cost pool and 2-costs mostly falling out of favor.
Where we struggled, we learned. In Galaxies we got a really good grasp of what kinds of champions should be in each cost bucket. At this point we’re quite confident we understand what it takes to make a good 1, 3, 4, and 5 cost champion. We’re still figuring out what a great 2-cost champion looks like, and their exact role in a team. That said, you can expect us to deliver carries at all cost levels that are fun and satisfying to play. (Satisfying to play being a big one here, though we think Kayle will go down as many players’ personal favorite TFT champs of all time!)
Major vs Minor Traits
We also wanted to ensure that investing deeply in major traits (6+ pieces) or spreading your comp more widely across different minor traits were both viable strategies. Some of the traits were very successful in this respect. Blademaster, Infiltrator, Mystic, Protector, Sniper, and others saw builds of all shapes and sizes. You could play flexibly around what you did or didn’t get, making for some fun decision-making challenges.
Other traits didn’t quite clear the bar. Cybernetic required an early 3 and then you were locked into 6 or you weren’t playing it, and Battlecast forced players into a similar spot. For most of the set, Chrono may as well have been a 2 or 4-piece origin only, and Star Guardian had some rough patches where it was 6 or bust. With all that said, overall we think Galaxies landed in a great spot regarding this. It’s likely we’ll never get every trait to be viable equally at all stages of the game, but expect us to continue to make sure that as many traits as possible are viable across all the stages of the game to open up as many unique builds as possible.
The Spatula
Finally, we promised to bring Spatulas back to the carousel in a more controlled and unique way. For the most part, Spatulas worked better in Galaxies than in Rise of the Elements. No longer could you force compositions because you were basically promised a Spatula. But when Spatulas did show up, new possibilities opened up in all sorts of ways, like with Infiltrator Irelia, Blademaster Zed, Protector Aurelion Sol, and more! These units were all strong, but none were oppressive.
That’s not to say Spatulas went down perfectly though. The Demolitionist Spatula in the first half of Galaxies certainly caused a few issues with Aurelion Sol and Kai’Sa. No, you couldn’t force the comp but when you did hit it, the power increase to your team was far beyond what we’d like to see. So, moving forward you can expect to see similar Spatula appearance rates, and unique combinations closer to the second half of Galaxies. We’re not looking to bring back Blender Nocturnes, but are happy with Protector Aurelion Sols!
Looking Forward
Theme & Fantasy
Compared to Rise of the Elements, we think Galaxies was much more successful on a thematic level. First, going for a more distinct theme compared to classic Runeterra vibes helped players feel like they were being taken on a journey to a different slice of the League of Legends Multiverse. The increased effort into theming more than just the Champions via Skins - things like the updated Carousel, more unique Arena Skins, and an update to the music track - also helped. We also prefer Galaxies’ approach to Champion Origins, particularly their focus on alternate Champion fantasies and grouping that you may not expect from having played Summoner’s Rift. Going forward, we want to keep expanding both the types of themes we explore, and how distinct they feel from one another. This doesn’t mean that we won’t return to a Runeterra focused Set someday or that you’ll never see a default skin again; but we would want to execute on that as the Set’s theme, rather than splitting the difference between feeling partially Runeterran, and partially a different piece of the Multiverse.
System Changes & Champion Counts
One lesson we learned is that changing our systems does a lot of harm to the stability of the game. When we launched Galaxies in 10.6 we not only started off at 52 champions, but also changed drop percentages, gold streaks, player damage, and some other minor adjustments. However these systems weren’t where we wanted them, and the result was some side effects like it being way too easy to force 3-star champions, early game damage not mattering enough resulting in everyone leveling to 8 as fast and safely as possible, and anyone who high rolled a win streak early snowballing into a dominating victory. Then we had to make adjustments in 10.7, 10.8, and 10.9 before we finally got those core systems stabilized. And even then it still wasn’t quite right as we saw players all forcing 1-cost 3-star champions in Patch 10.11.
THEN, when we launched the mid-set update in patch 10.12, we increased the champion pool to 57. All of a sudden things were off again, forcing us to adjust these core systems even further. It was almost too difficult to 3-star champs now with the extra champions, and sometimes finding a specific champ you needed for your comp became much harder. While we were able to stabilize these systems for the most part, we really don’t want everyone to have to go through that again, where all of the base rules shift in every patch.
So here’s what we’re changing. The next set will have 58 champions for the entirety of the set, and the core systems shouldn’t change after the initial set launch. This means if we add any champions to a set, we’ll remove others to keep that number at 58. This ensures that all the core systems in the game (like reroll chances, player damage, et cetera) can be kept stable. We know changing these systems is extremely disruptive since it changes when you level, when you roll down, and so on. Locking these in should allow you to focus on improving your strategy rather than keeping up with how to actually play the game.
The Galaxies Mechanic
The Galaxies Mechanic was one of our more ambitious system mechanics because it allowed us to learn more about the game, how all the systems work together, and which mechanics players preferred
That said, we certainly had our fair share of less popular galaxies. Binary Star might have been more interesting if Cybernetic didn’t drastically benefit from it. The same goes for Dwarf Planet which benefitted strong AOE comps like Super Mech and Protectors, and also reminded players what the launch set board felt like.
Other galaxies like Medium and Littler Legend gave us insight into player health adjustments, while Lilac Nebula and Star Cluster showed us how adjusting the champs on the carousel can alter the flow of the game.
But we also saw some really cool variants to the core game. Grabbing the extra item components from Galactic Armory, raking in the extra income from Plunder Planet or Treasure Trove, and getting a free reroll from Trade Sector all changed the game in really fun ways. Super Dense also opened up a bunch of new composition possibilities. All of these got players super excited and added to the life of Galaxies. It would be a shame to throw all that away. So with that in mind, sometime in the future, we will be exploring the idea of bringing back some of these variants back to TFT as a permanent part of the game. More details to be announced when they’re ready!
The Carousels
With Galaxies we introduced a lot more variety in the carousel system. This included things like full items, offense and defense components only, all the same component, and even all-Spatula carousels. The response was as expected… mixed. It certainly succeeded in creating varied games and memorable moments—things that you’d only see once in the entire set! But it also caused some problems by taking choices away from players.
So what did we learn from this? First, we still think variance is good in the carousel system, as it prevents certain strategies from being copy pasted from game to game. But, we went too far in the other direction, taking control of the game away from the players. No amount of strategic foresight can prepare you for the game where you get 3 full item carousels.
We learned a lot about the importance of player agency and decision-making in the item system. Give someone a full item and they’re left feeling helpless—you’ve pigeonholed them into a specific strategy. But give that same player item components, and suddenly you’ve opened up a world of possibilities, making it feel like you’re more in control of the game.
In the next set, you won’t see full item carousels (except the rare all Force of Nature ones) so you’ll always have control over your game. You’ll also see a more predictable style of carousels so that you can anticipate a more variant outcome at specific times. More details once the next set hits PBE, but you should feel like you can actually plan around the possible outcomes of the carousel from now on.
Item Flexibility
Making good TFT items means creating exciting interactions when you put them on certain champions, while also making sure that getting a certain item doesn’t lock you into a single build path. We’ve been pretty hit or miss in this area. Some item improvements like Giant Slayer allowed for more flexible uses, so caster comps weren’t in trouble if they ended up with bows and swords. But other items weren’t so forgiving.
There’s two big culprits here. The first is items that are too closely tied to specific champs. Take mana generation items Blue Buff and Spear of Shojin. These items take champions and SO DRAMATICALLY change their output, which makes it feel like you must have this item in order to play this champion. Cassiopeia is a great example—extremely weak without Blue Buff,extremely powerful with. The same goes for using Spear of Shojin with champs like Ashe or Janna.
Our second culprit is items that are too closely tied to specific traits. For example, using Warmogs with Protectors caused a larger power spike than should be allowed to exist. Battlecast and Red Buff/Morellos and Blasters with Red Buff are just a few other examples where the relationship between the two is just too impactful. We like traits having specific ties that make you want to optimize them, such as Blademasters with Statikk Shiv or Vanguards wanting Dragon’s Claw. We just want to make sure they can succeed without those key items.
When it comes to items, we’ll be working to ensure that you don’t feel locked into a particular build, or (vice versa) that you must have certain items to make builds function. Now it’s important to note here that this doesn’t mean all items will impact each build equally. There will still be cases where you want specific items more than others,but the difference between getting that item and not should not be as pronounced.
Quality of Life Changes
In patch 10.16 we added what we call item recipe hints where you can right-click a component to see all the items it builds into. This feature was a long time coming. Soon you’ll be able to see some new meter types including damage blocked and healing done, another long requested feature. The reality is we’d been neglecting these kinds of features as we prioritized getting the core game more stable, building the mobile version, and making sure our pass experience was ready. Now that we’re beginning to stabilize more, we’re going to shift some priorities back to these quality of life features. So, please let us know what features you’d like to see and expect us to continue improving the game with these kinds of features.
Balance & Patching Strategies
Finally, let’s talk about balance strategy. We had some great patches in Galaxies...and some not so great ones. With the two week patch cycle, we found ourselves often making too many changes too often, which would shift the meta on its head. Sometimes this went very right (like in 10.7) but other times it went very wrong (like in 10.14 and the Jarvan patch). Regardless of the outcome, we agree that such frequent, dramatic changes are too impactful to the core experience. However if we limited ourselves to making small changes every 2 weeks, it may take 3-4 patches before something reaches its desired state, which isn’t what we want either.
Expect to see patches shift to what you saw towards the end of Galaxies. There may be 1-3 high-level goals in a patch (For example in 10.16 one of those goals was “Improve the underused items”) and then some much smaller changes, only shifting 1-2 things per composition. From there, we may use the “B-Patch” option a bit more often to ensure things are moving the right direction. Ideally we won’t have to, but doing so will allow us to be more surgical in our changes so you don’t experience as much whiplash with each patch. You’ll see this change in balance and patching strategy continue into the new set, with lots of communication from us on how it’s going. If it doesn’t work out, we can shift back to our old cadence, but we think this will be healthy for most players' experience in the long run.
The End
And so we say goodbye to TFT: Galaxies. It’s been a fun ride, but let’s see what fate has in store for TFT with the next set. On behalf of the entire TFT development team, thank you for continuing to play and enjoy the game. It’s a crazy time right now, and if we can bring a little joy to your day, that brings us joy. Until the next mission, take it easy!