Teamfight Tactics Set Seven YouTube Videos Playlist
Set Six was a popular and successful revival for Teamfight Tactics after the series of missteps that plagued Set Five. Despite striking gold with the concept of the hextech augments introduced in Set Six, the developers went out of their way to replace as many units and traits as possible for the midset update that created Set 6.5, resulting in a muddled mess that left few players satisfied. The developers were much too ambitious in what they tried to achieve for the midset update and the best that could be said in their defense was that many of their resources were likely going into the forthcoming Set Seven. Of course Teamfight Tactics never stays still for long and barely three months later it was time for Set Seven to be unveiled to the public. The official name for this set was "Dragonlands" and Dragon units were the main feature for its duration.
The Dragon units proved to be a good theming mechanism that gave its own distinct feel to Set Seven and the continuation of the augments from the prior set resulted in gameplay that was mostly fun to experience. However, there were fundamental flaws in the Dragon units and the trait design for the set which made it nearly impossible to balance for most of Set Seven's duration. The developers only made things worse with an embarassingly bad series of patches which required constant hotfixes and emergency "B" patches that brought back all the worst aspects of balance thrashing from earlier sets. The fundamental core mechanics of Teamfight Tactics are great and for much of the set it felt as though Set Seven was fun IN SPITE OF the developers who were seemingly doing everything possible to wreck things. This set shone a harsh spotlight on the flaws with the current structure used by the design team which doesn't inspire much confidence for the future. Teamfight Tactics would greatly benefit from bringing in some new faces to break out of the stagnant pattern of new sets launched every 90 days with no room for alteration based on changing circumstances.
The main set mechanic for Set Seven was Dragon units and there turned out to be seven of them in total. Dragons counted as two units on the board in the same fashion as Colossus units from Set Six and each Dragon similarly provided three ranks of their unique trait. For example, Shyvana counted as three ranks of the Ragewing trait and the lightning Dragon Ao Shin gave three ranks of the Tempest trait and so on. Dragon units also gained additional health and higher stats than normal to offset the penalty of eating up two slots on each team. The drawback was that only one Dragon unit could be played on the board at a time or else they lost all of these advantages while still taking up two team slots. Outside of finding the uncommon Dragon Alliance/Dragon Horde augments, this limited each team to only a single Dragon unit since it wasn't worthwhile to play multiple Dragons without them gaining their extra HP and tripling the value of one of their traits. Dragons also cost double the gold of standard units and only appeared at the most expensive cost tiers, therefore being limited to 8 cost and 10 cost units at double the expense of the normal 4 cost and 5 cost units. The design goal was obvious: each player would build their team around a single powerful Dragon unit and then have those Dragons square off to determine the winner of the lobby.
We'll get into some of the problems with this design choice in a minute but first let's give the set design some credit: Set Seven truly did feel as though it was built around the Dragon concept. I mentioned last year that Set Five's "good versus evil" theming mechanism was a real dud and the "Gizmos and Gadgets" theme for Set Six didn't do much to inspire players beyond including a whole bunch of references to the Arcane Netflix series. By way of contrast, players really felt the "Dragon" aspects of Set Seven between the new artwork used for the loading/carousel screens and the presence of all the Dragon units. The developers had a real challenge on their hands because there are only two Dragon champions in League of Legends, Shyvana and Aurelion Sol, and they clearly needed a lot more Dragons than that for this set. They wound up doing a bunch of palette-swapped versions of those two champions and then pulled in Galio's character model as well to round out the group. They honestly did a nice job here at making the seven Dragons feel distinct and giving them their own gameplay mechanics which wasn't easy when creating entirely new characters like Sy'fen and Daeja out of thin air. There were a lot of comments during the first week of the set about how Shi Oh Yu resembled a ridiculously green-colored Galio and then everyone quickly got used to the new units and never mentioned that stuff again. This was a difficult feat and the developers did a fine job of creating new units from scratch that had never been playable in League of Legends.
Most of the development work clearly went into creating those Dragons which left less time available to work on trait and unit design. The trait design was adequate for Set Seven overall, better than stuff like the unbelievably-lazy Striker trait from Set 6.5 while not being as good as what players had experienced in Set Six. My picks for the best new traits were Guild and Jade, with Guild providing a unique stat bonus for each Guild unit placed on the team which grew in multiplicative fashion for each additional Guild unit added. Players could invest heavily into the trait and often do well or simply splash in a random Twitch for attack speed or Talon to grant everyone else additional AD. The Jade trait granted players the ability to place Jade statues on the battlefield which granted ticking health regeneration and attack speed to anyone who stood next to them at the start of combat. I'm a huge fan of anything that adds more depth to the positioning side of the gameplay like the Socialite spotlight from Set Six and Jade was a creative way to introduce real strengths and weaknesses to clustering around those weird green statues. And while I almost never played it myself, the concept of Shimmerscale as an economic trait that granted additional gold-generating items was a fun new spin on the Pirates/Mercenaries concept from earlier sets.
Other traits were less successful, either bringing back ideas from previous sets or trying concepts that didn't really work in practice. The Trainer trait that increased the power of the "Nomsy" unit over time was a neat idea that proved hard to balance in practice; poor Nomsy ate so many nerfs over the course of the set that it was down to something like 200 starting HP by the end. Set Seven brought back the double casting Mage trait which was responsible for its own repeated balance headaches, with Mages typically either oppressively strong with their double casts or unplayably weak. Having Mage Cap as a makeable spatula item only made this worse and probably wasn't something that the developers should have allowed. The Astral trait was a balancing nightmare in its own right as it made it far too easy to three star the various Astral units. There were long stretches of time where Nami, Varus, and/or Aurelion Sol were absolutely crushing lobbies because it was trivially easy to three star the non-legendary units and two star Aurelion Sol. The developers should probably stay away from any trait design that alters what units will appear in the shop; this is much more abusable than the Draconic/Yordles traits from earlier sets which could only provide their benefits once per round.
Aside from its clear theming mechanism, the best aspect of Set Seven was probably the augments which were retained from Set Six. The developers rebalanced the augments by thankfully cutting some of the least-interesting options from Set 6.5 while introducing new ones specifically for Set Seven. They also tweaked when the augments appeared by shuffling them slightly forward in terms of their timing and granted the player the option to reroll their augment choices one time over the course of the game. These were good decisions and definitely made the augments feel more fair as compared with Set 6.5; my personal experience was that the augments were back to feeling as though the player could almost always get something decent again after getting stuck with three bad choices way too often in the previous set. As I initially noted for Set Six, the augments continued to do an excellent job of providing variety from game to game, rewarding players who could think on their feet and pivot around their augment choices as opposing to forcing exactly the same team compositions in game after game. However, as great as the augments are from a gameplay perspective, I have to admit that they are starting to get somewhat stale after having them present for three consecutive sets. I hope that the developers are brainstorming ways to keep the same flexibility and variability that the augments provide while coming up with fresh gameplay mechanics for future sets.
Finally, I also have to mention the other new mechanic from Set Seven: the addition of the Treasure Dragon. This was a replacement for the minion round on Stage 4-7 whereby the player would get to choose from a list of different full items and item components rather than having the normal drops pop out from the minions. I was extremely negative towards this change before the set released as it felt like it would devalue the whole idea of playing the hand that the game deals and being flexible with item components, things that I view as fundamental to TFT's gameplay. After having the Treasure Dragon in place for a whole set, I have to say that it's... OK, I guess? I don't love it but I don't hate it either. I do wish that it was more expensive to reroll for a new set of items at the Treasure Dragon as the 1 gold cost feels too cheap. Then again, gold is precious in TFT and the player who can make strong items by spending 1-2 gold is gaining a major advantage over the dummy who rerolls 15 times.
I was also wrong about the Treasure Dragon giving an advantage to people greeding for best-in-slot items to place on their carries. It actually does the opposite by penalizing players who deliberately loss streak throughout the early game in order to secure first pick on the item carousels. By allowing the player to find one or two item components that they absolutely need via rerolls, the Treasure Dragon favors players who play strong boards throughout the whole game and find themselves getting last pick on most of the normal carousels. They still have a way to get that key item component that they can't get from being last pick and haven't been lucky enough to drop from the other minion rounds. Since I prefer to play from ahead and don't like sacking health for carousel priority, the Treasure Dragon has almost certainly helped me more than it's hindered me over the course of the set. I'm basically at the point where I'm fine if it remains and I won't shed any tears if it gets scrapped for a new gameplay mechanic. It's vastly better than pulling Radiant items out of a box though, that's for sure.
I've been trying my best to stay positive up to this point when discussing the Dragons since Set Seven did a great job of creating a world that felt as though it was inhabited by these mythical creatures. However, we've reached the point of the retrospective where I have to address the elephant (dragon?) in the room: the Dragon units were inherently flawed from a design perspective, flawed in ways that made it basically impossible for them ever to be balanced. There were two intrinsic problems to the Dragon units that couldn't be solved through clever tweaking of numbers: power and flexibility. The power of the Dragons was an inherently troublesome issue due to the fact that they ate up two unit slots and therefore had to be worth the cost of putting them on the board. At the start of Set Seven, the Dragon units were generally too weak to be worth playing because their numbers were set too low to justify requiring those two slots. Thus Xayah was the dominant unit throughout the early portions of the set, with truly absurd play rates and average finishing position in the data. The Dragons were so underpowered that people weren't even running Shyvana, the Ragewing Dragon, in their Ragewing team compositions featuring Xayah. It was better to add two more Guild units instead since the Dragons were too weak to be viable. Since so many of the set's 4 cost and 5 cost units were Dragons, there weren't enough options to build different team compositions and everyone was forced into some combination of Xayah, Corki, and Olaf. It was a frustrating metagame environment with too many people forcing the same teams for lack of choices.
This was a real problem for a set which was supposed to be geared around Dragon units. The developers therefore buffed the Dragon units across the board while nerfing the Xayah and Assassin Olaf teams which had been dominating the gameplay at the higher tiers of the ladder. This certainly shifted the metagame towards an emphasis on the Dragons but only created new issues which were equally problematic. As soon as the 8 cost and 10 cost Dragons became strong enough to be worth playing, they immediately became oppressive and players who were lucky enough to hit them started crushing the players who weren't as fortunate. It was common for someone to hit Sy'fen or Shi Oh Yu at 2% odds when they were Level 5 and suddenly inflict massive 10 damage losses because non-Dragon teams couldn't possibly deal with all of the stats on the Dragon units. The 10 cost Dragons were weak for much of the set but they were also inevitably buffed about halfway through Set Seven which resulted in a metagame of everyone trying to push to Level 8 and then roll for the 4% odds of the legendary Dragon units. The players who hit would win the lobby and those who missed would get knocked out with a Bottom 4 finish. This was no better than the Dragons being ignored for being too weak and may have even been worse due to the pure RNG luck involved.
Riot Mort, the lead developer for Teamfight Tactics, one compared the intended power level of the Dragon units to the combination of Jhin and Orianna from the previous Set Six. Jhin and Orianna shared the Clockwork trait in that set and were basically always played together which made this an appropriate comparison, with an 8 cost Dragon intended to equal the power level of two 4 cost normal units. However, there was a fundamental difference here: players had to find Jhin and Orianna separately in Set Six; they didn't magically come as a matched pair in the shop. Because Dragon units appeared in the store at the exact same shop odds as normal units - but were intended to have double the power of normal units - hitting them at low odds spiked the strength of a player's board to a much, much greater degree. It would be like finding Jhin *AND* Orianna at 2% odds back in Set Six rather than just one of them, a preposterous high roll that essentially never happened. This was like having all of the worst aspects of the Chosen units from Set Four back again without having the positive aspects to offset them.
The other intrinsic problem to the Dragon units was their lack of flexibility. Because only one Dragon unit could be played on the board at a time, players were severely constrained in terms of being able to pivot between different team compositions. There were only eleven 4 cost units and eight 5 cost units in the entirety of the set and seven of those units were Dragons (four at the 4/8 cost tier and three at the 5/10 cost tier). And as soon as the player had Idas on their team, all of those Daejas and Shi Oh Yus and Ao Shins became completely useless whenever they appeared in the store. Roughly a third of the most expensive units at the 4 and 5 cost tiers quite literally could not be played due to a single Dragon locking out all of the other Dragons from viability. Now that's always been an issue to some degree in Teamfight Tactics (it's not as though someone running a bunch of Arcanists in Set Six would suddenly want to play Fiora or Akali) but Set Seven only made things worse. Players couldn't even hold alternate Dragon units on their bench for a potential pivot because of their incredibly expensive 8 and 10 gold costs. This was a terrible set in terms of flexibility due to the very nature of how the Dragon units were designed.
The developers made other baffling design decisions in constructing the set that left the gameplay balance in a weird state for the entirety of its run. Riot Mort stated before the set's release that they deliberately tried to create a set based around 3 cost carries instead of the traditional 4 cost carries. This made very little sense as the design framework of Teamfight Tactics essentially forces 4 cost units to form the core of most team compositions. The 1 and 2 cost units are too weak to carry endgame boards most of the time, even when three starred, while the 5 cost units are too difficult to find on a consistent basis from game to game. Similarly, the 3 cost units are great for the midgame but aren't strong enough to win lategame teamfights unless they've been three starred which, again, isn't consistent from game to game. The 4 cost units have always been the core carries that most teams are designed around and this won't change barring major alterations to the economic structure of TFT's gameplay. The fact that TFT's own development team didn't seem to understand this was not encouraging, to say the least. And if the goal was to create a set designed around 3 cost carries, it was a total failure: Assassin Olaf and Astral Varus were the only units at the 3 cost tier that saw consistent use as carries. Once again, it was the 4 cost units that dominated most of the set's gameplay: Xayah, Corki, Shi Oh Yu, Sy'fen, Daeja, and so on.
More bizarrely of all, the TFT development team did not include any AP carry units at the 4 cost tier! This was utterly inexplicable and led to AD units and AD items dominating the gameplay for most of the set's duration - Corki and Xayah were the only non-Dragon 4 cost carries and both used AD itemization. The only units that could plausibly fit into the AP category were Sona and Daeja, with Sona being a support unit that lacked enough damage to be a main carry and Daeja more oriented around attack speed than ability power. This was mollified eventually when Daeja was reworked to make AP more useful on her and the set finally had something resembling an AP carry, albeit one that couldn't be played with other Dragons and had the variability of the Mirage trait shifting from game to game. Conversely, after there were no AP carries at the 4 cost tier, all three of the Dragons at the 5/10 cost tier dealt magic damage! Ao Shin, Aurelion Sol, and Shyvana all scaled off AP and shared related if not identical itemization patterns as different forms of burst casters. Ao Shin and Aurelion Sol in particular were so similar in their designs (magic dragons that blasted the board with flashy spellcasts) that I've never understood why both of them were placed at the same cost tier. Shouldn't one of them have been a 4/8 cost unit and the other one been a 5/10 cost unit? Why did two different Dragons have essentially identical functions at the same cost tier? And that wasn't even counting Pyke who was an AP Assassin at the same 5 cost level. Thus the set lacked AP carries at the 4 cost tier while having a pointless glut of them at the 5 cost tier - what was the logic here, again? 
There were a number of other weird design decisions that didn't make much sense when subjected to scrutiny. The Assassin trait was back again as in every previous set but this time seemed to be nothing but a bunch of utility units with no carries at all. Qiyana and Diana were clearly utility units while Talon's ult was too inconsistent to use him as the primary damage dealer. Talon was mostly played for his Guild AD benefit, not his actual ability. With Kayn arriving too early and Pyke arriving too late, this meant that the primary Assassin in the set was Olaf who didn't even have the trait at all! Olaf was so oppressive during the early weeks of the set that he had to be nerfed into the ground, making the non-Assassin version of Olaf virtually unplayable. Mages were another trait that seemed to lack any carry units outside of arguably Ryze who scaled up very slowly over the course of each round. This put Mages into the weird position of being stalling/healing comps which were extremely difficult to balance. Either the healing/shielding from Nami and Vlad and Sylas was set too high and they would win rounds with every unit surviving, or the healing/shielding was set too weak and Mages would simply get crushed without killing anything. Like Assassin Olaf, the best Mages turned out to be units that lacked the Mage trait entirely and thus Mage Sona and Mage Ao Shin/Aurelion Sol wound up spending a lot of time as problematic units.
The Dragonmancer and Legend traits were even more of a balance nightmare due to the ways in which they functioned. Both of these traits essentially tried to create a single super unit, with the Dragonmancer trait giving bonus health and AP to a single Hero unit while the Legend trait literally sacrificed some of the player's own units to give additional stats to the Legend units. These traits were combined together in the form of Volibear who had brief runs of being ludicrously overpowered before hotfix nerfs rendered him all but unplayable. While the Dragonmancer and Legend traits were cool ideas in theory, they didn't really work in practice because gearing the entire gameplay around one super unit turns out not to be much fun. The game is Teamfight Tactics, not Solo Unit Tactics after all. These traits were almost impossible to balance because the numbers tipped either towards utterly unkillable super units or uselessly weak units with little room for anything in between. The developers gave up on these traits pretty quickly and left them underpowered for most of the set which was honestly the best option since the alternative would have been worse.
A lot of these problems with Dragons and trait/unit design were fundamental to how Set Seven was constructed. Still, they might have been remedied with careful, patient balancing work carried out after extensive testing. What we got instead was...
Set Seven had the single worst patching process of any Teamfight Tactics set to date, a complete clown show that left an embarassing stain on the development team. I don't intend that as hyperbole and I'm well aware of how disastrously poor the patches were in some of the earlier sets. However, given the context of the fact that this was Set *SEVEN* and the development team has been doing this same process for three full years, there was absolutely no excuse for the train wreck of sloppy decision-making and obvious mistakes that this set stumbled through. I'm grading them on a curve because they should be getting better at this stuff with time, not falling into the same easily-spotted pitfalls over and over and over again. The only conclusion that I can draw is that the development team enjoys the neverending balance churn and deliberately breaks things in a misguided attempt to keep the game feeling "fresh" by rotating around different overpowered flavor of the week units. I continue to maintain that this is doing more damage to Teamfight Tactics than it is bringing in new players and Riot Games would be well served to bring in different individuals with less of a destructive mindset.
The problems with Set Seven began even before the set was released with the testing environment on the PBE server suffering through an endless sequence of bugs and broken champions. Anyone who plays on the PBE server should expect to come across unfinished gameplay elements and units with unbalanced abilities that need to be refined through further testing. What they should not expect to find, less than two weeks before the set's release date, was gameplay elements which would actually hard-crash the game back to the desktop. The Philosopher's Stone Shimmerscale item had to be removed from the game since it was locking up the TFT software in what was already a bad sign. Gameplay balance was even worse than normal for the PBE (and that's saying something) with Volibear extraordinarily overtuned for several days followed by the Mage Cap version of Aurelion Sol serving as an instant win button. Much like Set 6.5 before it, Set Seven clearly needed several more weeks of testing and polishing before it was ready to be released to the public. But the launch date for the set had been predetermined eight months in advance, for no clear reason, and therefore Set Seven had to be pushed out the door. Ready or not, here it was.
The release version of Set Seven immediately deployed with a hotfix patch to tone down the numbers on Aurelion Sol who had been crushing everyone on the PBE server during the last few days prior to release. The release version couldn't even make it one week before the developers released another "micropatch" that shipped out nerfs to Nami and other Astral units. (There were so many small patches released during this set that the patch tracker on the LoL Chess website doesn't even list them all.) Mages were very strong at the time and most players were running either a Mages board or something involving Xayah. The set could have stabilized at this point through modest adjustments to tone down the stronger units/traits and modest buffs to improve the weaker ones. But that's not the design philosophy of the TFT developers who never see a reason to make 5 changes when they could make 55 changes instead. The first full patch of the set contained this absolute disaster:
This was only a fraction of the total patch notes; the full thing had well over 100 lines of changes. That's not an exaggeration, click here to read the ridiculously long list of things that the TFT developers couldn't stop themselves from tinkering with. Furthermore, these were not the precise surgical adjustments of a careful artisan but rather the drunken stumblings of a college fratboy out on the town. The developers comically overbuffed Volibear in four separate areas, buffing his base attack speed, his bonus health on ability activation (by more than 50%), the damage from his ability, and the AP scaling from the Legend trait. They also buffed Anivia, Lee Sin, and Swain (units typically played alongside Volibear) while nerfing all of the units/traits that had previously been strong. The patch notes were previewed the day before they went live and everyone with any knowledge of TFT's gameplay immediately pointed out that Volibear would be stupidly overpowered if these changes went forward. The community knew this because we had already seen it in action on the PBE server: if Volibear's numbers were set too high, he became completely impossible to kill and crowded out every other team composition. Surely the developers would make last-minute changes in response, right? They wouldn't actually release a patch with a unit which was so obviously broken... right?
Well the patch was released with no alterations and - would you believe it - Volibear was hilariously overpowered. Who could have possibly seen that coming? 
The situation was bad enough that the TFT developers released an emergency hotfix patch less than 24 hours later which undid almost all of those buffs to Volibear and sent him right back into unplayable status again. It was a terrible look for the developers who went out of their way to ludicrously overbuff a unit and then immediately undo those buffs with a hotfix patch. It begs the question: why buff the unit in FOUR different ways simultaneously?! Why not make one of those changes, see what happens, and then make another change afterwards if the first one was too weak? It's not like Teamfight Tactics has a limited number of updates, the game gets patched at least every two weeks and more frequently on a weekly basis. There is absolutely no reason to pull this kind of nonsense unless it's an intended decision on the part of the developers.
Naturally there was another full "B" patch released the next week after the Volibear hotfix with close to 50 additional lines of changes and then the usual biweekly patch following seven days later. This time the developers inexplicably cut Nami's mana cost from 90 down to 75 which immediately pushed Astral and Mage teams into oppressive status. It was a classic example of a unit being tinkered with for no reason whatsoever and resulted in the third emergency hotfix patch of the set (and the second one involving Nami!) which reverted her mana cost back to the previous value of 90 two days later. At this point Set Seven had gone through seven different patch versions in six weeks, with three hotfixes and an offweek "B" patch, yet another dizzying sequence of balance changes which was nearly impossible to follow. Defenders of the developers will often praise them for being responsive and making changes quickly but after making these mistakes again and again across multiple sets I don't think they deserve the benefit of the doubt. Should we really be lauding a development team that can't even make it through two weeks without breaking something so badly that they have to do an emergency repair job?
The reader may think that I'm being too hard on the developers and maybe these were honest mistakes on their part. Stacking four simultaneous buffs onto Volibear was clearly excessive but perhaps the development team learned their lesson and would avoid those same problems in the future - right? Sadly, it's clear that they learned nothing at all because the next patch gave the exact same treatment to Aurelion Sol:
The developers buffed his base health, lowered his mana cost from 90 to 60 (!!!), massively increased the damage of his Black Hole ability (by more than 50% at two stars), lowered the time needed for his ascension to kick in, and increased the odds of Astral 9 trait producing an item from 30% to 75%. The minor reduction in the damage from the ascended Black Hole in no way compensated for the *FIVE* simultaneous buffs to the Astral Dragon. Aurelion Sol had barely been played throughout the set and needed some improvements but this was the stupidest possible way to handle the situation. Once again, everyone with any sense pointed out on TFT Reddit and Twitter that Aurelion Sol would be impossibly broken if released in this state. Once again, the developers released the patch with no changes and of course Aurelion Sol proved to be absurdly overpowered. The unit turned out to be even worse than anticipated because players were able to exploit a bug with the Astral trait, tossing in the other six Astral units in between combat rounds and then rolling for the 75% odds to drop an item component, then swapping out their unwanted Astral units before combat began. Aurelion Sol was so dominant that hitting a single one star version of the unit could carry a team into Stage 5, even with no other traits active or items placed on the unit. The developers then wanted to hotfix Aurelion Sol but couldn't do so because they had already used up their limit of three allowed hotfixes for the entire set, and thus the unit sat around in utterly broken state for a full week until they could release their next "B" patch.
WHO COULD HAVE POSSIBLY KNOWN AHEAD OF TIME?!?! 
So what's going on here? The TFT developers are not idiots and I refuse to believe that these repeated instances of overbuffing and overnerfing units are happening by accident. The endless gameplay churn has to be a situation that they are deliberately creating through intended choices on the development side. Some people in our local TFT group have floated the idea that the developers deliberately try to make every trait and every unit overpowered at some point in time across the length of the set because they think that this helps keep the gameplay fresh. Under this theory, Aurelion Sol had been weak and rarely seen for most of the set and therefore it was a deliberate decision to buff him into a laughably overpowered state so that he could get his turn at the top of the pile. That's a foolish way to try and increase replayability but it's hard for me to argue otherwise when we've seen the same pattern played out repeatedly across multiple sets.
Unfortunately this also ties into the economics behind Teamfight Tactics. As a Free To Play game, the developers have every incentive possible to keep pushing through rapid changes in order to sell more cosmetic doodads in the store. Of course there's no reason why this has to spill over into the gameplay and force the community to suffer through endless balance changes but clearly it does. We've seen that the TFT sets are clearly coming out too fast and that 90 days simply isn't enough time to get them ready for a wider audience. The development team desperately needs more time to test and polish each release rather than insisting on pushing out four different sets each year. However, the sets have to keep coming out so that new passes can be sold every 90 days as that's the biggest money maker for TFT. It's also why there are so many Little Legends and why the development team keeps creating rare drops out of the slot-machine eggs like the horrible Chibi Yasuo they added for this set. (That Yasuo is one of the outright worst things ever added to the game, forcing players to watch a humiliating cutscene when they get knocked out of the game. It's basically a "pay to bully others" feature built into the fabric of the game - ugh. 
)
It doesn't have to be this way, however. Just because Teamfight Tactics is a Free To Play game doesn't mean that the gameplay needs to go through an absurd amount of balance thrashing and release a new set every 90 days; those are deliberate choices made by the current development team. Riot Mort has said that TFT will always release a new set every 90 days because their internal data tells them that player retention drops off after three months and players are looking for something new. This is a classic example of misreading data and confusing correlation with causation; player retention drops off at the end of a set specifically because everyone knows a new set is about to release and there's little reason to keep playing the old version any longer. I know that I would make a more concerted effort to push for higher Ranked rating if the ladder didn't keep resetting at such a crazy fast pace. It doesn't help either that the development team has released unbalanced "4Fun" patches at the end of most of the sets telling everyone that they consider the remainder of the set's lifespan to be a joke. Of course players are dropping out at the end of each set - you're telling them there's no reason to stick around!
The notion that Teamfight Tactics must undergo constant change at a frenetic pace in order to be successful is belied by the experience of the many, many, many other games that do no such thing and retain a large audience for years on end. I pulled up the list of the most viewed games on Twitch while writing this article and the top two were Grand Theft Auto 5 and Minecraft, both of them among the most popular games of all time, both of them released more than a decade earlier, neither of them feeling the need to release patches at such a frenzied pace. Steam's list of the most popular games included Counterstrike, DOTA 2, PUBG Battlegrounds, Team Fortress 2, Path of Exile, and even Civilization 6 all among the top of the list when I checked to see what was there. Aside from demonstrating that the public loves shooters, I bring up these examples as evidence that it's entirely possible to have Free To Play games that don't release weekly patches and still retain huge audiences for years and years on end. Yes, developers need to release new content periodically but the pace of change used by Teamfight Tactics is wildly excessive. I continue to maintain that this is driving away more players than it's attracting - who wants to have to relearn the game every single week because there's been 75 patch changes since Tuesday? The current TFT development team spends too much time with the "extremely online crowd" that's playing a dozen games every day and fails to understand that this is a tiny minority that's not reflective of the average player.
This is why I believe that the game would benefit from putting some new people in charge with a fresh perspective on what Teamfight Tactics can look like. The same development team has been in place for three years now and they've been getting more and more rigid in terms of how the game is created. It's long overdue to stop insisting on the insane balance carousel and consider that there are other ways to drive player engagement beyond buffing and nerfing every unit and every trait on a weekly basis. We know that this is possible because it's already been done: a local Chinese developer created their own version of Teamfight Tactics (with the blessings of Riot Games) that brought back the units from Set One, plus added a bunch of extra features like a 2 vs 2 mode before it officially came out and a puzzle mode that has never been part of the main client. That was an awesome story - why can't we have that outside of China?! - and it shows the kind of things that are possible and the current development team is simply ignoring. Clearly the TFT development team needs more time between sets because 90 days isn't enough; why not bring back old sets for a couple of weeks to bridge the gap between them? I'd love to go back to Set Four for a couple of weeks while the developers iron out the issues in Set Eight. Riot could even sell more cosmetic stuff by making the between-set breaks into a mini-event. Wouldn't that be better than sticking with a release schedule determined a year in advance and never deviating from it as circumstances change? And maybe that would provide enough time for us to get replays (inexplicably still unavailable after three years), or custom Double Up lobbies, or a tutorial mode for new players, or a puzzle mode, or any of a hundred other things which would drive engagement and player retention. There are opportunities here that the current development team isn't seeing because they insist on doing everything the same way that it's always been done. They did a great job of getting TFT off the ground but the game is long past due for new leadership with fresh ideas.
This restrospective probaly gives the impression that I hated Set Seven but that's not true at all - I mostly liked the set even though it was a total mess from a gameplay perspective. Teamfight Tactics is a game with excellent fundamentals which are hard to screw up; the core gameplay mechanics of building your team and putting together the right combination of champions and items will always be satisfying to experience. I was mostly frustrated during this set by the development team which kept pulling me out of my enjoyment with their ridiculous insistence on changing everything at all times. If they would simply get out of the way and let me play the darned game I would have a much better time. As I've said before, my favorite TFT experiences have taken place during the winter holidays when the development team is out on vacation and physically barred from making changes because no patches can be released. This is a pretty damning indictment of them, that the gameplay functions the best when they do the least, but I firmly believe it to be true. Their misguided insistence that players have the attention span of a fruit fly and must be entertained through neverending changes will be the death of the game if they don't realize what they're doing.
I enjoyed Set Seven enough to play sufficient Ranked games to reach Diamond rating again, actually doing so even faster than in Set Six with a mere 36 games played and an average finish rating of 3.23. I also ran a bunch of Double Up co-op games with El Grillo and antisocialmunky until all of us reached Diamond rating in that mode as well. Without humblebragging too much, I wound up finishing Top 2 in every single Double Up game played, 27 of them in all. I'm definitely continuing to improve at Teamfight Tactics and sooner or later I'll like one of these sets enough to see if I can make the push to Master rating. I'm cautiously optimistic about the upcoming Set 7.5 where the developers have wisely scrapped some of the worst aspects of Set Seven; they are planning to add significantly more Dragons but allow them to be played together (with stacking bonuses for multiple Dragons) while also dropping the cost of the units down to 6/7/8 gold. In theory, this should allow for much more flexibility now that each Dragon isn't locking out all of the other Dragons and less expensive / weaker individual Dragons will reign in the problem of a single Dragon at low percentage odds spiking a player's board too hard. My faith in the balancing abilities of the current TFT development team can best be described using negative numbers but at least the fundamentals of Set 7.5 appear to be much improved.
This was a complete mess of a set which I managed to enjoy in spite of the developers, not because of their efforts. Let's all hope that they can learn from the experience or that the executives at Riot Games will bring in some new people who can do a better job.
