A Return to Form: Teamfight Tactics Set Six

Teamfight Tactics Set Six YouTube Videos Playlist

The various sets released over the three year history of Teamfight Tactics have proven to be pretty hit-or-miss. After the initial release set where the TFT developers didn't really know what they were doing yet, we saw Set Two's Rise of the Elements and Set Three's Galaxies which headed off in divergent directions. Set Two was based around elemental hexes and variable traits for Qiyana and Lux whereas Set Three assigned a new galaxy at the start of each match with its own rule changes. I enjoyed Set Two quite a bit and disliked Set Three due to poor gameplay balance and a bunch of the galaxies not being particularly fun to play; for whatever reason, the community opinion seems to be reversed with Set Three being more popular than Set Two. Set Four introduced the Chosen mechanic and resulted in an explosion of popularity for Teamfight Tactics, without which there's a good chance that the game would have been dropped by Riot Games. Then Set Five took a massive step backwards by focusing on shadow items which proved to be too complicated for most players and not particularly interesting as a gameplay mechanic. The designers essentially abandoned the second half of Set Five by providing minimal content updates to focus on the upcoming Set Six. This led our local Discord group to joke repeatedly that "Set Six will save TFT" since the playerbase was dropping the game in droves due to the unpopularity of Set Five.

Eventually Set Six did arrive and what do you know, it really did save TFT! The set was officially named Gizmos and Gadgets (a poor name choice which everyone ignored) and was centered around a new gameplay mechanic called Hextech Augments. Players were offered the choice to pick three augments over the course of the game which would each provide unique benefits for their team. Some augments granted additional ranks of traits like the old Chosen mechanic, or offered various economic benefits, or did all sorts of wacky things to help the team win combat rounds. Unlike the failed mechanic of the shadow items, the augments were a fantastic addition to the TFT gameplay by adding variation to each match, rewarding flexibility and player creativity, and also providing direction in team composition for players. The augments proved to be incredibly popular and the designers have already stated that they will be carried over into Set Seven. That was certainly a better result than what we saw in Set Five where the shadow items were completely discarded for the midset update.

With Set Six drawing to a close, I'm going to provide my traditional retrospective and take a look back at what worked and what didn't work in this set. This was arguably the best set of TFT ever produced so my comments will mostly be positive in nature. Before digging into the details of the set however, I'd like to start by providing a more theoretical framework on the philosophy used to design TFT gameplay. Set Six (and earlier Set Four) used a design philosophy better suited to TFT's gameplay while Set Five (and earlier Set Three) were fundamentally using the wrong approach for this style of game.

Optimization vs Variation

For any sort of competitive enterprize, players will work to get better over time and refine their approach to develop better tactics and strategies. This could be a traditional game like chess, it could be a video game like Teamfight Tactics, or it could be something random like a hotdog eating contest. If people are competing against one another, they're going to figure out what works and what doesn't work through practice and repetition. As players become really good at a game and approach mastery of the subject, their efforts will eventually lead them down one of two paths: optimization or variation. Optimization gameplay refers to trying to do everything as perfectly as possible to achieve the best possible outcome. Trying to get the highest score possible in a game is an optimization exercise and similarly so is the whole speedrunning community as they try to complete games in the fastest possible time. Optimization efforts typically are based around figuring out the best possible path and then replaying that same path over and over and over again to identify and then remove the most marginal errors. This can be a lot of fun and there are always a bunch of people who enjoy these optimization puzzles. At Realms Beyond, T-Hawk loves optimization exercises and has written about a bunch of them on his website as he's tried to win the fastest possible victories in Civilization 5 and max out the scoring in games like Terraforming Mars.

The alternative to optimization exercises in gameplay are variation exercises, what we've traditionally called variantism at Realms Beyond. Instead of focusing on the most ideal outcome possible like highest score or fastest finish, variation exercises deliberately explore alternate or suboptimal portions of the gameplay to introduce more variety. These would be the kind of challenges where players try to conquer the world in Civilization without ever capturing an enemy city or play through the whole game in Final Fantasy with only a single character using a single class. (Surely no one would be crazy enough to do that with every class in the game though, right?) Obviously this is what appeals to me and I've spent innumerable hours exploring these kind of zany variants through introducing extra challenges or deliberately ignoring the strongest options in favor of underpowered stuff. Now I want to be clear on one point: just because variation gameplay appeals more to me personally, that doesn't mean that it's inherently better or superior to optimization gameplay. The speedrunning community does some incredible stuff, and even though I personally don't want to play through Super Mario 64 eight thousand times in the hopes of shaving two seconds off my run time, more power to those who do enjoy that gameplay. Fun is where you find it and players should pursue whatever they personally enjoy doing.

However, some games are naturally more suited to one type of gameplay as opposed to the other, and players will tend to find them more entertaining if they lean towards a gameplay style that suits their strengths. For example, chess is innately geared towards optimization exercises due to its strict ruleset. There are only two players, the board layout and piece movement are always the same, there's perfect knowledge at all times with no hidden information, etc. That's not to say that chess doesn't have its own variants (there are tons and tons of them!) but none of them has ever really caught on and achieved much popularity. By contrast, Teamfight Tactics naturally lends itself towards variation in its gameplay. There are eight players who are all doing their own thing, the champions that pop up in the shop each round are random, the item components that pop out of the minion rounds are random, and which opponent the player will face each round is also a controlled random result. There's simply too much RNG innately baked into the gameplay of Teamfight Tactics to make optimization efforts bear much fruit. Sure, some players will sit there with a Mobafire guide trying to replicate an exact team compsition and item build, but that's not something that can be forced consistently or something that would be much fun to play longterm. Teamfight Tactics naturally lends itself towards variation gameplay and the set design should embrace this rather than fight against it.

Therefore it should come as no surprise that the most successful sets of Teamfight Tactics have been the ones that were built around variation gameplay instead of optimization gameplay. Set Four and Set Six both fall into this category and consequently have been the most popular sets to date. Neither of these sets was designed around building the absolute most strongest team compsition possible by optimizing the most perfect items for each champion. Quite the opposite, in fact, as both sets presented players with deliberate variations to shake them out of their comfort zone and encourage them to head off in different directions. Set Four did this with the Chosen mechanic; maybe someone was planning to run Warlords but then Chosen Ashe showed up in the store and suddenly Elderwood/Hunter looked a lot better. Set Six used the augments which were a larger and more fleshed out version of the Chosen units. For example, Yordles suddenly became much more viable with So Small granting them dodge chance while Innovators became infinitely stronger with additional hearts or souls to get the mechanical Dragon in play. The point is that rote memorization and repetition of idealized build paths were punished in these sets while adapatation and flexibility were rewarded.

By contrast, Set Five did the exact opposite by pushing in the direction of optimization gameplay. This was largely due to the introduction of the armory mechanic which was designed to give players more direct control over which item components they received. Shadow items were supposed to be the random element in the set but didn't present particularly interesting decisions or change up the gameplay in meaningful ways; a shadow Frozen Heart wasn't ultimately that different from a normal Frozen Heart. By removing Chosen units and replacing them with armories and shadow items, players were incentivized to build the same team comps in game after game. The challenge in the gameplay was an optimization exercise, seeing how perfectly the player could build the same team composition in each game, rather than flexing between many different teams. Of course players could still experiment with lots of different teams but it's inarguable that Set Five (and Set Three before it) were much more conducive to running the same setup in "20/20" games as they like to say on TFT Reddit. I'm reminded of how antisocialmunky in our local TFT group played reroll Forgotten Hecarim in something like 300 games and made it all the way to Challenger rating at one point in Set Five. This kind of optimization gameplay naturally appeals to some players and for them Set Five was a real hit.

But it's clear that this was a minority of players and for the bulk of the playerbase Set Five was a big miss. This was a natural result of Set Five's design leaning towards optimization gameplay instead of embracing the randomness inherent in TFT. I think that the designers were trying to cater towards the miniscule professional playerbase and wanted to strip as much randomness as possible out of TFT to make it better for competitive events. Amusingly, this was a total failure as the Set Five World Championship was dominated by reroll Kled as a 1 cost unit and the championship tournament was seen as a real dud heavily influenced by randomness anyway. We must always remember that TFT is a Free to Play game where the core goal is to keep people playing as long as possible so that they'll spend money on cosmetic doodads. Variation gameplay is fantastic for this because it keeps presenting players with new gameplay content that they want to try out. What if I find a Kalista Chosen in this game and can play 9 Cultist for the huge Galio? What if I get Prismatic Chemtech augment and now I can run Jinx and Tahm Kench as Chemtech units? These things are fun to try and incredibly satisfying when they work out in practice; they keep players coming back for more games since there's always something new and different to try. TFT works best when it accepts the randomness inherent in the system, doesn't try to make everything perfectly fair for TFT eSports (which is never going to be a thing, sorry!), and keeps the average player in mind rather than the streamers who play ten hours per day. Set Six bounced back in the direction of variation gameplay and its success proved that this is the way forward for the game as a whole.

Augments, Traits, and Units

I've been mentioning the augments a bunch so let's discuss the featured element of Set Six in a bit more detail. Players were offered a choice of augments at three different points in each match: the first augment on the final minion round before hitting other players (1-4), just before the Stage 3 carousel in the midgame (3-3), and then just after the Stage 4 carousel right on the cusp of the endgame (4-6). Some players wanted the timing of the augments to be shifted around (I saw a number of calls for the first one to be pushed back) but I thought that they were pretty much perfect in their location. The first one provided direction for the early game before any combat against other players and then two more augments offered additional opportunities to tailor the team composition in the midgame and lategame. The designers even had the courtesy to place the augments on stages that avoided the most common level timings (Level 6 on 3-2, Level 7 on 3-5 or 4-1, Level 8 on 4-5) to avoid overloading the player with too many things to juggle at once. I don't see any obvious place where the augments would have fit better than where they ended up.

To repeat what I stated above, the augments were essentially a larger and more fleshed out version of the Chosen units. Some of the augments were functionally identical to the old Chosen mechanic such as gaining a Mutant "heart" which would grant one additional rank in the Mutant trait for free. Every trait had some version of this augment and the rare Prismatic augments could grant multiple ranks of the same trait when they appeared. Many of the augments were specifically tailored to individual traits such as Assasins mana reaving with the Cutthroat augment and Arcanists getting a shield at the start of combat through the Runic Shield augment. These augments could be extremely powerful but of course did nothing if the player wasn't running that particular trait - is it worth the gamble of switching into that team composition just to take advantage of the augment? The specific augments were balanced against more general augments which would always be useful but typically had less power overall, things like Celestial Blessing providing percentage-based healing and Phony Frontline adding two target dummies. Augments always appeared at silver, gold, or prismatic tiers with the higher tiers appearing less often and having much more powerful benefits. Prismatic augment games in particular were pretty wild and not always particularly well balanced (getting a bad choice of Prismatic augments could feel horrible). However, the first augment was Prismatic only 6% of the time which mitigated against this being too much of a problem. The scarcity of the Prismatic stuff kept players coming back to TFT in the hopes of trying out some of the truly crazy augments like March of Progress and Built Different.

It's fair to point out here that the augments were another version of something that I've often criticized in the past: picking something from a list of choices. I was critical of the radiant armory added in Set 5.5 while I've been much more positive regarding the augments in Set Six. Why the difference in opinion? The key change is that the augments in Set Six provided a much more meaningful decision that legitimately changed the gameplay from match to match in a way that the radiant armory did not. The first augment choice in particular shaped the whole rest of the game and taking something like Knife's Edge (more attack damge for units in the front two rows) or Trade Sector (one free shop refresh each round) or Rich Get Richer (maximum interest increased from 5 to 7) could dramtically shift the remainder of the match. The radiant armory simply did not do this; players took the item which was best for their comp but essentially never rebuil their comp based on what they received. The radiant armory also arrived too late in the game (at the end of Stage 3) in comparison to the first augment which was present before the first PvP round. To make another comparison, the Galaxy mechanic in Set Three legitimately did change up the gameplay from match to match but had the huge problem of removing all player agency from the decision. Some galaxies benefited reroll team comps and some were geared towards lengthy matches featuring the legendary units, however players had absolutely no control over what would appear. This was the fatal flaw of Set Three fixed by both the Chosen units in Set Four and the augments in Set Six. Players always had the option to ignore Chosen units or the offered augments in Set Six in a way that they couldn't escape the altered rules of the galaxies.

Still, as positive of a development as the augments may have been, Set Six wouldn't have been a success without having good trait and unit design. The TFT developers obviously had way more design time to work on Set Six as compared with Set Five and came up with significantly better iterations of both the traits and the units. Gone were the heavy vertical scaling traits from Set Five in which nearly every team played into its own walled off group of units (Dawnbringers, Hellions, Redeemed, etc.) and the lategame was endlessly dominated by the cancerous "HIV" trio of Heimerdinger, Ivern, and Volibear. Set Five was also heavy on traits that added scaling stat bonuses with nothing else of interest going on; Hellions was about the only core trait that did something unique, maybe Coven and Dragonslayer as well before they were removed. Set Six featured much more interesting and better designed traits across the board. Challengers put a novel spin on the standard attack speed trait by adding a reset mechanic while Arcanist brought back the excellent Dusk trait from Set Four. Chemtech and Syndicate traits introduced novel healing mechanics that worked in different ways and didn't particularly resemble any traits from previous sets. Scrap trait took the boring "everyone gets a shield" mechanic from other sets and combined it with the number of items held by champions in a really creative way. Colossus units were massive tanks that could provide incredible frontline at the cost of eating up two unit slots on the team. While there were a number of old standbys from earlier sets like Assassins and Snipers and Bodyguards, the collection as a whole had a lot of exciting new ideas.

Three traits in particular stood out as truly exceptional from a design standpoint. The Mutant trait provided its units with one of seven different benefits that changed from game to game, and while this could be difficult to balance at times, it was another way in which Set Six benefited from incredible replay value. The TFT design team even made this easier for players to understand by making each of the seven Mutant abilities some kind of trait from earlier sets, like Blademaster or Predator or Cybernetics. Different Mutant units were stronger with different Mutant traits which was fascinating to experiment with. The Yordles trait was a truly unique spin on an economic trait, granding a free Yordle unit on the player's bench following each combat round while having no in-combat benefit. Players could try to 3 star all of the Yordles to unlock the secret Veigar unit or sell the free units for additional income. Yes, the Yordles trait was too strong economically for most of the set but it was still an amazing idea and tons of fun for players to chase after all those 3 star units.

Then there was the best trait of the whole set, Socialite. This trait placed a spotlight somewhere on the battlefield map which granted three scaling benefits to any unit that stood in the spotlight: 15% additional damage at Socialite 1, then additional mana per second at Socialite 2, then finally healing for all damage dealt at Socialite 3. The benefits from the Socialte spotlight were general enough that all carry units could make good use of them and therefore Socialite was always a trait that could potentially be flexed into a team composition. What made Socialite the best trait was the fact that the spotlight's position changed from game to game but always appeared in the same spot for everyone within the same game. In other words, it brought back a version of the elemental hexes from Set Two and added "terrain" to the battlefield map to play around, something that I've been requesting for the last two years in these TFT retrospectives. The presence of the Socialite spotlights added so many cool things to the dynamics of positioning units: is it worthwhile to play a ranged carry in the second row if that's where the spotlight appears? Should I move my carry off the spotlight if other players are targeting it with Zephyrs or Shrouds of Stillness? And of course these questions were in addition to the basic issue of how many ranks of Socialite to run, such as whether it was worthwhile to go up to Socialite 3 or hold at a lesser tier. I can't stress enough what a great design choice this was that helped to provide flavor for the whole set.

Of course not everything worked in terms of trait design for Set Six. The big failure here was the Imperial trait which never quite panned out as intended with the concept of the Imperial tyrant. The designers had to keep buffing the Imperial units as well as the trait itself until eventually they made it overpowered and then reverted back to the previous state. I think that the main fault here was more in the unit design as Imperial was supposed to provide this huge scaling damage bonus but lacked any 4 cost or 5 cost units outside of Sion who was a pure tank. This meant that Samira had to be the carry and she was inevitably either oppressively strong or more or less useless. It's no surprise that the trait was scrapped for the midset update. Nothing else was quite as much of a dud though I'll note that Protector also didn't really work as intended. It was only useful in a single reroll team composition and also proved hard to balance, several times becoming far too strong and particularly when Protector emblems were placed on units like Chogath. This was another casualty of the midset update that wasn't terribly missed.

As for the units, Set Six once again introduced about two dozen genuinely new abilities that players hadn't seen in previous sets. Every set is defined by its 4 cost and 5 cost champions which had much more interesting abilities in this set as compared with Set Five. Whether it was Yone's clone or Urgot's chaingun or Sion's colossal windup swing, these units had memorable abilities with strengths and weaknesses that could be played around. I singled out the legendary units in Set Five as having unusually bland abilities and by contrast Set Six delivered on creating powerful and entertaining 5 cost units. Yuumi could play into almost any team composition and be effective, Galio was both a massive tank and also a potential off-carry when given items with crit chance (Clapio!), while Tahm Kench brought back the concept of "unit scaling over time" from Ornn in Set Four with his hilarious ability to devour other units to gain stats. Powering up Tahm could be a tough decision due the cost of feeding him all those units but an early catfish with good itemization had the possibility of farming gold and item components for the team. My vote for best designed unit in the set was Jayce due to his flexibility of having both ranged and melee forms. Jayce could be an amazing frontline tank, or he could be played as a trait bot for his Enforcer ability and attack speed-buffing acceleration gate, or he could be a main carry in his own right from the back lines with AD items. This was a unit that could play into many team compositions and always felt great to have while usually not falling into oppressively overpowered status.

Just to wrap things up with the other side of the ledger, the worst-designed unit in Set Six was Akali. She constantly dashed around the map trying to execute targets and would always have innate healing thanks to running Syndicate trait. In practical terms, this meant that fights with Akali turned into massive RNG-fests where she might dash into a skillshot and die instantly or alternately heal her way back from near-death over and over again while killing the whole enemy team. Akali moved much too quickly for positioning to mean anything and these battles were basically dumb luck - not ideal for a strategy game. Outside of Akali, I don't think that Fiora ever quite found her place in this set either, typically used as the unit who received leftover items when Yone was fully itemized in a Challenger comp. I also have to list Janna here and not because she failed to find her place in this set - rather it was because she was way too good and played in virtually every single team! Janna had Enchanter and Scholar and Scrap traits which made her viable absolutely everywhere; she was a major reason why there was so much healing in Set Six and forced everyone to build Morellos in each game. Even though I absolutely loved Janna it was the right decision to take her out of the midset update because of how omnipresent she was.

One last note on champions in Set Six: the design team decided to use the base unit artwork from League of Legends for most of the champions and this was a huge help in terms of visual clarity. I've written before about how Set Three was difficult to play since it fell into the trap of every unit being "futuristic charater with sword/gun" and made all of the units look identical in the shop. Set Five tried to color-code the units by their respective traits but had the side effect of creating some hideously ugly portraits that came out looking nothing like their League of Legends counterparts (especially Udyr). By sticking with the base portrait art for many of the units, Set Six made it much easier to identify them and distinguish units from one another. This might sound like a small point but it really does matter: visual clarity in the unit design helps enormously when learning the set and also when rolling quickly in the shop while trying to piece together your endgame team in limited time. I appreciated this attention to detail even though the midset update looks to be going backwards with some of the new artwork, blah.

Ranking Up

This is the part of the retrospective where I always talk about my experience playing games of TFT over the duration of the set. The bulk of my games continued to be Normal games thanks to running weekly group sessions on Livestream as part of the Friday streaming sessions. However, I also played more Ranked games of TFT over the course of Set Six than in past sets and did much better than I was expecting, reaching all the way to Diamond rating which put me in the top 4 percent of the playerbase. I had never reached Diamond rating previously in Teamfight Tactics or in League of Legends and therefore this was a major milestone for me, confirming that I did indeed know what I was talking about when creating TFT content for YouTube. Sure it might not seem like much to those who reach Grandmaster or Challenger in every set but I was more than pleased to find myself at a ranking that exceeded 24 out of every 25 players. And in truth, I probably belonged in an even higher tier of players based on some of the observations I gathered while playing ranked games. Let me go into a bit more detail by way of describing my ranked experience in Set Six.

First of all, it's important to keep in mind that the matchmaking for TFT is extremely good at matching up players that it perceives as having the same skill level. The matchmaking does not work off the player's visible rating, of course, instead using the hidden matchmaking rating (MMR) which is never revealed to the player. I've known for a long time that my hidden MMR in TFT has always been higher than my official rating; this was taken to a comical extent back in Set Two when I was in Silver I and actually hit a Challenger and a Masters player in a ranked game. Set Six no longer contains anything that wacky but I still found myself routinely facing lobbies where all of the other players were several divisions higher than me, with everyone else being Platinum III or IV when I was in Gold II, that sort of thing. The matchmaking perceived that I belonged at a higher rating but hadn't won enough games to reach that point yet, resulting in me gaining many more League Points (LP) for winning matches than losing LP when I lost games. I've understood for a long time that this sort of thing balances out over time if the player runs enough ranked games to reach their "true" rating level.

However, I was missing an important piece of the puzzle that didn't click for me until this season. TFT's matchmaking system is so good that it pairs together players that it perceives as having similar skill levels even during the initial placement matches and when playing the first dozen or two games of the set. I realized this when I was something like five ranked games into Set Six, sitting at Silver I official rating, and everyone else in the game with me was building incredibly strong team compositions with excellent itemization on their carries. I think that I struggled my way to a second place finish and was curious enough to check out everyone's profiles on LoL Chess afterwards. Sure enough, everyone was in high Silver or low Gold just like me, and I'm thinking "there's no way that these are Silver players!" Then I looked at everyone's rating from past seasons and that's when it hit me: everyone in this game was in Diamond or high Platinum in previous seasons of TFT, plus there was a pair of Masters players in the group. TFT's matchmaking had correctly rated everyone in the game as being a high Elo player on their way up the ladder and matched us all together.

So no, this wasn't actually a Silver game at all, it was a game with Diamond MMR and I had held my own just fine against these other players. I started checking the past history of the players in all of my matches after this and I found the same pattern repeated over and over again. I was hitting nothing but Diamond and high Platinum players from previous sets with the occasional Masters and Grandmaster players mixed in. TFT's matchmaking was keeping all of us screened off from the players who were actually in Silver and Gold Elo so as to avoid ruining their games; this is why I kept getting queue times of 4 and 5 minutes which otherwise seemed inexplicable for someone in Gold III rating. This had been a puzzle that I couldn't figure out in past seasons: I would be sitting there in "Gold rating" and getting these extremely difficult games against opponents that felt vastly better than the players in Gold matches that I would sometimes watch on Livestream. I interpreted this to mean that I couldn't possibly compete with people further up the ladder; after all, if it was this hard to win in Gold Elo, how could I ever make it to Diamond?

But that wasn't the case at all! I was *ALREADY* playing people who were really Diamond players - I was never hitting the actual Gold and Silver players at any point in time! Once I realized this, I was determined to play some additional ranked games to see where I wound end up. And as it turned out, I kicked some serious ass when it came to TFT in Set Six:

It only took me 38 games to reach Diamond IV with an absolutely absurd Top 4 rate of 79% across those games. Even given a relatively modest sample size, finishing in the Top 4 in 30 out of 38 games with an average ranking of 3.26 is pretty insane. TFT's matchmaking had me pegged as the smurf account of a Masters or Grandmaster player and I kind of lived up to that incorrect assumption over the course of this season. Based on this data, I think that I would have had very good odds to reach Masters tier if I'd been willing to play enough games to make it there. I was continuing to do very well against the Diamond lobbies that I hit towards the end of this climb and I could drop off a long way from that 79% win ratio while still making progress. I mean, the general perception is that getting a Top 4 finish in 55% of your games is excellent and I was vastly higher than that. I also observed that there seems to be a slight inflationary pressure to the rating system in TFT; in other words, a win rate of 50% seems to result in a slight increase in rating over time, not remaining stuck in place. The whole system is deliberately designed to reward players for grinding out lots and lots of games which makes perfect sense for a Free to Play model. If I was willing to play 200 ranked games instead of fewer than 40 ranked games, the odds were pretty good that I could have made it another tier even higher.

There's just one problem there: I don't actually like playing ranked games very much! I've gotten better at avoiding ranked anxiety but I still have much more fun playing Normal games with lobbies full of friends where I can relax and enjoy myself. There's also the reality that I'm turning 40 years old later this year and there's a lot of stuff going on in my life outside of playing video games. In other words, the alternate history version of me that was born 20 years later and is still in college right now definitely played his way to Masters tier in Set Six, however don't expect to see me reach that point any time soon. I was extremely happy to hit Diamond rating and stop there without sweating things out further. Even better, this was the season where a whole bunch of people in our TFT group similarly broke through and reached Diamond rating for the first time. Mythradorri and XpL both hit Diamond (with XpL going on to Masters) to join longtime Diamond players like El Grillo and Headwinder and antisocialmunky (who returned back to Grandmaster status). The net result was that our TFT Friday group was much, MUCH higher in formal rating for this set as opposed to past ones. There was a time a couple of years ago where only a few of us had so much as Gold rating and now just about everyone was Diamond or high Platinum. Our practice games helped all of us grow and improve at the game which was super rewarding to experience. We even have some of the old games archived on YouTube making it easy to see how much better everyone has gotten over time! I had a very good Top 4 rate in our Friday lobbies and if everyone there was mostly Diamond in rating, well, clearly I belonged there as well.

Finally, I also need to mention the new co-op gameplay mode added in Set Six. Known officially as Double Up Mode, the two-player version of TFT allows a pair of teammates to work together against three other such teams. Players share a common health pool and can hop over to help their teammate if they win their own round quickly enough, with the turnaround that the same assistance can also help their opponent! Teammates also get a limited number of items that send a champion and any items they happen to be carrying over to their teammate's board, and this is where the crux of the strategy pops up for the gameplay mode. It's very important to be in constant communication with your partner so that each of you can be looking out for the key units and items that they need. Double Up teams tend to be significantly stronger than the teams in normal games and most lobbies are won by having a 3 star version of a 4 cost unit like Yone or Jhin or Lux. Players can work together to create a 3 star unit by passing champions back and forth across their boards; I remember one game where Cymbal found 4 Urgots and I found another 5 Urgots which we were able to combine together into a 3 star Urgot using a series of champion sends back and forth. Or there was the game I played with XpL where I was win streaking through the early game to preserve our team health as he loss streaked with Mercenaries, then XpL cashed out at the end of Stage 3 and eventually wound up with Kaisa 3 star (!!!) at the finish. There's a tremendous amount of strategy here above and beyond the normal base game and we're all looking forward to the addition of custom lobbies for 2 vs 2 vs 2 vs 2 Double Up matches. This has been an amazing addition to the TFT gameplay and it's already hard to believe that it didn't exist in any previous sets.

3 star Kaisa vs 3 star Akali - only in Double Up mode!


The truest test for whether a set of Teamfight Tactics has been a success is how much time its players spend with it. Set Six was a massive win in this regard as I found myself wanting to play more and more games because the matches were inherently lots of fun. Whereas playing even five ranked games on the last day of the set had been a chore for Set 5.5, I enjoyed challenging myself in dozens of ranked games for Set Six because the gameplay was inherently compelling on its own. I'm writing this as the midset update for Set Six has just been released on the Public Beta Environment (PBE) servers and my hope is that the second half of this season proves to be as entertaining as the first half. It's most likely going to be another version of Set 4.5 where the midseason update turns out slightly worse than the original incarnation while still being pretty darn good. I'll be satisfied if that's the outcome we ended up getting. Check back here in three months for some closing thoughts on the midset update. Until then, thanks as always for reading!