Microsoft now has a studio wholly devoted to the Age of Empires series, and is currently working on remastered 'definitive' editions of Age of Empires 2 and 3. Meanwhile, Relic Entertainment is busy creating Age of Empires 4. And now it sounds as though Microsoft may have plans to revive Empires' more fantastical spin-off, Age of Mythology.

Speaking to Eurogamer at E3, Age of Empires creative director Adam Isgreen hinted that Age of Mythology was something the team may be returning to once they've gotten the definitive editions are finished.

"It's great to see how much love there is for Age of Mythology, 'cause I love it too," he said. "It's more of a… less accurate history. It's super loose, people just have fun, and I love that too."

"After we get through the definitive editions for the three here, and [Age of Empires 4] is kind of rolling, then we're going to look back and see what we can do with Myth," Isgreen went on. "Because I love Myth, we're not going to leave it behind, we'll figure out what to do with it then."

Isgreen was noncommittal on what exactly he has in mind for Age of Mythology, although another definitive edition seems most likely. However, Age of Mythology: Extended Edition came out in 2014, and another remaster of the 2002 original might be a little underwhelming.

Isgreen said his team is planning a new patch for Age of Mythology: Extended Edition "really soon," which will fix some exploits and address various player concerns.

When Valve announced its Index VR headset back in April, it also shared screenshots of Aperture Hand Lab, a tech demo designed to showcase the capabilities of the Index controllers. The controllers are noteworthy because unlike those for other VR headsets, they track individual fingers rather than just clenched-fist hand movements. That tech demo is now available on Steam, if you want to try it for yourself, and you don't even need an Index to play it.

Aperture Hand Lab also runs on the HTC Vive headset, although aside from a bit of sightseeing that seems like it would be kind of a wasted effort, because the Vive doesn't support the Index controller. The main features, according to the Steam listing, are grabbing, shaking, and waving, all of it taking place within the confines of Aperture Science, the oddball setting for the Portal games.

If you don't have an Index headset (and you don't, because it's not actually released yet) or a Vive, you can still get a feel for what Aperture Hand Lab all about courtesy of Road to VR, which describes it as a "short game" that "passes you through several tests administered by a number of ‘personality cores’, the very same as seen in Valve’s landmark VR demo The Lab." Each core will give you a particular test to accomplish, but you can apparently freeform it if you like by throwing the horns, flipping the bird, or whatever else comes to mind, which will elicit different reactions.

Despite the Portal trappings, Aperture Hand Lab was actually developed by Cloudhead Games, which has been working with Valve on VR for a few years. It's free, which is a pretty big plus, especially since there's really not much to it; it is not, however, the "flagship" game mentioned in April. According to the Valve News Network video below, which is also embedded in the Aperture Hand Lab Steam page, that remains unannounced. VNN also said that while Valve is working on three big VR games (something that was first revealed in early 2017), only one of them—the flagship—will be out this year.

Rise and shine! It's been a long, long wait for Black Mesa's final chapters. When Black Mesa (the modder-made recreation of the original Half-Life in the Source Engine) was released in 2015, it ended with Gordon Freeman's leap into the portal that takes him to Xen. The actual Xen levels themselves weren't yet complete. But today, at long last (we've been hearing the Xen levels were almost ready since 2017) , you can get a healthy taste of Xen as the team behind Black Mesa have made three chapters of the alien dimension available in a technical beta.

"The purpose of this beta is to collect bugs and feedback on a range of different computers," a post on the Black Mesa Steam page reads. "We have made significant improvements and changes to the Source engine, and we want the game to run as smoothly as possible. If you want to be on the bleeding edge of testing, opt into this beta. If you want the polished, complete Xen experience, you should wait. It won’t be long!"

There are some known issues with the Xen beta, such as drastically lower framerates while playing in 4K, issues with ragdolls in water, missing collisions with certain plants and roots in the swamp, and a few other little issues, so don't expect it to be a blemish-free experience. Players are encouraged to report bugs on the Steam forum or the game's Discord.

To take part in the beta, find Black Mesa in your Steam library (you'll need to own it, and if you don't it's currently on sale) and right-click on it. Select Properties from the drop-down menu, then choose the beta tab. Opt-in to the public beta and close the menu. You'll see Black Mesa updating to provide you with the Xen chapters. When you launch the game, you can select New Game and choose chapter 15 to start playing the Xen beta.

Plagued by nightmarish visions of a drowned city and a colossal tentacled beast, private detective Charles Reed travels to the waterlogged town of Oakmont to find answers. Unnatural storms have been relentlessly battering this once thriving fishing community, cutting it off from the mainland, flooding parts of it, and leaving the rest sodden and dilapidated. And to make matters worse, the eccentric locals are deeply suspicious of outsiders. But with the source of, and potential solution to, these harrowing visions lurking somewhere in the city, Reed has no choice but to get his feet wet.

The Sinking City is a third-person detective adventure set in a fictional version of 1920s New England, inspired by the works of influential (and, yes, controversial) horror writer H.P. Lovecraft. Divers uncover an ancient cavern hidden beneath Oakmont filled with otherworldly statues of cosmic demons, and are driven mad by the discovery—a madness that begins to spread through the city and awakens strange, violent creatures the locals call wylebeasts. The game is divided between solving crimes and shooting monsters, although the gulf in quality between each flavour of play is pretty wide.

Ukrainian developer Frogwares has been making Sherlock Holmes games for over a decade now, including 2014's magnificent Crimes and Punishments. As such, The Sinking City is decent detective game—particularly in its use of the blatantly Sherlock-inspired Mind Palace system. As you explore a crime scene you collect clues that, while useless on their own, can be connected in the Mind Palace to open up new lines of investigation. And you have to make these connections yourself, without any hand-holding or hints, which makes a successful deduction especially satisfying.

But while Sherlock Holmes has to rely entirely on his intelligence and intuition to solve a mystery, Reed has a supernatural advantage. In certain locations a glowing blue portal will open up, through which he can witness whichever crime transpired there. These vignettes come in the form of shimmering silhouettes of the people involved and fragments of speech. Reed must then construct a narrative, figuring out the order these echoes of the past occurred in, which will usually result in a new clue appearing in the Mind Palace.

My biggest issue with The Sinking City, at least as a detective game, is that the cases aren't that interesting. One of the greatest strengths of Frogwares' Sherlock games is encountering a confounding mystery—say, a murder occurring in a room locked from the inside—and being overcome with a desire to solve it. But I rarely felt that compulsion here, and the solutions to many of the cases lacked a satisfying "Aha!' moment—the punchline that should round off any good detective story. In leaving Holmes behind, the developer's knack for writing a solid mystery seems to have suffered. Reed is also a completely charmless protagonist, with none of Holmes' wit, quirks, or nuance. He's the worst kind of grim, gravel-voiced private dick stereotype and I really struggled to love him.

As for the over-the-shoulder gun combat, it's perfectly functional, but feels like a box being ticked rather than a vital part of the game. Bullets being Oakmont's primary form of currency (it's a long story) adds an interesting element of ammo conservation to each wylebeast encounter. But mostly I was just eager for the shooting to stop so I could resume being a detective. There's a simple crafting system in there too, letting you make health kits, ammo, and other useful items from scrap gathered in the world. But, like the combat, this is all rather rote. I feel like The Sinking City would be a better, more consistent game if it focused purely on the detecting aspect.

The Sinking City's dreary setting is also hard to love. It's an open world game, letting you freely explore the streets of Oakmont. And while I love having to follow directions and pay attention to street names to find important locations, this bleak, muddy, perpetually rain-soaked city is a thoroughly depressing space to exist in. There's some nice world-building, including districts that have been almost completely submerged, with only the tops of lamp posts poking out of the murky water. But overall the city feels disappointingly lifeless, with robotic NPCs wandering aimlessly, repetitive scenery, and a forgettable, droning soundtrack adding to the sense of gloom.

There's a good, but not great, detective game lurking here—albeit one that doesn't quite stack up with the best of Frogwares' Sherlock series. But as a package, it doesn't quite work. The open world isn't interesting enough to compel you to explore it; the combat is basic and uninspiring; and the writing and construction of the mysteries are generally quite poor. If you were looking for a unique or surprising take on Lovecraft's particular brand of horror, you won't find it here. There are some moments of quality hands-on detecting to enjoy in The Sinking City. You'll just have to decide if you can be bothered picking through the seaweed and fish guts to find them.

The highly-anticipated Scorched Acres update for Risk of Rain 2 is now live, and that means you can take to the inhospitable alien wilderness as a scrappy houseplant named Rex. There's also a new biome, new enemies, and of course new gear to pick up in Hopoo Games' challenging rogue-lite shooter.

Rex has a set of abilities that are split between the plant and robot sides of his character. His passive is that all his abilities apply a weaken effect to enemies, which reduces their armor, speed, and damage. His primary fire shoots three syringes, and if the third of these hits an enemy, it'll apply weaken and, conveniently, heal Rex.

His secondary fire lets Rex sacrifice his own health to launch a seed-pod mortar, which can target faraway enemies and does massive damage. Rex also has a mobility move that boosts him backwards and shoves enemies, applying his passive weaken effect. And for an ultimate, Rex can spend health to plant a flower that pulls enemies toward it, rooting and weakening them.

Meanwhile, you'll be able to explore a lovely and forbidding new area called Scorched Acres, which is an alternative third stage. It's guarded by a new boss called the Grovetender, who fires hooks and seeker wisps and generally sounds quite scary.

There are also new enemies, new items and equipment, and new drones, and Hopoo Games has also implemented some community-suggested quality-of-life improvements as well. 

All in all, a pretty beefy update. Risk of Rain 2 is currently in Early Access, and the plan now is for it to launch a full version in the first half of next year, after a year's worth of updates, additions, and enhancements.

Detention, the ruminative Taiwanese horror game by Red Candle Games, is getting a film adaptation, and its set to hit cinemas in Taiwan this September. Set in Taiwan during the 'White Terror' period of martial law, the game is a cryptic and lonely exploration of political dissent and its ramifications, set entirely in an isolated mountainside high school.

The film adaptation is directed by John Hsu, who provided the following statement (via Taiwan News). "As a gamer, the original Detention game left me shocked and made a deep impact," he wrote. "Even then, I hoped the story could be adapted to film to reach even more people. My goal was to keep the film true to the spirit of the game, but to further develop a unique vision and emotion. I wanted to create a truly unique psychological thriller.”

The film stars Gingle Wang and is produced by 1 Production Film Co. and Filmagic Pictures. For anyone who has played the game, the set design in the trailer below will be very familiar: it's surprising just how true to the game it is. 

As for Red Candle Games' more recent game, Devotion, the game remains unavailable following it was unpublished due to in-game assets which were interpreted to mock Chinese president Xi Jinping. The studio has made no further statement on the state of the game since February.

Here's the trailer for Detention:

The gamers running at me have swords, because all gamers own swords, I guess. They also have Monty Python Dark Souls cosplay armor, but it can't save them from the shotgun blasts that reduce them to lumps at my feet. I'm out of ammo before I'm out of gamers, the last one charging with sword held high. But any object you kick is a one-shot kill in this game, and dead gamer lumps are kickable objects.

There are more impressive gun-fu stunts in My Friend Pedro than breaking somebody's neck by kicking their dead friend's torso at them, but this moment of improvisation is typical of what happens between gif-able highlights. Once or twice per level I'll do something like skateboard through a window and then kick that skateboard at someone while shooting someone else, or I'll throw a frying pan into the air and then ricochet bullets off it to clear a room. 

But the meat and potatoes of My Friend Pedro isn't these set pieces. It's jumping in and relying on the combination of a generous slow-mo meter, physics objects, and a lot of bullets to see you through.

This shit is bananas 

I should mention the talking banana. His name is Pedro and he's full of potassium and advice on how to kill people, whether retired mafiosi or Christmas-themed bounty hunters or gamers driven to madness by violent videogames. There's a story in My Friend Pedro but a minimal one—it feels like the plot of another hyperviolent Devolver game, Ruiner, only told on fast-forward. Pedro is there to explain things, but more importantly he appears in the corner of the screen when you pull off a high-scoring combo and tells you what rank you got at the end of a level.

It's a score-attack game, with bonus points for chaining together kills in combos. There are leaderboards, and I have replayed several levels just to get a higher rank than James. (Suck it, James.) Chasing high scores makes me look ahead at a level and think things like, "I should ride the barrel onto that guy then jump off while dual-wielding and split my fire so I can shoot that one and the other before I land." 

Split-fire looked neat in the gifs Pedro's designer showed off during development, but I was worried it would feel rough in play. I was super wrong. It's easy: right-click on one dude and hold to lock him in, and then for every left-click shot, no matter where it's aimed, the right-click guy gets a bullet as well. Levels that drop you down shafts with enemies on both sides, or have doors full of goons opening around you, are built for this. I never stopped enjoying it.

These are also fine moments for dodging by pressing W, which sends you into a bulletproof pirouette. Shooting while dodging sends bullets flying in whichever direction your guns happen to be pointing during that part of the animation, wild sprays flying around as you whirl about like a dog in a tumble dryer. Again: enjoyable.

Later levels reduce the number of physics objects like skateboards, basketballs, and knives, as well as goons to fling them at, in favor of mines, lasers, spinning platforms, and traps that require more perfectionist play. There's not as much freedom to rack up combos and a lot more falling to your death. They're less fun to replay, and I wish there were notes in the level select screen that said "this level is cool and has a skateboard" or "this one has too many lasers". There's one in the dream-like midsection with no enemies, a pure platforming challenge that doesn't play to My Friend Pedro's strengths.

So some levels are weaker, but before long there's always a refresher level that plays completely differently, a motorbike chase or a freefall or a boss fight. To pay My Friend Pedro the highest compliment, its boss fights didn't annoy me. They're bombastic, inventive, and they're over fast.

Along with the score, at the end of every level Pedro gives you a gif of a highlight to save or share. Postcards from murderland, they slice My Friend Pedro into a handful of enjoyable seconds. Its take on bullet-time is fun for more than seconds, though. Unlike Superhot or Max Payne I never got tired of its bullet-time thanks to short levels, high variety, and a storyline that's purest nonsense.

The title of Totally Reliable Delivery Service is a lie, because the delivery service in question is not actually reliable at all. And not because the guy says he rang the doorbell but clearly he didn't because you were there all day, literally waiting for your package. No, it's because there are so many other, more interesting things to do, like going for hot air balloon rides and trying to launch underpowered boats off of water-ski ramps.

Based on publisher TinyBuild's description of the game as "controlled noodly chaos," you probably already have a pretty good idea about whether or not you want to play it. If you're in anything but the "hard no" camp, then today is your day: For the next two weeks, the TRDS open beta is free for everyone on Steam.

I horsed around with the demo for a few minutes and it strikes me as kind of a more goal-oriented Goat Simulator. I didn't accomplish much in my first go-round: I delivered a package, completely missed the deadline for two others, got distracted, ramped a golf cart into a lake, and very briefly flew a plane. Online multiplayer looked like a good time and it was for a couple of minutes, until I guess I got too rowdy, at which point someone hurled a racist slur at me and then kicked me off the server.

That wasn't great, but then I hooked up on a different server and got into a brief, very stupid wrestling match with a couple of guys, after which joined them for a hot air balloon ride and a boat ride, both of which ended badly. (I fell out of the balloon after a biplane crashed into it, and the guy driving the boat tried to ramp it but just sort of bumped into the ramp instead, and we sank.)  

I'm not sure how it will hold up over the long-term, especially in its currently janky beta state, but that was fun. If you want to give it a spin yourself, the open beta is free for everyone until July 8.

Teamfight Tactics and other Auto Chess-style games have the same problem as their originating MOBA counterparts: a knowledge barrier that demands you know what the many, many individual champions do and how they synergize with each other. This is especially pernicious when TFT has over 50 champions to account for. You’ll quickly get blown up as everybody else’s armies turn into all-powerful meat grinders if you jump in blind, with your piddly one-star wonders as the sausage-in-waiting.

Luckily, there are a number of key standouts already. Here are the champions to look for, and why you want them on your team.

1. Aurelion Sol

Though probably first up to bat for a nerf in the near future, Autocannon Sol will likely remain one of the best DPS options in Teamfight Tactics. The hitbox from Voice of Light can cover most of the field on its own, and its raw damage tears through all but the hardiest tanks—usually fellow Dragons or units heavily adorned with Negatron Cloaks to negate spell damage.

But that alone isn’t why he tops the list. Other champions have area-of-effect nukes too, like Karthus or Brand. No other such champion, however, has Sol’s oppressive synergies. The Sorceror class passive, generating extra mana per attack, couples nicely with any sort of attack speed buff, whether from a Wild support team or Recurve Bow, giving Sol ample opportunity to fire off his cannon again and again. He is easily one of the best carries to build around.

2. Nidalee

Though not necessarily built as a carry in her own right, Nidalee makes for a strong early pick that opens up a huge range of options. Unlike Graves, for instance, she actually has a decently long auto-attack range and a relevant early-game ability. She also serves as the foundation of any Wild-centric strategies, amping up her team’s attack speed to generate more mana and spells. 

As attack speed is the single most important statistic in the game, determining how quickly and frequently game-winning spells will sling out, the Wild class in general often serves as the foundation of many teams. She will be a highly contested pick to start setting up the strategy.

3. Gnar

All the reasons that Nidalee is good also apply to Gnar, minus early game presence due to his high rarity. Late game, however, he often serves as the all-important fourth unique Wild character, capstoning the team strategy.

And all that is fine, but what really pushes Gnar over the top is when he goes full Kaiju and pushes the entire enemy team around. That can either buy crucial time for your ranged carries to do their thing, or—in a feat of utter hilarity—dump the enemy army on top of Kennen’s Slicing Maelstrom. 

4. Volibear

Bouncing attacks are already solid when TFT armies tend to be closely clustered together, so the bouncing attacks of a roaring bear of a tank that occasionally freezes his targets is why you prioritize drafting Volibear. 

On top of that, Glacial/Brawler is a hard-to-counter combination of classes. Volibear repeatedly disrupt enemy champions, mitigating their damage and slowing down their spells output. At the same time, stopping Volibear is harder when you have more than one Brawler in play due to their health-increasing synergy. 

5. Darius

Knight champions are already tanky and hard to kill. It’s unfair if they can also carry! Darius is about as independent a force of nature as you can get with the game: it only takes one other Knight to mitigate all damage against him and he heals himself. You’re free to lavish him with all the damage and attack speed items you can cram into his arms, and it only gets more absurd if you get his brother Draven or other Noxians into play.

6. Sejuani

You know what’s better than stunning one enemy champion? Stunning all of them. Though her damage output pales in comparison to Volibear, there’s no denying that having your entire team locked down by Glacial Prison is extremely demoralizing. 

7. Kayle

With the highest attack speed in the game and the ability to repeatedly turn allied champions invincible, the only reason why Kayle isn’t rated higher is because of how rare she is and how late she shows up. As the clincher to a six-Noble strategy, however, she turns an already tanky team into an incredibly resilient one, capable of shrugging off even Aurelion Sol’s Godzilla breath blasts with ease. 

8. Swain

Speaking of impossible-to-kill capstones, Swain is blasphemously good. Though he suffers the same relevance issue throughout most of the game (as in, he just ain’t there), he’s strong enough that it's worth preparing for him if you think you have a shot at snagging him. A combination of Imperial and Demon origin classes gives him a deadly double-dip in potential damage amplifiers, and the healing from Demonflare is nothing to sneeze at either. Watch the crow demon slide his way through the enemy ranks, and watch them all wither.

9. Brand

Whereas Swain gets up close and personal to devastate enemy ranks, Brand does the same from a safer distance. As Demons, both also amplify their respective damage with mana burn and true damage—and as an Elementalist, Brand shares the Sorcerer class’s ability to generate extra mana per attack. A few Recurve Bows, and the whole field’s on fire, just as Brand likes it.

10. Kennen

AOE crowd control is generally a high priority, and Kennen’s Slicing Maelstrom has it in spades—with his own bonus damage if he’s the only ninja you’ve got in the team. Better yet, he answers a common problem with the game: how do you deal with assassins that bypass the front line and attack your ranged carries?

Well, if they all get stunned and melt apart instead, there isn’t a problem anymore, is there?

Kennen’s main problem is a lack of easy synergies. To really juice up his impact, you either need all four ninjas or no others at all. To take advantage of the Yordle origin class’s dodge passive, you need pretty much all of them, and they’re spread out among almost the entire range of rarities. The “easiest” option then becomes drafting other Elementalists, which does have the benefit of churning out more Slicing Maelstroms, but two of the four are in the upper rarities, making it hard to get a functioning set early on.

There are about as many pithy or sardonic quotes about chess as there are chess plays, but Johannes Zukertort’s “chess is the struggle against error” is especially applicable to the burgeoning Auto Chess genre, and especially to League of Legends' new Teamfight Tactics mode. And it's fantastic fun.

The game mode—based extensively and deliberately off the mechanics invented for the original Dota 2 mod—is currently undergoing beta testing (it'll be live for all this week), so part of the struggle is against programming errors. Champions like Lulu, for instance, currently can’t get her spell to go off, rather ruining the point of drafting her in the first place. The AI can hang at points—and few things are more frustrating than, say, losing a 2v1 fight because Tristana decided she’s out of bullets for the day and won’t help in an engagement.

But Teamfight's biggest struggles aren’t with the occasional bug. You can always just quit out and start another match, as the current lack of a ranked mode makes that preferable to trying to tough it out for a comeback, and at least that way you’ve learned what to avoid until the next PBE update. 

As with the other Auto Chess variants, the design is simple: you begin by choosing a champion, with a countdown clock ticking away as you decide where to put them on a grid (for Drodo and Valve’s version, a traditional chessboard of squares; for Riot’s version, hexagons). You then get a combat phase, where your chosen champion fights creeps or another player's squad (there are eight players per game), either winning you extra gold or taking damage. Then your field resets, and gold is rewarded to you, calculated based on whether you’re on a winning or losing streak, and with interest based on how much gold you’ve carried over from round to round.

The rounds is which you fight NPC mooks instead of a fellow player’s army are a chance to get equippable items to beef up your squad, and that’s where you (may) hit a road bump.

The struggle is RNG

The items that bolster your army are a notable offender in design error—though Teamfight Tactics is less troublesome than its Drodo counterpart in this regard. Unlike other Auto Chess variants, you open the game with a free-for-all draft phase, letting you choose a character and equipment of your choice, floating before you in a rotating carousel. There are other draft phases throughout the game, allowing the worst-performing players a chance to catch up by giving them first pick of a new batch of champions and items.

The draft carousel phase is a great twist to the genre, but it’s just barely inadequate to make up for Teamfight’s frustrating RNG. Item randomness is particularly awful in TFT’s opening three combat phases, as it is entirely possible for the starting minions to be miserly and give you nothing more than a Negatron Cloak for your troubles—only for you to subsequently face down a two-star Tristana with two finished attack items. The problem persists even later on, when you’re fighting roaring bosses like Drake or Rift Herald, each capable of devastating even a late-game team on their own, only to be rewarded with the gear equivalent of a wet fart.

This might not be as much of a problem with other Auto Chess games, where a couple extra stats or a bit of health regeneration isn’t as important as hero and class synergies, but that isn’t so with Riot’s vision. Falling behind in Teamfight Tactics’ arms race can quickly put you in an untenable position through no fault of your own, as items can have devastating strategic consequences, whether by turning your fastest attacker into an AOE gatling gun, or temporarily banishing the enemy team’s most dangerous unit from the skirmish.

Whereas the draft carousel in Teamfight Tactics is better than Drodo’s purely random item drops, Valve’s Underlords stands above both with what is probably the fairest feature. You always get one item after an NPC round, regardless of whether you won or lost. Winning, however, means you can actively choose one out of three versions, whereas losing just sticks you with whatever the gods of fortune deign to grant you.

The struggle is me

Overall, however, these design and programming issues are forgivable as Riot works to polish an extremely fun game. The dopamine rush I get from sinking 20-30 gold into rerolls to unlock the last unit I need for a devastating six-Blademaster team of tree-chopping tyrants draws from the same well as gacha games and poker (without the financial consequences of either). The sound design (once Riot's patched it to work) is an entertaining cacophony of familiar League of Legends spell effects and character quips, the chiming of a successful kill, and the ka-ching of gold rolling into your carefully tended bank.

At the intermediate and advanced stages, Teamfight generates satisfying strategic complexity: scouting other players’ armies gives you a rough idea of what draft picks are available to you in future rounds, and what will end up trapping you into a war of attrition over dwindling resources with four or five other players. The draft carousel later on gives you the chance to either bounce back with a rare and powerful unit, or deny a leading opponent the missing link in their own endgame strategy.

Even unit placement matters in ways that aren’t immediately obvious at first glance. My favorite rounds have boiled down to positioning fights: scrambles to change my army formations and unit placements as I face down the sole remaining human player in the game, fully cognizant of what they want to do to my units and how they want to do it. If Blitzcrank is going to grab the farthest unit from it and destroy my most powerful source of damage, then what if I swap it with a tank? My assassins can’t get to their well-defended back line, but can I force them to move and open up space for my killers if I shove my entire army to the other side of the board?

Or should I throw my entire bank into rerolls for a chance at an upgrade?

The biggest struggle in Teamfight isn’t with the bugs and design problems, which are steadily improving patch to patch. It’s with myself, and the sins the game forces me to confront as I gamble and try as best as I can to avoid error.

• Greed: Drafting rare and expensive units early instead of investing in a solid baseline and letting bank interest accrue. 
• Gluttony: Cramming my bench with half-completed upgrades, even if I should have held off and saved money instead. 
• Wrath: Spamming rerolls because I’m frustrated at losing almost singlehandedly to a two-star Draven, wasting my bank as a result. 
• Sloth: Not wanting to rearrange my units, so of course Blitzcrank grabs the one I crammed all of the damage items onto. 
• Envy: Drafting Pirates after losing to some lucky bastard’s Miss Fortune and three-star Pyke, only to never actually unlock their gold-generating passive. 
• Lust: Burning through all of my bank and selling off my prepared upgrades just to grab Swain.

And the worst thing is that sometimes my sins work out. Sometimes throwing all my accumulated gold into rerolls gets me every single upgrade I needed to survive the next round. Sometimes banking on that rare gold-bordered unit means I unlock the final passive ability for its class, turning my entire army into unstoppable behemoths. Sometimes I end up right at the top, the sole survivor of an eight-way skirmish thanks to a bit of good planning and a lot of luck.

And then there's pride, of course: thinking that I’m a genius for winning last time, and surely can’t fail the next match.

Teamfight Tactics has all of League of Legends’ combat flashiness and none of the stress of being personally responsible for making sure Draven catches his damn axes. You’re a spectator and the team's strategic master, cheering your units on as they rout your foes. When they lose, it’s just a momentary and transient setback—easy enough to upgrade something, add a B. F. Sword to your Gunslinger, and try again next round, all grievances forgotten with that next clinking rain of victory gold.

Or perhaps it’s all part of your cunning plan to look weak, but have first-pick priority during the draft carousel to hog all of the Golden Spatulas and grow your bank to an enormous size, making you the first to field a massive nine or even 10-unit swarm of synergistic champions while everybody else still struggles to unlock their class passives.

Teamfight needs work, but it's already a struggle worth enjoying, complex and satisfying. The rapidly-forming Auto Chess battle has another serious contender.