Elegy of the Storm

It is the middle of the night, or close enough that the difference doesn't matter. My daughter is four months old. She has been fed, changed, and is now – for reasons that remain opaque to me and, I suspect, to her – simply awake, which means I am simply awake, which means we are on the couch together with nowhere to be and nothing to do until she decides otherwise.

I have learned, in four months, to find things I can do with one hand.

The HGC tournament is on – the Blizzard-subsidized Heroes of the Storm premiere esports league – and it has real production value, real casting talent, and real stakes. I watch two teams draft their way through the map selection and ban phases while my daughter silently makes her way through a bottle of milk. I should be tired. Hell, I am tired. I am also watching the draft phase with the focus of someone who has nowhere else to be, and I think this is, in its small way, one of the better nights I've had in a while.

I had no idea, sitting there in the dark, that I was watching something that only had about two years left.

The MOBA (multiplayer online battle arena) genre did not originate with Blizzard. That is worth saying plainly, because the history of MOBAs is one of the stranger genealogies in gaming: a modification of a modification, an improvised subgenre that grew out of a custom map for Brood War called Aeon of Strife, which was itself adapted into DOTA (Defense of the Ancients) – a custom map built inside the Warcraft III editor. When I wanted a break from ladder matches in the 2002-2004 era, I played DOTA. It was a decent game. I never needed to make it my primary game, but for a match or two at a time, it held up.

(It is worth pausing here to appreciate that Warcraft III shipped with the tools required to effectively build an entirely new game inside of it. Blizzard gave players an engine and then moved on, and the players made something the developers never anticipated. This is what happens when a company trusts its community.)

Over the next few years, League of Legends and DOTA 2 emerged as the genre's first real titans. I will be up-front about my limitations here: I never played much of either, and I am going to make assertions that are probably more impression than fact. What I can say with confidence is that both games followed the original DOTA formula faithfully. Five players to a team. A symmetrical map with three primary lanes, fog of war between them, jungle camps that rewarded independent play. Gold and experience accrued to whoever landed the killing blow – the last hit – on an enemy unit. Each hero leveled independently. Items were purchased from a shop. The formula was not complicated, and it worked: League is still going after seventeen years. The market had been established.

Which made Blizzard's absence from it all the more surprising, since they were arguably there when it was born. They had, in a meaningful sense, already built the primordial sandbox from which MOBAs emerged. But they missed the first wave entirely, or chose not to ride it, which amounts to the same thing from where the rest of us were standing. This is fine, historically speaking. First-mover advantage is real but it is not destiny, and waiting for a market to mature before entering it with a better version of the product is a legitimate strategy. Apple did not invent the smartphone, they just perfected it.

So when Heroes of the Storm arrived in 2015, ten years after Warcraft III began to fade into the background, and deep into League's stranglehold on the genre, I expected it to matter. Actually, expected is probably too weak a word. I assumed Blizzard would close the gap quickly, because Blizzard had everything required to do exactly that: the deepest bench of beloved characters in gaming, the production quality of a studio that treats visual and audio craft as non-negotiable, and – most importantly – the willingness to question the formula.

Every MOBA before Heroes had preserved the DOTA architecture with modest variation. Different heroes, different skins, the same fundamental mechanical philosophy. Last hits, individual levels, purchased items, one map. Blizzard looked at this and decided, apparently without much hand-wringing, to throw most of it out.

Last hits were gone. The incentive that forced players to obsessively watch their own attack animations and count down to the killing blow – one of the genre's least fun mechanical requirements – simply did not exist in Heroes. Experience was shared across the team. Individual hero levels became a team resource. Gold was replaced by a talent system that asked you to choose between playstyle options at fixed level breakpoints, creating meaningful variation without requiring you to track a shop. And rather than a single map repeated across thousands of matches, the game shipped with a rotating pool of maps that played differently from each other, some of them dramatically so.

These were not small design decisions. They were a coherent philosophical rejection of the genre's received wisdom, and they made Heroes a more welcoming, more strategically interesting, more fun game to play. They also made it, as it turned out, a less successful one.

I found the game engaging in a way no other MOBA had been – and I want to be precise about why, because it matters to what comes later. The map variety was the most immediate gift: it meant that the strategic layer of the game was never exactly the same twice, that there was always something worth studying, something the draft had to account for that it couldn't account for identically every time. The shared levels and the talent system made entry approachable without removing depth. But what I particularly loved was watching the professional scene.

Watching professional Heroes matches is an experience that deserves to be described properly. The matches were formatted as best-of-threes or best-of-fives, and the strategy began before a single hero was deployed. The drafting phase was elaborate and strategically deep: Team A selected the map. Both teams banned a hero. Team B picked a hero, Team A picked two, Team B picked two – then another round of bans, then Team A picked two, Team B picked two, and Team A rounded out their roster. The whole structure was a hostile negotiation conducted sequentially and under pressure, like a chess clock attached to a strategy session, with both teams simultaneously reading what the other was doing and adjusting in real time.

Do you take the map your team excels at, or do you prioritize banning the map your opponents are even better at? Do you pick your preferred composition early, signaling your intentions and losing the element of surprise, or do you pick something more ambiguous and live with the flexibility that buys you? These were the kinds of questions that defined high-level play, and the production around those questions was genuinely excellent.

Adding to this was an incredible roster of casters and commentators, both amateur and professional. Tasteless and Artosis brought over some of their chemistry from the Brood War era. Grubby, a former Warcraft III professional player, had deep strategic insight. And Gillyweed and Dreadnaught ¹⁰(himself a former HotS pro) were two of the tournament's best casters – enthusiastic, knowledgeable, with good chemistry and a real feel for the game's rhythms.

The tournament package they helmed in 2016 was the backdrop to those late nights on the couch, the thing I watched while my daughter made her infant burbles and obscure opinions known about the direction of events. I was sleep-deprived and holding a tiny person and watching professional players make decisions I found genuinely interesting, and that was enough.

In 2018, Blizzard announced the cancellation of the HGC, effectively ending the Heroes professional scene in one swift stroke. The development team at Blizzard was mostly redirected onto other projects. The game was not shut down, it continued to receive updates and even new heroes until 2020, but it was never the same. A competitive ecosystem, once dissolved, does not easily reconstitute, and the professional scene was the heart of the thing for me. The announcement felt abrupt. It probably was abrupt, though the business logic underneath it was legible enough.

Mike Morhaime – long rumored to be among Heroes' strongest internal advocates – departed Blizzard that same year. The game had not reached the viewership heights of League of Legends or DOTA 2, despite sustained investment. The calculation, made by people with more information than I had, was that continued investment was not the right use of resources. I don't even disagree. I have spent enough time thinking about resource allocation and portfolio prioritization to understand the logic.

And yet, Heroes of the Storm was, by any measure I care about, the best game in its genre. It was more welcoming, more strategically layered, more visually distinctive, and more mechanically creative than its competitors. It was, in many ways, a perfect encapsulation of Blizzard’s long-standing design philosophy of “easy to learn, difficult to master.” And on top of the design, it featured characters I already knew and cared about, maps with actual variety and strategic depth, a better spectator experience. It was, by my accounting, simply the superior product. Sadly, the better product does not always win.

This is not a novel observation. Markets do not run optimization algorithms. First-mover advantage, network effects, switching costs, inertia – all of these create conditions where an inferior product entrenches itself at exactly the moment a superior alternative arrives. League of Legends had a six year head-start before Heroes even came onto the scene, much less was trying to build a professional scene. That means enfranchised League players had six years of currency, heroes, cosmetic skins, and achievements already. I do not think there is a satisfying resolution to this. The game deserved better than it got.

I’ll occasionally go back and watch some of those VODs, nearly 8 years old now. The finals of the 2017 GSC is permanently etched in my brain. Tempest vs. MVP Black, set on the Towers of Doom map¹¹. Tempest went down in the early mid-game, their strategy was slow off the start and they took a while to get going, but once it did, they started to dominate every teamfight. In the end, they were able to secure all the mid-field towers, meaning shots would continue firing onto the enemy Core. Finally, MVP Black was down to one point of life left, shots continuing to fire, and we were treated to a thrilling conclusion as the last shot came out from Tempest’s Core and began its inevitable journey to the enemy base. The closest analogue I can think of is watching Steph Curry heave a shot from the half-court line as the buzzer sounds, except you know it’s going in already. And the soundtrack to it all: Dreadnaught absolutely losing his mind about what all of us were seeing.

That version of the game is still running somewhere, in the particular way that good things persist after their official ending. I'm glad I was watching when it was happening. I'm glad they made it at all.