What Can We Learn from Concord's Failure?

Probably not too much, actually

Screenwriter William Goldman famously wrote:

Nobody knows anything...... Not one person in the entire motion picture field knows for a certainty what's going to work. Every time out it's a guess and, if you're lucky, an educated one.

In the modern video game industry, by contrast, everybody knows everything. There are innumerable Youtubers and podcasters, circuit speakers and self-styled sales and marketing experts offering instant opinions on every topic.

A particular formula I’ve come to loathe is “What We Can Learn from Flappy Bird” - the idea that runaway successes and failures result in easily-digested takeaways like “keep it simple stupid.”

I don’t think there was much to learn from the success of Flappy Bird, as why it blew up remains a mystery. And I don’t think there’s much to learn from the failure of Concord.

Part 1: Broad Thoughts on Concord

Why Concord Failed - I Don’t Know (and Neither Do You)

Concord had some obvious problems and a tepid initial reception, with many immediately projecting it to flop. So in some sense everyone knows why it failed: misguided emphasis on story, unappealing characters, (apparent) been-there-done-that gameplay in a saturated genre, price, etc.

But the level of flop - a belly flop into the Grand Canyon that missed the water below entirely - is surprising¹. Out of line with the quality of the game, in the same way that Flappy Bird’s success was out of line with its quality.

Oppenheimer is a 3-hour historical drama and Barbie is a girl-power movie based on dolls, and the two have little in common beyond release date. But somehow they became the merged cultural phenomenon Barbenheimer, with Oppenheimer making a billion dollars worldwide despite being the sort of talky drama that foreign markets shy away from.

Barbenheimer benefitted from a positive snowball. Though detractors would be loathe to admit it, Concord suffered from a negative snowball. It became a cultural phenomenon like Barbenheimer, but as a must-skip rather than must-see.

The movie Ishtar was once regularly called one of the worst movies ever made. It bombed at the box office and was dogged by reports of troubled production. As a bloated expensive flop it was easy to pile onto, and became a poster child for common sins of the movie industry.

These days Ishtar lies mostly forgotten because, in truth, it’s merely a mediocre movie, not terrible enough to be memorable. Removed from the mania of its time, with critics tripping over each other to write the most scathing review, it’s just your typical kinda-bad movie. This can go the other way as well - The Shipping News was once lauded as one of the greatest novels. I’m not sure if anyone seriously believes that today, and I’d be a little surprised if the majority of people reading this have even heard of The Shipping News.

Everyone “knows” that E.T. for Atari is one of the worst games ever made, but the vast majority of people who “know” this have never played it. I like to point out that not only is E.T. not the worst game ever made, it’s not even the worst Atari game of 1982 based on a Steven Spielberg movie and designed by Howard Scott Warshaw: that would be Raiders of the Lost Ark.²

SwordQuest: FireWorld is an ugly, unfun, and totally incomprehensible 2600 game. At least E.T. has a nice rendition of the E.T. theme. But for whatever reason SwordQuest: FireWorld isn’t a game people love to hate.

Overreacting people hate to hear “calm down you’re overreacting” - it just makes them madder! I see that effect with Concord - someone claims it’s one of the worst games ever made and a $200 million flop, and at the suggestion they might be emotionally overinvested, revise upwards to the single worst game ever made and a $400 million flop.

Animal Crossing: New Horizons became a game people loved to love, with a success that probably outpaced its merits. It became Barbenheimer or Hawk Tuah Girl. Concord quickly became a game that people loved to hate; it became Ishtar or Bean Dad. It has real problems, but anyone claiming the magnitude of failure is entirely on merits (or lack thereof) is willfully deluding themselves.

How Much Did Concord Fail?

I’ve seen people claim that Concord is the biggest AAA failure of all time.

Here’s the question I’d pose in response: is it even the biggest AAA failure of this year?

Warner Brothers lost $200 million on Suicide Squad: Kill the Justice League. That’s a huge monetary loss, and a huge hit to the Arkham brand, a billion dollar franchise. And to Rocksteady’s reputation, a developer that regularly made Best-Of developer lists.

Suicide Squad: Kill the Justice League posed a huge opportunity cost - the game took forever to make and that’s time the team could have spent on another mainline Arkham game or more-successful spinoff.

Concord wasn’t an established brand and Firewalk wasn’t an established developer. There’s some damage to Sony’s reputation but Firewalk is not one of Sony’s marquee developers. The studio was formed specifically to make Concord so any opportunity cost is much more academic - Sony could have used that money on other ventures, but Concord isn’t clearly taking the spot of another game. “With that money they could have made 10 Soul Sacrifice sequels” isn’t a real point, Sony wasn’t going to make those games either way.

I’m skeptical of claims that Concord lost significantly more than $200 million. But even if it did, Suicide Squad seems like the bigger bust this year, let alone of all time. Suicide Squad (and, to a lesser extent, Gotham Knights) has significantly damaged a key franchise from a premier developer; Concord was a one-and-done flop from a new studio.

I’ve read “estimates” (aka people making up numbers from thin air) that Concord took 5 years and $100 million, then 8 years and $250 million, then 8 years and $400 million, then 10 years and $800 million. I laughed out loud typing out that last one, as that would put it at four times the cost of some of the most expensive games ever made.

It’s a game with a small set of levels and a small set of (admittedly expensive) character models - if it cost as much as massive open world games something went very wrong. I’m not going to dwell on this because I don’t know the budget, but nobody “reporting” on the cost knows either. The number continually changes (in particular, it grows with each retelling) and is never sourced.

For various reasons people desperately want this game to be a historical failure, and that requires the cost to be very high.

If Concord was somehow one of the most expensive games ever made I’m not sure what the resulting analysis could be beyond “don’t spend $2 million on each crate in a warehouse level.” Maybe AAA games should cost less to make - that’s a reasonable topic to broach. But AAA games of Concord’s scope don’t cost that much. Spider-Man 2 cost a lot but it’s a lot of game - you can see where that money went. If Concord had an obscenely out-of-proportion budget that says little about AAA gaming and more that something went horribly wrong.

Part 2: A Survey of Nonsense

The Case of the Incredibly Long Credits

Analysis like this is aimed squarely at people who want to believe and retweet whatever confirms their biases, but since I’m not that person I watched (well, skimmed) the credits.

The credits video is over an hour long, true. But the credits scroll incredibly slowly and the main credits end at around 9 minutes when they start listing production babies.

The rest of the credits lists basically everyone even tangentially involved in Concord through Sony - Sony’s legal team, the people who update and maintain the PSN store, the people who submit games to the ESRB, etc. These were not full-time employees who worked on Concord for a decade - many of them just sat in the same office as someone who “worked” on Concord for an hour by sending an email. The credits also print out the full text of some open source licenses.

The credits do contain many senior / lead / principle titles and the names sound like people who live on the US West Coast rather than Myanmar, so they don’t come cheap, but the credits are an hour long only in a purely technical sense. And the formula that translates each minute of credits to $5.7 million of budget is certainly new to me.

“Culture Killed Concord”

Here’s a video from someone who - I’m going to be blunt here - simply has no idea what they’re talking about.

The idea that “culture killed Concord” is based on a chain of plausible inferences. But it’s easy to generate explanations that sound plausible enough but aren’t true.

The logic goes like this: we know that the game didn’t sell well. Therefore we can surmise that players didn’t like the game. Therefore we can surmise that playtesters didn’t like the game, and that the developers ignored playtest feedback or didn’t run playtests at all. Therefore we can surmise that the team was arrogant and had a toxic environment, which further implies that top talent left, since top talent flees toxic environments. Etc etc.

The first link in this chain - the game didn’t sell which means players didn’t like it - is immediately suspect. Maybe players disliked what they thought the game was, based on poor marketing and rollout. The game’s biggest haters, after all, are the people who haven’t played it and aren’t familiar with the basic systems. Potential players thought the game lacked appeal, sure, but appeal is largely a function of marketing. Maybe the team rejected feedback - or maybe they were too open to bad feedback.

Famous slide from a Mark Cerny presentation

What characterizes this video is a total lack of legitimate information. The logic deployed in this video could be leveraged against any game that underperformed, starting at low sales and ending up at the conclusion that the team was arrogant and toxic. The argument here is, in essence, is that any game that underperforms indicates a culture problem, and that’s simply not true.

Hi-Fi Rush lost money - is that a culture problem? Does Remedy have a culture problem? Redfall bombed but Arkane’s previous games did well - am I to believe a huge cultural shift occurred during Redfall’s development? Did a similar cultural shift at CD Projekt Red explain the poor release state of Cyberpunk? But then that culture reverted to old form as the game was fixed up?

This video is one of many I’ve seen to the tune of “I’ve worked in the video game industry and here’s what happened!” But the authors of these videos don’t know anything, and they aren’t even making particularly solid suppositions.

Part 3: Hindsight Analysis and All-or-None Evaluation

In “Do Games Need Hooks” I wrote about the danger of hindsight analysis.

Do Games Need Hooks?

A major issue with hook analysis is that it’s mostly post-facto.…If you believe that games need hooks to sell and a game sells well you might skip over “does this game have a hook?” and right to “this game must have a hook, so what is it?” Conversely if a game does poorly it must be due to a weak hook, and your job is to identify that weakness.

Overwatch vs Concord character designs

We know Overwatch sold well and is therefore a good game, and we know that Concord sold poorly and is therefore a bad game, so it’s easy to say that Overwatch has great character design while Concord’s is lousy.

I’ve seen multiple videos comparing these two character specifically, where people tangentially involved in the game industry argue that the left figure is awesome while the right one is terrible.

There’s one huge flaw with this analysis: I have eyes.

The left figure is, to me, frankly, goofy-looking. It looks like a side character you’d see in a mid 2000s 4Kids Anime titled “Sky Warriors: Beast Fighterz.” It’s got the comically large pauldrons that so many Blizzard characters have, unmotivated lion theming³ (though he’s shaped more like an ape), and a kewl aesthetic. He looks a lot like this “Battle Armor He-Man” toy I once owned.

Action figure - definitely not a doll

Concord has character design issues in both conception and execution. I have a visceral negative reaction (for reasons I don’t consciously understand) to the blue lipstick. The character’s outfit looks blobby and inflated. Multiple Concord characters have odd tubing and their materials are often indistinct - not clearly metal or plastic or rubber or cloth. But the right character, with her football player appearance, is more clearly a tank than Reinhardt, the left character. Reinhardt apparently uses a hammer and an energy barrier - from the artwork above you get none of that.

I’ve seen so much analysis in the form of “when you look at this one picture of Reindhart you instantly know what his deal is” and that’s simply not true.

Video game analysis tends towards all or nothing. Some sites break scoring down into categories like “sound” and “graphics”, but usually the best games get high marks across the board, and bad games get low marks everywhere.

It’s easy - and a little lazy - to dwell on the negative and minimize the positive in a game that does poorly, and do the reverse for a game that does well. Most of the analysis I’ve seen of Concord works backwards from “we know the game is terrible”, and pegs every individual feature as terrible. But that the game didn’t sell doesn’t make every component terrible. There are some games where nearly every component is terrible - War Gods, for example. (Sorry War Gods, the truth hurts) But reality is often more complicated. The first game with cutting edge deferred rendering was the Shrek Xbox game of all things, with its 49 Metacritic.

Concord is a good-looking game. I have some quibbles with the character models (mostly in the materials) but they are undeniably well-done. The designs are a mixed bag and fairly weak overall, but that doesn’t mean the technical art is poor. People who’ve actually played the game tend to like the stacking buffs mechanic - I don’t think that mechanic is splashy enough to be a selling point, but existing in a game that failed doesn’t make it a failed mechanic.

I read on Japanse gamedev Twitter that the gem system in Capcom’s Marvel Super Heroes, in which gems fall from the sky for either player to scoop up, was inspired by the “Life Ball” in an NES Muscle wrestling game.

If you’ve watched Sakurai videos you’ve noticed he frequently references a variety of old, often fairly obscure games as inspiration. Picking out the good parts of bad games is how you end up with more diverse inspiration, and acknowledging the bad parts of good ones is how you improve rather than replicate. These are important skills for a game industry professional.

Part 4: What We Can Learn from Concord’s Failure

Live service games are fraught with danger, but also HoYoverse made $4 billion dollars.

“Make more Astrobots and less Concords” sounds nice, but if a knock on the industry is that it chases trends too often “make more Astrobots” is itself a variation on “do what was popular ten minutes ago.”

Smaller teams working on projects sometimes produce an Astrobot. But Concord had a pretty limited scope, with a modest number of maps and modes. Sometimes a small budget balloons into a big one. Sometimes a small, veteran team working on a modestly scoped project makes a Hi-Fi Rush then is shut down. Or they’re a Doublefine making a string of perfectly fine games. Or a FuRyu Corporation. (Google it)

I’m sure there are things Sony can learn from Concord. The fact that so many people immediately projected it to flop while Sony seemed somewhat⁴ unaware is concerning - there must be takeaways in terms of product evaluation, feedback collection, marketing strategies, etc.

But “here’s what the game industry can learn from Concord” is mostly the domain of engagement farmers, the people who charge for game consulting phone calls, and the guys making angry youtube videos aimed at low-information followers.

When a Flappy Bird blows up the lessons are “keep it simple, stupid”, “less is more” and “players crave simplicity.” When a Baldur’s Gate 3 blows up the lessons are more is more and that players can handle and crave complexity.

If the new Dragon Age does poorly people will point to the simplification of RPG mechanics as a culprit, but if it does well the lesson will be that people like that simplification. This sort of analysis is just stating what seems most obvious in hindsight, as part of a supposed larger trend that may or may not exist.

When Helldivers 2 blew up the devs were praised for sticking to their vision and their “a game for everyone is a game for no-one” philosophy. Now that Concord has flopped is the new takeaway “your vision might be suspect so run it by a group of 13-year-olds”?

A likely result of Concord’s failure is that Sony will rely more on focus testing, on “design by committee”, go through more “market validation” steps, involve more outside consultants - rely more on a formalized process that produces safer titles. A process that produces fewer Concords but also fewer Helldivers. A process that most gamers claim to dislike.

Executing on a unique creative vision is good, unless that vision is bad. Is that a useful lesson?

Maybe there isn’t always a pat takeaway. Life is not a string of Aesop’s Fables.

Maybe the lesson is they can’t all be winners. That sounds trite, but it’s no more trite than “the problem is it didn’t appeal to players.” Yeah - no shit!

If you’ve read my previous blogs you know I’m a fan of in-depth analysis - I spent hours running into rocks in Stellar Blade to understand the character controller. But understanding the details of a particular feature is very different from understanding what will sell. In that respect, as William Goldman wrote, nobody knows anything.

1. I don’t think you can credit the “go woke go broke” people with predicting Concord to flop, since they predict that every “woke” game will flop

2. This is, admittedly, not a widespread opinion, but Raiders also isn’t very good!

3. Maybe if you play the game it’s not unmotivated but these videos are about how the character design alone should communicate theme, playstyle etc - without any backstory this guy’s theme is just “loves lions for some reason.”

4. I say “somewhat” because the game had a quick rollout and limited marketing - it’s very possible Sony knew they were in trouble