Blog Archives
Cut the Concord
Statistically, you have never heard of it, but Sony is shutting down their new Overwatch-like hero shooter called Concord. After two whole weeks. On Steam, Concord apparently never broke 700 concurrent players at launch. The writing was probably on the wall from earlier when the open beta population was worse than closed beta – plus it launching as a $40 B2P in a sea of F2P competitors – but the sheer scale of the belly flop is shocking nonetheless. It is rumored to have cost $200 million, and for sure has been in development for eight (8!) years.
And now it’s gone.
You know, there are people out there that argue games should be more expensive. Not necessarily because of the traditional inflation reasons – although that factors in too – but because games costs more to make in general. Which really just translates into longer development times. And yet, as we can see with Concord along with many other examples, long development times do not necessarily translate back into better games. There is obviously some minimum, but longer isn’t better.
And yet, we have these industry leaders who suggest MSRP should be higher than the now-standard $70. To be “priced accordingly with its quality, breadth & depth,” as if any of that is really knowable from a consumer standpoint prior to purchase. We have reviews, sure, and Concord score a 70 from IGN. What does that tell you?
The overall way games are made is itself unsustainable, and an extra $10-$20 per copy isn’t going to fix anything. Indeed, there seems to be a blasé attitude in the industry that a rising MRSP will lift all the boats instead of, you know, causing the ones on the periphery to slide down the demand wave curve. Suppose GTA6 is released at $80. Is the argument that a consumer’s gaming budget will just indefinitely expand by an extra $10/year? Or will they, I dunno, just spend $10 less on other titles? Higher prices are certainly not going to expand the market, so… what?
As far as I can see it, the only reasonable knob to turn appears to be development time and nobody seems able to do it. I’m not trying to handwave away the effort and brute labor necessary to digitally animate mo-capped models in high fidelity. Or creating and debugging millions of lines of bespoke code. But I am asking how long it does take, how much of it is necessary, and how often these visually stunning games fall flat on their faces in the one function of their intended existence, e.g. being fun.
Throwing more money at the problem certainly doesn’t seem to be working.
Unsustainability
Senua Saga: Hellblade 2 recently came out to glowing reviews and… well, not so glowing concurrent player counts on Steam. Specifically, it peaked at about 4000 players, compared to 5600 for the original game back in 2017, and compared to ~6000 for Hi-Fi Rush and Redfall. The Reddit post where I found this information has the typical excuses, e.g. it’s all Game Pass’s fault (it was a Day 1 release):
They really don’t get that gamepass is unsustainable. It works for Netflix because movies and tv shows can be made in a year or less so they can keep pumping out content each year. Games take years to make and they can’t keep the same stream of new content releasing the same way streaming services do.
Gamepass subs are already stagnating, they would make more money if they held off putting new exclusives on gamepass like movies do with putting them in theatres first before putting them on streaming. (source)
Now, it’s worth pointing out that concurrent player counts is not precisely the best way to measure the relative success of a single-player game. Unless, I suppose, you are Baldur’s Gate 3. Also, Hellblade 2 is a story-based sequel to an artistic game that, as established, only hit a peak of 5600 concurrent players. According to Wikipedia, the original game sold about 1,000,000 copies by June 2018. Thus, one would likely presume that the sequel would sell roughly the same amount or less.
The thing that piqued my interest though, was the reply that came next:
Yeah, even “small” games like Hellblade and Hi-Fi Rush, which are both under 10h to complete, took 5/6 years to develop. It’s impossible to justify developing games like these with gigantic budgets if you’re going to have them on your subscription service.
I mean… sure. But there’s an unspoken assumption here that these small games with gigantic, 5-6 year budgets would be justified even without being on a subscription service. See hot take:
Hellblade 2 really is the ultimate example of the flaw of Xbox’s “hands off” approach to game dev.
How has a studio been able to take 5 years making a tiny game that is basically identical to the first?
How did Rare get away with farting out trailers for Everwild despite the game literlaly not existing?
Reddit may constantly slag off strict management and studio control, but sometimes it’s needed to reign studios in and actually create games…
Gaming’s “sustainability problem” has long been forecast, but it does feel like things have more recently come to a head. It is easy to villainize Microsoft for closing down, say, the Hi-Fi Rush devs a year after soaking up their accolades… but good reviews don’t always equate to profit. Did the game even make back its production costs? Would it be fiduciarily responsible to make the bet in 2024, that Hi-Fi Rush 2 would outperform the original in 2030?
To be clear, I’m not in favor of Microsoft shutting down the studio. Nor do I want fewer of these kind of games. Games are commercial products, but that is not all they can be. Things like Journey can be transformative experiences, and we would all be worse off for them not existing.
Last post, I mentioned that Square Enix is shifting priorities of their entire company based on poor numbers for their mainline Final Fantasy PS5 timed-exclusive releases. But the fundamental problem is a bit deeper. At Square Enix, we’ve heard for years about how one of their games will sell millions of copies but still be considered “underperforming.” For example, the original Tomb Raider reboot sold 3.4 million copies in the first month, but the execs thought that made it a failure. Well, there was a recent Reddit thread about an ex-Square Enix executive explaining the thought process. In short:
There’s a misunderstanding that has been repeated for nearly a decade and a half that Square Enix sets arbitrarily high sales requirements then gets upset when its arbitrarily high sales requirements fail to be met. […]
If a game costs $100m to make, and takes 5 years, then you have to beat, as an example, what the business could have returned investing $100m into the stock market over that period.
For the 5 years prior to Feb 2024, the stock market averaged a rate of return of 14.5%. Investing that $100m in the stock market would net you a return of $201m, so this is our ROI baseline. Can the game net a return higher than this after marketing, platform fees, and discounts are factored in?
That… makes sense. One might even say it’s basic economics.
However, that heuristic also seems outrageously unsustainable in of itself. Almost by definition, very few companies beat “the market.” Especially when the market is, by weight, Microsoft (7.16%), Apple (6.12%), Nvidia (5.79%), Amazon (3.74%), and Meta (2.31%). And 495 other companies, of course. As an investor, sure, why pick a videogame stock over SPY if the latter has the better return? But how exactly does one run a company this way?
Out of curiosity, I found a site to compare some game stocks vs SPY over the last 10 years:

I’ll be goddamned. They do usually beat the market. In case something happens to the picture:
- Square Enix – 75.89%
- EA – 276.53%
- Ubisoft – 30.56%
- Take Two – 595.14%
- S&P 500 – 170.51%
And it’s worth pointing out that Square Enix was beating the market in August 2023 before a big decline, followed by the even worse decline that we talked about recently. Indeed, every game company in this comparison was beating SPY, before Ubisoft started declining in 2022. Probably why they finally got around to “breaking the glass” when it comes to Assassin’s Creed: Japan.
Huh. This was not the direction I thought this post was going as I was writing it.
Fundamentally, I suppose the question remains as to how sustainable the videogame market is. The ex-Square Enix executive Reddit post I linked earlier has a lot more things to say on the topic, actually, and I absolutely recommend reading through it. One of the biggest takeaways is that major studios are struggling to adjust to the new reality that F2P juggernauts like Fortnite and Genshin Impact (etc) exist. Before, they could throw some more production value and/or marketing into their games and be relatively certain to achieve a certain amount of sales as long as a competitor wasn’t also releasing a major game the same month. Now, they have to worry about that and the fact that Fortnite and Genshin are still siphoning up both money and gamer time.
Which… feels kind of obvious when you write it out loud. There was never a time when I played fewer other games than when I was the in the throes of WoW (or MMOs in general). And while MMOs are niche, things like Fortnite no longer are. So not only do they have to beat out similar titles, they have to beat out a F2P title that gets huge updates every 6 weeks and has been refined to a razor edge over almost a decade. Sorta like how Rift or Warhammer or other MMOs had to debut into WoW’s shadow.
So, is gaming – or even AAA specifically – really unsustainable? Possibly.
What I think is unsustainable are production times. I have thought about this for a while, but it’s wild hearing about some of the sausage-making reporting on game development. My go-to example is always Mass Effect: Andromeda. The game spent five years in development, but it was pretty much stitched together in 18 months, and not just because of crunch. Perhaps it is unreasonable to assume the “spaghetti against the wall” phase of development can be shortened or removed, or I am not appreciating the iteration necessary to get gameplay just right. But the Production Time lever is the only one these companies can realistically pull – raising prices just makes the F2P juggernaut comparisons worse, gamer ire notwithstanding. And are any of these games even worth $80, $90, $100 in the first place?
Perversely, even if Square Enix and others were able to achieve shorter production times, that means they will be pumping out more games (assuming they don’t fire thousands of devs). Which means more competition, more overlap, and still facing down the Fortnite gun. Pivoting to live service games to more directly counter Fortnite doesn’t seem to be working either; none of us seem to want that.
I suppose we will have to see how this plays out over time. The game industry at large is clearly profitable and growing besides. We will also probably have the AAA spectacles of Call of Duty and the like that can easily justify the production values. Similarly, the indie scene will likely always be popping, as small team/solo devs shoot their shot in a crowded market, while keeping their day jobs to get by.
But the artistic AA games? Those may be in trouble. The only path for viability I see there is, ironically, something like Game Pass. Microsoft is closing (now internal) studios, yes, but it’s clearly supporting a lot of smaller titles from independent teams and giving them visibility they may not otherwise have achieved. And Game Pass needs these sort of games to pad out the catalog in-between major releases. There are conflicting stories about whether the Faustian Game Pass Bargain is worth it, but I imagine most of that is based on a post-hoc analysis of popularity. Curation and signal-boosting is only going to become increasingly required to succeed for medium-sized studios.
The Hopes of the Game Industry
Jan 20
Posted by Azuriel
In short: they hope GTA 6 will cost $100 so they can raise their own prices.
There is a ridiculous sort of myopia associated with seeing (and/or experiencing) high-profile commercial failures and escalating production costs, only to come to the conclusion everything would be better with higher unit prices. How about… *checks notes* … lower production costs? “But players demand AAAA-quality graphics!” Do they? I can appreciate the dilemma faced by developers, wherein the last game cost $400m and not wanting to gamble with a $350m (or lower) sequel. But if the acknowledgement is that the status quo of ever-increasing production costs is unsustainable, higher prices at best stems the bleed temporarily. At some point you need to address the root cause.
I was curious at this point as to what “the industry” actually thought about things, and if GTA 6 selling for $100 was all of it. So, the article I linked to above points to this VGC article, which then points to a 222-slide presentation by Matthew Ball, whom appears to be a “strategy advisor” to, presumably, the games industry (and others). If you have the time, I do very much encourage you to take a look yourself, as it is surprisingly straight-forward and facts-based. A summary:
The final section of the presentation includes thoughts on potential new growth engines. And it does include GTA 6, but also several others.
Again, I think it is worth looking at the presentation yourself, as each of the 11 bubbles there get multiple slides that introduce, justify, and even caution about the “solution.” Well, aside from GTA 6, which is noted would be the cheapest GTA ever (in real terms) if it comes out at $70. GTA 5 was released in 2013 at $60, which would be over $80 today, for example. Notwithstanding the billions of dollars GTA Online brought in, of course.
Overall, I did come away a bit more sympathetic to the plight of the games industry. Some of the headwinds I can personally attest to. For example, there have been multiple nights in which I found 2-3 hours of my “gaming time” consumed by Youtube Shorts scrolling. The network effect or “black hole” games are certainly a challenge as well, as anyone who has spent years playing MMOs can attest to. How do you compete against Fortnite, Minecraft, Roblox, and/or all the others?
“Raise prices,” of course!
Unfortunately, the actual solution is both pithy and hard to achieve: make fun games. Note how that solution did not include the words “spend 8 years painstakingly rendering every blade of grass.” Also note that I’m not saying that coming up with a fun game is easy either. But the industry seems stuck in this death loop of hiring more artists, programmers, marketers, and greenlighting enormously long development times… only for the game to fall flat because the fun wasn’t there. You can’t just hire more people to increase the fun quotient. And sometimes the fun that is achievable is only experienced by a narrow slice of the market, too small to be sustainable for the larger companies.
I don’t know the solution. If I did, I certainly wouldn’t be giving it out for free. But it might well be… decimation for the industry. I think a lot of publishers are just going to go bankrupt trying to spend their way out of the tailspin. AI could be a big disrupter, but disruption favors small indie shops, not the big guys. And while I do feel like longer development times is the obvious root issue for ballooning costs, I don’t see how the industry moves towards shorter development times and… then what? More releases? I mean, I wouldn’t be mad about a new Fallout every 2 years. If they keep the releases the same with a shorter development time though, that just means an implosion in the game jobs market. Not ideal.
…or maybe it is?
I dunno. I’m just a dude looking for fun games to play with my ever-decreasing amount of free time and eroding consumer surplus. When I look at my most-played games though, what I don’t see is full-priced titles with photo-realistic graphics and 8+ years in development. Well, I guess some Early Access titles were being worked on for that long, but it was like with three guys, not three hundred.
Anyway, Take Two can try and take $100 if they want and everyone raise prices as a result. Maybe it works, maybe it doesn’t. Nevertheless, my parsimony will abide.
Posted in Commentary
2 Comments
Tags: Consumer Surplus, Fun, GTA 6, State of the Industry, Unsustainable