This is not an essay about anyone’s bad judgment. It is a case study in what happens when listening to players turns, almost imperceptibly, into taking dictation.
It takes the better part of an hour to earn a locked door in Arc Raiders. Scavenging the rusted sprawl of the Rust Belt for a key by slipping past the ARC, the skittering, hovering machines that drove humanity underground. You cross a map in which every other human being is either a friend you haven’t met or a threat you haven’t identified. Then you reach the door, you begin to open it, and you hear footsteps.
In the game as it shipped, in the autumn of 2025, those footsteps carry a question, and the question has teeth. The three strangers behind you might wave. They might use proximity chat to ask, a little sheepishly, whether you’d mind company. Or they might let you finish the work and then make you a corpse. Whatever happens next is a negotiation between parties who can each hurt the other. That, as it turned out, is what made Arc Raiders so unexpectedly kind.
Now picture the same door in a world where nobody can be hurt. The strangers don’t need to ask. They lean against the wall, wait for the lock to cycle, and walk in beside you to empty the shelves you paid for. You would be furious. You might reach for your rifle. And your rifle would do nothing at all.
That world opened for business yesterday. From October 13th to October 20th, Embark Studios, the Stockholm developer behind Arc Raiders, is running a test of what it calls a PvE matchmaking toggle: a switch in the menu that lets players choose between the game’s usual player-versus-player-versus-environment sessions and a version in which players deal no damage to one another at all. Solos, duos, and trios share the same sessions, yet the ARC remain hostile, other people do not.
I should say where I stand. I have spent the past seven years as a user researcher in the games industry, much of that time measuring what players feel rather than what they say, sometimes with electrodes on their fingertips. I also play. And if Embark had sent me its survey, I might well have ticked the box. Being shot in the back by a stranger after an hour of patient scavenging is not a pleasant experience, and I am not immune to wanting it to stop. But the question in front of a researcher is never only what do players want? It is what will this do to the place they share? On that question, I would have argued against this feature, tooth and nail, in every meeting I could get into.
This is not an essay about anyone’s bad judgment. It is a case study in what happens when listening to players turns, almost imperceptibly, into taking dictation.
The Fifty-One Per Cent
Embark has been unusually candid about how it got here. In a post published on September 24th, the game’s executive producer, who signs himself simply Aleks, explained that the studio tries to weigh three things for every major decision: what the data says, what players say, and what the team’s instinct says. Then he shared the data. Sorting the player base by how people most like to play, Embark found that 37 per cent are friendly co-operative players, a group that has grown steadily since launch. 27 per cent are mixed, and 34 per cent are dedicated PvP players, a group that has held flat. More than seven in ten of the friendly players, and four in ten of the PvP players, said they would like fewer hostile encounters. And 51 per cent of surveyed players expressed interest in a dedicated PvE option. The studio also named the specific irritants “ratting,” the genre’s slang for hiding and ambushing, and camping.
The German games site GameStar, reporting the announcement, found a chorus of lapsed players promising to come back. The top comment on one Reddit thread, with more than a thousand upvotes, said, in effect, that the writer suddenly felt like playing again. Others described the sting of losing an hour’s loot to a stranger, or said they’d be content simply to take down large robots with friends.
None of this is fabricated, and none of it should be dismissed. Those players are telling the truth about how they feel. The question is what that truth licenses.
To be fair to Embark, the studio is not naïve about the stakes. The same post acknowledges that the original, unreleased Arc Raiders was a PvE co-op game, that the team reset it because that concept ran into trouble, and that a toggle risks draining the tension the game was built on. Some of the game’s highest-stakes conditions, among them Night Raids, the Hidden Bunker, the Locked Gate, and the Frigate, are excluded from the test precisely because, the studio says, they depend on not knowing whom you’ll meet. Loot value in PvE sessions has been lowered. In the classic mode, it has been nudged up. Downed players who crawl in PvE may find the machines re-acquiring them, an attempt to keep a little dread in the air. And the post states plainly that the test is not a promise. Running an experiment before committing is exactly the right instinct.
My disagreement is with the question the experiment is designed to answer.
The Gun as an Institution
In 2009, Elinor Ostrom became the first woman to win the Nobel Prize in Economics, for work that dismantled a piece of conventional wisdom. The “tragedy of the commons” held that any shared resource, such as a fishery, a pasture, or a groundwater basin, would inevitably be overused, because each individual gains from taking more and the cost is spread across everyone. Ostrom went out and looked at communities that had managed commons for centuries without collapse: Swiss alpine meadows, Japanese forests, irrigation systems in Spain and Nepal. What they shared was not saintliness but institutions. The users could monitor one another, and they could sanction one another, usually lightly at first and more heavily if the offense continued. The rules worked because breaking them cost something, and because the people who enforced them were the people who lived with the consequences.
An extraction shooter is a commons. The map’s loot is finite and shared. The work of earning it, finding keys, clearing machines, opening doors, is costly and visible. And in Arc Raiders, the monitoring is the sound of footsteps, while the sanction is the gun.
Most of the time, that gun never fires. This is the part that is easy to miss. The friendly culture that made Arc Raiders famous, the game players affectionately nicknamed the “Extraction (Don’t) Shooter,” grew up inside a world where betrayal was always possible. A wave from a stranger means something precisely because the stranger could have shot you instead. Economists call this a costly signal. It is credible because it forgoes something.
When I watch people play, the moments of highest arousal, the ones that show up in their skin conductance before they show up in their words, are rarely the firefights. They are the long seconds of approach, when two players are deciding whether to trust each other. Resolve that uncertainty kindly and you have a story people tell for weeks.
Remove the possibility of betrayal and you don’t get more kindness. You get less meaning. Friendliness stops being a choice and becomes a setting. Trust without the possibility of defection isn’t trust; it’s the absence of alternatives.
What Replaces the Gun?
Go back to the room. In the PvE test, the player who spends an hour earning access to a locked room bears the entire cost; anyone standing nearby captures the benefit for free and there is no sanction available to anyone. This is the textbook definition of a free-rider problem, and the textbook prediction is that the costly, shared work will get done less often. People will wait for someone else to open the door.
There is also a less obvious prediction. When communities lose a formal sanction, they rarely become sanction-free so they improvise informal ones. I would expect, and I offer this as a hypothesis to test, not a finding, to see players invent non-lethal ways to punish freeloaders: body-blocking doorways, deliberately dragging ARC aggression toward other players, spamming chat. The hostility relocates to channels the designers never built for it and cannot easily moderate.
Americans have spent the past few years arguing about a real-world version of this feeling. Viral videos of smash-and-grab thefts, crowds walking out of shops with armfuls of merchandise while staff stand back, prompted a fierce debate about enforcement. The statistics on whether retail theft actually rose, and why, remain disputed, and I make no claim about them here. But the videos captured something regardless, which is the sense that rules bind only those who choose to follow them. Whatever the numbers say, a shop where taking feels consequence-free changes how it feels to be the honest customer waiting in line. A loot room where taking is free changes how it feels to be the one holding the key.
Two Economies, One Stash
The second problem is economic in the narrower sense. Embark’s own post notes that friendly players, and by extension PvE sessions, extract more often and earn more. Its answer is to adjust the exchange rate: less loot in PvE, slightly more in PvPvE. That is a currency peg between two countries. If, as the promise of seamless switching suggests, the same stash travels between modes, the peg has to hold perfectly, forever, or one side becomes the farm and the other the casino. Every economy that has tried to run two price levels across one border has discovered how much effort that takes.
The third problem is the one I would have put in bold. The players most likely to flip the toggle are the friendly 37 per cent, the very people whose presence made the classic mode humane. When they leave, the remaining PvPvE population becomes more concentrated in exactly the behavior everyone complained about. For the 27 per cent of mixed players, who enjoy both the cooperation and the risk, the classic mode could get noticeably harsher, not gentler. Embark flags this itself, in a line about how PvE uptake might affect the quality of some PvPvE sessions. In other words, a feature designed to relieve the friction of ratting and camping could intensify it for anyone who stays behind.
What the Survey Can’t Tell You
Here is the thing every junior researcher learns, usually the hard way. Players are experts in their frustrations and unreliable designers of their fixes. “I’m tired of being ambushed” is a finding. “Give me a PvE toggle” is a proposed solution, and a proposed solution in a survey is a stated preference, which is to say a forecast about one’s own future behavior. We are not good at those.
Several things about the 51 per cent should give a team pause. It is 51 per cent of surveyed players, and people who answer game surveys are not a random sample of people who play. It measures interest in an option, and almost nobody says no to more options. Choice feels free when you are not the one paying for it. And the most enthusiastic voices in the coverage belong to people who have already left, promising to return. A promise to come back is not the same as a reason to stay. A one-week test, during a seasonal update, will capture novelty almost by design. If I were running it, I would care far less about how many people flip the toggle than about what happens to both populations a month later, how often doors get opened, how much of the classic mode’s sessions turn hostile, and whether returning players are still around when the toggle disappears on the 20th.
What I Would Have Written
A research report doesn’t end with findings. It ends with recommendations, and the recommendation is where the job actually lives. Mine would have read something like this.
Finding: A large and growing share of players want fewer hostile encounters. The pain is concentrated around specific behaviors, ratting and camping, rather than around the existence of other players.
Recommendation: Do not ship a toggle that removes player damage. It addresses the symptom by removing the institution that gives cooperation its value, it creates a free-rider problem at the game’s most rewarding moments, it splits the economy, and it risks making the classic mode worse for the players who remain.
Instead: Target the behaviors, not the possibility. Make playstyle-based matchmaking, which the game already uses, stronger and more legible, so friendly players can trust that they are likely to meet other friendly players. Give betrayal a memory, so that ambushing someone who waved at you leaves a visible mark. Build PvE spectacle by making the machines bigger rather than making the people harmless, the kind of large, communal events some players were asking for a year ago. And if the team still wants to test a toggle, test it against those alternatives, and measure the health of the community, not just the enthusiasm of the request.
That is not a refusal to listen but rather what listening looks like when it is done professionally. Yes, this is what players are asking for, and here, in detail, is why we don’t recommend giving it to them in this form.
Back at the Room
I keep returning to that locked room, and to the footsteps behind it. In the version of Arc Raiders that people fell in love with, the moment you hear them is the most alive you feel all evening. Someone has a choice to make about you, and you have one to make about them. Most of the time, it turns out, people choose to wave.
The test ends on October 20th, and Embark has said it will share what it learns. I hope I’m wrong. But I suspect the studio will discover what Ostrom found in the meadows: a commons stays generous not because nobody can take from it, but because taking has a price. Remove the price and you have not made the place safer. You have only made the kindness free, and free things are rarely treasured.
Sources:
- Embark Studios, “We’re testing a PvE Toggle: here’s what it means for you,” arcraiders.com, September 24, 2026. https://arcraiders.com/news/pve-toggle-beta-test
- Marie-Lena Höftmann, “Arc Raiders testet reines PvE und etliche Spieler kündigen schon jetzt ihre Rückkehr an,” GameStar, September 27, 2026. https://www.gamestar.de/artikel/arc-raiders-reines-pve-spieler-kuendigen-rueckkehr-an,3460195.html
- “The problem with PVE and why it won’t work,” ARC Raiders General Discussions, Steam Community, October 25, 2025. https://steamcommunity.com/app/1808500/discussions/0/599667160179405954/
- Benda Hofmeyr, “The tragedy of the commons revisited: Hardin meets Ostrom.” https://www.tandfonline.com/doi/abs/10.1080/02580136.2025.2513751
- Elinor Ostrom, “Tragedy of the Commons.” https://eclass.uoa.gr/modules/document/file.php/PHS542/Ostrom%20-%202008%20-%20Tragedy%20of%20the%20Commons.pdf