Snapshot Every number here is as of 21 September 2026, 09:05 UTC, 77 hours before launch. We will check it against the outcome on 26 September, misses included.
Revised on 21 September: following reader feedback we cut the text and reordered the sections. Later we added a check of the boredom finding by a different method (section 06) and corrected the note under the themes chart: happy reviewers mention weirdness one and a half times as often, not twice. The numbers, the benchmark and the seven predictions are exactly as recorded at the snapshot.
01The short version
Another masterpiece — or yet another boring, pretentious, incomprehensible thing? That is how people argue about Remedy games. CONTROL Resonant launches on 24 September at 14:00 UTC and has already walked into that argument: some reviewers call it the studio's finest work, others its weakest in a decade. This issue is a forecast: we write down what to expect before we know the answer.
We have two questions: how many people turn up in the first two days, and how many of them stay. The first is answered by pre-launch demand and six similar launches from this summer. The second, by 17,000 reviews of the studio's earlier games.
Six premium launches from this summer with similar demand peaked at 15,000 to 82,000 concurrent players. The width of the corridor is the spread among the cohort games; it does not measure our own error for Resonant.
How many turn up. Our benchmark is 35,000–55,000 concurrent players. Demand is strong and still rising: the game sits #19 on Steam's global Top Wishlists and is gaining followers at ten times the August pace. Pre-orders are more modest. Three things pull the benchmark down: fewer followers than the games at the same wishlist rank; the game is not sold in Russia; and Steam already files the sequel under action RPG.
How many stay. Remedy is hardly ever punished for being pretentious or incomprehensible. It is punished for being boring: unhappy players mention it two to three times as often as happy ones, for every game the studio has made. Its weirdness, by contrast, gets praised. And in Resonant, combat — where the studio lost players before — has moved to the centre of the game. Our bet is 78–85% positive reviews: between "masterpiece" and "boring again".
And the weak spot up front. We chose the boundary of the comparison group after seeing the outcomes: with a wider boundary the centre would have been about 25,000. Why we still took the narrow one is in section 05.
02Pre-launch demand
Demand has been accelerating for three weeks straight. Pre-orders, though, were moved by just two events — the launch trailer and the reviews.
CONTROL Resonant is the sequel to Control: a new lead, Dylan Faden, a paranatural Manhattan, and melee in place of gunplay. On Steam it is self-published by Remedy and costs $59.99. There is no demo, and we have not seen any playtests either.
So demand has to be read from three indirect series: the rank on the wishlist charts, community followers, and the sales chart, which counts pre-orders. We start with wishlists.
From late July and through August the game did not move on Steam's global Top Wishlists: #31–35. It shifted twice: with the Gamescom trailer on 26 August, and in the final week, which took it from #24 to #19.
Top Wishlists ranks the stock — how many wishlists have accumulated; the activity chart ranks how fast they are being added right now. The stock stood still in August, while the pace has climbed from #89 to #19 since early September.
Why a rank is not a wishlist count
How many wishlists the game has cannot be read off its rank. The only way to turn a rank into a number is the Steamworks dashboards developers publish themselves, and games in the top twenty do not show theirs. So "#19" in this issue only shows where the game stands relative to others.
Followers
The second series is community followers, visible from the announcement itself. Growth came in three waves: the announcement in December, the release-date reveal in early June, and the final month. Between the waves the count grew by 1,200–3,000 a month; over the last two days it grew by 959 and 775 a day. By 20 September it reached 48,253.
On announcement day, 12 December 2025, there were 7,309 followers. The history before 20 July is a third-party series, from SteamDB; where we have our own readings, it matches them to within a percent.
Pre-orders: two events
The third series is Top Sellers: Steam ranks games there by revenue and counts pre-orders. Over the ten days before launch Remedy published something almost daily, yet the US rank moved smoothly — except for two particular hours.
There are two steps: the launch trailer and the reviews. The bundle with the first game produced no step. The dip on the night of 20 September, down to #24, recovered by the evening.
- 17.09 · 14:05 UTC Launch trailerUS Top Sellers: 45 → 29 in four hours. In the previous 24 hours it had moved within #44–56.
- 18.09 · 13:00 UTC Review embargo liftsRank: 27 → 7 in three hours. At 15:09 the game enters the top-sellers block on the US Steam front page for the first time (#8, #4 by 16:38). Followers over the next day: +959.
- 18.09 · 14:00 UTC Launch-times postPublished an hour after the reviews, so its effect cannot be separated from theirs.
- 19.09 · 11:29 UTC Bundle with the first gameCONTROL Ultimate Edition and Resonant for $98.98 (−10%). Rank before and after: #8.
The link to the events shows in the timing: both climbs began in the same hour, after a flat stretch. But a rank only says that pre-orders in those hours ran faster than for the chart neighbours. It does not say how many there were.
Every Steam storefront at the snapshot
| Storefront | Rank | Where | Comment |
|---|---|---|---|
| Popular Coming Soon | #4 | in all 40 panel countries where the game is sold | Unchanged at #4 since 17 September |
| Top Wishlists | #19 | global chart | Up from #34 over eight weeks |
| Top Sellers | #10 | US and UK | China #24, Germany #26, Brazil #36, Japan #127 |
| Global Top Sellers | #15 | global chart | The only unreleased game above it is EA SPORTS FC 27 (#3) — out the same day |
| Front page: top sellers | #2 | 10 countries in total | In the US only on the evening of 18 September, #4–8; at the snapshot Norway, Switzerland, New Zealand |
"All countries" in this issue means 40 of the 41 countries we observe: the game is not sold in Russia.
The "More like this" block gives the game 1,339 pages from which Steam sends a US visitor its way. The most visible are the first of six visible slots in the "Coming soon" section on the pages of Black Myth: Wukong, Cyberpunk 2077, The Witcher 3, Elden Ring, Monster Hunter: World, Fallout 4 and Hogwarts Legacy. This describes the shelf. We have not yet shown for any game that such placements bring in buyers.
03The series warms up
The first Control has picked up players again ahead of the sequel — even without a discount. But a noticeable part of its audience will not be able to buy the sequel.
Demand for the sequel also shows in the first game. CONTROL Ultimate Edition — 54,500 reviews, 85.8% positive — held a daily peak of 684–1,200 players from 19 July to 25 August. On 26 August, the day of the Gamescom trailer, it went on a −90% sale, down to $3.99, for twelve days.
After the sale the player count did not return to the old base but settled at about two and a half times July. And on 20 September, four days before the sequel and with no discount at all, the first game hit 3,638 — the highest point of the whole observation window.
Reviews show the same warm-up: before the sale the first game drew 12–32 reviews a day, during and after it 28–68. The effect of the sale cannot be separated from the trailer and the bundle here: they fell into the same window.
Russia and China
Now for those who cannot follow the sequel. Of the reviews of the first game since December 2025, 18% are in Russian (political ones excluded) and 11% in Simplified Chinese. Resonant is unavailable in the Russian store: a noticeable part of the series' current audience cannot buy the sequel officially.
The backlash came long before launch. On 3 June, the morning after the date announcement, a wave of negative Russian-language reviews hit the first Control: 41 in a day in our sample against the usual one to five, 29 of them about region locks and politics. In September such reviews are almost gone: the wave was a one-off. The theme analysis in section 06 excludes them.
China, by contrast, got a lot of attention: since 25 August the page has had localised banners and links to bilibili, weibo and Xiaohongshu. The game is #24 on the Chinese Top Sellers chart right now.
04Six premium launches from this summer
To turn demand into a player count we need neighbours — games that came to launch with similar numbers.
We took every release from 26 July to 17 September 2026 priced at $39.99 or more that sat no lower than #60 on Top Wishlists in its final week before launch. There are eight. Two multiplayer shooters — WARDOGS and Hell Let Loose: Vietnam — were excluded. That leaves six single-player games, all measured the same way: from the moment of launch.
With similar follower counts the cohort peaked very differently: from 0.37 to 1.43 players per follower. Beast of Reincarnation reached launch with almost as many followers as Resonant has now, yet peaked at 15,000 — the lowest ratio in the cohort. Its follower count is an estimate; see the table below.
The wishlist rank orders the outcome only roughly: the two best-ranked games, Dawnwalker and Onimusha, landed in the top three by players; the two worst, Halo and Beast, in the bottom pair. In the middle the order does not hold. On this measure Resonant sits near the top: #19 against #9–41 for the rest.
The cohort table: prices, ranks, followers, retention
| Game | Price | Top Wishlists, 3 days before launch | Followers, 3 days before launch | 48-hour peak | Day-7 retention | Positive, week one |
|---|---|---|---|---|---|---|
| Onimusha: Way of the Sword | $69.99 | 18 | 56,885 | 81,559 | 68% | 88.1% |
| STAR WARS Zero Company | $49.99 | 26 | 70,667 | 69,046 | 57% | 78.5% |
| The Blood of Dawnwalker | $69.99 | 9 | 100,465 | 64,438 | 74% | 82.3% |
| Mortal Shell II | $49.99 | 25 | 50,102 | 44,486 | 53% | 72.5% |
| Halo: Campaign Evolved | $49.99 | 41 | 46,339 | 25,371 | 34% | 58.2% |
| Beast of Reincarnation | $59.99 | 38 | ≈40,500 | 14,995 | 35% | 55.6% |
| CONTROL Resonant | $59.99 | 19 | 48,253 | ? | ? | ? |
"3 days before launch" means 72 hours before launch, the same point where Resonant is now; for followers it is the latest known value 2–5 days before launch. "Retention" is the day-seven peak as a share of the first-48-hour peak. Beast of Reincarnation's followers are interpolated between SteamDB's points: 37,509 on 27 July and 41,725 on 3 August.
The final week of pre-orders
The cohort does have one thing in common. All six climbed to #1–3 on US Top Sellers by launch — even those that sat tenth or fifty-first three days before release.
The cohort converges on the top in the final day: pre-orders on the eve of launch and the first hours after it. Resonant stands at #9–10 at the snapshot, 77 hours out. This chart alone is not enough for a forecast: Halo spent the whole week in the top five yet peaked at only 25,000.
05The benchmark: 35,000–55,000
Two simple estimates from the cohort give 54,000 and 37,000. The benchmark stands on them.
There is nothing to fit a model to for an unreleased game: it has no player counts yet. So we take two anchors from the cohort — both are visible in the charts above.
Adjustments whose direction we know but whose size we do not:
- Down — fewer followers than the games at the same wishlist rank: 48,000 against 57,000 for Onimusha and 100,000 for Dawnwalker.
- Down — the game is not sold in Russia.
- Down — a 48-hour window from Thursday 14:00 UTC ends on Saturday at 14:00, before Saturday evening. The two Friday releases in the cohort peaked precisely on day two.
- Up — the final-week acceleration in three series at once: wishlist pace, followers, pre-orders.
- Unknown — whether the shooter audience follows the game into action RPG.
The weak spot of the forecast Before seeing the outcomes we wrote down a wide cohort boundary: Top Wishlists rank of #150 or better. The "#60 or better" threshold and the $39.99 price floor were chosen later, with the list of peaks already in front of us. The reason: Resonant is #19, while the excluded games were #104–147. But the choice was not made blind, and it has a price. With the wide boundary the group would have included Resonance: A Plague Tale Legacy (5.5k), STEINS;GATE RE:BOOT and METAL GEAR SOLID: Master Collection — and the centre would have dropped to about 25,000. If the outcome lands near the bottom of the corridor, this is the first thing we will check.
Benchmark recorded on 21 September (refined at 13:30 UTC) Peak concurrent players in the first 48 hours — from 24 September, 14:00 to 26 September, 14:00 UTC — 35,000–55,000. Cohort corridor: 15,000–80,000. Landing in 35–55k confirms the benchmark; 15–35k or 55–80k is inside the corridor but off-centre; outside 15–80k the benchmark is wrong, and we will work out why. We refined the second anchor at 13:30 UTC, before launch: adding Beast of Reincarnation's followers to the cohort lowered it from 43,000 to 37,000, and the lower bound of the benchmark from 40,000 to 35,000.
06Will the players stay
How many people turn up depends on demand. How many stay depends on the reviews. And what spoils reviews of Remedy games is not the weirdness — it is boredom.
First, why reviews matter so much. In the cohort the link shows even without a model.
Games at 78% positive or better kept 57–74% of their peak by day seven, and the two below 60% kept about a third. The dashed line is the trend across these six points: roughly +0.12 of peak for every 10 points of positive share. There are only six points, but the direction matches what we have seen across thousands of indie releases.
So the main question is what the reviews will look like. Remedy has a peculiar reputation: the same games get called both masterpieces and pretentious bores. The review embargo lifted on 18 September, six days before launch, and the reviews of Resonant split the same way.
"Masterpiece"
- GameSpot — 9 out of 10
- VICE — 10 out of 10: "A Mind-Bending Action Masterpiece"
- Empire: "for those who persevere, this emerges as another brilliant slice of high concept sci-fi gaming"
- Rock, Paper, Shotgun: "a superbly strange action-RPG"
- Skill Up, as a curator: "Remedy at their absolute finest"
"Boring again"
- ComicBook — 2 out of 5: "Its melee combat is bafflingly hollow, relentlessly repetitive, and poorly balanced across the board"
- Same review: "not even the unexplainable is intriguing enough"
- Eurogamer — 3 out of 5
- CGMagazine: the story and the gameplay "seem at odds"; as a curator — "the weakest Remedy game in over a decade"
Taken together, the press gives the game neither a flop nor a breakthrough. The top-critic average on OpenCritic is 84 from 66 reviews, exactly the same as the first Control's; on Metacritic the PC version sits at 83. What differs is the split: the weirdness is praised, the combat is panned. Even Empire's praise opens with a caveat: "for those who persevere".
What Remedy's players write
The press argues, but players will decide. To find out which of the three charges — boring, confusing, pretentious — actually sets unhappy players apart, we took 17,000 Steam reviews in English and Russian across five of the studio's games, from Alan Wake to Firebreak, and counted how often each theme shows up in negative and in positive reviews.
"Boring" and "repetitive" show up three times as often in unhappy reviews as in happy ones. "Confusing" and "pretentious" are rarer and only twice as frequent. And weirdness and atmosphere are mentioned one and a half times as often by people who liked the game. Unhappy reviewers write longer, but the gap holds among reviews of the same length too.
And this is not a quirk of Control alone — it holds for all five games.
For every game the studio has made, unhappy reviewers mention boredom two to three times as often as happy ones, and for Control and Alan Wake it appears in a third of all negative reviews. Only the pair of bars within one game can be compared, not the games with each other. This is a word search; the check by meaning is below.
A word search is a blunt tool, so we re-checked the boredom finding another way. We pulled the reviews of the same five games afresh and labelled 2,000 of them by meaning: 200 negative and 200 positive per game. The labelling was done by the Claude language model, blind — it saw neither the game nor the review's verdict. 200 reviews were labelled twice, and the labels matched in 96.5% of cases.
The finding held, with room to spare. By meaning, unhappy reviewers mention boredom 3.5–7 times as often as happy ones, and for all five games the lower bound of the 95% interval is at least 2.5. The word search errs both ways. It misses about a third of the complaints: it cannot see "My smartwatch thought I was taking a nap". And in positive reviews every third hit is false: it counts "Still I did not get bored yet" as boredom. Both errors squeeze the gap, so the numbers in the chart above are a lower bound. By meaning, boredom comes up in 43% of the negative reviews of Control and 54% of Alan Wake's, not a third.
The check by meaning: numbers, causes of boredom and caveats
| Game | Boredom by meaning: neg. / pos. | Gap | 95% interval |
|---|---|---|---|
| Alan Wake | 54% / 16% | ×3.5 | 2.5–5.2 |
| Quantum Break | 32% / 5% | ×7.1 | 4.1–16.6 |
| CONTROL Ultimate Edition | 43% / 9% | ×4.8 | 3.2–8.2 |
| Alan Wake's American Nightmare | 59% / 16% | ×3.7 | 2.7–5.5 |
| FBC: Firebreak | 38% / 10% | ×4.0 | 2.6–6.8 |
We tuned the comparison word list to reproduce the issue's shares. On this sample it has 84% precision and 67% recall — 58% for Russian. In negative reviews 8% of its hits are false; in positive ones, 36%. The sample was pulled afresh through Steam's public API and is not the same as the issue's sample, so we do not replace the issue's numbers; we show these next to them.
We also recorded what the complaint was about. For Alan Wake and Quantum Break boredom is most often about combat: 38% and 36% of complaints. For Control combat is still the first named cause, but at 22%, with almost as much, 17%, going to running around the map. For Control, "combat — where the studio lost players before" says more than the data allows.
The prediction about "boring" in reviews of Resonant from section 08 will be checked the way it was recorded — by words. By meaning the share is higher by construction, and holding it against the one-third threshold would not be fair.
All three themes across the five games
| Game | Positive on Steam | "Boring, repetitive" | "Confusing, pretentious" | "Weird, atmospheric" |
|---|---|---|---|---|
| Alan Wake | 89% | 35% / 14% | 6% / 2% | 14% / 23% |
| Quantum Break | 86% | 17% / 7% | 7% / 3% | 6% / 7% |
| CONTROL Ultimate Edition | 86% | 33% / 10% | 15% / 6% | 16% / 25% |
| Alan Wake's American Nightmare | 74% | 34% / 16% | 5% / 2% | 14% / 13% |
| FBC: Firebreak | 66% | 24% / 11% | 5% / 3% | 7% / 9% |
The theme columns show the share in negative / in positive reviews. Confusion is noticeable only for Control, and at half the rate of boredom. Weirdness comes up more often among happy reviewers of Control and Alan Wake, and evenly for the other games.
And the disappointed leave early. 38% of negative reviews of Control were written before the fifth hour of play, against 14% of positive ones. The median unhappy reviewer had played 7.5 hours, the median happy one 19.3.
Combat has moved to the centre
That is Resonant's risk. Combat used to support Remedy's stories, and even then boredom came up in a third of negative reviews. In the sequel combat is at the centre: gunplay has given way to melee in an open city. It is also what critics pan most, while the weirdness is praised here too.
Steam has already taken note. The first game leads with Action, Female Protagonist, Supernatural, Story Rich, with Third-Person Shooter further down. Resonant leads with Action RPG and Hack and Slash, and has neither a heroine nor a shooter tag at all. Its "More like this" block lists Zenless Zone Zero, Mortal Shell II, Final Fantasy VII Rebirth and God of War, with the first Control only at #20. So the sequel is shown to action RPG players. Whether those who loved the first game's gunplay follow it will show in the reviews.
Two hypotheses, recorded on 21 September "Another masterpiece": by 26 September the share of positive Steam reviews is 85% or higher, like the first game's; "boring" and "repetitive" appear in no more than a third of negative reviews; by day seven the game keeps 0.68 of its peak or more. "Boring again": the positive share is below 75%; "boring" and "repetitive" appear in more than 40% of negative reviews, with nearly half of those written before the fifth hour; retention is half the peak or less. Our bet is in between: 78–85% positive. The press puts the game level with the first one, but pans the very combat that the sequel is built around.
07What decides the outcome
Five forks that will show, within the first days, which edge of the corridor the launch is heading for.
- The final day of pre-orders. All six cohort games were at #1–3 on US Top Sellers by launch. If Resonant stays around tenth, it is weaker than all of them, and the benchmark should be read closer to its lower edge.
- The first reviews. In the cohort, a positive share below 60% meant keeping about a third of the peak by the end of the week and a sag as early as day two. Among the game's 53 curators are two with a political agenda, so the first hours of reviews should be read with possible organised negativity in mind.
- Carrying over the series audience. The sequel changed its lead and its combat. Whether shooter fans follow will show in the reviews.
- Russia. How much of the audience dropped out will show in the share of Russian-language reviews compared with 18% for the first game.
- Genre neighbours are still in play. In the three weeks before Resonant, Onimusha, Dawnwalker and Mortal Shell II came out, and their players are still busy: on 20 September Onimusha peaked at 34,000 and Dawnwalker at 30,000.
The calendar around launch
- 23.09 · $49.99 SILENT HILL: TownfallTop Wishlists #75. Story-driven single-player, out the day before.
- 24.09 · $59.99 CONTROL ResonantTop Wishlists #19.
- 24.09 · $69.99 EA SPORTS FC 27Top Wishlists #329. Different audiences, but the same day — a fight for the top of Top Sellers and New Releases; FC is fifth in US pre-orders.
- 01.10 · $69.99 ACE COMBAT 8Top Wishlists #39. Premium single-player a week later.
- 06.10 · $69.99 Gears of War: E-DayTop Wishlists #40. Premium third-person action.
- 28.10 · $59.99 Phantom Blade ZeroTop Wishlists #6. A direct competitor by genre and price.
Resonant has a clean window of about three weeks: in October the genre wave arrives, from Valor Mortis and Castlevania: Belmont's Curse to Phantom Blade Zero. A Thursday 14:00 UTC release catches a full European and American evening on day one, and the first weekend falls on days three and four.
There is no money in this issue. We estimate sales from reviews, and an unreleased game has none. A first estimate is possible no earlier than a week after launch — and only with an interval.
08How we will check ourselves
Seven predictions recorded on 21 September at 09:05 UTC. We will check them in the next issue, misses included — just as we checked The Sorting Bureau forecast.
- 24 September, 14:00 UTC In the final 24 hours before launch Resonant climbs to #1–3 on US Top SellersAll six cohort games did.
- 26 September, 14:00 UTC First-48-hour peak of 35,000–55,000 concurrent playersTwo anchors on six cohort games: a 54k median and 37k from followers (refined at 13:30 UTC).
- 26 September The share of positive Steam reviews is 78–85%: neither "masterpiece" (85% and up) nor "boring again" (below 75%)The press puts it level with the first game, but pans the combat, which is now the game's core.
- 26 September If the positive share over the first two days is 72% or higher, the day-two peak is at least 95% of day one0.97–1.10 for the four cohort games in that range; 0.68–0.73 for the two below 60%.
- 1 October "Boring" and "repetitive" will appear in more than a third of negative reviews of Resonant — more than for the first gameBoredom is what separates Remedy's unhappy players from happy ones, section 06.
- 1 October Day-seven retention matches its quality band: 80%+ positive — 0.68–0.74 of peak, 72–79% — 0.53–0.57, below 60% — about 0.35Six cohort games, section 06.
- 1 October The share of Russian-language reviews comes in well below 18%, the first game's figureThe game is not sold in Russia.
09Limitations
- Review themes are counted by words, not meaning. We searched for words and their forms in English and Russian, so "not boring at all" counts as "boring". For big games we analysed recent reviews: since 20 December 2025 for Control, since October 2025 for Alan Wake, since July 2023 for Quantum Break — these are today's players. Unhappy reviewers write half as long again, so any word shows up in their reviews more often; the conclusions were checked within buckets of equal length. The check by meaning (section 06) showed that the word list's errors squeeze the gap rather than inflate it: it misses a third of the complaints, and its false hits land mostly in positive reviews.
- The cohort boundary was chosen after we saw the outcomes. This is the benchmark's main weak spot; the details are in section 05.
- Six games is few. Every number derived from them should be read as approximate. The cohort covers only summer 2026; we did not observe the first Control or Firebreak at launch.
- Onimusha lifts the top of the cohort with China. 43% of its sampled reviews are in Chinese against 11% for the first Control; its 82,000 cannot be carried over to Resonant literally. The bias is upward.
- Top Sellers is one market. The US rank overstates the Western share of demand: in Japan the game is #127.
- Followers are as of 20 September. The anchor uses 48,000; at the current pace it will be about 51,000 by launch. The bias of the benchmark is downward.
- Russia. We observed Resonant's unavailability directly but did not check the cohort games. If they were sold in Russia, the benchmark is too high.
- The social accounts belong to the whole studio, so we barely use them here.
Disclaimer. The material on this page is informational and exploratory. These are observations from open sources and estimates built on them: samples are small, the method is imperfect, calculations are approximate, and forecasts and interpretations may turn out to be wrong. We do not observe sales or revenue. Nothing here is business, financial or investment advice; any decisions you make on this basis are your own.
Observations: Steam storefronts across 41 countries (the game is sold in 40), the upcoming, wishlist-activity and sales charts, community followers, concurrent players, the store listing and store page, the recommendation graph, curator reviews, prices and bundles, and the game's news. Data snapshot — 21 September 2026, 09:05 UTC; storefront ranks are live and have almost certainly moved by the time you read this. The comparison set is six releases from 26 July to 17 September 2026: priced at $39.99 or more, #60 or better on Top Wishlists in the final week, single-player; the first-48-hour peak is measured from the moment of launch. The follower history since the announcement is from SteamDB, checked against our own readings from 20 July. The launch time comes from Remedy's "Global Launch Times" post. Review quotes are from Rock, Paper, Shotgun and CGMagazine; the other quotes are from ComicBook, VICE and Empire, checked against the pages. The game page is CONTROL Resonant on Steam, and its news is in the game's feed. Press scores are from OpenCritic and Metacritic as of 21 September; player review themes come from 17,000 Steam reviews in English and Russian across five Remedy games in our database. Previous issues: The Sorting Bureau — also a forecast made before launch, and Valheim — the latest teardown.
The ten researches and datasets behind this issue were prepared on the PlayRecon v0.5-beta platform.
Figures from studio posts and third-party publications are cited with a link to the original and remain the property of their authors. If you quote or reuse this analysis, please link back to the original.