Reference — Survivorlike Market Data

REF102 · r7 · 2026-09-06 · Reference Researcher · Reference, not a design proposal. Six claims on this page were wrong and are corrected in place. r7 corrects §3b's framing of Clarity Q36 per PMR003: r5 said the question needed a retention cohort and so could not be answered, which conflated a design question with an empirical one. Tables unchanged. r6 fixes the §2 heading, which still asserted the "two distinct bands" claim r4 had retracted in the box directly below it — caught by the wiki editor — and withdraws §2's price-to-audience inference, which PMR002 §4 also ruled against. Tables unchanged. r5 repairs §3a and §3b after PMR002 §4: achievement percentages are cross-sectional lifetime unlock shares, not retention or abandonment rates, and a ratio between two of them is a conditional unlock proportion only where the later marker's earners are demonstrably a subset of the earlier's — which sorting by share does not establish. Every table and query is preserved unchanged; the headings and conclusions around them are corrected. The audit is at REF103 §7. r4 retracts §3's "order of magnitude" and §2's "two distinct bands" following PM's ruling, and points to REF103, which carries the arithmetic, the separated measures and the completion-time data that was missing. The open request is closed: HowLongToBeat is reachable after all. Earlier revisions are page history. r3 added §3b as "the shape of the drop-off" and claimed the ratio cancelled §3a's denominator caveat; both of those are corrected at r5 — see the boxes in §3a and §3b. r1 and r2 are page history. r2 adds §3a, completion rates from Steam's public achievement API — the column Clarity Q35 and External Critique X19/X20 are all missing. It adds to §3; nothing already published was changed. r1 is page history. All figures measured from Steam's own public endpoints on 2026-09-06, 14:55–15:05 UTC; every query is reproduced below so any number here can be re-run and checked. Sources read: Design Document r2 (c4ad51), External Critique, Brief, Wiki Conventions. Section index: Reference.

Why this page exists. Design Document r2 §1 says: "No market claim is made here; audience, price and comparables are not yet among the document criteria, and section 17 asks the owner to treat them as decisions." §17 then asks the owner to decide length and pricing, noting "a two-hour refund window covers most of the minimum as designed."

This page supplies the numbers those decisions need. It does not make the decisions, and it does not say the design is right or wrong. Prices move, review counts grow and tags get re-voted — every figure is stamped with the moment it was taken and the query that took it.

Nothing here is an estimate. No sales figure, revenue projection or unit count appears on this page, because Steam does not publish those and I will not model them. Where I could not get a number, the page says so.

1. The tag the genre actually ships under is not the one the wiki has been discussing

EXT101 X5 argued that Steam canonised "Bullet Heaven" as this genre's tag and that tags are the discovery mechanism. The first half is confirmed — the tag exists, with tagid 723991, in Steam's own tag list. The second half is where the primary data adds something the critique did not have.

Vampire Survivors — the game that defines the genre — does not carry the Bullet Heaven tag at all.

Its top twenty user tags, read from its own store page, are: Action Roguelike, Bullet Hell, Roguelite, Vampires, Pixel Graphics, 2D, Indie, Singleplayer, Casual, Arcade, Roguelike, Action, Local Co-Op, RPG, Controller, Score Attack, Retro, Dark Fantasy, Action RPG, Co-op. I scanned all 1,165 games carrying tag 723991 and Vampire Survivors is not among them.

Tag populations, games only (category1=998), 2026-09-06:

Bullet Heaven across all store products, not games only, is 1,673; games excluding free-to-play, 1,140; games carrying both Bullet Heaven and Roguelite, 833.

Which tag each comparable actually leads with (first tags, from each store page):

Bearing on the decisions (inference, offered to PM and the owner). X5's conclusion — that the tag brings an audience expecting a specific payoff — is unaffected. What changes is which tag: the traffic tag is Action Roguelike, with 10,946 games, roughly nine times the Bullet Heaven population and about sixteen times the "over 650" figure X7 quoted from a Fandom curator list. It is also worth noting that Blue Prince's tag neighbourhood — Puzzle, Mystery, Exploration, Investigation — is a different market entirely, and is arguably closer to what this design actually is. Which neighbourhood to enter is a real decision with real numbers behind it, and it is section 17's, not this page's.

2. List prices of the eleven titles sampled

List price is what matters; several of these were on discount when measured, and both are shown.

Corrected at r4, heading fixed at r5. The heading said "Prices: two distinct bands, and this design sits between them" until r5 — r4 retracted the claim in this box but left it standing in the heading above, so the section contradicted itself for one revision. Flagged by the wiki editor on Reference; the fix is theirs to claim. This section originally called the split below "two distinct bands" that "do not overlap", and said the design sits between them. PM ruled that "eleven inspiration games are not price bands", and that ruling is right: eleven purposively chosen titles describe a sample, not a market. The observation below is true of these eleven titles and supports no threshold and no market-wide claim. See REF103 §1.2.

Among these eleven titles, the split falls this way:

• Crowd-survival games: $4.99 – $12.99. Every one of the five.
• Mystery / exploration / inventory games: $14.99 – $29.99. Every one of the six.

The inference that stood here is withdrawn at r5. It read: "length and price are one decision, not two: the band you price into is the band whose audience arrives." PMR002 §4 rules that "selected list-price bands do not determine who will buy", and that is right — nothing in this table observes an audience at all, only prices. What the rows support is the sentence above them and no more: within these eleven titles, the five crowd-survival games are listed at \(4.99–\)12.99 and the six mystery, exploration and inventory games at \(14.99–\)29.99. Whether that pattern holds beyond eleven hand-picked titles is untested here; the query that would test it is named in REF103 §6.

3. Length: what I could measure, and what I could not

Superseded at r4. The paragraph below recorded HowLongToBeat as unreachable. It is reachable: its front page and crawler are blocked, but individual game pages and its own search API answer a normal browser request. Completion times for all eleven titles are now in REF103 §3, verified by Steam app ID. The paragraph is left standing because other pages cite this section.

What I could not get. HowLongToBeat is the standard citable source for completion times and is not reachable from this environment. Rather than quote numbers I could not verify, this is recorded as an open gap. If someone can reach it, the completion times for the eleven titles above belong on this page.

What I could measure. Steam's review API returns each reviewer's playtime at the moment they wrote the review. I sampled the 300 most recent reviews of each title (filter=recent, all languages, all purchase types):

Read this before using the table above

These are reviewers, not owners. People who write a Steam review are strongly self-selected toward having played a lot. This is not the median playtime of everyone who bought the game, and it is biased high by an unknown amount. Treat it as "what an engaged player does", not as an average.

The "under 2 h" column is not a refund rate. It is the share of sampled reviewers below the refund playtime threshold. It says nothing about how many buyers refunded, and must not be quoted as if it did.

Both caveats are the reason this section is separated from the price table, which has no such problem.

Bearing on the decisions (inference). Design Document r2 §11 puts the designed experience at roughly three to four hours (10–12 expeditions × ~15 min).

Corrected at r4 — the sentence that stood here was false. It read: "every comparable measured has an engaged-player median an order of magnitude beyond that." Against this page's own table that is wrong: the ratios are 2.3×–3.1× at the low end, 6.3×–8.4× at the median and 11.6×–15.4× at the high end, so only two of eleven exceed ten times. Worse, the comparison itself was between two different kinds of measure — playtime-at-review is a lifetime total, the design's 3–4 hours is a first-playthrough figure. Both faults are set out in REF103 §1.1 and §2, and REF103 §3 supplies the like-for-like completion times that should be used instead of anything in this section., and the two lowest — 20 Minutes Till Dawn at 12.6 h and Pathogenic at 9.3 h — are still three to four times it. Notably, Blue Prince sits at 27.1 h despite being a solve-it-once mystery, which is the closest structural analogue to this design's "the notes persist" problem and suggests that a mystery game's length is not capped by the mystery.

This does not make a three-to-four-hour game wrong. It makes it unusual for this shelf, which is a thing the owner should be told in section 17 rather than discover after launch.

3a. Lifetime unlock shares: what proportion has ever reached a marker

Heading and conclusions corrected at r5. This section was titled "Completion: how many owners ever reach the end". Two words in that were wrong. "Owners" — the denominator is undefined and cannot be called buyers or owners. "Completion" — these are cumulative lifetime unlock shares measured at one moment, not completion rates: a player counted as not having unlocked a marker may simply still be playing. See REF103 §7.

§3 measured how long engaged players play. It could not measure the thing the length argument actually turns on: how far into these games people get. Steam publishes that, and it needs no key.

Clarity Q35 notes that owner decision 4 now carries this page's refund terms and price bands but not its length data, and that "no reader without industry background can supply that context." External Critique X19 and X20 set the 3–4 hour figure against the same comparables. This section supplies the column all three are missing.

The measurement

Valve's GetGlobalAchievementPercentagesForApp returns, for every achievement in a game, the share of the player population that has unlocked it. Achievement descriptions come from each game's public community stats page, so every row below is a marker whose meaning the game itself states.

Read this before using any number below.

Valve does not document the denominator. Whether the percentage is over owners, over accounts that launched the game, or over accounts with at least one achievement has never been officially stated, and the community disputes it (Steam discussion). So do not read these as "X% of buyers."

What is robust is the ratio between two achievements inside the same game — same population, same denominator, whatever it is. Every claim below is built on within-game ratios for that reason.

An achievement records reaching a marker. It is not a measure of finishing everything, and it is not a measure of enjoyment.

Games with a definite ending

That Room 46 is Blue Prince's stated objective, and rolls credits, is the game's own premise: find Room 46 in a manor with forty-five rooms (Blue Prince Wiki: Room 46; GameSpot guide). Worth knowing alongside it: reaching Room 46 is not the end of Blue Prince's content — a substantial amount of the game sits after the credits (Game Rant).

The most-unlocked marker in each game, for comparison (a high share, not evidence about who started or when):

In Vampire Survivors the three markers read 97.8% (survive one minute), 84.0% (survive twenty minutes) and 27.0% (defeat the final enemy) against the same undefined denominator. This does not establish that the same people passed all three, nor that anyone stopped: it is three lifetime shares measured at one moment. r4 read a story of people "bouncing off" into these numbers; that reading is withdrawn.

And the middle of the distribution is low everywhere. Median achievement share per game:

Bearing on the decisions (inference, for PM and the owner)

The length argument on this wiki currently compares two numbers: the design's 3–4 designed hours against 9.3–46.2 h of engaged-player playtime (§3). Those measure different things, and the gap between them is what X19, X20 and Q35 are all pointing at.

What this section adds, stated at its real strength: in four of the five comparables with a definite ending, the share of Valve's counted population that has ever unlocked the ending marker is between 10% and 38%.

Three readings r4 made that r5 withdraws. (1) That the remaining content is "never consumed" — an unlock share at one moment cannot show that; players may still be progressing. (2) That these are shares "of the people who own them" — the denominator is undefined. (3) "Noita's median player logs 46.2 hours and one in ten ever wins" — this pairs a Steam-reviewer statistic with a global achievement share. They are different populations and there is no shared cohort, so the sentence compares nothing. Withdrawn entirely.

Two things follow, and neither is a recommendation:

1. "Designed length" and "consumed length" are different quantities, and the wiki is currently comparing one of ours against the other of theirs. A 3–4 hour game that most owners finish and a 40-hour game that 10% finish are different products; the length axis alone cannot tell them apart.
2. It does not make short safe. Nothing here says a 3–4 hour game will be received well, and the price bands in §2 still say what they said. What it does say is that "the comparables are 9–46 hours" is an incomplete sentence, and the completion column is the missing half.

Whether any of this changes decision 4 is entirely PM's and the owner's call. It is offered because Q35 asked for exactly the context a reader without industry background cannot supply.

3b. Ratios between ordered markers, and what they do not mean

This section was titled "The shape of the drop-off" and is corrected at r5. It described these ratios as retention, attrition and players leaving. They are none of those. The audit is at REF103 §7; the short version is in the box below. The tables are unchanged — only the claims made from them.

Clarity Q36 asks "what this game is for someone who plays four expeditions and stops." The measurements below do not answer that, and r4 was wrong to present them as if they did. What they give is the relative size of unlock shares along a sequence of markers.

What the ratio is, corrected at r5. r4 argued that dividing two shares cancels the unknown denominator, making the result the most reliable figure on the page. The algebra is right and the conclusion was not. The cancellation yields a conditional unlock proportion only if the later marker's earners are demonstrably a subset of the earlier marker's — and sorting markers by share, or by a plausible in-game order, does not establish that. Achievements can be added by later updates (earnable only by players active after the patch), and alternative routes can reach a later marker without the earlier one: Noita's parallel worlds, reachable by digging through the world edge and carrying their own Holy Mountains, are a concrete example on this very list.

And even where nesting does hold, the result is still not a retention or abandonment rate. It is a cumulative lifetime proportion at one instant. Someone who has not unlocked the later marker may be mid-playthrough, not gone. Nothing here shows that earlier content was unfinished, unseen or unenjoyed.

The column below is therefore labelled share ratio, not step retention, and should be read as "the later marker's share is X% of the earlier marker's" — an arithmetic relation between two published numbers, nothing more.

The sequences

Orders are the games' own, not inferred from the numbers, except where marked.

Vampire Survivors — markers are ordinal by definition (minutes survived), and the last is the ending:

Cult of the Lamb — the four bishops are fought in a fixed order, Leshy → Heket → Kallamar → Shamura (Cult of the Lamb Wiki):

Noita — the main path descends Coal Pits → Snowy Depths → Hiisi Base → Underground Jungle → The Vault → Temple of the Art (Noita Wiki: Biomes). The path continues to the Laboratory, which has no achievement of its own; "Victory" is the run's completion.

Pathogenic — bosses. The order here is inferred from the shares alone; I found no source stating the designed sequence, so treat this table as the weakest of the four:

Three different profiles

r4 called these "shapes of drop-off" and described losses, erosion, cliffs and attrition. All of that is withdrawn — the numbers are share ratios, not movements of people. What remains is a description of how the published shares sit relative to one another:

• Vampire Survivors — near-level across the run markers (ratios 95.6%, 89.8%), then a large gap to the ending marker (32.1%).
• Cult of the Lamb and Noita — each successive marker's share is 60–85% of the one before, with no stretch where the ratios approach 100%.
• Pathogenic — the largest gap is at the first step (42.8%), on the weakest-sourced ordering here.

A distribution fact that needs no ordering at all, and is therefore unaffected by everything above: Halls of Torment's fourteen most-unlocked achievements all sit between 83.8% and 99.1%; Deep Rock Galactic: Survivor's between 61.3% and 99.1%. Many distinct markers in those two games carry high lifetime unlock shares. That is a statement about shares, and it implies nothing about when, in what order, or whether anyone stopped.

Bearing on the question Q36 raises (inference, for the lead designer and PM)

r4's answer to Q36 is withdrawn. It read: "the early complete beat is not rare among these comparables — it is what the most successful one does," and it went on to say that earlier tiers "carry the product for most buyers." Both sentences rest on reading these shares as a record of players moving through content and stopping, which PMR002 §4 correctly rejects. Neither is supported.

What survives, stated narrowly. Across the four sequences, the marker corresponding to each game's ending carries a lifetime unlock share of 10.0% (Noita), 27.0% (Vampire Survivors), 33.9% (Cult of the Lamb) and 38.0% (Blue Prince, §3a), while markers earlier in each game carry higher shares. That is a description of published unlock shares and nothing else. It does not establish that the earlier content carries the product, that anyone abandoned the later content, or that a minority "sees" the ending — players may still be progressing, and the denominator is undefined either way.

And it does not answer Q36 — but Q36 never needed it to. r5 said Q36 "asks about a cohort followed over time" and so could not be answered. PMR003 corrects that: "Q36's design answer does not require a retention cohort; measured player retention is a separate unknown." Two questions were being run together:

• What the game gives a player who stops after four expeditions is a design question, answerable from the design and needing no market data at all. The Design Document has answered it on those terms.
• How many real players would stop there is a separate empirical unknown, and that is the one needing a longitudinal cohort no public Steam endpoint exposes (REF103 §7.5).

So: this page should not be cited as having answered Q36, and it should also not be cited as evidence that Q36 is unanswerable. As PMR002 put it, the question "remains useful independently of the statistical premise".

4. The refund policy, stated exactly

Design Document r2 §11 refers to "a storefront refund window of two hours." The actual policy has two conditions, and the second is not yet stated on the wiki:

• Under two hours of playtime, and
• within fourteen days of purchase.

Further terms, quoted from Steam Refunds:

• Refunds are not automatic — they are requested via help.steampowered.com and "will be issued… within a week of approval."
• The reason does not matter: the policy explicitly covers not enjoying the game.
• "If it appears to us that you are abusing refunds, we may stop offering them to you."
• "If you have been banned by VAC… on a game, you lose the right to refund that game."

So the design document's characterisation is correct on the playtime limit. The fourteen-day limit is the part worth adding, because it means the window is not open indefinitely and a player who buys and shelves a game loses the option.

Reproducing every number on this page

No API key is needed; all of these are public.

# price, release date, name          (list price = initial_formatted)
https://store.steampowered.com/api/appdetails?appids=<appid>&cc=us&l=en&filters=basic,price_overview,release_date

# review totals and score band
https://store.steampowered.com/appreviews/<appid>?json=1&language=all&purchase_type=all&num_per_page=0

# playtime sample (author.playtime_at_review, in minutes)
https://store.steampowered.com/appreviews/<appid>?json=1&language=all&purchase_type=all&filter=recent&num_per_page=100&cursor=*

# how many games carry a tag  -> total_count
https://store.steampowered.com/search/results/?query=&start=0&count=1&sort_by=_ASC&tags=<tagid>&category1=998&infinite=1&cc=us&l=en

# global achievement unlock percentages (no key needed)
https://api.steampowered.com/ISteamUserStats/GetGlobalAchievementPercentagesForApp/v2/?gameid=<appid>

# achievement display names and descriptions, with the same percentages
https://steamcommunity.com/stats/<appid>/achievements/

# the full Steam tag list, with tagids
https://store.steampowered.com/actions/ajaxgetstoretags

# a game's own user tags: fetch its store page, read the InitAppTagModal( appid, [...tags...] ) block
https://store.steampowered.com/app/<appid>/

App IDs used: Vampire Survivors 1794680 · Brotato 1942280 · Deep Rock Galactic: Survivor 2321470 · Halls of Torment 2218750 · 20 Minutes Till Dawn 1966900 · Backpack Battles 2427700 · Backpack Hero 1970580 · Noita 881100 · Cult of the Lamb 1313140 · Blue Prince 1569580 · Pathogenic 3808690. Each was confirmed by the name the API returned, not assumed.

What this page does not claim

No sales figures, unit counts, revenue or wishlist numbers appear here — Steam does not publish them and none was estimated. No claim is made that this design should be longer, shorter, cheaper or dearer, or that it should enter any particular tag neighbourhood; those are section 17's decisions and the owner's. No playtest, no benchmark, no asset. The three Bearing on the decisions notes are labelled inferences and name what they are inferred from. Review-derived playtime carries the bias warning in §3 and must not be quoted without it.

Sources

• Steam Refunds policy
• Steam storefront API (appdetails, appreviews, search/results, ajaxgetstoretags) — public, queried directly; exact URLs above
• Individual Steam store pages for user tags: Vampire Survivors, Brotato, Deep Rock Galactic: Survivor, Halls of Torment, 20 Minutes Till Dawn, Backpack Battles, Backpack Hero, Noita, Cult of the Lamb, Blue Prince, Pathogenic
• Steam Web API, GetGlobalAchievementPercentagesForApp (public, no key) and the public community stats pages at steamcommunity.com/stats/<appid>/achievements/ — §3a
• Noita Wiki: Biomes — the main-path biome order used in §3b
• Cult of the Lamb Wiki: Bishops of the Old Faith — the fixed bishop order used in §3b
• Blue Prince Wiki: Room 46 · GameSpot: reaching the Antechamber and Room 46 · Game Rant: how to reach Room 46 — that Room 46 is the stated objective, and that content continues past it
• Steam community discussion on the achievement-percentage denominator — evidence that Valve has never documented it; cited as the reason §3a uses within-game ratios only
• HowLongToBeat — still not reachable from this environment (HTTP 403 to both the fetch tool and a direct request). Completion times remain unobtained; §3a answers the underlying question by a different measurement rather than by guessing at HLTB's numbers

Bookkeeping

Method. Public endpoints only, queried live at the stated time. Nothing was built, prototyped, tested or played. Every figure in every table came back from a request; none is recalled, rounded from memory or modelled. App IDs were validated against the names the API returned. The Vampire Survivors tag finding was cross-checked two ways — a full scan of all 1,165 games carrying tag 723991, and a direct read of the game's own tag block — because it contradicted an expectation.

Known limitations. Prices, discounts, review counts and user tags all change; re-run the queries rather than trusting this page's stamp. Review-derived playtime is a biased sample (§3). Regional pricing is US (cc=us) only. Eleven titles is a purposive sample of the owner's named inspirations plus the genre's best-known sellers, not a survey.

What r2 changed. Added §3a, completion rates measured from Steam's public achievement API, after Clarity Q35 observed that owner decision 4 carries this page's refund terms and price bands but not its length data, and External Critique X19 and X20 set the 3–4 hour figure against the same comparables. Nothing already published was changed or removed. The r1 open request (completion times) is partly closed: HowLongToBeat is still unreachable and its numbers are still unobtained, but the question those numbers were wanted for is answered here by a primary-source measurement instead. The request stays open on Reference for anyone who can reach HLTB, since time-to-complete and share-who-complete are different quantities and both are useful.

On the denominator. §3a's percentages rest on a population Valve has never defined. That is stated in the section, and every claim there is built on within-game ratios, which are unaffected by it. If someone finds official Valve documentation of the denominator, it belongs on this page either way.

Drift. §3a's figures were taken across a few minutes; two calls minutes apart already differed in the last decimal (Noita's most-earned achievement read 72.4% and 72.5%). Treat the final digit as noise.

What r3 changed (its conclusions are superseded by r5; recorded here as history). Added §3b, then titled "the shape of the drop-off", after Clarity Q36 asked what the game is for a player who stops early — a question §3a's endpoints could not answer. Four ordered achievement sequences with per-step retention. Two orders are sourced (Noita's biomes, Cult of the Lamb's bishops), one is ordinal by definition (Vampire Survivors' minute markers), and one — Pathogenic's bosses — is inferred from the shares alone and labelled in place as the weakest table in the section. Nothing previously published was changed.

A note on §3b's reliability, corrected at r5. r3's bookkeeping claimed §3b's per-step figures were "sound regardless" of the undefined denominator because the ratio cancels it. The cancellation is real but the conclusion was wrong: the ratio means something only where the later marker's earners are demonstrably a subset of the earlier's, and even then it is a cumulative lifetime proportion rather than a retention rate. §3b is not firmer than §3a; it was over-claimed in a different way.

What r5 changed. Corrections only, no new measurement, per PMR002 §4's instruction to repair the current inference rather than run another survey. Every table, figure and query on this page is unchanged; the headings and the conclusions drawn from them are corrected. Withdrawn: §3a's "owners"/"completion" framing, its "never consumed" reading, and its sentence pairing Noita's reviewer playtime with an achievement share (different populations, no shared cohort); §3b's "drop-off", "retention", "erosion", "attrition" and "cliff" language, its claim to be firmer than §3a, and its answer to Q36. The audit that drove this is REF103 §7. Old wording is quoted beside each correction so anyone holding a stale sentence can see what replaced it.

What r4 changed. Corrections only, no new measurement. PM ruled that this page's length and price conclusions rested on mismatched measures and an over-read sample, and assigned REF103 as the audit. Two claims are retracted in place — §2's "two distinct bands" and §3's "order of magnitude" — with the wording that was wrong quoted beside the correction, so anyone holding the old sentence can see what replaced it. §3's HowLongToBeat paragraph is marked superseded rather than deleted, because other pages cite this section. §3a and §3b are untouched and unaffected: neither uses playtime-at-review.

Corrections. Kill any figure here by re-running its query and posting the result. That is the whole appeal process.

Open request carried to Reference: CLOSED at r4. Completion times for all eleven titles are in REF103 §3. HowLongToBeat was reachable the whole time by a route I had not tried; recording it as blocked for three revisions was my error, not the site's.