The Index

Every operator in the pay-to-rank wave publishes what their board claims. Nobody publishes what it earns. We have run a sensor on the category since it started — hourly board parses, a two-minute tap on the leader's own API, a category-level detector that finds new boards on its own, and twenty-minute decay curves on the second tier — so these are measured numbers, not reported ones.

Realized revenue

A blank is "the operator publishes no money field" — never $0. A dead board and an unreadable board must not look alike. Excluded from every figure on this page because we cannot source them: timebid.lol, millionpixelworld.com.

1 · The distribution is degenerate, not long-tailed

outbid.lol holds 91.2% of the $152,482.06 we can source across 12 instruments. An independent directory, leadingboards.lol, listed 52 boards in this category when we read it at 2026-09-20T15:10:00.397Z — so the 12 instruments above are the ones we can source, not the ones that exist. That is a directory's view rather than a census, and it is the denominator our concentration figure is missing. This is one winner, a thin middle, and a long tail we can only partly read.

91.2% is a ceiling, not a point estimate: $5,959.06 of that denominator is itself a lower bound, and if every bounded figure were understated 2× the leader share would be 87.77%. It cannot be much higher either — a 99% leader share is arithmetically impossible while second place alone is 5.30% of first. That one-line check is what caught our own earlier 99% figure.

2 · The discriminator is not novelty — it is WHEN the buyer gets value

Every instrument that earns lets the buyer act immediately and unilaterally, and books cash at the moment of payment. Every zero defers or conditions it.

lamborghini.lol displays $25,000 claimed and has realized $0. Its own FAQ: "Everyone gets a full refund, 6–8 weeks after the round closes." Payout requires a $200,000 cap to fill; it is stalled at 12.5% with one sponsor. lastspot.lol (values decay 5%/day) and thehundred.space (sequential fixed ladder) are both at 0/100 claimed, $0 earned.

3 · Concentration is a property of the mechanic, not of scale

pixelraid.lol: a single $626.80 raid is 86.7% of everything we have observed it earn (6 money events, $723.30) — the whale shape of the category leader, on a board ~192× smaller. vibewar.lol has 3× topchat.lol's participants and 11× its money. Participants and revenue are different axes. A board that lets one buyer take a large, visible, contestable position gets a whale; a board of $2 confessions structurally cannot.

4 · A mature instrument earns a floor punctuated by single buyers — and defending is free

Measured hour by hour from h+46 to h+63 of the leader's life (519 samples, 345 priced events; 5 rows carry no timestamp and are excluded, not defaulted):

Median hour $205.00, mean hour $856.65 — the mean is 4.2× the median. The biggest single hour (h+61) is 27.2% of the entire window, and $1,306.00 of that hour is one buyer (myworkoutlogs.com). Any mean $/hr quoted without naming that buyer is a statement about luck, not about the instrument.

Over the same window, $14,793.00 of new money arrived. Defensive re-bids were $3,500.00 — 23.7% of it, across 93 events, of which 79 of 93 were exactly $1.00. The longest of the 22 defence chains in the window, jagsim.com, paid 11 times over 77 minutes to climb 500 ranks for $31.00 of new money. A rank nobody contests costs nothing — and is worth what it costs.

Decay is peak → collapse → floor, not peak → zero: leader concurrency went 957 → 466 (-51.3%) → 466 (+0% off the floor). New paying participants import attention at roughly the rate old attention decays.

The leader's own revenue counter is not strictly monotone. 2 down-step(s), printed rather than smoothed: 2026-08-22T06:46:01.073Z $129072 → $129070 (-2); 2026-08-22T13:20:00.575Z $138058 → $137902 (-156). Small, but it means the authoritative figure above is the operator's number and not a law of nature.

5 · A deterministic ladder publishes its own revenue history

warmap.lol prices every steal at ceil(1.5× the last price) rounded up to the nearest dollar — verified on 109 of 110 owned countries against the live board. Each country also publishes its opening floor and its flip count, so replaying the ladder reconstructs every payment ever made on it: $1,962 against the operator's own published $2,034, a 3.5% gap fully explained by the ten buyers who chose to pay above the minimum. Two independent routes to one number is the strongest verification anywhere in this table.

It is not one board's quirk. pixelraid.lol is the same shape: claims at exactly 1.0000¢/cell across four independent events, raids at 1.7¢/cell. If you ship one of these, your lifetime revenue is a public fact — recomputable by a stranger in an afternoon. That should be a choice. Right now it is an accident.

6 · The picks-and-shovels layer has formed

A "kit" at $100. Boilerplate at $79 and $199. A directory of the clones. Productised entry means commoditised entry — a mid-to-late marker, not an early one.

Bounds, stated rather than buried

Every figure has a source row, or it is not a figure. Each number above prints its file, its row's timestamp and how old that row is. Anything we cannot source is named and excluded. This is a script, not a habit — it is how we found that our own published warmap figure had no sensor behind it at all.

Never sum an activity feed. The leader's feed publishes each entry's cumulative bid, not the amount paid; summing it inflates money 3.4×. New money is a per-entry delta keyed on the entry's URL — eight different advertisers on that board link to x.com. The identity test: a cumulative field must never decrease. Host → 5 decreases, name → 2, url → 0 across 163 events.

Lower bounds are labelled. pixelraid's floor omits 5 money events that predate our first sample; the completeness gap is measured, not guessed.

Our derivative count is a proxy. We see boards that advertised on their source: ~21% capture against an independent enumeration of ~85. The shape holds; the level is understated ~4.7×.

Visitor counters are rolling windows, not lifetime totals, on the leader and on its clones — so any per-visitor figure derived from them is overstated by an unknown factor. We publish none.

Where we have corrected ourselves

Every measurement defect we have found in our own sensor is published rather than quietly fixed — we do not have a verified count of them and will not publish one. The most useful: a clone we reported at $41 had actually earned $3,601 — the parser summed every dollar token on a page denominated in SOL, including advertiser ad copy. A concentration figure we published at 99% was really 94.9%; the error ran in our own favour, which is exactly the class that survives review. And the warmap figure in the table above was, twelve hours before this page existed, a number nobody had ever measured.

Most recently, our own detector published a landing-page figure as a board's lifetime total on 3 boards (outbidception.lol, outbid-directory.lol, charitybid.lol) when the number it had read was a single row — the top listing on one board, one charity on another. Those figures are withdrawn and annotated in place rather than deleted. The rule that separates the two needs no knowledge of any particular site: a lifetime total appears once on a page, and a per-row amount appears once per row, so two or more labelled amounts means we are looking at rows and the honest output is a refusal that states the count.

recomputed 2026-10-01T01:50:01.116Z  — every figure below has a source row

instrument                  realized  bound         row age  source
outbid.lol               $139,058.00  exact          56830m  pulse.jsonl
maxbid.lol                 $3,341.76  lower_bound       45m  maxbid.lol.jsonl
warmap.lol                 $7,373.00  exact             10m  tier2.jsonl
                           cross-check: independently rebuilt at $6,830.00 (7.36% below the operator figure, 47 overbids / 9 moved floors)
pixelraid.lol                $723.30  lower_bound       10m  tier2.jsonl
                           MISSED 5 of 11 money events (they predate our first sample) — cumulative_minor is a floor
vibewar.lol                $1,894.00  lower_bound       10m  tier2.jsonl
topchat.lol                   $51.00  exact             10m  tier2.jsonl
                           cross-check: operator total agrees with our own row sum to the cent
timebid.lol                        —  —                 10m  tier2.jsonl  << NO MONEY RAIL FOR PARTICIPANTS — rank is bought with seconds of attention
millionpixelworld.com              —  —                 10m  tier2.jsonl  << no money field exposed; pixelsClaimed is the only quantity published
lamborghini.lol                $0.00  exact             45m  lamborghini.lol.jsonl
outbidception.lol              $1.00  exact             65m  instruments.jsonl
                           operator text: "cement TopShopifyApp $6.01 new placement outbidception.lol +$1 raised to $6 sponsors · rotating become one →"
afternic.lol                   $5.00  exact          56945m  instruments.jsonl
                           operator text: "rs powered by datafast visitors all time 204 listings 1 bid revenue $5 top bid $5 1 domains for sale Include s"
outbidcreators.lol            $11.00  exact             65m  instruments.jsonl
                           operator text: ". Be the first creator on the board. This bidding board has raised $ 11 since launch Outbid Creators is a bit o"
feedbid.lol                   $21.00  exact          56991m  instruments.jsonl
                           operator text: "nk for $2 11 videos on the board Refresh feedbid in numbers $21 paid to rank 11 videos on the board 3,462 cl"
youbid.lol                     $3.00  exact          56991m  instruments.jsonl
                           operator text: "uFl2k 1 day ago • on the board until 2026-09-26 • 21 clicks $3 Paid placement, ranked by current verified b"

concentration: outbid.lol holds 91.2% of $152,482.06 across 12 sourced instruments
  $5,959.06 of the denominator is a LOWER BOUND, so 91.2% is a CEILING (double every bound and it is 87.77%)
  UNSOURCED, excluded from every figure above: timebid.lol, millionpixelworld.com

now 2026-10-01T01:50:01.412Z

host                  launch (UTC)            age       basis
outbid.lol            2026-08-19T23:08:00Z    h+1010.7  operator-stated
maxbid.lol            2026-08-21T14:47:00Z    h+971.1   operator-stated
topchat.lol           2026-08-19T22:01:59Z    h+1011.8  earliest row (age is a floor, $/hr a ceiling)
vibewar.lol           2026-08-21T09:14:00Z    h+976.6   earliest row (age is a floor, $/hr a ceiling)
pixelraid.lol         2026-08-21T13:26:51Z    h+972.4   earliest row (age is a floor, $/hr a ceiling)
lamborghini.lol       2026-08-21T03:00:00Z    h+982.8   estimated
warmap.lol            2026-08-21T10:53:28Z    h+974.9   earliest current-owner claim (age is a floor, $/hr a ceiling)

pending milestones (absolute — do NOT resolve a prediction before its timestamp):
  outbid.lol h+96         2026-08-23T23:08:00.000Z   SETTLED: VOID — outbid.lol unobserved since 2026-08-22T14:05Z (429, breaker L5); P30/P31 unresolvable
  outbid.lol h+168        2026-08-26T23:08:00.000Z   SETTLED: P32 NOT FALSIFIED on 5 sensored derivatives (max warmap $6,839 exact); P33 VOID (no sensor on lastspot/thehundred)
  maxbid.lol h+72         2026-08-24T14:47:00.000Z   SETTLED: P13 v2 VOID — bar invalid per C-024 (board sum is not lifetime gross); 28.32 SOL at 08-26
  P36 anchor holds (>=60% top / <3% defence)  2026-08-23T12:00:00.000Z   SETTLED: VOID — no outbid.lol events after 2026-08-22T14:34Z
  P37 largest payment > 100x floor            2026-08-23T23:08:00.000Z   SETTLED: VOID — no outbid.lol events after 2026-08-22T14:34Z