Unbroken Receipts
Conviction, on chain, with proof
An audit of the Solana trenches · June 2026

Until There Are More Unbroken Hawks

Why we built a machine that audits the trenches instead of selling them.

0.43%
1,380 wallets examined · 6 real winners found
Published by the $UNBROKEN community · Data: Helius, Birdeye, DexScreener, Rugcheck, Dune Analytics

There is a story this market tells about itself, and it goes like this: somewhere out there, a regular person put grocery money into a meme coin, held it through hell while everyone laughed, and came out the other side with their life changed. That story is the entire emotional engine of the trenches. It is why people show up. It is what every influencer, every launchpad, every "next 100x" thread is actually selling.

We wanted to know if the story was still true. So we built a machine to find out. Not by asking anyone. By reading the chain itself.

The machineWhat it does

Unbroken Hawks is an on-chain conviction scanner for Solana. It starts from coins that already proved they were not rugs: thirty days alive, real liquidity, authorities revoked, LP locked, real holder counts. Time itself filters the scams, because a rug, by definition, is dead within the month. Then, for each surviving coin, it reconstructs the complete trading ledger of its wallets. Every buy and every sell, priced in dollars, including the messy parts most tools skip.

Then it asks one question, the only one that matters: who actually lived the story? Who bought with relatable money, not a whale, not a bot in the launch block, not the dev's cousin wallet. Who went genuinely underwater on their own cost basis by 30, 50, 90 percent. Who held through the shakeout, and then sold that same coin for real, realized profit.

Not paper gains. Not "still holding and praying." Closed the trade. Won.

We call that wallet an Unbroken Hawk, and every filter in the machine exists to protect the meaning of the name. Snipers die by timing analysis. Insiders die by funding traces back to the creator. Burner wallets die by age checks. Wallets whose tokens appeared out of nowhere, from airdrops and transfers with no verifiable cost basis, die by ledger integrity checks. The machine would rather find nothing than lie to you.

The landscapeWhat the trenches actually are

Before the machine ran, the public data already sketched the shape of the problem. An academic study of a single month of pump.fun activity counted 655,770 tokens created. Of those, 4,338 reached graduation, roughly 0.63 percent[1]. And graduation is not survival: of the tokens that do graduate, very few trade for more than a month, and many end their arc within a day[2].

Exhibit AarXiv study, Sept 2025 · CryptoRank
The death funnel: one month of pump.fun
655,770
Tokens createdin 30 days, by 243,123 creator wallets
4,338
Graduated (0.63%)survived the bonding curve to reach a real DEX
"very few"
Alive after 30 daysmost graduates end their trading arc within days
99.4% of launched tokens never make it out of the bonding curve. Survival past one month is the rare event our qualifier selects for. Sources [1], [2].

The wallet side is harsher. Across the platform's life, out of 13.55 million wallet addresses, only 55,296 ever realized more than $10,000 in profit. About 0.048 percent cleared $100,000. Roughly 293 wallets, 0.002 percent, ever realized more than a million dollars[3].

Exhibit BDune Analytics via Yahoo Finance
All-time realized profit, pump.fun wallets
All wallets ever13,550,000
Realized over $10K55,296 · 0.4%
Realized over $100K~6,500 · 0.048%
Realized over $1M293 · 0.002%
Bar lengths are square-root scaled for visibility; true linear bars for the bottom three rows would be invisible at this ratio. That is the point. Source [3].

In an ordinary month it is worse than the lifetime numbers suggest. In March 2026, roughly 96 percent of wallets trading pump.fun tokens either lost money or made less than $500, with just over half posting an outright loss[4].

Exhibit CDune Analytics via Protos, March 2026
One month of trader outcomes
50.6% LOST
45.6% MADE <$500
Posted a loss Profit under $500 3.8% made more than $500
Two wallets cleared $1M that month. 671,376 wallets lost under $500 each. The house, meanwhile, has collected over $500M in fees since 2024. Sources [4], [5].

And the human toll is measurable. Between May and December of 2025, the platform lost roughly 65 percent of its active participants, from 5.2 million wallets to 1.8 million. A near-total flush of the retail cohort, most of whom experienced losses severe enough that they simply stopped showing up[6].

The honest counterweight Early 2026 actually improved: by April, around 70 percent of active traders posted a positive month, the best stretch ever recorded[6]. But look inside the number. The largest group of winners, about 65 percent of all active wallets, made between $1 and $500. The market got better at paying out lunch money to the survivors of the flush. The life-changing outcome, the one the entire culture advertises, is a different species. That species is what we went hunting for.

The huntWhat the machine found

We pointed the scanner at the healthiest part of the casino: every coin we could find that passed the survival test. Twenty-two qualified. Across them, the machine examined 1,380 top wallets, rebuilt their ledgers, and ran the kill filters.

Exhibit DUnbroken Hawks scanner · June 12, 2026
The Receipt Rate
22
Coins qualified30+ days alive, real liquidity, locked LP, authorities revoked, real holders
1,380
Top wallets examinedfull ledger reconstruction: every buy, every sell, in USD
1,374
Rejectedsnipers, dev-funded insiders, burners, whales, unverifiable ledgers, paper hands, paper gains
6
Unbroken Hawks confirmedrelatable entry · deep drawdown survived · sold the same coin green
0.43%
Receipt rate
99.57%
Everything else
2x / wk
Scans, forever
Live results, methodology, and every verified receipt: unbroken-hawks.vercel.app. The scanner reruns automatically; this number updates whether it flatters anyone or not.

Sit with that. In the best coins the trenches currently produce, the story this entire culture is built on, the one being sold to you daily in screenshot form, exists at a rate of less than half of one percent.

We also pointed it at the blue chips, the all-time legends everyone assumes are full of glory stories. The holder scans came back empty. Not because nobody ever won, but because the winners who sold are no longer holders, and the wallets still sitting there are either whales with bottomless histories or bags that never closed. We are building cycle-tracing to chase even those ghosts, because the number deserves to be exact.

The technology did not fail. The door is real. The crime ate the culture standing in front of it.

The disclosureThe math nobody says out loud

Part of the audit is showing how the sausage is made. We published the bonding curve mathematics behind every fresh launch, live and interactive. The short version: a pump.fun coin "graduating," the moment everyone celebrates, takes roughly $12,000 of real inflow to manufacture an on-paper "market cap" of about $60,000. The first $1,000 into a fresh coin can capture roughly 15 percent of the entire supply[7]. Market cap is a story. Inflow is a fact. The gap between them is where most of the trenches' pain lives.

And we made the machine a mirror. Drop any wallet into the profiler and the chain calls it what it is, by behavior, not by bio: chad, OG, jeet, bot, dev, extractor, whale, exit liquidity[7].

The thesisTwo ways to read 0.43%

The cynical read is available to anyone: the trenches are a casino where the marketed outcome is statistically near-extinct, where 13 million wallets have churned through and 0.002 percent ever cleared a million, where in an ordinary month half of everyone loses, and where the house collects regardless[3][5]. That read is true. We publish the evidence for it ourselves.

But there is a second read, and it is the one we are building on: six is not zero.

The story is rare, not dead. The door, a market where $200 and conviction can still genuinely change a regular person's life, with no accredited investor gate, no broker, no permission slip, is still open. It exists nowhere else in the financial world. It has just been buried under bundles, snipers, vamp launches, and recycled tickers, while everyone profiting from the burial keeps selling the dream so you keep digging in the wrong spot.

The dollar itself is a belief system with a printer attached. Brands, flags, markets: belief systems that survived. The question was never whether a meme can hold value. The question is whether the belief survives pain, and whether the game around it is honest enough for surviving to matter.

The missionWhy a token publishes an audit

$UNBROKEN's brand was always conviction surviving pain. A real hawk, a real arrow, still flying. The scanner turns the slogan into an instrument. The community does not have to claim the trenches can be real. It publishes the receipts, on a timer, forever.

The scanner runs itself twice a week. Every Hawk it confirms gets a verified receipt card: money in, drawdown survived, days held, profit out. Every wallet anyone profiles gets judged by its behavior, not its bio. The Receipt Rate, coins scanned, wallets examined, Hawks found, becomes a public metric for whether this market is healing or rotting.

We are not selling hope for engagement. Hope-as-product is exactly what produced the 0.43 percent. We are doing the opposite: proving, in public, with falsifiable on-chain evidence, exactly how rare the real thing is, and existing as the community that wants it common again.

Until there are more Unbroken Hawks, the trenches don't deserve the door. We're building the reason they should keep it.
Until death, all defeat is psychological.
View live receipts Profile a wallet $UNBROKEN

Sources

  1. Pasotti, P. et al., "Predicting the success of new crypto-tokens: the Pump.fun case," arXiv, Feb 2026. 655,770 tokens created in Sept 2025; 4,338 graduated (~0.63%). arxiv.org/html/2602.14860v1
  2. CryptoRank, "Pump.fun graduating tokens break to 1.15% of new launches," Feb 2026. Historical graduation range ~0.5% to ~2%; few graduates trade beyond a month. cryptorank.io
  3. Dune Analytics (Adam Tehc) via Yahoo Finance, "99.6% of Pump.fun traders have not realized over $10,000 in profits," Jan 2025. 13.55M wallets; 55,296 over $10K; ~293 over $1M. finance.yahoo.com
  4. Dune Analytics (@oladee) via Protos, "Over 50% of Pump Fun token traders lost money this month," March 2026. ~96% of wallets under $500 or at a loss; 50.6% posted losses. protos.com
  5. Coinpedia, "Over 50% of Pump.fun traders lost money this month, while 2 wallets made over $1M," March 2026. Platform fee accumulation over $500M since 2024. coinpedia.org
  6. CoinGecko Research, "Pump.fun Traders Are Making a Comeback," May 2026. Active wallets fell from 5.2M (May 2025) to 1.8M (Dec 2025); early-2026 profitability recovery concentrated in the $1 to $500 tier. coingecko.com/research
  7. Unbroken Hawks, launchpad disclosure: interactive bonding-curve mathematics and wallet behavior profiler, computed from pump.fun's published curve constants and live chain data via Helius and Birdeye. unbroken-hawks.vercel.app/launchpad