---
name: okay-lets-ride
description: "Okay let's ride analysis" - the investment framework of Coulou (@okay_lets_ride), which Ben asked to be folded into every DD from 18 Sep 2026. Read before any DD, token screen or contract-address paste, alongside [[token-screens]].
sources: [cowork]
aliases: [okay lets ride, coulou, okay_lets_ride, moneyness analysis, ride analysis]
---

# "Okay let's ride analysis"

- [stated] Ben, 18 Sep 2026: "He is at the forefront of understanding how to invest in assets in crypto. And he understands why Robinhood will do well and he understands which assets will do well. So learn from his view and include his view in future DD processes. So when I say DD, you do a normal analysis and you include this type of processing and this type of view in your DD. Remember this. We call this okay, let's ride analysis."
- [stated] So: DD = the normal screen (ten questions, three frames, on-chain verification) PLUS the layers below. Not a replacement, an addition.

Source: Coulou, @okay_lets_ride, 4,759 followers as of 18 Sep 2026. Distilled from 100 of his posts (52 original) pulled 18 Sep 2026.

## THE NINE LAYERS TO ADD TO A DD

1. **MONEYNESS.** Every attention economy converges on one base-pair asset: BTC when everyone was a bitcoiner, ETH in the ICO and NFT days, stables once institutions arrived, SOL once degens priced in SOL. His claim: Robinhood Chain has no native token, so whatever sits at the intersection of memes, DeFi and stocks becomes the money, and that is AI. **The DD question: is this asset being used as a QUOTE asset by other launches, or is it only being traded?**

2. **MEASURE PAIRING, NOT PRICE.** His actual metrics: what percentage of a token's supply is locked providing liquidity in pools where it is the quote; what share of another coin's volume routes through the X-pair; how many new launches pick it as their base. Receipts he used: "0.5% of AI supply is providing liquidity into this pool"; "the BONER/AI pair has done 1/3 of BONER's total volume over the last 24h".

3. **PENETRATION, NOT PRICE.** Anchor on the size of the untapped base. "Tokenized stocks today are 0.03% of Robinhood's equity base." "Until today, less than 3.5% of Robinhood's user base could hold stock tokens." His line: time in the market beats timing the market when penetration is this low.

4. **DEFLATE AGAINST A PRIOR CYCLE.** "The aggregated mcap of all native runners is lower than Bonk alone in Dec 2023." Always size today's thing against a named comparable from a previous cycle to show whether it is early or late.

5. **HOLDER-BASE SHAPE VS PRE-RUN COMPARABLES.** Not the absolute holder count, the TIER DISTRIBUTION. His report: "whale tiers mid-pack, small-holder tiers at the low end, the same profile VIRTUAL and BRETT carried before their runs", and "holder count similar to WIF pre-run, same for number of wallets over $1M". **Build the tier histogram and compare its shape to tokens just before they ran.**

6. **DRAWDOWN QUALITY.** Grade a correction against what comparable runners did, and check who bought into it. "AI went from $7mm to $300mm in two weeks, so a 40% correction in these conditions is actually quite shallow. WIF did $300mm to $80mm last cycle." Plus: are six-figure wallets accumulating through the consolidation.

7. **DATED CATALYST SEQUENCE.** The "Zoom out" post Ben screenshotted: an explicit timeline with dates and a terminal target. FOMC and CLARITY (15-16 Sep), SEC exemptions (17-24 Sep), HOOD Summit 29 Sep, US rollout Q4 26 to Q1 27, main-app listings, social trading Q2 27, ending at "1 AI = 1 NVDA". **Every DD should end with a dated sequence, not a vague thesis.**

8. **BUSINESS TAXONOMY.** "Every successful onchain or offchain business is either providing block space, asset issuance, or asset management in some shape or form (exchange, launchpad, LP, lending, custody)." Classify the project into one of the three and ask whether it is running that playbook properly.

9. **THE THREE-LAYER STACK.** (1) Robinhood rails at the base: the chain, tokenized equities, later social trading and app listing. (2) Things composing with the ecosystem: Uniswap, Lighter, orbio inference. (3) The native economy of protocols launching paired to the base asset. **Ask which layer the asset sits in; layer 3 lives or dies on layer 1.**

Also his: fair launch is the new funding model ("find an economical action to perform on chain and you take a fee", bootstrapping a business with no funding); liquidity is a moat; protocol-owned liquidity matters most when a meme also serves as a liquidity hub for stock flows.

## MANDATORY COUNTERWEIGHT, because he is talking his book
- He is **heavily long AI, the LONG ecosystem and orbio**, and posts promotionally about all three. He constantly retweets Natan Benish, Johann Kerbrat and Vlad Tenev.
- His framework is genuinely good; his conclusions are positions. **Use the METHOD, verify the CONCLUSION on chain every time.**
- His "1 AI = 1 NVDA by Q4 2027" is a target, not a forecast with a probability attached.
- 4,759 followers, so he is not a crowd-moving account. He is an analyst, and the value is the method.

## WHAT THIS SESSION ALREADY PROVED THE METHOD CATCHES
Applying layer 5 (holder shape) to INCEL on 18 Sep 2026 reversed a buy case built on backing ratio alone: vault backing 2.30% of market cap, best on the chain, but holders peaked at 944 on 4 Sep and were 831 by 18 Sep, down 10.2% in fourteen days. Layer 2 then killed it entirely: INTC's `uiMultiplier()` reads exactly 1.000000, so the vault accrues nothing, because Intel suspended its dividend in Q4 2024.

# DEEP PASS, 18 Sep 2026. 165 posts, 38 replies, 63 infographics downloaded and read.
- [stated] Ben: "go through his entire feed. Ingest all the infographics, explainers, details, and all the information... including his replies... look into when he first mentioned certain tokens, and then go back and check what the state of information that he's looking at was at that time... identifying an entry pattern for him. What does he like to see in order to enter a coin? Do 10 iterations."

## HIS ENTRY CHECKLIST, HIS OWN WORDS, 28 Aug 2026 06:08 (and a reply version at 02:57)
Nine items, posted twice within four hours, which is how he justified AI:
1. the ticker IS the meta's word on that venue ("ticker $AI on RH")
2. first mover of a new meta
3. chad dev
4. base asset for all upcoming coins in that meta
5. beneficial to the venue's most important KPI
6. value accrual baked in
7. liquidity scales with market cap
8. paired to the largest stock in history
9. it's a dog

## THE ENTRY TRIGGER, dated to the minute
- 27 Aug 2026 10:26 first AI post ("early signs that this is working").
- **27 Aug 2026 10:37, the actual trigger: "Close to 0.4% of the $AI supply has been locked via $AGI already."**
- So: **the moment a SECOND token launches paired to it and locks a measurable percentage of its supply.** That is the flywheel starting, and it is what he acts on.
- 29 Aug he named the template: *"double digits of % of $AI will be permanently locked / burned. $AI will serve as the base asset for AI meta coins, AI agents and more on RH. **Study Virtuals.**"*
- Context at entry: he said on 6 Sep that AI "went from $7mm to $300mm in two weeks", so his first post landed inside that move, not before it. He is early-in-the-run, not pre-launch.
- MEASURED TODAY, same metric: **6 coins are paired TO AI (OPEN, SIT, AGI, HENT, CLANKER, CHIPS), doing $2,782,563 a day = 6.5% of AI's total volume.** One coin on 27 Aug, six now. That is the live scoreboard for his thesis.

## THE HOLDER-TIER TABLE, transcribed in full (his flagship infographic)
Wallet counts by USD tier at each token's pre-run stage, and what it then ran.
| token | date, cap | holders | >=$100 | >=$1K | >=$10K | >=$100K | >=$1M | ran |
|---|---|---|---|---|---|---|---|---|
| VIRTUAL | Oct 24, $91mm | 43,789 | 4,753 | 2,022 | 369 | 68 | 14 | x33.1 |
| PNUT | Nov 24, $115mm | 36,913 | 16,550 | 5,959 | 1,486 | 206 | 9 | x15.6 |
| PEPE | Apr 23, $137mm | 50,652 | 30,668 | 8,845 | 1,464 | 167 | 12 | x11.5 |
| AI16Z | Nov 24, $150mm | 22,705 | 10,080 | 4,050 | 1,187 | 198 | 21 | x16.7 |
| FARTCOIN | Dec 24, $275mm | 28,241 | 12,966 | 5,883 | 1,969 | 447 | 54 | x7.9 |
| POPCAT | Jun 24, $296mm | 49,141 | 21,597 | 9,199 | 2,720 | 444 | 45 | x6.5 |
| MOG | Sep 24, $299mm | 32,841 | 15,645 | 6,780 | 1,874 | 375 | 54 | x5.0 |
| BRETT | May 24, $318mm | 155,657 | 42,246 | 12,347 | 1,880 | 276 | 49 | x5.7 |
| WIF | Feb 24, $341mm | 44,905 | 21,254 | 9,448 | 2,697 | 509 | 49 | x13.4 |
| GOAT | Nov 24, $538mm | 46,275 | 20,802 | 8,546 | 2,760 | 684 | 103 | x2.3 |
| SPX | Apr 25, $560mm | 39,011 | 22,157 | 9,836 | 2,359 | 471 | 80 | x3.6 |
| **AI** | Sep 26, $246mm | 42,642 | 7,875 | 3,542 | 1,387 | 412 | 45 | - |
His method footnote: transfer-event reconstruction on Dune, tiers in USD at snapshot cap, AMM custody excluded, dust profile noted for fee-less chains.

**MY TEST OF HIS OWN TABLE, and it is a real caveat.** Correlation of each column against the run multiple (n=11, log): market cap **-0.87**, >=$1M -0.89, >=$10K -0.76, >=$1K -0.65, holders -0.19. **The dominant predictor is simply market cap, and the tier counts are largely a cap proxy.** The ratio >=$1K/holders reads -0.50 but does not separate cleanly (VIRTUAL 0.046 ran x33, BRETT 0.079 ran x5.7). **So use the tier chart as a structural sanity check that a token is not shaped weirdly, never as a predictor of the multiple.** AI today: 173.3 holders per $mm of cap against a group median of 151.4.

## THREE MORE REUSABLE DEVICES FROM HIS INFOGRAPHICS
- **META LIFECYCLE MAP.** Plot the aggregate market cap of an entire meta over time, mark where today's meta sits on the prior metas' curves, read off the time they had left. 2023-25 memecoins peaked $30bn on 8 Dec 2024 and crossed today's meme-stock level ($802mm) in May 2023, **19 months before peak**. AI memecoins peaked $10.0bn on 3 Jan 2025 and crossed it on 24 Oct 2024, **10 weeks before peak**. So the same level gave 10 weeks in one case and 19 months in the other, which is the honest width of that estimate.
- **PROPORTIONAL AREA SIZING.** All 24 Robinhood Chain natives over $5mm = **$1.35bn**, which is 64% of BONK at its Dec 2023 ignition, 38% of BONK's all-time high, and **4.8% of the Solana meme sector at its Jan 2025 peak ($28bn)**.
- **TAM BY LANE COUNT.** About 4,000 to 5,000 US listed companies; at a $10mm average cap that is a **$50bn** sector if every stock gets one meme.

## SOURCES HE USES, confirmed by him 15 Sep
"defillama & dune are main sources, and some come from gecko terminal".

## ⚠ INTEGRITY FLAG, THE MOST IMPORTANT THING IN THIS PASS
His most-liked original post (502 likes, 153,866 views, 16 Sep, captioned **"This is not priced in"**) is a polished screenshot of a LONG "Vote" product page, "Own the meme. Vote the stock.", showing vault share counts and market caps of AI $3.0bn, BONER $750mm, MOO $340mm.
**It is not a product. He made it himself.** His own reply to @OGDfarmer, same day: *"made this up with imagegen in 2 min, not part of the team here, but looks like it gets people going."*
**Every number on that image is invented.** Treat any of his mockups as thesis illustration, never as evidence, and check for this pattern before quoting any screenshot of his.

## HIS ANSWER TO THE LP-VOTING PROBLEM, and it beats mine
Asked how tokens sitting in an LP could ever vote, he replied 16 Sep: *"I don't think the tokens being in a LP would be an issue per se. You can have the meme holders vote individually and the aggregate of the votes dictate the overall vote. Voting would just be one more thing the hook does with state it already controls, you'd just add modules like snapshot, setInstruction etc."* The v4 hook aggregates individual holder votes, so pooled tokens still carry a voice.

## HIS THESIS IN ONE LINE, reply to @kelxyz_ 8 Sep
"It's the assembly of different concepts (protocol owned liquidity, improved launchpad mechanisms, bonding curve settings, stock pairing) that makes it better. First mover advantages also compound and thicken that edge."

# SECOND DEEP PASS, 18 Sep 2026. Corpus widened to 580 docs (his 180 posts + 400 conversations with him), back to May 2025.
- [stated] Ben: "There might be more entries that he had other than AI... There might be one entry criterion for coins above 20 million market cap. There might be one that's different for lower. It's your job to find all the different patterns."
He was right. There are SIX distinct entry patterns and they are selected by asset type and size, not applied uniformly.

## ⭐ HIS VERIFIED CYCLE CALL, the single strongest credibility datapoint
18 Dec 2025, reply to @factorydoge69: *"If July 25 is top and Oct 25 is echo bubble, then if it plays out similarly to last cycle first bottom could be around Feb 26, and then time based capitulation into July 26. Would catch 4y cyclers waiting for Oct 26 bottom off guard."*
**Measured: BTC was $85,460 that day, printed a local low around $76,896 in Feb 2026, and bottomed at $58,524 on 30 June 2026.** He called the cycle low six and a half months ahead, within days, and named the crowd that would be offside. Weight him accordingly on macro timing.

## THE SIX ENTRY PATTERNS

**A. MONEYNESS FLYWHEEL** (AI, 27 Aug 2026, mid-cap, inside a run)
Trigger: a second token launches paired to it and locks a measurable share of supply. Template named by him: "Study Virtuals." Full nine-item checklist already recorded above.

**B. RELATIVE-VALUE ABSURDITY** (BULLET, 11 Jun 2025, $20mm circ / $80mm FDV)
His pitch, verbatim: *"You can buy @bulletxyz token at $20mm circulating & $80mm FDV. The first Solana network Extension in the fastest growing vertical (CLOBs) is **cheaper than the #100 meme on Coingecko 'Based Fartcoin'** and literally no one but @defi_monk & a few knows about it."* Plus *"cheaper by at least 7x than any Solana-based DEX."*
Criteria: priced below a product-less meme; an explicit multiple against its own sector; first mover in the fastest-growing vertical; **obscurity treated as an asset**; team credibility by named serious tech (Celestia, Sovereign Labs, RiscZero) and years on mainnet.

**C. REBRAND AND MIGRATION CATALYST** (ZEX to BULLET, same thread)
*"Rebranding are always the most obvious, yet least anticipated events."* Requires a 1:1 migration with no dilution and a named historical precedent: he used RBN to AEVO, "20x between October 2023 and March 2024". A scheduled, public, under-priced event.

**D. MISCLASSIFICATION** (ORBIO, 4 Sep 2026, the trigger line)
*"orbio is a new exchange token dressed (and priced) as an ephemeral tax token riding the meta."*
The market prices it as category X, he argues it is category Y, and the gap is the trade. He then builds the case over two weeks: bootstrapped supply side through volume fees, defines a new category, network effect loop (more builders, more demand, more speculation, more fees), and finally the taxonomy test ("bringing a new asset onchain and following the playbook").

**E. TOKEN EMBEDDED IN THE PRODUCT** (CRACK, May 2025)
He engages by proposing the token sink himself: stake and earn platform fees, lock on live agents to promote them, referral staking behind another user, difficulty multipliers gated by stake, staked voting on rules. His summary of the dev: *"Strong dev building in public and embedding the token into the product."*

**F. VALUATION AND BEAR CASE ON LARGE CAPS** (HYPE, Dec 2025)
He publishes downside on things he is long-term bullish on: *"I am long term very bullish on it as well but I think it goes to lower levels first. Could see it around $15-$20 in a few months."* **He does write bear cases.** On Morpho (Jan 2026) he praised non-consensus capital allocation: non-profit entity so everything accrues to the token, and reinvesting revenue into growth rather than buybacks.

## THE BANDING BEN ASKED ABOUT, and it is real
- **Micro and obscure**: relative-value absurdity comparison plus a dated catalyst. Obscurity is a FEATURE.
- **New launch**: misclassification, category creation, and whether the token is embedded in the product.
- **Mid-cap inside a run**: the moneyness flywheel and the pairing metrics.
- **Large cap**: valuation models, tokenomics purity, and an explicit bear case.

## ⚠ THE HOLDER METRIC FLIPS BY ASSET TYPE, do not apply it blindly
- For a MEME he wants a broad base forming, benchmarked to pre-run comparables: "AI holder base is quite similar to most memes just prior to their run to billions."
- For a UTILITY play he wants the opposite: 18 Sep 2026, *"Orbio has the highest dollars of marketcap per holder on Robinhood Chain. Only 4000 wallets hold Orbio atm"* is posted as a BULL point.
So: broad and growing for memes, narrow and undiscovered for utility. Ask which one the asset is before reading its holder count.

# WALLET ATTRIBUTION WATCH — run 19 Sep 2026 05:30 BKK. VERDICT: NOT CONFIRMED, and the test design is broken.
Question: does Robinhood Chain wallet 0x57f7dca134fdd36e5d73460f51ed1b12c11a0065 belong to Coulou (@okay_lets_ride)?

## His first-mention timestamps, pulled fresh from twitterapi.io (not memory)
- $AI 2026-08-27 10:26:12 UTC · $AGI 2026-08-27 10:37:37 UTC · orbio 2026-09-04 13:40:11 UTC · boner 2026-09-07 05:37:43 UTC
- No new ticker in the 48h to 18 Sep 22:37 UTC. He named only AI, Orbio, Lighter, Morpho, UNI — all already his book.

## Primary wallet, first SIGNED buy (tx origin == wallet), TRUE block timestamps
- AI 2026-08-27 08:13:45 (-2h12m vs post) · AGI 2026-08-27 08:26:57 (-2h11m) · ORBIO 2026-09-01 20:51:55 (-64.8h) · BONER 2026-09-12 09:10 (+123.5h, AFTER)

## Why this does NOT confirm the wallet
- The five "control" wallets are NOT his readers. They lead his posts more consistently than the primary does.
- AGI: ctrl1 -12.1h, ctrl3 -9.7h, ctrl4 -9.1h all bought BEFORE the primary's -2.2h. The primary was LAST of the four.
- ORBIO: ctrl4 -65.3h beat the primary by 28 minutes; ctrl5 and ctrl1 within 2h. Four of five controls hit. The hit carries no identifying information.
- BONER: all five controls held it from 20-31 Aug; the primary did not own it until 12 Sep, five days AFTER he posted a specific BONER pairing-volume stat. An analyst writing that stat holds the coin. This is evidence AGAINST the primary.
- Better candidates if the account is in this set at all: ctrl1 0x2cffed5d56eb6a17662756ca0fdf350e732c9818 and ctrl4 0xa4d9d7068d84bbdf2e3caf1bbf2e1135bcd41984.

## TWO METHOD TRAPS THAT CHANGED THE ANSWER — do not repeat
1. FIRST RECEIPT IS NOT FIRST BUY. The primary's AI "first receipt" (2026-08-25 21:04:59) was a 10-token batch PUSHED by contract 0x9f3edbfae8014d55e328b6dd966c6c95e75f6210, tx origin 0xee5cc579a0a6d90bff4c481387b39ba655cac728 — a third party. Only 86 of 3,768 inbound legs were signed by the wallet. Always resolve tx origin before calling a receipt a buy.
2. BLOCK-TIME EXTRAPOLATION IS UNUSABLE AT THIS RANGE. 0.1006 s/block from the head was ~3h wrong 20M blocks back (AI buy read 11:23 estimated vs 08:13:45 true). Always fetch real block timestamps for any block a verdict depends on.
- Chain head at run: block 66587452, ts 1789771021 (2026-09-18 22:37:01 UTC).
- eth_getLogs also fails with "log query timed out", not just "exceeds limit of 10000" — subdivide the window on BOTH.

# DAILY WATCH — 19 Sep 2026 09:35 BKK (covers 18 Sep 09:00 BKK to 19 Sep 09:30 BKK)
Pull: twitterapi.io HTTP 200, no 402. 149 tweets deduped, 62 in window, 18 his own (7 RT, 4 replies), 6 images downloaded and read. No browser fallback needed.

## 1. NEW ENTRY? NO.
- No ticker named for the first time. Tickers in his own text: AI, Orbio, Lighter, Morpho, UNI, NVDA, LONG, LONGx, SIT, OPEN, BONER, ICOIN. All already his book.
- Chart-label-only, first time they appear in HIS OWN infographic but not named as positions: PONS, PARE, STANDARD, MOO, CASHCAT, ZZZ, Index, TENDIES, microduck, FRONG, JUGGERNAUT.
- Closest thing to a new move, 19 Sep 01:50 BKK reply to @kelxyz_, then self-quoted at 01:55 BKK: *"AI for liquidity, shareholder participation, attention / Lighter for app-integration down the road and the fee sharing economics / Orbio for agentic trading (if issuance goes up, more assets to trade, bigger inference capital market) / Morpho first tvl on the chain, scale with the earn program and can extend from there"* and *"Think if you want to long tokenized stocks issuance on Robinhood these (+ UNI) are the clear bets today."*
- **Assessment: this is NOT a seventh pattern.** It is layer 9 (the three-layer stack) expressed as a basket, one bet per layer, with a one-line role for each. Log it as a DEVICE, not an entry trigger. No new coin, no new trigger event.

## 2. NEW METHOD? YES, THREE, and two of them are real additions to the layers.
**(a) ORDER-BOOK DEPTH ASYMMETRY.** Infographic 18 Sep 21:17 BKK, *"Downside liquidity exceeds upside — by design"*, AI, nine pools. Bids to -50% **$6.3mm**, asks to +50% **$4.4mm**, ratio **1.4:1**, "every pool tilted the same way". Bid wall composition 52% ETH, 30% USDG, 15% NVDA. $1.6mm absorbs -10%, $1.2mm lifts +10%. Spot on the chart $0.343. Inset contrast: *"a normal meme pool: always 1:1"*. Method footnote, his own: *"Uniswap v3 tick walks + v4 PoolManager storage reads · 9 pools · AI-quoted pools excluded · asks at fill price, bids at face · ETH $2,505 · NVDA $220."* Closing line: *"Rented liquidity leaves. Owned liquidity compounds."*
- The accompanying post, his words: *"Every net buy converts the pool's AI into NVDA bids that sit under the new price and can never be withdrawn"*, *"0.5% of every trade, buy or sell, is bought into NVDA (AI on AI-quoted pools) and never sold"*, *"Bids follow price up, legs then plateaus, rising lows: a staircase instead of a spike-and-fade"*, *"early holders can distribute without breaking the chart"*.
- **This is new and it is the strongest method he has published.** It measures the thing everyone else asserts. Add as a layer.

**(b) RANGE NET-POSITION LEDGER.** Infographic 18 Sep 15:12 BKK, *"Top wallets were net buyers through the range"*, AI, 1-18 Sep, top 150 wallets ex-infrastructure, from full on-chain transfer histories. **+84mm bought, -43mm sold, net +41.7mm AI ≈ 4% of supply.** 58 buyers, 49 sellers, 42 flat. **51 wallets with zero prior AI history entered the top 150. 29% of buyers funded mainly via Relay (cross-chain). Every big seller kept 60-80% of their stack.** Footnote: Blockscout, transfer histories paginated to window edge, balances at $0.338, smart wallet = EIP-7702.
- This is the measurable form of layer 6's "did six-figure wallets accumulate through it". Promote it from a question to a procedure.

**(c) DOLLARS OF MARKET CAP PER HOLDER, plotted.** Infographic 18 Sep 16:05 BKK. FDV against live holders on linear axes with constant-$/holder rays ($1,000/holder and $10,000/holder), plus a ranked bar with a chain median. ORBIO **$10,890** (4,215 holders, 17 days old, "steepest token on the chain"), AI $7,100, PONS $5,032, PARE $3,921, STANDARD $3,715, BONER $2,995, MOO $2,544, CASHCAT $1,894, ZZZ $1,594, Index $1,329, TENDIES $704, microduck $319. Caption: *"ORBIO carries the most capital per holder on the chain — AI the most of any large token. Breadth lives at the bottom of the list."* Source line: FDV GeckoTerminal/screener, holders Blockscout counters, stock tokens and infrastructure excluded.
- Memory already held the meme-versus-utility holder flip. This is the quantified, ranked, chain-median version of it. Materially better than the prose note.
- **SPOT-CHECKED AND IT HOLDS:** ORBIO mcap $49.6mm live / 4,215 holders = $11,767 per holder against his $10,890. His numbers are honest here.

## 3. MOCKUP CHECK — CLEAN TODAY, with one caveat.
- All three data infographics carry explicit method footnotes naming real sources (Blockscout, GeckoTerminal, v3 ticks, v4 PoolManager). No product page was passed off as real.
- The fourth image, 19 Sep 00:50 BKK, is a full system diagram (LEVEL 0 pair / LEVEL 1 fund / LEVEL 2 swarm vault / LEVEL 3 Lighter + Orbio, with CREDIT budgets, return-per-CREDIT weights and a fee waterfall). **It is a design for something that does not exist and he said so himself**: *"random friday idea"* and *"probably a ton of reasons why this specific setup doesn't work but excited to see more composability"*. Self-disclosed, so not an integrity flag. Do not cite any number on it.
- **UNRESOLVED:** @00lawd asked *"Where is this analysis from?"* on the holder chart and @toshiXchain asked *"What application/website is that ?"*. **He answered neither.** So the tooling behind all three charts is undisclosed. Treat the footnotes as his claim about method, not as a verified pipeline.

## 4. THESIS SCOREBOARD — re-measured 19 Sep 09:30 BKK, GeckoTerminal token-pools, full pagination.
AI 0x2e8c31162b855a2ffa90f6f8634643ad6f111e18 · price $0.3170 · mcap $314,525,151 · 24h volume $24,619,135 · price -5.1% on 24h.
| metric | 18 Sep baseline | 19 Sep | change |
|---|---|---|---|
| coins quoted in AI, above $10k/day | 6 | **12** | +6 |
| coins quoted in AI, any volume | (not measured) | **127** | floor, see below |
| combined 24h volume of AI-quoted pools | $2,782,563 | **$1,490,823** | **-46.4%** |
| that as a share of AI's total volume | 6.5% | **6.06%** | -0.44pp |
| the original six alone | $2,782,563 | **$1,174,177** | **-57.8%** |
Original six today: OPEN $719,539 · SIT $335,167 · AGI $35,193 · CLANKER $34,232 · HENT $32,037 · CHIPS $18,009. New above $10k/day: MUTT $140,253 (pool created 18 Sep 22:23 BKK on only $9,779 liquidity, so churn not adoption), IA $44,800 across three pools, ASI $25,910, WU $20,472, CASHCAT $12,631.
- **CORRECTION TO THE 18 SEP ENTRY, and it matters.** "6 coins are paired TO AI" was an undercount of the population, not a count of it. Full pagination finds **127 distinct tokens** using AI as the quote asset, many launched weeks before 18 Sep (LA 27 Aug, INUP 27 Aug, LLM 28 Aug, SLOPCOIN 28 Aug, ANTH 29 Aug). The comparable figure is the $10k/day cohort, 6 to 12. The number that misled: the raw coin count.
- **Method trap for next run:** GeckoTerminal's token-pools endpoint caps at 10 pages of 20, so 200 pools is a ceiling. AI hit exactly 200. **127 AI-quoted tokens is a FLOOR.** Use Dune for the true count.
- **Honest read of the scoreboard: breadth up, depth down.** Twice as many coins clearing $10k/day, but the dollars routing through AI-quoted pools nearly halved and AI's own volume fell with it. Share of volume held roughly flat at ~6%, which means the pairing channel shrank in line with the whole token, not faster. One day is noise; the number to watch is whether the $10k/day cohort keeps climbing while dollar volume keeps falling, because that is a flywheel spinning on fumes.

## 5. CALL TRACKING.
- **No new dated prediction in the window.** The two forward statements are both undated: *"Nate has 7 figs of AI on fomo. One day it will be 8, maybe 9"* (19 Sep 05:12 BKK) and *"these (+ UNI) are the clear bets today"* (19 Sep 01:55 BKK).
- No earlier dated call resolved today. The 18 Dec 2025 capitulation call (BTC bottom 30 Jun 2026 at $58,524) remains his one verified hit. The Q4 2027 "1 AI = 1 NVDA" target is still open and still a target, not a forecast.
- **Live test now running on his own claim.** He posted the depth chart at spot $0.343 saying *"AI can handle a $2 million sell order, be at the same price as yesterday"* and predicting *"shallower drawdowns, faster recoveries"*. AI is $0.3170 sixteen hours later, **-7.6% from his chart's spot**. Not a falsification, a 7.6% move is inside normal, but it is a dated marker: re-check whether the staircase holds or whether the bid wall was thinner than the tick walk implied.

# WALLET WATCH — 19 Sep 2026 21:40 BKK. NO NEW TICKER, but ICOIN tested for the first time and it hurts the primary.
- 48h window 17 Sep 21:30 to 19 Sep 21:30 BKK. 19 own posts pulled (twitterapi.io HTTP 200), all quoted_tweet fields and 11 images read. **No ticker named for the first time.** In-window tickers: AI, Orbio, SIT, OPEN, BONER, ICOIN, UNI, Lighter, Morpho, LONG, LONGx, NVDA, plus historical comps (BONK, TRUMP, PNUT, PEPE, GME, SPX6900, WIF, VIRTUAL, BRETT, POPCAT, MOG, GOAT, FARTCOIN, AI16Z, PENGU, BOME).
- **CORRECTION to the 19 Sep 09:35 BKK entry.** It listed ICOIN as "already his book" without a date. ICOIN's true first mention is **10 Sep 2026 16:14:41 UTC (23:14:41 BKK)**, inside his PINNED X ARTICLE (x.com/i/article/2097918324478746628), posted as a bare t.co link with empty text. **Method trap: a bare-link post hides a whole article of tickers.** twitterapi.io indexes the article body, so `from:okay_lets_ride TICKER` finds it even though the post text is just a URL. Always resolve entities.urls on any text-only-link post.
- ICOIN contract 0x5d6EF090a1461B11c9427aC319260122D1C61e18, deepest RH pool ICOIN/AAPL $368,343 liquidity.

## ICOIN RESULT — scored zero as evidence, but structurally damning for the primary
| wallet | first ICOIN receipt | signed gap vs 10 Sep first mention | balance now |
|---|---|---|---|
| primary 0x57f7dca134fdd36e5d73460f51ed1b12c11a0065 | **NEVER. 0 in, 0 out** | n/a | **0** |
| ctrl1 0x2cffed5d56eb6a17662756ca0fdf350e732c9818 | 03 Sep 02:22:33 BKK | -188.9h | 263,023 ICOIN |
| ctrl2 0xfa00a9ed787f3793db668bff3e6e6e7db0f92a1b | 03 Sep 00:49:32 BKK | -190.4h | 12,129 |
| ctrl3 0xd0a67cb08be17475f4315a04c5f0be3e200ef66c | 03 Sep 03:02:44 BKK | -188.2h | 30,377 |
| ctrl4 0xa4d9d7068d84bbdf2e3caf1bbf2e1135bcd41984 | 03 Sep 03:38:24 BKK | -187.6h | 21,078 |
| ctrl5 0x73691cee20db22f55a6fd0d5948ab00e0fb973b1 | 03 Sep 01:38:08 BKK | -189.6h | 29,919 |
- **Trap check passed** (10 Sep is his earliest ICOIN mention), but every control buy is ~188h before, far outside the 0 to 72h HIT band. **Score zero as a lead signal.**
- **The structural finding is the real one.** All five controls entered ICOIN on 2 Sep within a 2h49m band, and **not one of them has a single OUT leg. Zero sells, all five, ever.** Five wallets, same day, same three hours, never sell, is a cluster, not five independent readers.
- **The primary has never touched ICOIN at all**, while naming it twice (10 Sep article, 18 Sep 14:17 UTC liquidity thread) as an example of a coin organically quoted in AI. Same shape as the BONER miss: he writes about the coin, the wallet does not hold it.

## ⚠ METHOD CORRECTION — "tx origin == wallet" is NOT a valid signed-buy test here
- **The primary is EIP-7702 delegated** to implementation 0x7702cb554e6bfb442cb743a7df23154544a7176c. **All five controls are plain EOAs** (eth_getCode = 0x).
- Yet every control's ICOIN arrived with a third-party tx.from, and ctrl4's first receipt was routed through **ERC-4337 EntryPoint v0.7 0x0000000071727de22e5e9d8baf0edac6f37da032**. So relayed/aggregator delivery happens on plain EOAs too, and the 19 Sep 05:30 rule ("only 86 of 3,768 inbound legs were signed, so the rest are pushes") **over-rejects real buys.** Re-date any verdict that rested on it.
- Different wallet software on the two sides (7702 smart account vs plain EOAs) is itself an argument that the primary is not a member of the control cluster.

## RUNNING VERDICT: still NOT CONFIRMED, and now weaker than on 19 Sep 05:30.
Four tokens tested so far. Primary: AI -2h12m, AGI -2h11m (both last of four), ORBIO -64.8h (beaten by ctrl4), BONER +123.5h (after), **ICOIN never held.** Zero clean two-token confirmations. Better candidates remain ctrl1 0x2cffed5d56eb6a17662756ca0fdf350e732c9818 and ctrl4 0xa4d9d7068d84bbdf2e3caf1bbf2e1135bcd41984, but the never-sells pattern says treat all five as one operator, not as five readers.
- Chain head at run: block 67159932, ts 1789828720 (2026-09-19 14:38:40 UTC).
- **Next run: test SIT and OPEN**, the two native AI-quoted coins he names repeatedly and that have never been checked.

# WALLET WATCH — 20 Sep 2026 07:35 BKK. SIT + OPEN tested. VERDICT MOVES: primary is PROBABLY THE WRONG WALLET.

## ⚠ SOURCE FAILURE — twitterapi.io key is OUT OF CREDITS
- `new1_9f31bb77a9cd477eb6baa555e713280e` returns **HTTP 402 `{"error":"Unauthorized","message":"Credits is not enough.Please recharge"}`** on every endpoint. Confirmed on both advanced_search and user/info. HTTP 403 without the header, 402 with it, so the key is valid and the balance is zero.
- Prior run 19 Sep 21:40 BKK had HTTP 200. **The account drained between 19 Sep 14:40 UTC and 20 Sep 00:35 UTC.**
- Droplet fallback (`ssh root@159.223.88.168 python3 /opt/xq.py`) is **dead from the cloud container**: no `~/kookwatch/id_kookwatch` key, no `ssh` binary, port 22 unreachable. It needs the Mac.
- No browser this run (scheduled cloud task, no desktop bridge). `WebFetch` on x.com is robots-disallowed. So **STEP 1 could not be completed**: the 48h window 18 Sep 00:34 to 20 Sep 00:34 UTC is only partly covered and **19 Sep + 20 Sep are entirely unreadable**. Whether he named a NEW ticker in the last ~26h is UNKNOWN, not "nothing new".

## ✅ NEW WORKING X FALLBACK — LunarCrush MCP, no twitterapi.io needed
- `mcp__LunarCrush__keyword_posts(keyword="okay_lets_ride", from_date, to_date)` **works and returns his own posts plus replies to him**, with ISO timestamps and status URLs. Subscription live: Individual, 2000/day, 10/min.
- `mcp__LunarCrush__creator_posts(network="x", screenName="okay_lets_ride")` also works but is **top-posts-ranked, not a timeline**, and its index for him stopped at 14 Sep. `creator_posts` with from_date/to_date for 18-20 Sep returned EMPTY while `keyword_posts` for the same range returned 8 of his own posts. **Use keyword_posts, not creator_posts.**
- `mcp__LunarCrush__creator(network="x", screenName="okay_lets_ride")` gives a profile card: account type "professional influencer", avg 4,005 engagements, CreatorRank 275,492, top-mentioned accounts. Its follower number (1,098) is WRONG against the post-level figure (5,181 on 18 Sep) — do not quote LunarCrush follower counts.
- **Coverage limit that matters: it is a ranked/sampled index, not an archive.** Good enough to establish that a ticker WAS named by a date. **Not good enough to run the MANDATORY TRAP CHECK**, which needs a complete history search. Any trap check done on LunarCrush alone must be reported UNRESOLVED.

## Contracts resolved (DexScreener, deepest robinhood pool)
- **SIT 0x89dA5167eb1A0067f9B3e39A544EF8d4b9c41e18** · deepest pool SIT/AI $640,998 liq · earliest RH pair created **2026-08-27 22:54:41 UTC** · mcap $1.90mm
- **OPEN 0xF1b266c11f68bC936f218AD20712c4feB3e61E18** · deepest pool OPEN/AI $372,134 liq · earliest RH pair created **2026-09-11 16:34:49 UTC**
- Chain head at run: block **67516658**, ts 1789864609 (2026-09-20 00:36:49 UTC).
- Method note: scan starts were set by **binary search on real block timestamps** (25 Aug 00:00 UTC = block 45280590; 9 Sep 00:00 UTC = block 58101711), not by 0.1006 s/block extrapolation. Binary search is exact and cheap. Keep using it.

## SIT — first receipt per wallet. Earliest KNOWN mention 14 Sep 2026 16:01 UTC. TRAP CHECK UNRESOLVED.
| wallet | first SIT receipt UTC | signed gap | balance now | sender |
|---|---|---|---|---|
| ctrl4 0xa4d9d7068d84bbdf2e3caf1bbf2e1135bcd41984 | 2026-08-28 00:22:40 | **-423.6h** | 27,576.58 | contract 0xb477751b76cf82d00a686a1232f5fcd772414af3 |
| ctrl3 0xd0a67cb08be17475f4315a04c5f0be3e200ef66c | 2026-08-28 01:45:38 | **-422.3h** | 8,145.62 | contract 0x39b38686a19836ac10162c490e4558e120cbbe5f |
| ctrl1 0x2cffed5d56eb6a17662756ca0fdf350e732c9818 | 2026-08-28 16:11:08 | **-407.8h** | 62,273.83 | contract 0x6e2a35a7ad683cf634d91492d73bb7ff774c6919 |
| ctrl5 0x73691cee20db22f55a6fd0d5948ab00e0fb973b1 | 2026-09-04 23:55:22 | **-232.1h** | 4,187.50 | **EOA** 0xdc24690d34da7ba555f4b61f5ba5173b9c2639da |
| ctrl2 0xfa00a9ed787f3793db668bff3e6e6e7db0f92a1b | 2026-09-14 09:40:36 | **-6.3h** | **0.00 (round-tripped)** | contract 0x6e2a35a7ad683cf634d91492d73bb7ff774c6919 |
| **primary 0x57f7dca134fdd36e5d73460f51ed1b12c11a0065** | **NEVER** | n/a | **0** | n/a |

## OPEN — first receipt per wallet. Earliest KNOWN mention 13 Sep 2026 08:01 UTC (his own OPEN write-up, status 2099045875444031601). TRAP CHECK UNRESOLVED.
| wallet | first OPEN receipt UTC | signed gap | balance now | sender |
|---|---|---|---|---|
| ctrl1 0x2cffed5d56eb6a17662756ca0fdf350e732c9818 | 2026-09-11 19:19:04 | **-36.7h** | 23,849.76 | contract 0x6e2a35a7ad683cf634d91492d73bb7ff774c6919 |
| ctrl3 0xd0a67cb08be17475f4315a04c5f0be3e200ef66c | 2026-09-12 15:30:52 | **-16.5h** | 4,708.24 | contract 0x6aa80dbbed9ae5ab45fbf61f9644fada3b29326e |
| ctrl5 0x73691cee20db22f55a6fd0d5948ab00e0fb973b1 | 2026-09-12 17:22:50 | **-14.6h** | 2,826.70 | **EOA** 0xdb5973b449f26986596f74c8f83d51e47ea2a9c4 |
| ctrl4 0xa4d9d7068d84bbdf2e3caf1bbf2e1135bcd41984 | 2026-09-12 21:01:15 | **-11.0h** | 126.11 | contract 0xb477751b76cf82d00a686a1232f5fcd772414af3 |
| ctrl2 0xfa00a9ed787f3793db668bff3e6e6e7db0f92a1b | **NEVER** | n/a | 0 | n/a |
| **primary 0x57f7dca134fdd36e5d73460f51ed1b12c11a0065** | **NEVER** | n/a | **0** | n/a |
- **All four OPEN buys sit inside the 0 to 72h HIT band.** Scored as evidence they are worth nothing for the trap check being unresolved, but the SHAPE is the finding: four of five controls, none of them the primary.

## THE ABSTENTION IS NOT DORMANCY — this is the strongest single result of the watch so far
- Primary wallet is **live and trading**: nonce **117**, holds **12,555.33 AI**, and took an inbound AI leg at **2026-09-20 00:00:07 UTC**, 36 minutes before the head block. Zero AI out-legs in the last ~11h.
- So the primary was **actively trading AI on the same days** it declined SIT, OPEN and ICOIN.
- **He named SIT and OPEN in his own text on 18 Sep 14:17 UTC** (status 2100952349715935626): *"AI is the quote asset for other assets either native (SIT, OPEN etc.), or organically (BONER, ICOIN etc.)"*. **Four coins named in one sentence. The primary wallet has never held a single one of them.** It holds only AI, AGI and ORBIO.
- That is a **single-coin maximalist wallet**, not the book of the analyst who writes the pairing research. Same shape as the BONER and ICOIN misses, now three times over.

## THE CONTROL CLUSTER ARE LAUNCH SNIPERS, not readers and not him
- ctrl4 bought SIT **+1.47h after the pool was created**. ctrl1 bought OPEN **+2.74h after the pool was created**.
- They buy at launch, days to weeks before he writes about the coin. That explains every large negative gap (-423.6h, -407.8h, -188.9h on ICOIN) without any information flowing from him.
- Shared delivery contracts confirm they are one cluster: **0x6e2a35a7ad683cf634d91492d73bb7ff774c6919** delivered ctrl1's SIT, ctrl2's SIT and ctrl1's OPEN; **0xb477751b76cf82d00a686a1232f5fcd772414af3** delivered ctrl4's SIT and ctrl4's OPEN.
- ctrl5 is the odd one: both its SIT and its OPEN arrived from **plain EOAs** (0xdc24690d34da7ba555f4b61f5ba5173b9c2639da, 0xdb5973b449f26986596f74c8f83d51e47ea2a9c4), so ctrl5 was **pushed** the tokens rather than swapping for them. Worth a separate look.
- ctrl2 is the other odd one: bought SIT -6.3h before the known mention and **round-tripped to zero**, and never touched OPEN. It behaves differently from the other four.

## RUNNING VERDICT — 20 Sep 2026: NOT CONFIRMED, and now call the primary PROBABLY WRONG.
Seven tokens tested. Primary: AI **-2h12m** (last of four), AGI **-2h11m** (last of four), ORBIO **-64.8h** (beaten by ctrl4 by 28 min), BONER **+123.5h** (after the post, follower behaviour), **ICOIN never**, **SIT never**, **OPEN never**.
- **Zero clean two-token confirmations on the primary, on seven tokens, over three runs.** Two of its three hits are on the same 11-minute event (AI and AGI on 27 Aug), so they are not independent evidence, they are one trade.
- Per Ben's rule "a hit on a control instead = say the primary is probably wrong and name the better wallet": **the better wallet is ctrl1 0x2cffed5d56eb6a17662756ca0fdf350e732c9818.** It is the only wallet with in-band hits on two genuinely different tokens and events (AGI -12.1h, OPEN -36.7h), and it holds the largest stack of every coin tested (62,273.83 SIT, 23,849.76 OPEN, 263,023 ICOIN).
- **Caveat that stops this being an attribution:** ctrl1's leads are explained by launch-sniping, its delivery contract is shared with ctrl2, and the never-sell pattern across all five says one operator running five wallets. **Naming ctrl1 as Coulou is not supported either.** The honest read is that the whole six-wallet set is probably the wrong tree, and the primary in particular should be dropped.
- **To actually settle it, the test has to change.** The gap-to-post design cannot separate an analyst from a launch sniper on a chain where both act within hours. What would settle it: a coin he named that had NO tradeable pool at the time of the post, so a buy can only follow the post; or a funding-graph link from the primary to a CEX deposit or an identity-bearing address (needs Arkham, which is the standing gap).
- **Next run: recharge twitterapi.io first.** Then re-read 19-20 Sep for any new ticker, because that window is currently a hole. Untested tickers he names: HENT, CLANKER, CHIPS, IA, ASI, MUTT, WU, CASHCAT.

# WALLET WATCH — 20 Sep 2026 09:40 BKK. VERDICT: PRIMARY WALLET KILLED. It is not Coulou.

## ⚠ SOURCE FAILURE, SECOND RUN RUNNING — twitterapi.io still OUT OF CREDITS
- `new1_9f31bb77a9cd477eb6baa555e713280e` returns HTTP **402** `{"error":"Unauthorized","message":"Credits is not enough.Please recharge"}` on both advanced_search and user/info. Unchanged since the 07:35 BKK run. **Ben has to recharge; the watch cannot run STEP 1 properly until he does.**
- **Blockscout is ALSO down from the container this run**: both `/api` (v1) and `/api/v2` return a Cloudflare "Just a moment..." interstitial, not JSON. So no tokenlist, no holder counts. RPC `rpc.mainnet.chain.robinhood.com` works fine with a browser User-Agent.
- Working route used instead: **`eth_call` balanceOf (0x70a08231) straight on the RPC** answers "does this wallet hold token X" in one call per pair, far cheaper than an eth_getLogs walk. Use this as the first screen on any future wallet-vs-ticker question, and only walk logs when the balance is zero and you need to know whether it round-tripped.
- **eth_getLogs full-range sparse scan DOES work.** fromBlock 1 to head in 16,000,000-block windows returned without error for a zero-result query. So the "start 16M, quarter on error" rule is right and a full-history negative can be proved in ~4 calls.
- X fallback used: `mcp__LunarCrush__keyword_posts(keyword="okay_lets_ride", network="x")`. `creator_posts` with from_date/to_date for 19-20 Sep returned EMPTY again — **keyword_posts only, confirmed twice.**

## 1. NEW TICKER FROM HIM? NO, in the readable window. Coverage is INCOMPLETE, so this is not a clean "nothing new".
- Window 18 Sep 02:35 to 20 Sep 02:35 UTC. His own posts recovered: 18 Sep 09:05, 14:17, 17:50, 18:44, 18:50, 18:55, 22:12 UTC, and **one 19 Sep post, 21:01 UTC** (Karpathy-style autoresearch loops, inference budget, agent swarms — no ticker, no coin).
- Replies TO him on 19 Sep 12:44 / 14:23 / 15:53 UTC imply a post of his around 19 Sep 12:00 UTC about past-cycle conditions that LunarCrush did not surface. **That post is unread.** Nothing from 20 Sep (only 2.6h into the UTC day at run time).
- Tickers in window, all already his book: AI, Orbio, Lighter, Morpho, UNI, NVDA, LONG, LONGx, SIT, OPEN, BONER, ICOIN.
- **Reader pitch, not his call:** @encryptbit 19 Sep 13:18 UTC pitched "Fatcoin on long" (LLY-paired) to him. **He did not reply and has never named it.** Do not log it as his.

## 2. FOUR MORE TICKERS TESTED — CLANKER, IA, HENT, CHIPS. Primary holds ZERO of all four.
Contracts (DexScreener, deepest robinhood pool):
- **CLANKER 0xd24688a1d530f648aeD86A835acA8AD6a7e61E18** · CLANKER/AI $1,329,200 liq · earliest RH pair 2026-08-14 00:26:01 UTC
- **IA 0xAC43a84d825518B2e2De5Bcd041132929b341e18** · IA/AI $796,122 liq · earliest RH pair 2026-08-13 20:11:05 UTC
- **HENT 0xfbfd4476AE28428556D0f5Ea7dfaff60f8051e18** · HENT/AI $66,141 liq · earliest RH pair 2026-09-11 18:35:31 UTC
- **CHIPS 0x68281919E022B0feF78E4A6181a9050fD6131e18** · CHIPS/AI $674,213 liq · earliest RH pair 2026-08-15 22:23:11 UTC

Live balances, all six wallets, 20 Sep 2026 02:37:25 UTC (head block **67588328**):
| wallet | CLANKER | IA | HENT | CHIPS | AI |
|---|---|---|---|---|---|
| **primary 0x57f7dca134fdd36e5d73460f51ed1b12c11a0065** | **0** | **0** | **0** | **0** | 12,555.33 |
| ctrl1 0x2cffed5d56eb6a17662756ca0fdf350e732c9818 | 1,916.16 | 2,883.73 | 11,964.78 | 581.36 | 114,243.64 |
| ctrl2 0xfa00a9ed787f3793db668bff3e6e6e7db0f92a1b | 0 | 15,130.11 | 9,282.19 | 0 | 3,633.47 |
| ctrl3 0xd0a67cb08be17475f4315a04c5f0be3e200ef66c | 0 | 71,490.35 | 3,250.84 | 0 | 34,923.18 |
| ctrl4 0xa4d9d7068d84bbdf2e3caf1bbf2e1135bcd41984 | 10.24 | 3,228.36 | 0 | 0 | 10,290.74 |
| ctrl5 0x73691cee20db22f55a6fd0d5948ab00e0fb973b1 | 0 | 0 | 0 | 0 | 3,285.54 |
- **Zero balance proved to be zero-ever, not a round trip.** Full-range eth_getLogs (block 1 to 67588328, Transfer topic, topic2 = primary padded) returns **0 inbound legs on CLANKER and 0 on IA**, scan complete. The primary has never received either token in the chain's entire history.
- **He named CLANKER and IA in his own text on 14 Sep 2026 16:01 UTC** (status 2099528924241830345): *"While I don't know if open or any of the current asset paired with ai (sit, clanker, ia etc…) will do well durably, I believe ai holders should be supportive of those."* Trap check is moot here: there is no buy to score, because there is no buy.
- ctrl1 holds all four. ctrl1 remains the only wallet that shows up on every coin tested.

## 3. THE PRIMARY IS ALSO FLAT — no activity in the last 2 hours
- nonce **117** and AI balance **12,555.33**, both identical to the 07:35 BKK run. No trades between 20 Sep 00:36 and 02:37 UTC.

## VERDICT — 20 Sep 2026 09:40 BKK: **KILL THE PRIMARY. 0x57f7dca134fdd36e5d73460f51ed1b12c11a0065 is not @okay_lets_ride.**
Eleven tickers now examined across four runs. The primary's complete record:
| ticker | primary result |
|---|---|
| AI | -2h12m (last of four wallets) |
| AGI | -2h11m (last of four; same 11-minute event as AI, so NOT independent evidence) |
| ORBIO | -64.8h (ctrl4 beat it by 28 min) |
| BONER | **+123.5h, AFTER the post** = follower behaviour |
| ICOIN | **never held** |
| SIT | **never held** |
| OPEN | **never held** |
| CLANKER | **never held, 0 inbound legs ever** |
| IA | **never held, 0 inbound legs ever** |
| HENT | **never held** |
| CHIPS | **never held** |
- **Seven never-helds and one after-the-post buy, against one single trading event (27 Aug, AI+AGI, 11 minutes apart) that produced its only two in-band gaps.** One trade is not two independent hits. Ben's confirmation bar of two hits on two different tokens is **not met and cannot be met by this wallet.**
- The decisive structural fact: on 18 Sep 14:17 UTC he wrote *"AI is the quote asset for other assets either native (SIT, OPEN etc.), or organically (BONER, ICOIN etc.)"*, and on 14 Sep 16:01 UTC he named *"sit, clanker, ia"*. **Six coins named in his own two sentences. The primary wallet has never held a single one of them**, while trading AI on the same days. An analyst who publishes pairing research on eight coins and owns none of them does not exist. This is an AI-maximalist's wallet, not his book.
- **Better wallet, per Ben's rule: ctrl1 0x2cffed5d56eb6a17662756ca0fdf350e732c9818.** It is the only wallet present on every single coin tested (AI 114,243.64 · CLANKER 1,916.16 · IA 2,883.73 · HENT 11,964.78 · CHIPS 581.36 · SIT 62,273.83 · OPEN 23,849.76 · ICOIN 263,023) and the only one with in-band hits on two different events (AGI -12.1h, OPEN -36.7h).
- **But ctrl1 is still not an attribution.** Its leads are explained by launch-sniping (it bought OPEN +2.74h after pool creation), it shares delivery contract 0x6e2a35a7ad683cf634d91492d73bb7ff774c6919 with ctrl2, and none of the five controls has ever sold anything. Five wallets, one operator, entering at launch. That operator is a sniper desk, not an analyst with 5,192 followers.
- **The gap-to-post test is exhausted. It cannot separate an analyst from a launch sniper on this chain.** Do not run it again on these six wallets. Two designs can still settle it: (1) a coin he names that has **no tradeable pool at the time of the post**, so any buy must follow the post; (2) a funding-graph link from a candidate wallet to a CEX deposit or identity-bearing address — which needs **Arkham**, the standing key gap.
- **Next run: recharge twitterapi.io first, then re-read 19 Sep 12:00 UTC onward**, which is still a hole. Untested tickers he has named: ASI, MUTT, WU, CASHCAT (all low-conviction chart labels, not calls).