---
name: rh-infra-bottom-screen
description: The 6 Sep 2026 screen of every Robinhood Chain infrastructure token for "has it bottomed and could it be a key chain player" - the corrected PONS base case, the routed-cash test that replaced Ben's usage-divergence hypothesis, launch-cohort seller ledgers for 21 tokens, and the ranked board. Read before answering "is X the next PONS", before any RH infra pullback question, and before reusing the cohort method.
sources: [cowork]
aliases: [infra-bottom-screen, bottomed-board, rh-pullback-infra, next-pons]
---

# RH Chain infrastructure bottom screen, 6 Sep 2026
Artifact: https://claude.ai/code/artifact/2a0b3f2a-dfcb-4f52-9ba1-dec67defed1f
Companions: [[pullback-comeback]], [[delta-liquidity]], [[rh-ve33-dexes]], [[slvr-token]], [[rh-chain-infrastructure]].
- [stated] Ben's ask: use the insider / seller-exhaustion / extended-and-pulled-back framework from the Insider Selling Ledger artifact to find RH-chain tokens that have bottomed or are close and could be key infrastructure players. He named PONS as the ideal example, remembering that "it pulled back while its usage numbers did well, and insiders weren't selling anymore at the bottom", and asked whether DELTA is next.

## ⚠ THE CORRECTION THAT CHANGED THE SCREEN: PONS's usage did NOT hold up through its pullback
Measured from GeckoTerminal daily candles on PONS/WETH plus DefiLlama Pons V1+V2 daily revenue.
- True intraday ATH **$0.043934 on 30 Jul**. Base low **$0.015145 on 5 Aug, -66%**. Base ran 5 to 22 Aug (~18 days). Real leg started 23 Aug. Now $0.8686, **19.7x off that low**.
- **Revenue FELL through the base**: ~$145,832 (10 Aug) down to **$54,986 on 20 Aug, its lowest print of the month**, while the price sat flat-to-up. Revenue and price then rose TOGETHER from 23-24 Aug. Usage did not lead; it was coincident.
- **WHAT WAS ACTUALLY TRUE, and it is a better and more testable claim: at the 5 Aug low PONS was a $12,224,774 market cap earning $93,228/day = $34.03M annualised = 278% OF ITS OWN MARKET CAP PER YEAR, with 80% of that doc-claimed to TWAP-buy and burn the token.** You were not buying rising usage. You were buying a live cash pipe pointed at the token at 0.36x its annual output. (Supply 700,211,659, mcap = FDV.)
- **STANDING: do not screen post-dump infra on "usage rising while price falls". Screen on CASH ROUTED TO THE TOKEN as a share of market cap, and verify the route on chain.** Usage divergence is a nice-to-have; the routed pipe is the mechanism.

## THE THREE TESTS used, and the answer
1. **Extended then pulled back** - drawdown from the TRUE INTRADAY high (GT daily candles, highest-volume USD-quoted pool, partial day dropped), never closes.
2. **Launch cohort has stopped** - every first-hour receiver + one hop in the first 24h, contracts and the v4 PoolManager removed by `eth_getCode` (7702 wallets KEPT, they are people). A sale counts only into a DexScreener-listed pool for that token. Live `balanceOf` for what is left.
3. **Cash pointed at the token** - last three COMPLETE days of protocol revenue, annualised, over market cap; then read on chain where it goes (holders / lockers only / nowhere), including a burn-and-mint count from Transfer events to and from 0x0 and 0xdead.

**ANSWER DELIVERED: nothing on the chain is a clean PONS repeat. DELTA passes tests 1 and 2 outright and scores ZERO on test 3.**

## THE BOARD, 6 Sep 2026 (mcap / % off true high / days since high / revenue last-3-complete-days / trend vs prior 7d / annualised revenue as % of cap / cohort still holds / cohort sold 7d / sold 24h)
| Token | contract | mcap | off high | rev/day | trend | routed %cap | route | holds | sold 7d | sold 24h |
|---|---|---|---|---|---|---|---|---|---|---|
| DELTA | 0xe8ffd7e24187F72afB08d75B1bb13088A989a791 | $11,649,041 | -69% | n/a | n/a | **0%** | **nothing** | 0.19% | **0.000%** | 0.000% |
| UP | 0x57C0E45cB534413D1C20A4240955d6bB250BB4F1 | $14,608,499 | -50% | $96,839 | 1.37x | 242% | lockers only | 0.02% | **0.000%** | 0.000% |
| PROLOGUE | 0xb9972ca7188e511174947e3936a5315ac7073277 | $8,235,530 | -59% | $0 | n/a | **0%** | **nothing** | 6.43% | 0.001% | 0.000% |
| SLVR | 0x791229E3EbD6CFdC3D8157f48722684173C29aD9 | $1,861,624 | -69% | $68,974 | 2.41x | **1,352%** | holders | 0.99% | 0.006% | 0.000% |
| HKRS | 0x833153ecb2c183702907e1142317a707137af954 | $1,245,829 | -38% | $1,462 | 1.53x | 43% | **holders, burn verified** | 16.72% | **0.000%** | 0.000% |
| BOW | 0x451b42A15100C340CA12F7c66DE06fac5EA2D751 | $9,187,497 | -29% | no adapter | - | - | unverified | 5.44% | 0.006% | 0.000% |
| HOOKR | 0x18e674231a58c239dc7daedcffe15ec3a24cff5c | $17,835,865 | -42% | no adapter | - | - | unverified | 7.93% | 0.093% | 0.000% |
| INJOH | 0x2cC0FAC44B8252f6B10208B091aFf2c94B4da77D | $4,820,165 | -32% | no adapter | - | - | holders | 7.53% | **2.132%** | 0.100% |
| MANCER | 0xc72F232a6869e6CF34dC06129AfFD07F8a2a246A | $2,804,199 | -68% | no adapter | - | - | unverified | **18.79%** | **3.324%** | 0.276% |
| RIP | 0xe4b8723d202FDD0298B1D7496A86656b2986C532 | $1,631,469 | -59% | $4,700 | 3.56x | 105% | unverified | 13.41% | **14.348%** | 0.091% |
| LUTE | 0xD1e861CC5Eee7eA88649206b74504D78CCD7AEeA | $4,247,446 | -52% | $7,646 | 1.34x | 66% | unverified | 0.00% | 0.000% | 0.000% |
| POOLS | 0x385b36Ff682Ab4C76E7c37A66b96aABC466471d5 | $1,320,129 | -70% | n/a | - | - | unverified | 0.39% | 0.008% | 0.000% |
| RVH | 0x96765066f6a040a21EB027167D2315B707c82633 | $696,698 | -67% | $1,404 | 1.16x | 74% | unverified | 0.14% | 0.001% | 0.000% |
| STONX | 0x570c5aa79c798e7a418412cc8399ae5bcce570c5 | $851,385 | -42% | $1,116 | 0.81x | 48% | unverified | 0.30% | 1.890% | 0.278% |
| WTH | 0xb8Fa8010833463Aac5595b55B9045479239EfF79 | $4,004,365 | -21% | $887 | - | 8% | nothing until 4 Sep | 2.47% | 0.633% | 0.031% |
| VAULT | 0xFdae23CE76018Da62507bB5ef20e6ef5450e8312 | $453,920 | -88% | n/a | - | - | unverified | 0.30% | 0.005% | 0.000% |
| SELECT | 0xFc39e99e78524d7c891669F70d5aa92aB1041eb6 | $163,320 | -80% | $867 | 0.54x | 194% | unverified | not measured | | |
| DEEP | 0x1DA24f6Bb623b9d1aFEae3F3146659A2662D6d27 | $1,175,338 | -74% | $1,483 | 0.32x | 46% | unverified | not measured | | |
Also on the board with no revenue adapter and no cohort resolved: TWO -85%, WIRE -86%, CLAN -82%, WALL3 -89%, NET -56%, DTF -41%, GWOOD -74%, VEX -84%, YARD -57%, RAXOL -60%, BACKED -60%, SEED -61%, OPTIMUS -56%, PEEPS -82%, VIM -89%, FROTH -75%, DERP -92%, SLUICE -89%, VYNEX -93%, MAST -88%, ACRE -100%, BRICKS -99%.
Reference rows, not pullbacks: **PONS -11% off high, $608,203,847, rev $1,402,183/day, trend 2.16x, 84% of cap routed/yr, cap/annRev 1.19x.** **PAIR +34% ABOVE its prior high, $23,897,902, rev $19,600/day.**

## PER-NAME FINDINGS
### DELTA - passes the seller test and the usage test, fails the mechanism test
- **Cohort finished: 93 EOA/7702 wallets, 0.19% of supply ($22,154), 0.000% sold into a pool in 24h, 72h AND 7d.** The fall from $29.3M (31 Aug) to $11.6M was NOT launch-insider distribution. All-holder selling into pools was 133% of supply over 7d, i.e. general rotation and churn.
- **Product is growing hard: own endpoint `https://deltaliquidity.app/ix/pnl/tvl` returns TVL $1,989,012, 203 pools, 12,822 open rungs (4,651 v3 + 8,171 v4)**, against $1,633,818 / 116 pools on 4 Sep. **+21.7% in two days while the token fell 45%.** MC/TVL **5.86x**, down from 16.3x on 31 Aug, the cheapest it has ever been on its own metric and now inside PROLOGUE's 5.7x band.
- **THE KILL: nothing points at the token.** 1% claim fee lands in `0x266311a44e70363b3d168bcb38cf38cbf2ccc51f`, zero bytes of code. **Burns over 7d: 53 tokens in 2 dust events. Mints: 0.**
- TRIGGER TO FLIP: any non-dust burn, or the fee recipient becoming a contract. Unchanged from the 31 Aug screen and still unannounced.
### UP - the biggest pipe, wrong claimants
- Cohort 15 wallets, 0.02% held, **0.000% sold in 7d**. -50% off the 1 Sep high of $1.4470, +35% off yesterday's low.
- Revenue **$96,839/day (last 3 complete), 1.37x the prior week, still below the $150,880 peak of 1 Sep.** 242% of circulating cap per year.
- **89.1% of supply locked in VotingEscrow `0x5d321dE36F0bf98D92b291280514F3878582B7B6` and 100% of voter fees go to those lockers. Circulating holders receive nothing.** Like-for-like FDV/annualised-revenue ~10x against AERO ~20x (AERO mcap $535,602,557, ~50% float, rev30d $4,299,997).
### SLVR - clears PONS's bar on the numbers, fails on customers and dilution
See the appended block in [[screen-log-outcomes-4]] for the full 6 Sep bettor and staker measurement. Cohort 340 wallets, 0.99% held, 0.006% sold in 7d.
### HKRS / Hookers - the only name passing all three tests, at a size where that is easy
- **Live buy-and-burn verified on chain: 4,848,070 HKRS burned over 7d (0.485% of supply) in 23 events, 27 events over 14d, ZERO minted. Every burn's source is the Uniswap v4 PoolManager singleton `0x8366a39cc670b4001a1121b8f6a443a643e40951`, i.e. bought out of the pool and destroyed.**
- **But the burn is SHRINKING with revenue: ~2,032,031/day 13 days ago -> ~400,000/day now.** At $0.001246 that is roughly $498/day of bid against $1,462/day of revenue, so ~34% of revenue is routed.
- Cohort has sold **0.000% in 7d but still holds 16.72% ($214,389) across 60 wallets. Frozen, not exhausted** - the distinction that matters.
### The kill list
- **RIP is the trap**: revenue up 3.56x, 105% of cap routed per year, and the cohort sold **14.348% of supply in 7 days** and still holds 13.41%. Every screener will light up on it.
- **MANCER**: largest live overhang measured, 4,250 cohort wallets holding 18.79% ($1,249,692) and still selling 3.324%/7d.
- **INJOH**: best fee mechanism in the launch-tooling tier but still trickling, 2.132%/7d and 0.100% in the last 24h. Revisit when 24h reaches zero.
- **LUTE**: cohort clean and revenue up 20x since early Aug, but 66.7% of supply sits in two team SafeProxy multisigs with no published lock ([[rh-ve33-dexes]]).
- DEEP (rev -68% trend 0.32x), SELECT (0.54x), STONX (0.81x) are all cheap because they are shrinking.

## CHAIN CONTEXT, 6 Sep 2026 (mandatory block)
DeFi TVL **$908,649,407**, +29.6% 7d, **+108.8% 30d**. DEX volume 24h $1,612,673,902, 7d $11.30B, 30d $24.12B, **+176.5% on the month**. Chain fees 24h $18,621,632, 30d **$188,156,495, +421.7% on the month**. 152 protocols on the fee board, 85 chain-native. Largest single earner is still **GMGN at $18.9M/30d with no token**. Every token above sold off INTO an accelerating chain.

## SOURCES THAT FAILED THIS RUN, both billing, neither a bug
- **Dune: HTTP 402 on execute, "would exceed your configured datapoint limit per billing cycle".** Same wall as the morning run. Cost the chain-wide holder-arrival / holder-velocity layer; the cohort work was rebuilt on direct RPC `eth_getLogs` instead and is arguably better, just slower (~2-10 min per token).
- **twitterapi.io: HTTP 402 Payment Required on every query. Out of credit.** No X claim audit, no tier-one coverage check in this screen. **ACTION FOR BEN: top up twitterapi.io and raise the Dune datapoint limit; two of the four standing sources are currently dark.**
- No Mac bridge on this session, so no fomo theses and no purealpha.

## METHOD NOTES worth reusing
- **DefiLlama's daily series final row is a ROLLING 24h window, not a complete day. Drop it before averaging.** This is the same class of error as scoring the partial UTC OHLCV day and it corrupted two live falsifier legs today.
- **Cohort building on this chain: blocks are 0.100843s (~857,000/day), so a block number can be computed arithmetically from a timestamp instead of binary-searching.** `eth_getLogs` on the RH public RPC times out above ~200,000 blocks with only a topic0 filter; 400,000 works once topic2 is constrained to a pool set. Chunk and halve on error.
- **The cohort MUST be filtered by `eth_getCode` before anything is counted.** A first pass without it reported DELTA insiders as having sold 62.5% of supply in 7 days; the sellers were routers and the v4 PoolManager. With the filter the true figure is 0.000%. Keep 7702 delegated wallets (`0xef0100` prefix) - they are people.
- **A GT daily series' first bars can be dust (PONS closed at 6.0e-18 on 16 Jul). Filter the price series to `p >= max(p)/3000` before charting or computing anything on a log scale.**
- `pkill -f <name>` in the same bash invocation as other commands containing that string kills those too. Send the pkill on its own.
- Three parallel ledger workers against the RH RPC is the practical ceiling; more and the 429 backoff eats the gain.