TL;DR
This memo reviews the Arcus research brief against refreshed live data. Arcus is the dYdX-team DEX on Robinhood Chain, with live perps, zero-fee tokenized-stock spot trading, and an early pToken primitive.
Two findings stand out.
1. The brief's baseline numbers hold up. I refreshed every headline metric against DefiLlama and Arcus primary sources (2026-09-05, ~05:15 UTC). TVL, 30d spot volume, and 30d revenue land within a few percent of the embedded baseline. Cumulative perp volume is higher ($2.457B vs ~$2.38B), while 24h perp volume is lower ($75.4M vs ~$93M). Nothing in the brief is stale enough to invalidate it.
2. The brief has one serious blind spot that changes the answer. It never names Lighter, which runs $5.86B in 30d perp volume on the same Robinhood Chain versus Arcus' $2.03B, holds 3x the TVL, and is listed alongside Arcus, not instead of it, in Robinhood's own launch announcement. The brief's Section 3 asks whether Robinhood Chain gives Arcus a "distribution moat." Live data says no: the distribution is non-exclusive, and Arcus is currently the #2 perp venue on its own home chain.
My verdict on the underlying asset: WATCHLIST, not "high-conviction market infrastructure."
Refreshed Project State Matrix
Arcus went live on Robinhood Chain around 2026-07-10 (first DefiLlama TVL print: $5.25M). Volume has compounded fast: 83% of all-time perp volume occurred in the last 30 days. But the absolute base is still small, and the growth is concentrated in perps, not the tokenized-equity spot product that the brief treats as the core thesis.
| Product | Status | Volume (30d) | TVL | Revenue (30d) | Confidence |
|---|---|---|---|---|---|
| Arcus Perps | Live, scaling | $2.029B | $21.57M | $338.1K | CONFIRMED (DefiLlama) |
| Arcus Spot (Stock Tokens) | Live, 0% fee | $206.3M | n/a (RFQ) | $0 | CONFIRMED |
| Arcus pTokens | Live ~2026-08-25 | not tracked | $926.5K | n/a | CONFIRMED |
| Cross-margin / stock collateral | Partial rollout | — | — | — | OFFICIAL ROADMAP |
| Arcus token | Not launched | — | — | — | SPECULATIVE |
The curve is real growth, not a single deposit: TVL is +13.8% over 30 days and 4.1x since launch, all denominated in USDG on Robinhood Chain.
Supporting cuts, all as of 2026-09-05: perp volume 24h $75.4M / 7d $459.7M, cumulative $2.457B; spot cumulative $324.5M; 1,621 active addresses / 145 new addresses / 159,072 transactions in 24h. Open interest reads ~$13-15M on DefiLlama's cached protocol page. I could not re-fetch it live, so treat that one as MEDIUM CONFIDENCE.
The economics are healthier than the size suggests. Fees of $366K on $2.029B of perp notional imply a ~1.8 bps blended take rate, and $338K of 30d revenue against $21.6M of TVL annualizes to ~19% revenue-on-TVL, better capital efficiency than Lighter's ~12.7% on the same chain.
DefiLlama's trailing annualization: $4.46M fees / $4.11M revenue. Note that spot generates zero revenue by design (0% fees), so 100% of protocol income is perps. The brief's Section 20-style "protocol revenue" question has a blunt answer: Arcus is a perp exchange with a free stock-token loss leader attached.
The Robinhood Question: Dependency Without a Moat
This is where the brief's framing needs correcting. Robinhood Crypto has invested in Arcus and Stock Tokens are issued on Robinhood Chain infrastructure. But Robinhood's own mainnet announcement names "Uniswap, Rialto, Lighter, Arcus, and 1inch" as the DEXs serving Stock Token spot trading in the Robinhood Wallet. Arcus is one of five, and Uniswap was named a day-one liquidity partner.
That is a distribution channel, not a distribution moat.
The competitive scoreboard on Arcus' home chain, per DefiLlama:
| Metric | Arcus | Lighter (RH Chain) | Read |
|---|---|---|---|
| Perp volume 30d | $2.029B | $5.861B | Arcus holds ~26% of the two-venue split |
| TVL | $21.57M | $61.80M | Lighter TVL +462% in 30d |
| Fees 30d | $366K | $870K | Arcus take rate is higher (1.8 vs 1.5 bps) |
| Derivatives TVL rank | #14 | #7 | Both sub-scale globally |
Against the wider field, Arcus is 0.95% of Hyperliquid's $213.9B 30d perp volume, but already 1.76x its own predecessor, dYdX v4 ($1.152B). That pairing is the honest read on the dYdX lineage: the team's new venue has outgrown the old one in two months, and still barely registers versus the category leader.
Robinhood Chain itself is credible: $865M TVL, the #10 chain globally. But Arcus is only the #6 protocol on it (2.5% of chain TVL), behind Morpho Blue ($505M), Steakhouse ($472M), Uniswap ($232M), Lighter ($61.8M) and Spark ($31.1M).
Scoring the brief's requested metrics:
| Metric | Score |
|---|---|
| Robinhood Dependency Score | ~85/100 |
| Robinhood Distribution Advantage Score | ~45/100 |
| Ecosystem Beta | Extreme |
Architecture and the pToken Primitive
The brief asks whether pTokens are "a genuinely new and valuable financial primitive." The design is genuinely clean; the adoption is not there yet.
Arcus deliberately runs two execution models: a CLOB for perps, where price discovery happens on the book, and RFQ for Stock Tokens, because a token's fair value is set on the underlying exchange, not on-chain. Resting orders go stale outside NYSE hours, and makers quote defensively wide. Traders sign an intent with a hidden minimum; makers return firm signed quotes; settlement is atomic.
This is a more sophisticated answer to tokenized-equity market structure than the AMM approach most competitors use, and it is the strongest single argument in the bull case.
pTokens wrap an entire Arcus perps account as an ERC-20: a leveraged-ETF structure with threshold-triggered rebalancing rather than a daily close, NAV oracle-signed and verified on-chain. First generation: pBTC (1x), pBTC3x, pHOOD3x.
Ten days after launch, total pToken TVL is $926.5K, ranking #57 by TVL among 420 derivatives protocols.
Architectural inference: the primitive is sound, and the composability story is coherent on a chain that has both a DEX layer and lending venues. Users can hold it in a wallet, trade it on Uniswap, or eventually post it in Morpho.
Current observation: demand is negligible so far. Do not underwrite pTokens as a value driver on 10 days of sub-$1M data.
Red Flags the Brief Should Surface
Running the brief's own falsification engine against live evidence:
| Red flag | Why it matters |
|---|---|
| Jurisdiction exclusion is structural, not temporary. | Arcus states directly that it "is not available in the U.S., Canada, United Kingdom and other restricted jurisdictions." The venue built on US-equity exposure cannot serve US traders. Robinhood's 27.7M funded accounts are largely in exactly the markets Arcus is fenced out of. |
| Stock Tokens are not equity. | Per Arcus' own disclosure, they are a "contractual claim against the Issuer for a cash Redemption," carrying tracking divergence, redemption limits and issuer credit risk. This caps how far the "tokenized equity exchange" narrative can be pushed. |
| Spot volume is lumpy and unmonetized. | 24h spot of $5.28M sits ~59% below the trailing 7d daily average of $12.9M. Zero-fee volume on an RFQ venue is also the easiest kind of volume to inflate. |
| Arcus is losing the home-chain perp race. | Lighter is growing TVL faster (+462% vs +13.8% in 30d) on the same distribution. |
| No token, no way to express the thesis. | No TGE or public sale exists in any dataset I queried. Reporting that dYdX holders receive a future Arcus token allocation is SPECULATIVE (Tier 4/5 only), and DYDX fell ~45% on the launch announcement. The market read Arcus as value leaving dYdX, not accruing to it. |
| Fee history is 30 days deep. | Cumulative fees exactly equal 30d fees, meaning the tracking adapter is new. There is no multi-quarter revenue trend to underwrite. |
Scoring and Scenarios
Applying the brief's weighted model to refreshed data:
| Dimension | Weight | Score | Note |
|---|---|---|---|
| Technology & Architecture | 15% | 4.0 | RFQ/CLOB split, pTokens, dYdX pedigree |
| Product / PMF | 15% | 3.0 | Perps working; spot and pTokens unproven |
| Liquidity / Market Structure | 15% | 2.5 | ~$22M TVL, ~$13-15M OI, thin books |
| Protocol Economics | 10% | 3.0 | $4.1M annualized revenue, 1.8 bps take |
| Distribution | 15% | 3.0 | Robinhood-backed but non-exclusive, US/UK excluded |
| Market Opportunity | 10% | 4.0 | Tokenized equities ~$2.3B market cap, doubling |
| Network Effects | 10% | 2.0 | Ceding share to Lighter on home chain |
| Risk / Security | 10% | 2.5 | Jurisdiction, issuer credit, unverified audits |
Protocol Quality 68/100 · Market Structure 48/100 · RWA Infrastructure 62/100 · Competitive Moat 38/100 · Investment Quality 61/100.
| Scenario | Prob. | Annualized perp volume | TVL | Revenue | Trigger |
|---|---|---|---|---|---|
| Bull | 20% | >$50B | >$150M | >$25M | Takes perp share back from Lighter; stock-collateral cross-margin ships; pTokens get real DeFi integrations |
| Base | 50% | $20-30B | $30-60M | $5-10M | Differentiated but sub-scale niche venue; token launch is the main catalyst |
| Bear | 30% | <$15B | flat/down | <$4M | Spot stays lumpy and unmonetized; Lighter/Uniswap absorb Robinhood Wallet flow |
Leading indicators to watch weekly:
- Arcus vs Lighter perp volume share on Robinhood Chain.
- Spot volume floor: does the 7d daily average hold above ~$7M?
- pToken TVL crossing $10M.
- Jurisdictional expansion, the single item that would re-rate the distribution score.
Conclusion
The original brief is well-constructed, and its embedded baseline is accurate as of today. Where it needs an edit is the premise: it asks whether Arcus can become "the dominant Robinhood Chain financial market" while assuming the Robinhood relationship is a moat.
It is not. The chain hosts five competing DEXs, and the direct perp competitor is running 2.9x Arcus' volume with 3x the TVL. Section 3 should be rewritten around contested distribution, and a Lighter comparison belongs in Sections 27-29.
On the seven final verdict questions:
| Question | Verdict |
|---|---|
| PMF in tokenized stock trading | Not yet proven: unmonetized and lumpy |
| Serious perp competitor | Yes, regionally; no, globally |
| Durable Robinhood advantage | No, it is shared |
| pTokens genuinely new | Yes as design, no as adoption |
| Sustainable revenue | Real but small at ~$4.1M annualized |
| On-chain prime brokerage | Architecturally plausible, empirically distant |
| Global 24/7 capital markets layer | SPECULATIVE |
Bottom line. Classify Arcus as WATCHLIST, not high-conviction infrastructure. It is the second-best perp venue on the tenth-largest chain, with the best market-structure design in tokenized equities and no token to buy. Re-underwrite if Arcus takes perp share back from Lighter, or if a token launch gives the thesis an instrument.
Data refreshed 2026-09-05 ~05:15 UTC from DefiLlama, arcus.xyz, and Robinhood's newsroom.
