On September 11, the XRP Ledger held $1.126 billion in stablecoin value. The pools connecting that value to XRP itself held $4.6 million.
That ratio is 0.41%. It is the only number in this dataset that matters. Not the 22.2% month-over-month expansion of XRPL stablecoin supply. Not the 152% quarter-over-quarter surge in decentralized exchange volume. Not the 29.7% monthly appreciation in XRP's price, which has already surrendered 8.7% over the trailing seven days.
I have watched this exact pattern across three cycles. A network reports extraordinary growth in an adjacent category. The market extrapolates that the native asset must be capturing that value. Nobody reconciles the two figures until the position is already underwater. In May 2022 I pulled Anchor Protocol deposit flows at 3 a.m. CET and watched pool composition invert before a single headline appeared. The lesson was never that a protocol was fraudulent. The lesson was that aggregate metrics are lagging indicators of a story, while pool-level data is a leading indicator of a fact.
The XRP Ledger publishes both. Most participants are reading the wrong one.
Context: What This Network Actually Is
The XRP Ledger has been in continuous operation since 2012 โ twelve years of uptime without a consensus failure. It does not use proof-of-work or proof-of-stake. It runs a federated consensus protocol in which validators are selected by a Unique Node List, currently comprising roughly fifteen independent operators, a subset of which are controlled or influenced by Ripple Labs. Throughput sits near 1,500 transactions per second with three-to-five-second finality. Solana advertises 65,000. XRPL does not need to win that race.
The architectural trade-off is deliberate. The base ledger is not Turing-complete. There is no virtual machine executing arbitrary bytecode. Functionality arrives through native transaction types and amendment voting โ a governance process that requires validator supermajority. Complex logic is pushed to sidechains. This narrows the attack surface considerably, and it means every line of logic on the base layer has been reviewed by thousands of eyes over more than a decade.
What the ledger does have is a native order book that predates almost every competitor, and an automated market maker introduced only in 2023 โ eleven years after genesis. That recency matters, and I will return to it.
The mechanism that anchors the entire bull case is auto-bridging. XRPL's pathfinding engine compares a direct order book between two issued currencies against a two-hop path routed through XRP. If the composite XRP path yields a better rate after fees, the engine takes it. USDT to EURC may execute as USDT to XRP to EURC. The design intent is that XRP becomes a universal intermediary in stablecoin settlement โ a monetary switchboard.
Here is my methodology, and its boundary. I pulled monthly DEX volume from ledger aggregates, stablecoin TVL from registry dashboards, single-pool depth from AMM state, and reconciled Ripple's published RLUSD attestation against ledger-reported supply. What I could not pull โ because it does not exist in public form โ is path-level attribution, meaning which proportion of executed trades actually moved through XRP, and inventory duration, meaning how long intermediaries hold the asset between legs. That absence is not a footnote. It is the thesis.
Core: The Evidence Chain
Auto-bridging is a fallback, not a mandate. The default path is written into the pathfinding parameters, and almost nobody reads them. XRP is selected only when it produces a better composite rate than a direct pair. That is an engineering decision with a counterintuitive consequence: XRP's routing share is inversely proportional to the maturity of the stablecoin pairs themselves. When a direct USDT/EURC book is thin, XRP wins the route. When that book deepens, XRP is bypassed. The mechanism that supposedly guarantees demand for XRP degrades as the ecosystem it serves succeeds. The alpha is in the silenced code, and the code says XRP is a liquidity subsidy for pairs that have not yet attracted enough liquidity.
In 2020 I wrote a Python script that tracked oracle update latency between Uniswap and SushiSwap and surfaced a $2.4 million mispricing. The fund earned 15% in 48 hours. Nothing in that trade showed up in any public volume dashboard. Path-level inefficiency is real, measurable, and structurally invisible to the metrics most people quote. XRPL has the same blind spot, and its community quotes the same metrics.
The fee burn is arithmetically irrelevant. Every XRPL transaction destroys a minimum of 10 drops โ 0.00001 XRP โ and the fee is burned rather than paid to validators. That design makes the asset deflationary in proportion to usage, which sounds like a value capture flywheel. Run the numbers.
Monthly DEX volume is $253.1 million. Assume an average trade notional of $10,000 โ generous for a network at this stage. That yields roughly 25,310 transactions. At the minimum fee, those transactions destroy 253,100 drops in total. That is 0.2531 XRP, for the entire month. Even if ledger load escalated fees by a factor of one hundred, the figure lands near 25 XRP burned per month. Against that, Ripple's escrow schedule releases up to one billion XRP monthly, though the majority has historically been returned to escrow unspent. Scarcity is an algorithm, not a belief system, and the algorithm here is a constructor argument specifying a 100 billion cap. The effective float is a function of escrow discretion โ a policy, not an invariant. Anyone modeling the release schedule as fixed code is modeling a policy as though it were a constraint. I learned in May 2022 that models fail precisely at the moment policy changes.
The RLUSD reconciliation gap is the least discussed and most consequential number in this dataset. Ripple's published figure for RLUSD in circulation sits near 2.396 billion tokens. The ledger reports approximately 1.032 billion. The gap is 1.364 billion โ 57% of the attested figure cannot be located on-chain under a straightforward reading.
There are benign explanations. Multichain issuance with lagged attestation cycles. Different definitions of circulating supply, where minted-but-treasury-held inventory is counted differently from outstanding float. Custodial accounting conventions that reconcile quarterly rather than continuously.
But for a stablecoin, the attestation is the product. In 2017 I audited the token distribution contracts of fifteen pre-sale offerings. I found a reentrancy vulnerability in one distribution mechanism that delayed its launch. The bug was not in the transfer logic. It was in the distance between what the marketing documentation promised and what the constructor function actually permitted. The gap between an attestation and a ledger is the only place worth looking, in every cycle, without exception.
In 2025 I helped architect a validation framework that used zero-knowledge proofs and decentralized oracle networks to verify AI-generated data before it reached execution logic for institutional clients. That project raised $50 million precisely because 'trust the report' does not scale to size. The cryptographic tooling to prove reserve composition on-chain exists today and is production-grade. Its absence from a regulated stablecoin's disclosure is a design choice, not a technical limitation.
Pool depth is the metric that determines usability, and it is not the metric anyone quotes. The XRP/RLUSD AMM pool holds $4.6 million against $1.126 billion in network stablecoin value. On a constant-product curve, that implies roughly $2.3 million on each side. A $250,000 swap against that reserve produces a price impact approaching 10%. A $1 million swap approaches 30%. No treasury desk executes a settlement transfer at a 30% haircut. The corridor is not functional at institutional size, and the 0.41% capture ratio tells you the market already knows this.
Compare the competitive frame. Tether operates across a stablecoin base near $120 billion. Circle's USDC sits near $35 billion. Ripple's own stablecoin float is roughly $240 million. This is not a contest of technology. It is a contest of depth, and depth compounds.
In 2021 I built a rarity scoring model that analyzed more than 50,000 Bored Ape traits against historical sales. The signal was not in the rare traits โ it was in twelve traits the market classified as common but which were statistically significant for floor price stability. Depth of bid, not frequency of occurrence, predicted resilience. The same asymmetry applies here. The $253.1 million in monthly DEX volume is a frequency metric. The $4.6 million pool is a depth metric. Only one of them survives a stress event.
The velocity numbers contain a contradiction worth naming. Quarterly DEX volume grew 152%. Weekly volume fell 25.4% to $30.6 million. Both are true, and they describe different things: a structural expansion in participation across the quarter and a sharp deceleration in the most recent window. Without pair-level attribution, the aggregate cannot distinguish between stablecoin-to-stablecoin activity and activity that genuinely routes through XRP. Volume is a throughput measurement. Utility is a routing measurement. They are not the same number, and no public dashboard on this network separates them.
The AMM itself is eleven years younger than the ledger. Introduced in 2023, it is a relatively untested liquidity venue inside a highly tested consensus system. Hybrid order book and AMM routing creates fragmentation: the same stablecoin pair can trade at different effective prices depending on which venue the router selects. That fragmentation is an arbitrage surface, which is genuinely interesting, and it is also a slippage risk that compounds the depth problem. Arbitrageurs will close the spread. They will not manufacture the depth.
Contrarian: Correlation Is Doing All the Work
Correlations are the lie; liquidity is the truth.
Three claims carry the current narrative, and each is a coincidence wearing a causal costume.
The first claim: stablecoin supply grew 22.2%, therefore XRP demand grew. Both events occurred in the same window. If XRP were the mandatory bridge, pool depth would scale with network stablecoin supply. Supply expanded by $204 million month-over-month. The XRP/RLUSD pool did not scale proportionally. It sits at 0.41% of the base it supposedly intermediates. A bridge that captures four-tenths of one percent of the traffic crossing it is not infrastructure. It is decoration.
The second claim: DEX volume grew 152%, therefore XRP utility grew. Volume measures throughput. Utility measures routing. A network can process enormous stablecoin-to-stablecoin volume while XRP sits entirely idle on the sideline. Absent path-level data, this claim is unfalsifiable, and unfalsifiable claims get repriced violently when they finally meet evidence.
The third claim, and the one that deserves the most scrutiny: RLUSD adoption creates bridge demand for XRP. This inverts the actual design. RLUSD is engineered as a settlement asset โ a token whose purpose is to eliminate the need for a volatile intermediary in a payment corridor. If a bank pairs USD directly against a local currency, XRP has no seat at the table. The stablecoin's success is not automatically the bridge asset's success. In many corridor configurations, it is the bridge asset's replacement.
I have to acknowledge the counter-argument, because ignoring it would make this analysis dishonest. Ripple's distribution network is a real moat that no dashboard can measure. Banking relationships with institutions of the scale of Standard Chartered, Mizuho, and Sumitomo Mitsui represent off-chain commitments with onboarding costs that dwarf anything visible in AMM state. My 2025 institutional work taught me that the most commercially significant data is usually the data that has not reached a public ledger yet. On-chain analysis is structurally blind to a signed term sheet.
So the defensible position is not that XRP captures nothing. It is that the value capture mechanism is currently unmeasurable at the resolution required to price it. Assets with unmeasurable value capture are priced by narrative. And narrative-priced assets get repriced by narrative, at narrative speed, usually between a Thursday and a Friday.
Takeaway: The Signals That Will Settle This
Four measurements will resolve the question, and none of them require a new thesis.
Watch XRP/RLUSD pool depth against a $20 million threshold. At that level, each side holds roughly $10 million, and a $250,000 institutional swap costs about 2.4% in slippage instead of 10%. That is the first point at which a treasury desk can plausibly execute. A 4.3x move in pool depth is the single cleanest tell that routing demand is real rather than advertised.
Watch for third-party path-level attribution โ any disclosure showing what share of executed DEX volume actually transited XRP. If this data does not appear within two quarters despite the ecosystem's growth claims, treat its continued absence as the answer.
Watch the monthly escrow return ratio against actual burn. If net float expansion continues while burn remains sub-hundred-XRP, the supply dynamics are policy-driven, and price is therefore a policy outcome, not a protocol outcome.
Watch RLUSD's attestation-to-ledger reconciliation over a full thirty-day cycle. Convergence to under one percent closes the credibility question. Persistent divergence at 57% means the most regulated asset in this ecosystem is also its least verifiable.
The ledger remembers what the marketing forgets. Due diligence is the only hedge against chaos โ and in a sideways tape, where positioning is decided by which signal you trusted first, the difference between $1.126 billion and $4.6 million is the whole trade.