DEV Community: Nyx Lesende

Measuring Order Book vs AMM Share on the XRP Ledger

Nyx Lesende — Mon, 21 Sep 2026 20:00:18 +0000

When the AMM amendment activated on the XRP Ledger in 2024, the expectation was that liquidity would migrate into pools the way it did on every smart-contract chain before it.

You do not have to speculate about whether that happened. The ledger publishes both numbers.

Getting the split

const { H24 } = await (
  await fetch('https://api.xrpl.to/v1/stats')
).json();

const ob  = H24.globalVol24hOrderbook;
const amm = H24.globalVol24hAMM;

console.log('order book share:',
  (ob / (ob + amm) * 100).toFixed(1) + '%');

Right now that prints 89.8%.

Venue	Volume (XRP)	Trades	Avg trade (XRP)
Order book	8.43M	40,758	206.77
AMM pools	957.2K	78,057	12.26

The interesting number is the average trade size

The order book takes a minority of the trade count but the large majority of volume. Big orders route to the book; small ones are content in pools.

That is the whole story in one ratio, and it inverts what you see on most chains.

Why the order book survived here

The XRPL has had an on-chain central limit order book since 2012. AMMs exist elsewhere largely as a workaround: where posting and cancelling quotes costs meaningful gas, continuous market making is uneconomic, so liquidity gets parked in a passive curve instead.

On the XRPL a cancel costs a fraction of a cent and confirms in three to four seconds. Professional quoting is viable on-chain, so the book stays deep.

Market structure follows transaction economics. The industry generalised from expensive-blockspace chains and concluded AMMs had won outright. This is a live counterexample.

Tracking it over time

async function sample() {
  const { H24 } = await (await fetch('https://api.xrpl.to/v1/stats')).json();
  const ob = H24.globalVol24hOrderbook, amm = H24.globalVol24hAMM;
  return {
    at: new Date().toISOString(),
    obShare: ob / (ob + amm),
    traders: H24.uniqueTraders24H,
  };
}

// append to a file on an interval and you have a time series

Both figures are 24-hour rolling windows, so sampling more often than hourly mostly gives you overlapping data.

Practical takeaway

If you are routing orders, optimise for total depth rather than picking a venue — the XRPL consumes book and pool liquidity in the same trade path. For liquid pairs the book usually gives the better fill because it honours price-time priority instead of filling along a curve.

Live prices, holder counts, trustlines and order-book depth for every XRPL token are at xrpl.to.

Trustlines: Why XRPL Holder Counts Are Hard to Fake

Nyx Lesende — Sun, 20 Sep 2026 20:02:24 +0000

Most chains let anyone push a token balance into any address for free. That is why "holders" is a nearly meaningless metric almost everywhere — an airdrop to 100,000 addresses costs the sender gas and nothing else.

The XRP Ledger works differently, and the difference is worth understanding if you are building anything that reads token data.

There is no token contract

XRPL tokens are not contracts. Issuance, transfer, trading and freezing are protocol features. A token is a balance denominated against an issuer account, and to hold one you must first create a trustline to that issuer.

That trustline is an explicit, on-ledger opt-in. Nobody can push a balance at you without it.

Trustlines cost the holder money

Each trustline raises the holder's XRP reserve requirement. The reserve is not a fee — you get it back when you close the line — but it is capital locked for as long as you hold the token.

The consequence: every holder made a small economic commitment. You cannot manufacture a million holders cheaply, because each one costs the holder, not you.

Current leaders by holder count:

Token	Holders	Trustlines
SOLO	215,509	283,994
RLUSD	67,347	98,052
CORE	62,412	71,659
VGB	41,757	47,013
PHNIX	28,881	29,677
ARMY	22,006	22,549

Reading the holders-to-trustlines gap

Both numbers are published, and the ratio is more informative than either alone:

const { tokens } = await (await fetch(
  'https://api.xrpl.to/v1/tokens?limit=50&sortBy=holders&sortType=desc'
)).json();

for (const t of tokens) {
  const stickiness = t.holders / Math.max(t.trustlines, 1);
  console.log(t.name, stickiness.toFixed(2));
}

Ratio near 1 — almost everyone who ever opened a line still holds a balance
Ratio near 0 — accounts opened, traded out, and never closed the line

A low ratio is not automatically bad. It describes a token people passed through rather than held, which is exactly what you would expect from something traded actively.

What this means if you are building

Treat holders as a real signal on the XRPL in a way you should not elsewhere. But pair it with two others:

uniqueTraders24h — is the volume many accounts or a few?
Order-book depth — could you actually exit at the quoted price?

Holder count tells you about distribution. It says nothing about liquidity, and the two are frequently unrelated.

Live prices, holder counts, trustlines and order-book depth for every XRPL token are at xrpl.to.

Build an XRPL Token Screener in About 50 Lines

Nyx Lesende — Fri, 18 Sep 2026 20:00:19 +0000

There are 20,387 tokens issued on the XRP Ledger and only 1,530 of them traded today. A screener is how you tell those two groups apart.

Here is a complete one. No dependencies, runs on any Node 18+.

The whole thing

const API = 'https://api.xrpl.to/v1';

async function get(path, attempt = 1) {
  const r = await fetch(API + path);
  if (r.ok) return r.json();
  if ((r.status === 429 || r.status >= 500) && attempt < 4) {
    await new Promise(s => setTimeout(s, attempt * 4000));
    return get(path, attempt + 1);
  }
  throw new Error(`HTTP ${r.status}`);
}

function score(t) {
  // Volume per unique trader — high values mean few accounts moving a lot,
  // which is what wash trading looks like from the outside.
  const perTrader = t.vol24hxrp / Math.max(t.uniqueTraders24h, 1);
  // Trustlines that no longer hold a balance are historical interest.
  const stickiness = t.holders / Math.max(t.trustlines, 1);
  return { perTrader, stickiness };
}

async function screen() {
  const { tokens } = await get('/tokens?limit=100&sortBy=vol24hxrp&sortType=desc');

  return tokens
    .filter(t => t.holders >= 1000)          // ignore dust
    .filter(t => t.uniqueTraders24h >= 20)        // ignore one-account markets
    .map(t => ({ name: t.name, issuer: t.issuer, ...score(t),
                 holders: t.holders, vol: t.vol24hxrp }))
    .filter(t => t.stickiness > 0.15)             // most holders still hold
    .sort((a, b) => a.perTrader - b.perTrader);   // broadest participation first
}

screen().then(rows => {
  for (const r of rows.slice(0, 15)) {
    console.log(
      r.name.padEnd(12),
      String(r.holders).padStart(7),
      r.perTrader.toFixed(0).padStart(9),
      r.stickiness.toFixed(2)
    );
  }
});

What each filter is actually doing

holders >= 1000 — on the XRPL, holding a token requires an explicit trustline, and a trustline locks an XRP reserve. Every holder paid something to be counted, so this threshold means more here than a wallet count would on a contract chain.

uniqueTraders24h >= 20 — volume with three accounts behind it is not a market. This one filter removes most of the noise.

stickiness > 0.15 — the ratio of current holders to total trustlines ever opened. A very low value means people opened a line, traded, and left.

Sorting by perTrader ascending — counterintuitive, but low volume-per-trader means many accounts each trading modest size. High values mean a handful of accounts churning, which is the pattern you want to look at twice.

What it deliberately does not do

It does not read issuer flags. Two things matter and are not in this endpoint: whether the issuing account is blackholed (no further supply can be minted) and whether the issuer kept freeze authority. Both are on-ledger and worth adding via a rippled account_info call before you trust any output.

It also does not measure order-book depth. A token can pass every filter above and still have no bid you could actually sell into. Depth is the check that matters most and the one people skip.

Live prices, holder counts, trustlines and order-book depth for every XRPL token are at xrpl.to.

Query Every Token on the XRP Ledger With One Endpoint

Nyx Lesende — Thu, 17 Sep 2026 20:00:23 +0000

The XRP Ledger currently has 20,385 issued tokens, and 1,300 of them traded in the last 24 hours. Getting at that data usually means running your own rippled node or stitching together several exchange APIs.

There is a simpler path. Here is the whole thing in one fetch.

The endpoint

const res = await fetch(
  'https://api.xrpl.to/v1/tokens?limit=5&sortBy=vol24hxrp&sortType=desc'
);
const { tokens, total } = await res.json();

console.log(total);                    // 20,385
for (const t of tokens) {
  console.log(t.name, t.vol24hxrp, t.holders);
}

No API key, no signup. Output right now looks like this:

Token	24h volume (XRP)	Holders
RLUSD	3.29M	67,326
USDC	316.3K	10,363
FUZZY	103.5K	11,023
589	29.2K	8,743
ARMY	27.7K	21,994

Useful query parameters

sortBy — vol24hxrp, marketcap, holders, trustlines
sortType — asc or desc
start and limit — pagination over all 20,385 tokens
tagName — filter to a category such as stablecoins or memes

Ledger-wide stats

A second endpoint gives you the 24-hour aggregates:

const { H24, exch, total } = await (
  await fetch('https://api.xrpl.to/v1/stats')
).json();

console.log('tokens traded:', H24.tradedTokens24H);   // 1,300
console.log('trades:', H24.transactions24H);          // 74,805
console.log('unique traders:', H24.uniqueTraders24H); // 6,627
console.log('XRP per USD:', exch.USD);

Note that exch.USD is XRP per USD, not the other way round. To get the XRP price in dollars, invert it:

const xrpUsd = 1 / exch.USD;   // ≈ 1.2956

That inversion catches almost everyone the first time.

Fields worth knowing

vol24hxrp — 24h volume denominated in XRP, not USD
marketcap — also denominated in XRP
holders vs trustlines — the gap tells you how many accounts opened a line and then emptied it
uniqueTraders24h — the cheapest wash-trading filter you will find
pro24h — 24h percentage change

One gotcha

Token names are not unique. Multiple issuers can publish under the same ticker — there are two different tokens called USDC on the ledger right now. Always key on the issuer field, or on md5, rather than on name:

const byIssuer = new Map(tokens.map(t => [`${t.issuer}:${t.currency}`, t]));

Rate limits are real but generous. Back off on a 429 and you will be fine:

async function get(url, attempt = 1) {
  const r = await fetch(url);
  if (r.ok) return r.json();
  if (r.status === 429 && attempt < 4) {
    await new Promise(s => setTimeout(s, attempt * 4000));
    return get(url, attempt + 1);
  }
  throw new Error(`HTTP ${r.status}`);
}

Live prices, holder counts, trustlines and order-book depth for every XRPL token are at xrpl.to.

Five Programmatic Safety Checks for XRPL Tokens

Nyx Lesende — Wed, 16 Sep 2026 20:00:21 +0000

20,379 tokens are issued on the XRP Ledger. 1,531 traded in the last 24 hours. Anyone can issue one in seconds without permission, so the entire burden of screening sits with whoever is reading the data.

Here are five checks you can automate, in rough order of how much they save you.

1. Depth, not market cap

Market cap on a thin token is the last trade price times supply, and the last trade may have been for a handful of dollars. It is the most quoted number and the least useful.

The question that matters is: if you sold a position this size, where would it fill? Anything else is decoration.

2. Volume per unique trader

const { tokens } = await (await fetch(
  'https://api.xrpl.to/v1/tokens?limit=100&sortBy=vol24hxrp&sortType=desc'
)).json();

const suspicious = tokens.filter(t => {
  const perTrader = t.vol24hxrp / Math.max(t.uniqueTraders24h, 1);
  return t.vol24hxrp > 10000 && perTrader > 50000;
});

High volume divided by very few distinct traders is what wash trading looks like from the outside. This single ratio removes most of the noise.

3. Holders against trustlines

const stickiness = t.holders / Math.max(t.trustlines, 1);

Because XRPL trustlines cost the holder a locked reserve, holder counts are meaningfully harder to inflate than wallet counts elsewhere. A very low stickiness ratio means most accounts that ever opened a line have since emptied it.

4. Issuer flags — the one not in the API

This check needs a rippled call rather than the token endpoint, and it is the one that prevents the worst outcomes:

const res = await fetch('https://xrplcluster.com/', {
  method: 'POST',
  headers: { 'content-type': 'application/json' },
  body: JSON.stringify({
    method: 'account_info',
    params: [{ account: ISSUER_ADDRESS, ledger_index: 'validated' }],
  }),
});
const { result } = await res.json();
const flags = result.account_data.Flags;

Two things to read out of that:

Blackholed — the issuing account's master key is disabled and its regular key removed, so no further supply can ever be minted
Freeze authority — whether the issuer can freeze individual balances, or globally freeze the token

Neither is automatically disqualifying. Regulated stablecoin issuers legitimately need freeze powers. But you should know which situation you are in before you hold the token.

5. Concentration

Check how much of the float sits in the top handful of accounts. A token where a few wallets hold most of the supply is one decision away from a much lower price, and concentration quietly distorts every other metric you might compute.

What none of this does

It does not predict a price. A token can pass all five checks and still go to zero, and plenty that fail them went up first.

What it removes is the category of loss that was knowable in advance: the exit that was impossible because there were no bids, the supply that expanded because the issuer was never blackholed, the volume that was three accounts trading with themselves.

Those are the ones worth refusing to take.

Live prices, holder counts, trustlines and order-book depth for every XRPL token are at xrpl.to.

Measuring Order Book vs AMM Share on the XRP Ledger

Nyx Lesende — Tue, 15 Sep 2026 20:00:21 +0000

When the AMM amendment activated on the XRP Ledger in 2024, the expectation was that liquidity would migrate into pools the way it did on every smart-contract chain before it.

You do not have to speculate about whether that happened. The ledger publishes both numbers.

Getting the split

const { H24 } = await (
  await fetch('https://api.xrpl.to/v1/stats')
).json();

const ob  = H24.globalVol24hOrderbook;
const amm = H24.globalVol24hAMM;

console.log('order book share:',
  (ob / (ob + amm) * 100).toFixed(1) + '%');

Right now that prints 82.4%.

Venue	Volume (XRP)	Trades	Avg trade (XRP)
Order book	8.33M	56,372	147.68
AMM pools	1.77M	85,197	20.83

The interesting number is the average trade size

The order book takes a minority of the trade count but the large majority of volume. Big orders route to the book; small ones are content in pools.

That is the whole story in one ratio, and it inverts what you see on most chains.

Why the order book survived here

On the XRPL a cancel costs a fraction of a cent and confirms in three to four seconds. Professional quoting is viable on-chain, so the book stays deep.

Market structure follows transaction economics. The industry generalised from expensive-blockspace chains and concluded AMMs had won outright. This is a live counterexample.

Tracking it over time

async function sample() {
  const { H24 } = await (await fetch('https://api.xrpl.to/v1/stats')).json();
  const ob = H24.globalVol24hOrderbook, amm = H24.globalVol24hAMM;
  return {
    at: new Date().toISOString(),
    obShare: ob / (ob + amm),
    traders: H24.uniqueTraders24H,
  };
}

// append to a file on an interval and you have a time series

Both figures are 24-hour rolling windows, so sampling more often than hourly mostly gives you overlapping data.

Practical takeaway

Live prices, holder counts, trustlines and order-book depth for every XRPL token are at xrpl.to.

Trustlines: Why XRPL Holder Counts Are Hard to Fake

Nyx Lesende — Mon, 14 Sep 2026 20:00:19 +0000

The XRP Ledger works differently, and the difference is worth understanding if you are building anything that reads token data.

There is no token contract

That trustline is an explicit, on-ledger opt-in. Nobody can push a balance at you without it.

Trustlines cost the holder money

Each trustline raises the holder's XRP reserve requirement. The reserve is not a fee — you get it back when you close the line — but it is capital locked for as long as you hold the token.

The consequence: every holder made a small economic commitment. You cannot manufacture a million holders cheaply, because each one costs the holder, not you.

Current leaders by holder count:

Token	Holders	Trustlines
SOLO	215,825	284,488
RLUSD	67,299	97,717
VGB	41,772	47,024
PHNIX	28,896	29,673
CSC	28,540	60,217
ETH	25,829	94,329

Reading the holders-to-trustlines gap

Both numbers are published, and the ratio is more informative than either alone:

const { tokens } = await (await fetch(
  'https://api.xrpl.to/v1/tokens?limit=50&sortBy=holders&sortType=desc'
)).json();

for (const t of tokens) {
  const stickiness = t.holders / Math.max(t.trustlines, 1);
  console.log(t.name, stickiness.toFixed(2));
}

Ratio near 1 — almost everyone who ever opened a line still holds a balance
Ratio near 0 — accounts opened, traded out, and never closed the line

A low ratio is not automatically bad. It describes a token people passed through rather than held, which is exactly what you would expect from something traded actively.

What this means if you are building

Treat holders as a real signal on the XRPL in a way you should not elsewhere. But pair it with two others:

uniqueTraders24h — is the volume many accounts or a few?
Order-book depth — could you actually exit at the quoted price?

Holder count tells you about distribution. It says nothing about liquidity, and the two are frequently unrelated.

Live prices, holder counts, trustlines and order-book depth for every XRPL token are at xrpl.to.

Build an XRPL Token Screener in About 50 Lines

Nyx Lesende — Sun, 13 Sep 2026 20:00:22 +0000

There are 20,374 tokens issued on the XRP Ledger and only 1,275 of them traded today. A screener is how you tell those two groups apart.

Here is a complete one. No dependencies, runs on any Node 18+.

The whole thing

const API = 'https://api.xrpl.to/v1';

async function get(path, attempt = 1) {
  const r = await fetch(API + path);
  if (r.ok) return r.json();
  if ((r.status === 429 || r.status >= 500) && attempt < 4) {
    await new Promise(s => setTimeout(s, attempt * 4000));
    return get(path, attempt + 1);
  }
  throw new Error(`HTTP ${r.status}`);
}

function score(t) {
  // Volume per unique trader — high values mean few accounts moving a lot,
  // which is what wash trading looks like from the outside.
  const perTrader = t.vol24hxrp / Math.max(t.uniqueTraders24h, 1);
  // Trustlines that no longer hold a balance are historical interest.
  const stickiness = t.holders / Math.max(t.trustlines, 1);
  return { perTrader, stickiness };
}

async function screen() {
  const { tokens } = await get('/tokens?limit=100&sortBy=vol24hxrp&sortType=desc');

  return tokens
    .filter(t => t.holders >= 1000)          // ignore dust
    .filter(t => t.uniqueTraders24h >= 20)        // ignore one-account markets
    .map(t => ({ name: t.name, issuer: t.issuer, ...score(t),
                 holders: t.holders, vol: t.vol24hxrp }))
    .filter(t => t.stickiness > 0.15)             // most holders still hold
    .sort((a, b) => a.perTrader - b.perTrader);   // broadest participation first
}

screen().then(rows => {
  for (const r of rows.slice(0, 15)) {
    console.log(
      r.name.padEnd(12),
      String(r.holders).padStart(7),
      r.perTrader.toFixed(0).padStart(9),
      r.stickiness.toFixed(2)
    );
  }
});

What each filter is actually doing

uniqueTraders24h >= 20 — volume with three accounts behind it is not a market. This one filter removes most of the noise.

stickiness > 0.15 — the ratio of current holders to total trustlines ever opened. A very low value means people opened a line, traded, and left.

What it deliberately does not do

It also does not measure order-book depth. A token can pass every filter above and still have no bid you could actually sell into. Depth is the check that matters most and the one people skip.

Live prices, holder counts, trustlines and order-book depth for every XRPL token are at xrpl.to.

Query Every Token on the XRP Ledger With One Endpoint

Nyx Lesende — Sat, 12 Sep 2026 20:00:20 +0000

The XRP Ledger currently has 20,374 issued tokens, and 1,207 of them traded in the last 24 hours. Getting at that data usually means running your own rippled node or stitching together several exchange APIs.

There is a simpler path. Here is the whole thing in one fetch.

The endpoint

const res = await fetch(
  'https://api.xrpl.to/v1/tokens?limit=5&sortBy=vol24hxrp&sortType=desc'
);
const { tokens, total } = await res.json();

console.log(total);                    // 20,374
for (const t of tokens) {
  console.log(t.name, t.vol24hxrp, t.holders);
}

No API key, no signup. Output right now looks like this:

Token	24h volume (XRP)	Holders
RLUSD	1.30M	67,259
USDC	94.7K	10,358
Opulence	21.6K	3,787
FUZZY	20.8K	10,995
PHNIX	20.8K	28,908

Useful query parameters

sortBy — vol24hxrp, marketcap, holders, trustlines
sortType — asc or desc
start and limit — pagination over all 20,374 tokens
tagName — filter to a category such as stablecoins or memes

Ledger-wide stats

A second endpoint gives you the 24-hour aggregates:

const { H24, exch, total } = await (
  await fetch('https://api.xrpl.to/v1/stats')
).json();

console.log('tokens traded:', H24.tradedTokens24H);   // 1,207
console.log('trades:', H24.transactions24H);          // 69,519
console.log('unique traders:', H24.uniqueTraders24H); // 5,432
console.log('XRP per USD:', exch.USD);

Note that exch.USD is XRP per USD, not the other way round. To get the XRP price in dollars, invert it:

const xrpUsd = 1 / exch.USD;   // ≈ 1.3624

That inversion catches almost everyone the first time.

Fields worth knowing

vol24hxrp — 24h volume denominated in XRP, not USD
marketcap — also denominated in XRP
holders vs trustlines — the gap tells you how many accounts opened a line and then emptied it
uniqueTraders24h — the cheapest wash-trading filter you will find
pro24h — 24h percentage change

One gotcha

const byIssuer = new Map(tokens.map(t => [`${t.issuer}:${t.currency}`, t]));

Rate limits are real but generous. Back off on a 429 and you will be fine:

async function get(url, attempt = 1) {
  const r = await fetch(url);
  if (r.ok) return r.json();
  if (r.status === 429 && attempt < 4) {
    await new Promise(s => setTimeout(s, attempt * 4000));
    return get(url, attempt + 1);
  }
  throw new Error(`HTTP ${r.status}`);
}

Live prices, holder counts, trustlines and order-book depth for every XRPL token are at xrpl.to.

Five Programmatic Safety Checks for XRPL Tokens

Nyx Lesende — Fri, 11 Sep 2026 20:00:19 +0000

20,369 tokens are issued on the XRP Ledger. 1,327 traded in the last 24 hours. Anyone can issue one in seconds without permission, so the entire burden of screening sits with whoever is reading the data.

Here are five checks you can automate, in rough order of how much they save you.

1. Depth, not market cap

Market cap on a thin token is the last trade price times supply, and the last trade may have been for a handful of dollars. It is the most quoted number and the least useful.

The question that matters is: if you sold a position this size, where would it fill? Anything else is decoration.

2. Volume per unique trader

const { tokens } = await (await fetch(
  'https://api.xrpl.to/v1/tokens?limit=100&sortBy=vol24hxrp&sortType=desc'
)).json();

const suspicious = tokens.filter(t => {
  const perTrader = t.vol24hxrp / Math.max(t.uniqueTraders24h, 1);
  return t.vol24hxrp > 10000 && perTrader > 50000;
});

High volume divided by very few distinct traders is what wash trading looks like from the outside. This single ratio removes most of the noise.

3. Holders against trustlines

const stickiness = t.holders / Math.max(t.trustlines, 1);

4. Issuer flags — the one not in the API

This check needs a rippled call rather than the token endpoint, and it is the one that prevents the worst outcomes:

const res = await fetch('https://xrplcluster.com/', {
  method: 'POST',
  headers: { 'content-type': 'application/json' },
  body: JSON.stringify({
    method: 'account_info',
    params: [{ account: ISSUER_ADDRESS, ledger_index: 'validated' }],
  }),
});
const { result } = await res.json();
const flags = result.account_data.Flags;

Two things to read out of that:

Blackholed — the issuing account's master key is disabled and its regular key removed, so no further supply can ever be minted
Freeze authority — whether the issuer can freeze individual balances, or globally freeze the token

Neither is automatically disqualifying. Regulated stablecoin issuers legitimately need freeze powers. But you should know which situation you are in before you hold the token.

5. Concentration

What none of this does

It does not predict a price. A token can pass all five checks and still go to zero, and plenty that fail them went up first.

Those are the ones worth refusing to take.

Live prices, holder counts, trustlines and order-book depth for every XRPL token are at xrpl.to.

Measuring Order Book vs AMM Share on the XRP Ledger

Nyx Lesende — Fri, 11 Sep 2026 00:01:26 +0000

When the AMM amendment activated on the XRP Ledger in 2024, the expectation was that liquidity would migrate into pools the way it did on every smart-contract chain before it.

You do not have to speculate about whether that happened. The ledger publishes both numbers.

Getting the split

const { H24 } = await (
  await fetch('https://api.xrpl.to/v1/stats')
).json();

const ob  = H24.globalVol24hOrderbook;
const amm = H24.globalVol24hAMM;

console.log('order book share:',
  (ob / (ob + amm) * 100).toFixed(1) + '%');

Right now that prints 88.2%.

Venue	Volume (XRP)	Trades	Avg trade (XRP)
Order book	3.03M	26,418	114.71
AMM pools	406.7K	32,707	12.43

The interesting number is the average trade size

The order book takes a minority of the trade count but the large majority of volume. Big orders route to the book; small ones are content in pools.

That is the whole story in one ratio, and it inverts what you see on most chains.

Why the order book survived here

On the XRPL a cancel costs a fraction of a cent and confirms in three to four seconds. Professional quoting is viable on-chain, so the book stays deep.

Market structure follows transaction economics. The industry generalised from expensive-blockspace chains and concluded AMMs had won outright. This is a live counterexample.

Tracking it over time

async function sample() {
  const { H24 } = await (await fetch('https://api.xrpl.to/v1/stats')).json();
  const ob = H24.globalVol24hOrderbook, amm = H24.globalVol24hAMM;
  return {
    at: new Date().toISOString(),
    obShare: ob / (ob + amm),
    traders: H24.uniqueTraders24H,
  };
}

// append to a file on an interval and you have a time series

Both figures are 24-hour rolling windows, so sampling more often than hourly mostly gives you overlapping data.

Practical takeaway

Live prices, holder counts, trustlines and order-book depth for every XRPL token are at xrpl.to.

Trustlines: Why XRPL Holder Counts Are Hard to Fake

Nyx Lesende — Wed, 09 Sep 2026 20:02:21 +0000

The XRP Ledger works differently, and the difference is worth understanding if you are building anything that reads token data.

There is no token contract

That trustline is an explicit, on-ledger opt-in. Nobody can push a balance at you without it.

Trustlines cost the holder money

Each trustline raises the holder's XRP reserve requirement. The reserve is not a fee — you get it back when you close the line — but it is capital locked for as long as you hold the token.

The consequence: every holder made a small economic commitment. You cannot manufacture a million holders cheaply, because each one costs the holder, not you.

Current leaders by holder count:

Token	Holders	Trustlines
RLUSD	67,202	100,514
PHNIX	28,940	29,652
CSC	28,603	60,284
ETH	25,831	94,304
ARMY	22,015	22,530
XRdoge	17,546	25,487

Reading the holders-to-trustlines gap

Both numbers are published, and the ratio is more informative than either alone:

const { tokens } = await (await fetch(
  'https://api.xrpl.to/v1/tokens?limit=50&sortBy=holders&sortType=desc'
)).json();

for (const t of tokens) {
  const stickiness = t.holders / Math.max(t.trustlines, 1);
  console.log(t.name, stickiness.toFixed(2));
}

Ratio near 1 — almost everyone who ever opened a line still holds a balance
Ratio near 0 — accounts opened, traded out, and never closed the line

A low ratio is not automatically bad. It describes a token people passed through rather than held, which is exactly what you would expect from something traded actively.

What this means if you are building

Treat holders as a real signal on the XRPL in a way you should not elsewhere. But pair it with two others:

uniqueTraders24h — is the volume many accounts or a few?
Order-book depth — could you actually exit at the quoted price?

Holder count tells you about distribution. It says nothing about liquidity, and the two are frequently unrelated.

Live prices, holder counts, trustlines and order-book depth for every XRPL token are at xrpl.to.