The Trader’s Manifesto

1. Ten years on the other side of the screen

My name is Dylan. I’ve been trading for over ten years, and for most of them I was the product.

I’ve made my living from the markets, moving between forex and crypto depending on where the opportunity was, and I’ve taught tens of thousands of traders around the world how to do the same. Alongside that I got obsessed with the things that eventually turned into this protocol: blockchain, programming, AI, and the machinery underneath trading that almost nobody looks at.

I started trading forex long before FTMO existed, before anyone used the words “prop firm” to mean what they mean now. Back then you had a broker, and that was it. If the broker was regulated, the rules were absurd. FIFO, so you couldn’t manage your own positions the way you wanted. Leverage capped low enough to make the strategy you’d spent two years building unusable. A pile of restrictions written by people who had never held a position.

So I went offshore, like everyone else did. Offshore brokers took crypto deposits, which meant no banks, no wire forms, no three-day holds, no institution deciding whether I was allowed to move my own money. I loved that part. I still love that part. It’s most of the reason this protocol settles on-chain the way it does today.

But I hated what I traded against. An offshore broker calling itself ECN is very often B-book underneath, which means it takes the other side of your position. Your loss is its revenue. It also owns the price feed, the execution, the spread, the server, and the record of what happened. I heard the same rumors everyone hears about stop hunts and spread widening and orders that fill strangely at exactly the wrong moment. Then I saw the software. There are products built and sold for this, with interfaces, with settings.

So the industry offered two doors.

Behind the first: a regulated broker with rules that make good trading harder, and a license that has never once stopped a firm from doing the exact thing the license exists to prevent.

I want to be specific here, because this is the part people wave off as trader folklore.

FXCM, 2011. The CFTC found that from June 2008 to December 2010, the way FXCM handled customer orders “prevented its customers from receiving the benefit of price movements in customers’ favor, but allowed its customers to suffer detrimental price movements.” Read that again. If the price moved your way between placing the order and filling it, the broker kept the difference. If it moved against you, you ate it. More than 57,000 accounts. FXCM paid $8,261,937 in restitution and a $6 million penalty, and had to retain an independent monitor for three years.

FXCM again, in Britain, 2014. The FCA fined FXCM’s UK entities £4 million for the same practice, running from August 2006 to December 2010, roughly £6 million of client profits kept instead of passed on. The FCA also found FXCM UK had failed to mention that US authorities were investigating the group for identical conduct.

FOREX.com, 2010. The NFA’s Business Conduct Committee brought a complaint against GAIN Capital Group, which operates FOREX.com, and its chief executive, alleging margin, liquidation and price-slippage practices that benefited the firm at its customers’ expense. It settled for a $459,000 fine and refunds to affected clients.

FXCM, one more time, 2017. This is the one that should end the argument. The CFTC found that from September 2009 through at least 2014, FXCM advertised a “No Dealing Desk” model and told customers it had no conflict of interest with them, while holding an undisclosed interest in the market maker that consistently won the largest share of its order flow and took the other side of its customers’ trades. That market maker rebated roughly 70% of its revenue back to FXCM, about $77 million between 2010 and 2014. The CFTC also found FXCM willfully made false statements to the NFA to conceal the relationship. Seven million dollar penalty. FXCM and its two founding partners were permanently barred from CFTC registration and left the US market.

Sit with what that last one is. It isn’t sloppy paperwork. It’s a firm telling its customers it had no reason to want them to lose, while owning a piece of the company that profited when they did.

And the tooling was commercially available the whole time. The MetaTrader platform most of this industry still runs on ships with a Virtual Dealer plugin that delays order execution, which is precisely the window asymmetric slippage lives inside. Not a rumor, not a hack, not something someone built in a basement. A product, with a settings panel. I’ve seen it demonstrated.

Behind the second door: an offshore broker that might be honest, might be regulated somewhere that means nothing, and is probably your counterparty.

The whole industry is toxic. One door is a higher grade of toxic than the other, and that’s the entire choice.

2. Then prop firms arrived and made it worse

Prop firms are not brokers, legally. That means no regulator, no oversight, and no standard anyone has to meet.

So a prop firm can invent whatever rules it likes, and then grade you against them itself. It writes the rules, applies the rules, decides whether you broke the rules, and profits when the answer is yes. There is no other industry where we’d accept that arrangement for a second. You would not accept a referee who gets paid by one team.

And the enforcement record is close to empty. The most prominent attempt to regulate a prop firm was the CFTC’s 2023 action against My Forex Funds. In May 2025 a federal judge dismissed that case with prejudice, adopting a Special Master’s findings that CFTC attorneys had made false statements to the court and withheld exculpatory evidence for months. The CFTC was sanctioned, ordered to pay the defendants’ fees, and placed several staff on administrative leave.

I’m not relitigating that case and I’m making no claim about whether that firm did anything wrong. The point is narrower, and it’s about you. The one time a regulator came for a prop firm at scale, the case collapsed under the regulator’s own conduct. So if your firm decides you breached, there is no agency coming. There is no arbitration panel. There is a support ticket, and there is whatever the firm feels like doing about it.

And the business model points exactly where you think it does. The firm’s revenue is your failure. Every incentive it has, at every decision point, is to find a reason. Not necessarily by lying. It doesn’t need to lie. It just needs tight rules, a slow review on large payouts, a risk-management clause that could mean anything, and a support queue that moves slower the more you’re owed.

You already know how this ends, because you’ve either lived it or watched someone you know live it. You pass. You screenshot the dashboard. You tell one person, maybe two. You request the payout.

And then the tone changes. There’s a review you didn’t know existed. A clause you’d skimmed as boilerplate turns out to be the entire document. A reply arrives four days later answering a question you didn’t ask. Nobody ever actually tells you no. They just stop being in a hurry, and somewhere in week three you work out that the hurry was never coming.

I watched this get worse for ten years. I watched it happen to people I taught.

3. The question I couldn’t put down

The whole time, one thought kept coming back.

The loophole that lets prop firms operate outside every rule is also the loophole that would let someone build one that operates outside the conflict. You could give a trader an account size they will never access any other way, with a clean experience and no bank in the middle, if you could make two guarantees credible:

We can actually pay you. Not “we intend to.” Provably, with reserves anyone can check.

We cannot touch your trades. Not “we promise not to.” Structurally, with no mechanism that could.

The second one is the hard part, and it’s the part nobody in this industry has ever offered, because until recently it wasn’t possible to prove anything to anyone.

After enough years of trading, teaching, writing code, and living in crypto, I realized I had accumulated the exact combination of things this required. So I built it.

4. I put a bounty on catching the house

This is the center of everything, so I want to be precise about it.

Every prop firm on earth grades your account in a room you cannot see. You get a verdict. You do not get the work. And when you dispute the verdict, you are arguing with the same party that wrote the rules, applied them, and profits from the answer. That arrangement only functions because nobody outside the building can check the arithmetic.

Here, the arithmetic is published.

When your evaluation concludes, the firm has to commit its grading on-chain as a transcript: every step of the math that carried your account from its opening balance to its result. Then a window opens, 72 hours, and during that window anyone on earth can take a single step of that transcript and prove it wrong. Not you. Not a regulator. Anyone, with no permission from anybody, using nothing but public data.

If one step is wrong, the result is void, the firm’s bond is slashed, and half of that bond goes to whoever caught it. Every firm posts 50 SOL against its own honesty before it can sell you anything. No money moves anywhere until the window closes clean.

That’s a bounty on catching the house, paid by the house, claimable by strangers.

4.1 The watchers are other traders

There’s a hole in “anyone can check the work,” and I fell into it right after I was proud of closing it.

I built the watchtower first as software. A program that pulls a settlement off the chain, replays every step, and tells you whether the house lied. It works. It’s permissionless. Anyone can run it.

Anyone who can run a Node process, keep it online, point it at an RPC endpoint, and understand what it prints. Which is not traders. That’s a few hundred people on earth who’d ever bother, and not one of them is the person whose payout is actually on the line. A system that is permissionless in theory and only usable by people with a terminal has a very small public, and I had built exactly that and called it done.

So I moved it into the browser.

Every trader’s terminal now carries the verifier itself. It runs with nothing to install, no page to keep open, and no button to press. When a settlement is proposed anywhere on the platform, a handful of traders’ browsers are drawn to check it. Each one reads the commitment straight off Solana by RPC, not from any DecentralProp server, pulls the published transcript, replays the entire thing locally using the browser’s own cryptography, and signs the verdict it reached.

You cannot pick your own jury. The draw is seeded from the blockhash of the exact block your settlement was proposed in, which was already fixed and public before anyone could act on it, so nobody can stall and re-roll for a friendlier panel. Anyone can recompute that draw from public inputs and confirm the right people were picked. Verifiers the integrity graph links to the trader being checked are excluded automatically. And a verifier who blesses a settlement that later turns out to be fraudulent is permanently disqualified from ever being drawn again.

Four to six independent replays of every settlement, running on machines the platform doesn’t own or control, asking nothing of the people whose browsers are doing it.

The honest boundary: a browser that disagrees doesn’t slash anyone. It raises the alarm. The slashing still runs through the on-chain fault proof, which stays permissionless and which anyone can still submit the hard way. What the browsers changed is the odds that a lie gets noticed at all, and by whom.

And the rules can’t move underneath you. Your rulebook is frozen on-chain the moment you buy in. Whatever it said that day is what you are judged against, permanently. Not the operator, not the platform, not me, and not a lawyer six weeks later reading section 8.3 out loud to you.

Here is what each of those proofs actually kills:

Every one of those is a working instruction in a deployed program, not a roadmap item.

I could never prove I was cheated. That sentence cost me ten years, and it is the reason this window is open to strangers instead of only to you.

5. The other half: silence

The way you actually don’t get paid by a prop firm is not a denial. It’s that nobody answers.

So silence is a provable offense here. If an operator never settles your account and never files the record, you can’t build a proof, and a firm that wanted to stall would sit exactly there forever. There is a second door for that. Once your account sits unsettled past its deadline, a dispute opens with no proof required at all. The operator either cures it by settling, or you get paid out of insurance.

You cannot be ghosted into giving up.

6. The house is paid when you get paid

Now the money, because a structure only holds if the incentives hold.

At every prop firm you have ever used, your payout is a loss on their books. Here it’s an invoice they collect on. On a standard payout the operator receives about 6% of your gross profit, paid out of the payout itself, in their own firm’s token. That channel does not exist when you fail. There is no line anywhere in this protocol that pays an operator for your loss.

And it goes further in the direction you’d want. When a payout has to reach past the firm’s own treasury into the shared backstop or the protocol-wide pool, the operator receives nothing at all. The full amount goes to the trader. The alignment tilts hardest toward you exactly when things are worst, which is precisely when every other firm’s alignment tilts away.

7. Where the money actually is

I made two promises back there and I’ve only kept one. Honest grading is worthless if the account is empty when you win, so here is the other half.

Your payout doesn’t come out of a firm’s good intentions. Four sources stand behind it, in order. First the firm’s own treasury, funded by its launch fee and a share of every evaluation it sells. Then a token reserve minted at launch and held back for exactly this. Then a backstop escrow, staked by investors who take that risk on deliberately and are paid a premium for carrying it. And last, a pool that every other firm on the protocol pays into out of every deployment fee and every evaluation fee, which exists for the case where one firm’s own reserves fall short.

Three of those four are somebody else’s money standing behind your firm.

The protocol will not let a firm pay out more than it actually holds. Payout caps scale with the real treasury balance, and no single account can drain a disproportionate share of it inside a short window. All of it is on-chain: the balances, the caps, the draws, and every payout every firm has ever made. In this industry a firm’s payout record is a marketing claim. Here it’s an address.

You should also know what actually lands in your wallet, because this document would be dodging if it left that implied. A payout buys the firm’s token off its curve and sends you that. You can keep it, stake it, or sell it straight to SOL, and you can split it across all three in proportions you choose in one signed transaction. That’s deliberate: it’s the mechanism that makes your payout something the firm’s holders want to happen rather than something they’d rather delay. It also means the asset you’re paid in has a price that moves. That’s a real tradeoff, and I’m not going to bury it in a footnote.

8. What else is in there

Nothing can be aimed at you. No template, execution profile, operator, or administrator can change the price or fill speed of an individual account. The fill path never reads who you are before deciding what you get. If the integrity system flags an account, it acts on payouts and account status, never inside execution.

Your fills can be rebuilt. Prices come from a feed aggregating around 90 independent publishers, and no one at a firm sets or adjusts them. Every account gets its own seeded random stream, so any fill you have ever received can be reconstructed later and checked against what should have happened. Every fill leaves a signed receipt.

Your position size is your business. No lot caps, no position-count limits. You’re bounded by margin and by the drawdown rules, and by nothing else.

A machine sits where the risk manager used to. At every prop firm there’s a human whose job is deciding how much you’re allowed to make. Here that role is an engine reading live protocol health, and it can only ever tighten from a platform-fixed baseline it isn’t permitted to loosen for anybody, including a firm’s biggest customer.

You touch your wallet twice. SOL in when you buy, tokens out when you’re paid. Everything in between is instant, with no gas and no popups. None of the proof machinery costs you a single click.

9. The thing I didn’t set out to build

I want to be honest about how the bigger idea arrived, because it wasn’t a vision. It was something I noticed while solving a smaller problem.

To make the grading provable, the rulebook had to go on-chain and lock at purchase. Once the rulebook was on-chain, the judge was code instead of a person hunting for a reason. Once the judge was code, the treasury could be code too, because a payout is just another rule. And then I looked at what was left.

A prop firm is a treasury, a rulebook, a judge, a payout rail, and a name. Four of those five are software.

Which means the firm itself becomes something you deploy. Anyone can launch one in an afternoon: it mints its token, opens its curve, creates its vaults, and starts selling evaluations. The protocol holds the rules and the money. The operator holds the brand and one economic dial, the profit split. They cannot touch grading. They cannot touch execution. They cannot touch payouts. There is no human in the middle of the part that matters, because I removed the chair that human sits in.

The loophole that lets prop firms invent any rule they want is the same loophole I used to build one that can’t.

10. Why I think this is where trading goes

Receipts arrived everywhere except here. You can verify a bank transfer, a package delivery, a signature, even a lottery draw. You cannot verify the two numbers that decide a trader’s life: the price you were filled at, and whether the house says you failed. Trading is the last major industry still running on trust me.

Once proof is possible, refusing to show your work becomes information. That’s the real shift. Not that the whole industry migrates next year, but that opacity stops being the default and becomes a visible choice a firm is making about you.

And the counterparty problem was never really a morality problem. For thirty years there was no way to run a market where the house couldn’t quietly tilt it, so the industry built regulation instead, and regulation produced a label that firms wore while doing it anyway. The label was a patch on a technical impossibility. It stopped being impossible.

11. What isn’t finished

This document is worthless if it overclaims, since its whole subject is distrust of people who overclaim. So here is the honest edge of it.

Evaluation fees are revenue whether you win or lose. An operator earns 6% to 20% of every evaluation fee either way. The honest claim is narrower than “the firm only makes money when you win.” The firm makes money both ways. The payout channel is the only one that grows with your success instead of shrinking.

A stressed firm pays a smaller split, 70% at Warning and 60% at Critical. That’s solvency management, and it applies to accounts sold during the stressed window, never retroactively to a split you already locked in.

The engine is a pure simulation. No external counterparty takes the other side of your trade. The alignment I described is economic. It is not a claim that software is rooting for you.

Not all of this is on mainnet. Large parts are proven on devnet with real transactions you can look up, and devnet is not production. Some features are deliberately gated off until an audit exists. No independent security firm has reviewed the programs yet.

I still hold the upgrade keys today. That’s the real one. Right now you would still be trusting me, and that is not the deal I want to offer anyone. The keys move to a multisig before real value, then behind a timelock, then get revoked permanently. After that nobody can change the programs, including me.

The last thing I ship is the instruction that makes me powerless over it.

12. If you want to try it

Connect a wallet and buy an evaluation. Trade it in the browser, with no gas and nothing to sign on any individual trade. Pass, and the grading is published where you and four to six strangers can replay it. Request your payout, and it moves when the window closes clean.

If you’d rather check the thing before spending anything on it, do that instead. Every settlement every firm has ever made is already public, and the verifier is already in the page.

13. One more thing, and it isn’t technical

I fight constantly with the question of whether I’m doing what God wants, or whether I’m just doing what I want and calling it purpose. I don’t have that resolved. I didn’t have it resolved while I was building this, and I don’t have it resolved now that it’s built.

What I decided is that I’ll give a tenth of what I earn from this protocol to God’s work.

I tried to put that in the code first. I wanted the protocol itself to give it, automatically, forever. Sitting here writing this, I realize why that was wrong, and it took me until now to see it: the protocol’s money isn’t mine to give. A tithe taken out of other people’s fees isn’t a tithe. It’s a tax with a halo on it. What’s mine is mine to give.

So I tried again, smaller and correctly, to program it out of my own share, skimmed automatically into a separate wallet. And I couldn’t. The smart contract has a hard data limit and I’d already reached it. There was physically no room left for it.

I think that was the point. It was never the machine’s to carry.

So I’ll carry it. I’ll take that tenth out of my own earnings by hand and send it to things that serve God, and I’ll write those transactions on-chain, or give through places that accept crypto donations, so the record is public and you can check it the same way you can check everything else here.

I built a machine so that nothing between a trader and their money would have to be taken on faith. Then I tried to put the most important thing I own inside it, and it wouldn’t fit.

That part is mine to prove, the same way the rest of it is.