Probatio is an open proving ground for Solana. Trade simulated capital against live on-chain liquidity. Real slippage, real latency, a record anyone can verify.
$PROB is now live.
CA: CzSDyFGHgZQP6HB1f32xuZybn8gtSSYfHj23xVgpump
Probatio is building an open trading simulator for Solana.
Today, most paper trading fills at the mid price. It ignores slippage, latency and depth, so it teaches a market that does not exist, and the track record it produces is whatever the platform says it is.
Probatio prices every fill against the real market.
Traders can trade any https://t.co/oY1HhlLjhE launch with paper money, take the slippage their size actually causes, wait out a real delay, compete in ranked seasons, and build a record from real outcomes.
The project is fully open source and already includes a live launch terminal, a movers board, ranked seasons, a public record for every trader, a verification SDK, and a CLI on npm.
Website: https://t.co/rK4TYDPpCf
GitHub: https://t.co/yJxWW183vX
Thanks to everyone who has supported the project so far. This is only the beginning.
Season 1 has closed. Four entrants, a 0.20 SOL pool built entirely from entries, and one paid place.
The winner was paid fifty seven seconds after the bell, straight out of the prize wallet, without being asked to claim anything or fill anything in. You do not have to take my word for that part, which is the whole point of doing it this way. The transfer is on chain and you can open it yourself.
https://t.co/ghPKcUw4TU
Season 2 is up next.
https://t.co/PKbIhUMatC
The reason this is open source is not generosity. It is that the entire claim rests on the engine, and you have no reason to believe a word about it from me.
So read it. The whole fill engine is 2,655 lines in one folder, and you can see for yourself where the quote comes from, that the pool is read twice, where the latency is applied, and why a take profit fires on what a sell would have fetched rather than on where the line went. There are 1,903 tests around it, and the interesting ones are not the ones proving it works. They are the ones proving it refuses: unknown counted as unmet, a stale price never used, a result that cannot be measured never reported as if it were.
If you find something wrong with it, that is the point. A simulator nobody can check is just a nicer-looking screenshot.
https://t.co/gCPcawHBNf
Your take profit works. It has worked every time you tested it, because you tested it against a line on a chart, and a line will always let you out at exactly the level you drew.
Replay it here instead. It runs your take profit, stop loss and timeout over one token's actual recorded swaps, and prices every exit at what a sell would really have fetched out of the reserves at that moment. Not the mid, not the close, the number the pool would have paid you.
The gap between those two answers is the whole reason strategies that backtest beautifully lose money live. Better to find it now than during a season.
https://t.co/MsRloC2qVG
Eleven hours and twenty two minutes left in Season 1. Here is what happens when it ends.
Open positions are marked one last time, the standings become final, and the season is closed and sealed. Winners are paid straight out of the prize wallet, split by the ruleset that was published before the season started, and nobody has to claim anything or ask for it.
Ten hours after that, Season 2 opens. That gap exists on purpose, so one season is settled and paid before anybody is asked to enter the next.
Whoever is holding, is holding. Whatever it is worth at the bell is the season.
https://t.co/8ywYenH8Qi
The terminal is the good way to trade this, but nobody wants to open a tab to check one position.
So there is a bot. Link your wallet once and you can buy, sell, watch a token and see where you stand from Telegram, on the same account and through the same engine, with every fill landing on the same sealed record. Nothing about a trade placed there is different from one placed on the site.
Watch a token and it tells you when it moves, which is the part that actually stops you missing things.
https://t.co/BZczlFFKj1
You are already paying to learn this. Every position that taught you something about liquidity, or about buying a launch where the dev still held half the supply, was tuition, and you paid it in SOL.
The lesson does not get cheaper by being repeated. It gets cheaper by being taken somewhere it costs nothing, on prices that are real and fills that behave the way the market behaves, so what you learn transfers instead of flattering you.
That is all this is. Paper money, real prices, and a record honest enough to be worth something afterwards.
Free to start.
https://t.co/zx6EbWuxBA
Most places that let you paper trade treat an algorithm as a novelty, put it in its own leaderboard, and quietly give it a faster lane. Probatio does not have a bot division, because the whole point of a record is that it can be compared.
You write the rules, the site runs them for the whole season on its own clock, and the orders go through the same engine a click goes through. Same six hundred millisecond latency, same slippage, same price impact ceiling, same sealed record. It lands on the same board as everyone else, and a small tag says how the row was traded, which is a fact about method rather than a handicap.
If your algorithm beats people, it beats people. That is the only interesting version of the question.
https://t.co/dlbm6cP4PR
A screenshot proves nothing. Everyone knows it, and everyone posts them anyway, because there has never been an alternative that a stranger could check.
Probatio is that alternative. Every fill is sealed as it happens and committed to Solana, so a record here is not a claim you make about yourself. It is something anybody can verify without trusting you, or me, or the site. The whole thing is open source, so you can read exactly what the engine did and why.
That is the point of the seasons and the board and the duels. Not the prize, though there is one. The record. Being good at this is currently unprovable, and it should not be.
Start free. https://t.co/zx6EbWtZM2
The interesting part of building an honest simulator is not what it does. It is what it refuses to do.
When live prices cannot be read, trading stops. It never quietly falls back to a cached price, because a fill quoted against a stale number is a fill that never could have happened. When a strategy asks a question the system cannot answer, whether the creator still holds half the supply, whether a launch was bundled, how many wallets actually hold it, the answer counts as unmet rather than clean. A condition that fails open is worse than no condition, because it looks like it worked.
When a token's price drifts too far from what this engine thinks it is worth, the token is suspended rather than traded through.
None of that makes the numbers bigger. All of it makes them real.
https://t.co/zx6EbWtZM2
Every paper trading app lies to you.
It fills you at the price on the chart, which is a price nobody could have got, and then congratulates you on a return that would not have survived contact with the pool. You learn a habit that costs you money the first time you use real SOL.
Probatio fills you the way the market would have. Quoted out of real reserves, with the same latency and the same slippage, and a take profit that fires on what a sell would actually have fetched rather than on where the line went. Every fill is sealed as it is made, so a record here is checkable by a stranger instead of screenshotted by its owner.
On top of that there is now a whole arena. Ranked seasons with a real prize pool, algorithms you write yourself that trade beside humans on the same board, backtests priced at a real exit, and duels where one trader takes on another over a window they both agreed to.
Paper money. Real prices. Records kept honestly. All of it open source.
https://t.co/zx6EbWtZM2
Season 1 is running with two days left, and the board says something worth sitting with. The leader is on 0.00% with zero trades. Everyone who has actually traded is down.
That is not a bug in the scoring. Equity is marked at the current price, so a trader holding nothing is holding exactly what they started with, and right now that beats every position anybody opened.
Prize pool is 0.20 SOL and one place gets paid. Sixteen more entries and the top three all get paid instead.
Two days to prove the market wrong.
https://t.co/8ywYenH8Qi
Phase one made it possible to prove you are good. Phase two gives that proof somewhere to go.
Guilds: Trade as a squad and pool your records, so the board ranks groups as well as people and one good week stops being only yours.
Tournaments: Duels prove you can beat one person. A bracket asks whether you can do it four times running, seeded by record, with a real prize pool.
Streamer tools: Overlays and embeds, so you broadcast the sealed record live instead of screenshotting it afterwards.
Rivalries: A running head to head that follows two traders season after season, so the eleventh time you play someone you both know the score.
Same engine underneath. Same clock, same fills, same record.
https://t.co/7URJiflVP8
Strategies can now bet what they actually believe.
Every condition was a gate, so a token that cleared all seven comfortably and one that scraped past the last of them placed the exact same bet. Now the size scales with how much you cleared by.
The catch is the good part. Exits are priced out of real reserves and a take profit fires when your position is biggest against its pool, which is when leaving costs most, so the cap is pool depth rather than your balance. Without it, confidence sizing turns your best entries into your worst exits.
Leave the new box empty and nothing changes.
https://t.co/dlbm6cOx0j
In a few hours we are releasing the thing strategies have been missing.
Every entry condition is a gate right now, so a token that clears all seven comfortably and one that scrapes past the last of them produce exactly the same bet, because the position size is a single number you set once and never revisit.
Tonight it scales with how much you cleared by, so conviction and stake finally move together.
The catch is the interesting part, because exits here are priced out of real reserves and a take profit fires precisely when your position is largest against its pool, which is the moment leaving costs the most, so the cap has to be against pool depth rather than against your balance, or confidence sizing quietly turns your best entries into your worst exits while looking like an upgrade.
https://t.co/dlbm6cOx0j
Phase one is complete.
Strategies, backtesting, spectating, and now duels. All four live, and the roadmap only says shipped when the last one is.
Phase two, in four parts.
Guilds: Trade as a squad and pool your records, so the board ranks groups as well as people. A season your friends are in is one you come back to.
Tournaments: Duels prove you can beat one person. A bracket asks whether you can do it four times running, seeded by record, with a real prize pool and a final anyone can watch.
Streamer tools: Overlays and embeds, so a trader broadcasts the sealed record itself instead of screenshotting it afterwards.
Rivalries: A running head to head that follows two traders season to season, so the eleventh time you play someone, you both know the score.
https://t.co/7URJifmtEG
Head to head duels are live.
One trader against one, over a window you both agreed to. It is scored off the account you already entered the season with, so there is nothing new to fund and nothing new to learn. Trade exactly as you would have. At the end the two returns are compared and both are sealed.
The clock starts when they accept, not when you offer, so sitting on a challenge while a position runs buys nothing. It closes itself when the window runs out, whether or not anybody is watching, because settling when somebody opens the page would let the loser choose the moment they are measured.
Paying a winner is the claim path the seasons use, and it will arrive with the code that pays it.
https://t.co/Sq3l7OqkWc