Why I built a preparation engine that can't flatter you.
Almost everything sold to a retail trader is an outstretched hand telling a story that isn't true — a screenshot of a 900% month, a "91% win rate", a signal group, a course. The story sells; the losses are never in it. That's not an accident, it's the business model.
Stratsemble is the opposite trade. I took the liar's tools away — no profit claims, no cherry-picked windows, no hidden losing years — and built an engine that can't flatter you: costs on, no look-ahead, every result measured against simply buying and holding, and the losses shown in full. What's left is the one thing a con can't copy: the true story, drawdowns and all. Less exciting, far more useful.
“Chance favors the prepared mind.” — Louis Pasteur. That's the whole idea, and I mean it in the narrow way: I can't hand you an edge — no one honest can. What preparation buys you is this — if luck runs against you, you already knew it could; and if it runs for you, you won't mistake it for skill. A prepared mind gets fooled less. It doesn't win more.
Who's behind it
One engineer. I build quantitative trading systems — it's the work and the obsession — and I'm keeping my name off it on purpose. Not because there's anything to hide, but because I'd rather you didn't have to trust me at all. On a site about honesty, a résumé you can't verify is just another story asking for faith. So here's a different deal: judge the work.
I've spent years building the unglamorous machinery behind trading, end to end: data pipelines that survive real-world gaps, bad ticks and split/dividend adjustments; point-in-time backtesting engines that refuse to peek a single bar into the future; risk-governor and position-sizing libraries that decide how much to bet and when to cut a loser; signal research and machine-learning pipelines with independent validators built to kill their own results; and live execution, where being wrong costs real money the same day. The strategy is the easy part. The hard part — the part almost nobody does — is the plumbing that keeps the numbers honest.
And that's what building all of it taught me: once you stop cheating — closed-bar evaluation, costs and slippage on every fill, survivorship handled, everything measured against just holding — most strategies, including most of mine, don't beat the benchmark. That's not defeatism; it's the whole reason this exists. Instead of selling you an edge I can't honestly promise, I took the honest measuring instrument I use on my own capital and put it in your hands.
So don't take my word for any of it. The method is fully open — every number is computed exactly the way it says. The engine is the same one I hold my own decisions to. And the scoreboard shows most strategies failing, mine included. That's the proof: not a name, not a track record — a tool that can't flatter you and doesn't try.
— A.P.
There's no fund to sell you, no signal to upsell, no sales team paid on conversions. The only thing that earns anything is you finding the tool genuinely useful — so the only incentive anywhere in the building is for the numbers to be true, not impressive.
Why I built it — and what it does for you
I built it because nothing on the market could answer one question honestly: does this actually work? So I took the honest measuring instrument I use on my own capital and turned it into something you can use on yours. Here's what that gives you:
Run a real backtest on a famous stock or coin in seconds — costs and losses included — and find out if an idea holds up before you risk a cent.
Not a yes/no. The whole equity curve, the worst drawdown you'd have sat through, and every result measured against simply buying and holding.
Forward-test a strategy from today and build a track record time actually lived — the one kind a screenshot can't fake — then get a mechanical alert when it signals.
The cheapest place to learn a strategy doesn't work is here — not with real money in a live account. Most don't beat buy & hold, and it's better to find that out now.
A no-code builder, dozens of famous assets, and free calculators for the math behind a trade — position size, risk of ruin, expectancy, and more.
Your strategies and tests are yours; a free backtest needs no account at all, and nothing you build gets taken away.
Commitments that never bend
Non-custodial. No funds held, no orders placed — ever. You execute on your own broker.
No strategy here is sold as a way to make money. Results are hypothetical; the honest test is the value.
The whole equity curve, the drawdown, the losing trades — measured against buy & hold, every time.
A listing or an alert is a mechanism you test and decide on — never a recommendation, never a trade I route.
Closed-bar, no look-ahead, costs on — enforced in code, on every run. See the full method.
I don't sell your data, and a free backtest needs no account at all.
The full recipe — data sources, cost model, survivorship, what a backtest can't tell you — is on the methodology page.
Where this is right now
Stratsemble is in free, open beta — everything is free while I find out whether it's genuinely useful. Paid plans will come later, and when they do, nothing you've built, saved, or forward-tested gets taken away, and early users keep founder pricing. I'd rather earn the subscription than trap you into it.
What this is — and isn't
It is: an educational, analytical backtesting & paper-forward-testing tool with an honest engine and transparent data.
It isn't: a broker, a portfolio manager, an investment adviser, a signal service, or a get-rich scheme. It gives no personalized advice and makes no recommendations. Past performance is not a reliable indicator of future results, and you make every decision yourself.
Contact
Questions, corrections, or found something that isn't honest? A contact address goes live at launch.