Imagine you are buying a used car. A salesman shows you a gorgeous photo: the car gleaming at sunset, freshly washed. He says "look how beautiful." But you do not want a pretty photo: you want to know whether the engine survives a real highway, in the rain, over potholes, with 200,000 km ahead of it. AlphaLab is the mechanic who puts the car on the test bench instead of staring at the photo.
Most trading products you will see online are exactly that: pretty photos. Charts full of green arrows saying "buy here" right where the price went up. The problem is that those arrows are painted by looking at the past, when we already know what happened. In the real market, going forward, those strategies die.
The problem that fed up
Trading marketing lives off the "green arrow." They show you a backtest (a simulation over historical data) that looks perfect and tell you that you will make money. But a pretty backtest is incredibly easy to manufacture: just tweak the strategy until it fits the past perfectly. That is called overfitting, and it is the number-one trap in the industry.
We were tired of seeing profit curves that shot up like a rocket in simulation and collapsed the moment they touched real money. So we built the opposite of an illusion-selling machine.
What AlphaLab IS (and is NOT)
AlphaLab is a Windows desktop app (€59/month or €590/year) that works as a quantitative research lab. It runs 100% on YOUR own computer.
Let us be clear from the start:
- It is NOT a signals service that tells you "buy now."
- It is NOT financial advice.
- It is NOT a money machine and it does not promise profits.
It is a tool to research strategies and, above all, to discard the false ones. Trading always carries risk of loss; what AlphaLab does is help you avoid fooling yourself.
How it works, in 4 simple steps
- An idea. We start from a hypothesis: "maybe when X happens in the market, it pays to do Y."
- Discover candidates. The lab explores many variants of that idea and finds the ones that look promising on the past.
- Try to DESTROY them. This is the heart of AlphaLab. Instead of celebrating the candidates, we put them through hostile tests designed to knock out the impostors. Most fall. Good: that is what the tests are for.
- Keep only the survivors. Only the strategies that withstand that punishment deserve your attention. And even then, surviving the exam does not guarantee making money tomorrow: the market changes.
The arsenal of tests
To separate skill from luck we use a set of serious statistical tests, the kind professional funds use:
- Deflated Sharpe Ratio (DSR) — punishes you for having tried many strategies.
- PBO — estimates the probability that your "best" strategy is pure chance.
- Walk-Forward / CPCV — examines it with data it never saw.
- Monte Carlo — shuffles thousands of alternative scenarios.
- Realistic costs — spread, slippage and commissions, because real money costs.
Each one has its own beginner-friendly article; do not worry if they sound like gibberish right now.
How AlphaLab handles this (and why it protects you)
Three things make us different, and all three are on your side:
- Honesty. We also publish the strategies that fail. An honest mechanic tells you which cars NOT to buy.
- Auditable, no black box. The code can be inspected. We do not ask for blind faith.
- Local and private. Your data and your strategies stay on YOUR PC. They never travel to a server of ours. It is also multi-broker, and you trade with the same broker you researched with, so what you tested resembles what you will live.
Who it is for and who it is not
It is for curious people who are honest with themselves and want to understand before risking, and who accept that many ideas will be discarded. It is not for those looking for a magic "make me rich" button or promises of returns.
How is it different from the rest?
Against flashy LuxAlgo-style indicators (gorgeous on screen, but never proving they hold up live), AlphaLab bets on statistical proof. Against StrategyQuant-style strategy generators (which churn out mountains of candidates), our focus is not to generate more, but to eliminate the false ones.
Key takeaways
- A pretty photo (backtest) is not a reliable car (real strategy).
- AlphaLab discovers strategies and then tries to destroy them to keep the survivors.
- It is honest (publishes failures), auditable and private (everything on your PC).
- It is not signals, not financial advice, not a profit promise. Trading carries risk.
If you want to see the lab from the inside, you can try it free for 14 days (a card is required, but nothing is charged if you cancel before day 14). Start your free trial here and judge for yourself.