Your backtest is lying to you — and we can prove it.

Hey — if you've spent any time on trading TikTok, you've seen the clip. Mark the Asia and London session high and low. Wait for New York to sweep one of them. Drop to the 1-minute chart, find the fair value gap, enter on the inversion, 1:2 risk-reward. Clean rules, and a backtest screenshot underneath with a win rate that looks like a printing press.

Maybe you traded it live for a few weeks. Maybe the results didn't look anything like the screenshot. If that's you, you're not bad at this. There's a good chance the backtest itself was never real to begin with, and the setup was never given a fair shot at proving whether it actually works.

The one line of code that fakes a strategy into looking like a moneymaker

Here's the mechanism, and it's not a conspiracy, it's just sloppy backtesting, and almost everyone does it by accident. Say a single 1-minute bar's price range touches both your stop loss and your 2R target. In real trading, price only goes one direction first, and you never know which until it happens — that's the entire risk of the trade. Most hobby backtesting tools don't know which happened first inside that bar. So they guess. And the lazy guess almost every tool defaults to is: the target got hit first. The win.

Do that across a few hundred trades and you don't get a slightly-too-generous number. You get a strategy that looks like a cheat code, because on every single coin-flip bar, the tool flipped it in your favor.

We built a rule engine that runs the exact same rule both ways on the exact same bars, so the gap between "looks good" and "actually good" is a number you can see, not a guess. No real ES/NQ tick data exists locally yet (more on that below), so we proved the mechanism the honest way instead: on a simulated price series built with zero real edge on purpose, so the only place any apparent profit can come from is the tie-break rule itself, not the strategy.

A profit factor of 3.06 is the kind of number that gets a strategy a following. There is no edge in this data by construction — we built it to have none. But the collapse from 3.06 to 0.22 isn't one mistake, it's two, and crediting one for the other's effect is exactly the failure mode this report exists to catch. Fix only the coin-flip tie-break, holding costs at zero, and the strategy is still profitable: profit factor 1.60, expectancy +0.33R. It's real-world costs stacked on top of that — slippage and commission eating into stops that are often just a few ticks wide — that finish the job and drag it down to 0.22. Same bars, same rule, same setup throughout. The tie-break alone accounts for roughly a quarter of the naive-to-corrected swing; trading costs account for the other three-quarters.

One hand-built example makes it concrete: a single later bar's range touches both the stop and the 2R target. Naive resolution scores that bar a win, +2.00R. Fixing just the tie-break — still no costs — scores the identical bar a loss, −1.00R. Add realistic slippage and commission on top and it's −1.21R. On this one hand-picked trade the tie-break is nearly the whole story; at scale, across all 1,897 simulated signals, costs turn out to matter three times more.

We also sanity-checked the tie-break correction against the true underlying price path, something only possible because the data is synthetic and we can see it: it tracked the real outcome roughly 2.5x more accurately than naive did. True expectancy on this zero-edge series comes out to +0.16R — not exactly zero, small but real (we dig into why in the full report card), and nowhere near the 3.06 the naive convention prints. Full methodology and every number above ships in the card itself, not just this page.

What Report Card #001 actually is

We took the exact rules from the clip — the literal transcript, nothing added — and did three things almost nobody selling a course does:

• Wrote down every place the clip is genuinely ambiguous (session clock times, what counts as a "sweep," touch vs. close-through on the FVG inversion) instead of quietly picking one and hiding the choice.
• Built and validated a rule engine that resolves trades the naive way and the corrected way, so the manufactured edge is a number you can see, not a guess.
• Told you plainly what it isn't: a real backtest on ES or NQ. We went looking for the data and the infrastructure to run one and neither cleared our bar — 12 days of one expired contract in an undocumented binary format, and a trading add-on that only runs inside a desktop app with no headless backtest path. Rather than force a number out of that, we showed you the mechanism instead and said so.

Why the Founding Setup Audit exists

Everything above proves the mechanism on simulated bars, on purpose — that's the only honest way to isolate it. The Founding Setup Audit is the same naive-vs-corrected engine, pointed at your setup's actual rules, run against real ES/NQ tick data. A corrected profit factor under 1 isn't just a bad number on paper — it's what happens to your odds on a 50K prop-firm combine when the setup's real expectancy is working against you before you've placed a trade.

Why free, and why $27

Report Card #001 stays free, permanently — it's done, it cost us the time to build, and we'd rather build trust with a real trader for free than sell one more overhyped course. The $27 is different: it isn't paying for our time, it's paying for the one thing standing between "we proved the mechanism" and "we proved it works on your setup with real money on the line" — an actual market-data subscription, paid for out of our own pocket, not the pre-order's. First 50 only, because that's the batch we can honestly commit to auditing by hand within 2 business days each.