TJR’s strategy — does it actually work?

TJR markets this ICT-derived intraday strategy with a brokerage track record he says totals $874,782 this year, at a ~64% daily win rate (~1.23 reward-to-risk). The pitch: follow his rules and you copy the result.

“If we see this year I’ve made $874,782 … my daily win rate is around 64% or 64.29%.”— TJR, “My UPDATED Day Trading Strategy (2026)” · sources: $874,782 ↗ 1:27 · win rate ↗ 2:09

So we did exactly that: we coded his method rule-for-rule from his own videos, then tested it honestly — every trade, real tick-level costs and slippage, across seven years of NASDAQ and S&P data.

The strategy, exactly as he teaches it

Every rule links to the exact moment he states it — check any of them.

We coded it exactly

Before testing whether his strategy holds up, we had to be certain we had coded it faithfully: his method exactly as he teaches it, not our interpretation of it. So we took a trade TJR presents as his own in his strategy video, the 29 May 2026 ES short, and ran his rules through our engine. The swept level, the reversal FVG, the inverse FVG, the 1-minute break-of-structure entry, the stop and the target are all placed by the code, not drawn by hand. It reproduces his trade to the tick, confirming the strategy tested below is genuinely his. Flip through the tabs for more setups the same engine takes on his rules — a London-high short and an Asia-low long — covering both retrace mechanisms he teaches: a 1-minute break of structure and a 5-minute fair-value-gap retrace. ↗ TJR: “I took this exact same trade… on the S&P 500” — 31:46

Every variation we tried

TJR leaves several steps open to interpretation, so we didn’t pick one version — we coded 22 readings of his rules, one for every reasonable interpretation: which reversal confirmation (BOS or inverse FVG), which continuation (1-min BOS, 5-min FVG, or equilibrium), which entry trigger, which index, which targets (nearest draw, fixed 1:1, 1.5R, scale-out), where the stop sits (1st or 2nd swing), which entry window, each filter on or off — plus a “loose” version that takes every setup. Each one ran across seven years of NASDAQ and S&P tick data, with real commissions and slippage, measured against a matched random-direction null — the honest bar for “is this better than a coin flip.”

Not one is a real edge. Not one of the twenty-two readings is statistically significant — the best, SMT-divergence-required at profit factor 1.34, reaches only t +1.34, short of significance and exactly what you’d expect from the luckiest of twenty-two tries. Ten lose money outright; the rest cluster near break-even. His own strategy, on the unswept-draw exits he describes in his own words, is a 32% win-rate coin flip (PF 1.06, t +0.44). Nowhere in the interpretation space is there anything close to his advertised 64%. Each row below is one reading, sorted by t-statistic.

Win rate is per-trade on his draw exits, which sit far away — so a low win rate with a high payoff, by design (his daily win-rate claim is a different number). Push the win rate up with a closer target and the profit factor drops below 1: the 1:1 and nearest-draw readings reach 48% but lose money. You can have his win rate or a profit, not both.

How we made sure this is a fair test

A debunk is only worth anything if the test is honest. Here is exactly what we did, so that “it doesn’t work” can’t be waved away as “you tested it wrong.”

What about the $874,782 and the 64%?

We coded his strategy exactly and ran it on seven years of data. Every faithful reading of his rules lands on the same result: a coin flip at best, with a win rate nowhere near what he advertises.

We take his brokerage numbers at face value — the point isn’t that he didn’t make the money, it’s that his published rules don’t reproduce it. His 64% is a daily win rate from hand-picked, discretionary trades; mechanized exactly as taught, the rules give 29–48% — a coin flip, not an edge — in every one of the twenty-two ways we read them.

Coded faithfully, his strategy produces no significant edge.

Coded rule-for-rule from his own videos and tested across all twenty-two readings of his rules, TJR’s strategy has no real edge. Not one variation is a statistically significant winner — the results swing from a 136% loss to a 40% gain purely on how you read his vague rules, and the best of the twenty-two reaches only t +1.34, short of significance. Deflate for the fact that we tried 22 things and none clears the bar — his own strategy’s deflated Sharpe is 0.09, indistinguishable from luck.

Coded faithfully and tested honestly, TJR’s published rules have no edge — not one of the twenty-two readings beats a coin flip, across seven years of data. Split the seven years in half and test the recent, out-of-sample years on their own and nothing carries: the variations that looked best in-sample fall apart, and the loose high-frequency version — the only one with enough recent trades to measure — comes in dead flat.

A research result on the mechanical rules as taught. TJR trades this by hand and by feel; our test isolates whether the rules themselves carry an edge, and finds none.