- London only. 03:00 to 06:30 New York time, and nothing outside it. “Outside of this, you're not interested in price action.”
- One chart: the 30-minute. The daily is used for bias and for the target, never for the entry.
- Above the daily 200 EMA, longs only. He is candid that he does not short well and that almost all his money came from longs.
- The 5, 9, 13 and 21 EMAs must be stacked. “If they were intertwining I wouldn't be interested in any price action” — he calls that a low-probability condition.
- A fair value gap printed inside the window. He rates the 02:30 and 03:00 candles highest.
- The trident: a doji that wicks through the 50% of that gap — what ICT calls the consequent encroachment. The doji is the point: sellers pushed down and were driven all the way back.
- The next candle must close BELOW the doji's high. “If it closes above the high, I'll invalidate the trade.” Counter-intuitive for a long, and he says it twice, so it is coded as stated.
- Entry: a limit at the 50%. “We simply set a limit here.” If the confirmation candle already closed inside the gap, “there's no need to set a limit” — in on the close.
- Stop below the doji's low. On gold, no hard stop: “I will wait for a close below this candle.”
- Exit: ride the trend until the EMAs cross — or, in his words, “if you want to just solely utilise a 1 to 20, you can.” Both are tested.
His London window helps on gold. The same set-up at any hour makes −$97.31 a trade with the 1:20 exit; inside his window it makes $289.40. He says the timing is the edge, and on gold this data agrees with him.
The Nasdaq depends on a rule he never gives. He gives a stop for his forex pairs and for gold, not for the Nasdaq. With a hard stop under the doji every one of 18 trades was stopped out (−$23.00 a trade). With his gold rule — out only on a close below the doji — the same trades make $56.96 with the 1:20 exit. Both are shown because he said neither.
The result also depends on what counts as a doji. He draws one; he never gives a size. So every size is shown, not one of ours:
| market | doji body up to | trades | 1:20 exit | ride the trend |
|---|---|---|---|---|
| gold | 20% | 14 | −$5.00 | −$506.07 |
| gold | 35% | 25 | $289.40 | $2,650.21 |
| gold | 50% | 37 | $89.59 | $1,695.43 |
| Nasdaq | 20% | 7 | −$14.25 | −$14.25 |
| Nasdaq | 35% | 18 | −$23.00 | −$23.00 |
| Nasdaq | 50% | 24 | −$24.46 | −$24.46 |
On gold, a strict doji loses riding the trend and a looser one makes a lot. With this few trades, that is the honest answer: the test cannot tell.
| market, exit and setting | trades | win | a trade | 95% range |
|---|---|---|---|---|
| gold, 1:20 · his model | 25 | 28.0% | $289.40 | −$379.61 to $1,191.05 |
| gold, 1:20 · any hour | 173 | 16.8% | −$97.31 | −$455.99 to $313.07 |
| gold, 1:20 · without the EMA filters | 38 | 31.6% | $260.26 | −$218.82 to $906.45 |
| gold, 1:20 · without the doji | 55 | 20.0% | $26.73 | −$369.93 to $508.93 |
| gold, 1:20 · without his invalidation | 28 | 28.6% | $299.64 | −$307.14 to $1,125.45 |
| gold, 1:20 · none of his filters | 836 | 16.0% | −$104.87 | −$236.78 to $36.63 |
| gold, ride the trend · his model | 24 | 12.5% | $2,650.21 | −$513.34 to $8,582.31 |
| gold, ride the trend · any hour | 152 | 13.2% | $587.43 | −$253.26 to $1,735.31 |
| gold, ride the trend · without the EMA filters | 35 | 11.4% | $1,776.29 | −$416.72 to $5,882.04 |
| gold, ride the trend · without the doji | 53 | 7.5% | $980.09 | −$510.10 to $3,743.33 |
| gold, ride the trend · without his invalidation | 27 | 14.8% | $2,482.04 | −$395.56 to $7,856.89 |
| gold, ride the trend · none of his filters | 647 | 18.7% | $138.28 | −$112.99 to $456.17 |
| Nasdaq, 1:20 · his model | 18 | 0.0% | −$23.00 | −$30.32 to −$16.28 |
| Nasdaq, 1:20 · any hour | 136 | 8.1% | $35.59 | −$13.47 to $100.04 |
| Nasdaq, 1:20 · without the EMA filters | 32 | 6.2% | −$10.44 | −$26.87 to $14.53 |
| Nasdaq, 1:20 · without the doji | 37 | 0.0% | −$23.16 | −$28.55 to −$18.22 |
| Nasdaq, 1:20 · without his invalidation | 21 | 0.0% | −$23.10 | −$29.21 to −$17.29 |
| Nasdaq, 1:20 · none of his filters | 587 | 9.0% | $32.47 | $6.65 to $64.09 |
| Nasdaq, ride the trend · his model | 18 | 0.0% | −$23.00 | −$29.97 to −$16.40 |
| Nasdaq, ride the trend · any hour | 126 | 6.3% | $43.10 | −$12.07 to $114.69 |
| Nasdaq, ride the trend · without the EMA filters | 32 | 0.0% | −$23.59 | −$29.09 to −$18.14 |
| Nasdaq, ride the trend · without the doji | 37 | 0.0% | −$23.16 | −$28.41 to −$18.26 |
| Nasdaq, ride the trend · without his invalidation | 21 | 0.0% | −$23.10 | −$29.35 to −$17.25 |
| Nasdaq, ride the trend · none of his filters | 527 | 9.5% | $21.95 | $0.89 to $45.00 |
The same table, live. This one follows the switches above.
Everything else worth saying
What an earlier version of this page got wrong. It entered on the confirmation close, used a hard stop on gold, forced the doji to be the very next candle, and used a fixed 1:20 as the only exit. None of those is what he says. On those rules it concluded that his London window cost him money. On his rules the opposite is true on gold.
He is long-only here, as he trades it. He says he takes shorts below the 200 EMA but is not good at them, and every example he shows is a long. This tests his A+ long set-up.
What this test does not do
We tested his stated rules: the window, the EMA filters, the gap, the doji, the invalidation, his limit entry, his stop and both of his exits. He is explicit that some of it is not mechanical: “no system is purely mechanical. There's always some discretion.”
- A candle-strength indicator he uses is not included. He says a set-up is invalid when its daily candles are not green on a third-party script. It could not be reproduced, and it only removes trades.
- His forex pairs are not tested. They are not in this archive, and nothing was substituted for them.
- Three numbers he never gives are shown across their range: doji size, the Nasdaq stop type, and how long a limit waits to fill.
- The replay is a sample: 25 gold set-ups and 18 Nasdaq set-ups, scored over the whole period.
If you use a pair that is not in the grid, or a filter you think matters, say so and we will test it the same way and publish it.
Close to $400,000 in payouts in three months, and a single trade of $51,000 risking $1,000 — a 1:51, which he says was the largest payout in that firm's history. He is explicit that the timing is the whole edge: “without time, the price action means absolutely nothing.” He is equally explicit that the model is not fully mechanical — “no system is purely mechanical. There's always some discretion.”
Read it in their own words first: