The Pharos desk · paper · simulated fills
On the always-on stream my terminal runs a paper book that trades what Pharos calls. In the cash session, from 09:40 ET, it rests a limit one point in front of the morning call wall and put wall — but only a wall the overnight session traded inside, never one the tape has already run through — target four points, stop twelve through the wall, one contract, up to two touches per wall, flat at 15:55 ET; on positive-gamma days it does the same at the overnight Globex high and low. From 03:00 ET, the London session, it fades the Asia-session high and low and yesterday’s close the same way with a three-point target, nothing new after 07:55, flat at 09:25. It holds one position at a time: while a trade is open the other orders come off the ladder. Every fill is simulated on the live tape and lands in the terminal’s own journal with a note. This page shows all of it, ten minutes behind the tape like the picture, misses included.
The ledger opened on 8 September 2026, so these are the numbers of a very short record and not a verdict on anything. Its first four trades ran on the original rule; the tightened rule described below has been live on the desk since 13:09 UTC that day.
SIM · armed · london opens at 03:00 ET
| date | session | in | out | trade | level | result | points | 1 ES |
|---|---|---|---|---|---|---|---|---|
| 2026-09-08 | london | 08:32 | 09:08 | short 1 @ 7707.75 → 7714.75 | yesterday's close 7708.75 | stop | -7.00 | -$350 |
| 2026-09-08 | london | 07:10 | 07:16 | long 1 @ 7695.75 → 7698.75 | the Asia low 7694.75 | target | +3.00 | $150 |
| 2026-09-08 | london | 06:53 | 06:54 | long 1 @ 7695.75 → 7698.75 | the Asia low 7694.75 | target | +3.00 | $150 |
| 2026-09-08 | london | 06:21 | 06:33 | long 1 @ 7709.75 → 7702.75 | yesterday's close 7708.75 | stop | -7.00 | -$350 |
| 2026-09-08 | london | 05:59 | 06:11 | short 1 @ 7707.75 → 7704.75 | yesterday's close 7708.75 | target | +3.00 | $150 |
I swept the recorded ES tape from April to September 2026, one-minute bars, with the morning walls the levels page published from July on (47 sessions). A limit one point in front of the wall was counted filled only when the tape traded to it, never when the market had already opened through the level; stops were assumed filled at price; 0.6 points of costs came off every round trip. Fading the walls with a four-point target and a twelve-point stop won 88% of 42 trades for +1.38 points a trade; the overnight range on positive-gamma days won 95% of 44 for +2.63. Together: 82 trades, 91% wins, +1.95 a trade, worst day −13.6, and 7 losses in the whole sample. The London leg came from the same tape over 95 nights: fading the Asia-session high and low from 03:00 with a three-point target won 90% of 130 trades for +0.93 a trade, every month positive; yesterday’s close, 84% of 62 for +0.79. Asia itself got no rule: nothing faded profitably over 90 nights, so the desk stays quiet there.
A second pass over the same tape tightened three things, and the desk has run on the tighter rule since 8 September. London stops taking new trades at 07:55 rather than 08:25, because the last ninety minutes were 36 trades at 78% for a small loss; on the rest, 146 trades at 91% and every month green. A wall is only faded when it sits inside the overnight range, because walls the night had already run through were ten trades for nothing; the ones inside won 93% of 70. And a level that has been traded through now needs a whole minute six points clear before it rests again, where a single tick used to be enough. Breakeven stops, time stops and one-sided fades all tested worse and were left out.
Seven losses is a small number to build a claim on, so I am not building one. The rule was the best cell of a large grid on two months of data, the overnight fade lost money in April and June before the walls existed, and a twelve-point stop means three wins pay for one loss. That is why the desk runs on paper, on the air, with the record kept in the open. The rule changes when the record says so, not before.
Hypothetical performance results have many inherent limitations. No representation is being made that any account will or is likely to achieve profits or losses similar to those shown; simulated results do not account for slippage, commissions, or the impact of the trades themselves. Nothing on this page is advice.
Today’s walls, free →·The walls’ graded record →·The terminal, five days for $0 →
