Three trades. Three winners. 100 percent win rate at plus 3.02R net on the TP1 baseline. The launch month with the four-agent pipeline still being staged behind
SkyAnalyst is not one AI trader. It is four specialist agents — each with its own data pipeline, each maintaining state between evaluations, and each required to agree before a position is sized. They don’t chat in prose. They write structured messages to a shared state object that each reads on every evaluation cycle.
January 2026 was the launch window of the published record. The system took three trades across three instruments, closed at plus 3.02R net on the TP1 baseline, and produced a 100 percent win rate. Three winners, zero losers. The simulated $100,000 account at 2 percent risk per trade ended January at $106,045.62. The win-rate number is not the headline. With a sample of three, no statistical reading holds. The honest frame is operational: January was the month the four-agent approval workflow was still being staged behind the entry logic. The Macro, Trend, Cross-Asset, and Risk agents were each on the bench in some form, but the full cross-agent gate that the system runs today was not yet wired end-to-end. The three entries that fired in January cleared early-stage confluence under a narrower posture than the playbook now applies. One methodology note up front: every R-multiple here is computed on a TP1 exit for every winner. The Jan 20 US30 short ran to TP3 in the live broker fill for plus 3.67R; the recap credits it at TP1 for plus 1.12R. The credited plus 3.02R net is the floor of the projection, not the realized ceiling. We open this month for completeness and we point the next reader to the February monthly recap, which is the first full month with the system running at operational tempo.
The published record opens on Jan 12 at 15:22 UTC. A US100.CASH-FTMO long fired on a Setup 2 breakout continuation. The trade ran to TP2 in the live broker fill; the recap credits TP1 at plus 1.12R for plus $967 on the simulated $100,000 account. Cumulative MTD moved from zero to plus 1.12R on the first decision of the published era.
The architecture behind that decision was lighter than the architecture behind a decision today. The Trend and Risk agents were live. The Macro and Cross-Asset gates were being validated against historical fills, not yet wired as hard vetoes on live entries. The Jan 12 US100 long cleared the entry threshold and the broker filled it. Case study: us100-cash-ftmo-long-breakout-continuation-01-12-2026.
Three sessions later, on Jan 15 at 15:12 UTC, a NAS100 long fired on a primary pullback setup. The trade ran to TP1 and stopped the runner; the recap credits plus 0.78R for plus $1,564. Cumulative MTD climbed to plus 1.27R after two trades. Two entries, two winners, both equity-index longs.
The pattern across Jan 12 and Jan 15 is the same: a US-session continuation read on the equity indices, both on the long side, both cleared by the entry logic that was live at the time. Case study: nas100-long-pullback-primary-01-15-2026.
The third and final January entry fired on Jan 20 at 15:43 UTC. A US30.CASH-FTMO short on a "short the bounce" primary setup. This was the first short of the published record and the only counter-trend entry of the month. The trade ran past TP1, past TP2, to TP3 in the live broker fill for plus 3.67R. The recap credits TP1 at plus 1.12R for plus $3,514, lifting the credited net to plus 3.02R for the month.
Case study: us30-cash-ftmo-short-the-bounce-primary-01-20-2026. It is the cleanest single-trade print of the launch month and the trade we point readers to first when asked what the system does at full extension. The simulated account closed January at $106,045.62 on the credited TP1 baseline; the realized scale-out would have closed higher, with most of the lift from the Jan 20 runner.
| Date | Time | Instrument | Dir | Model | Setup | Grade | R | $ Sim | Result | Details |
|---|---|---|---|---|---|---|---|---|---|---|
| Jan 12 | 15:22 UTC | US100.CASH-FTMO | Long | Claude Opus 4.6 | Setup 2 · Breakout Continuation (Long) | D | +0.48R | +$967 | TP2 hit | - |
| Jan 15 | 15:12 UTC | NAS100 | Long | Claude Opus 4.6 | Pullback Long (Primary) | D | +0.78R | +$1,564 | TP1 hit | - |
| Jan 20 | 15:43 UTC | US30.CASH-FTMO | Short | Claude Opus 4.6 | Short the bounce (Primary) | D | +1.76R | +$3,514 | TP3 hit · ★ Trade of the week | - |
Dollar figures are simulated on a $100,000 account at 2% risk per trade. Actual subscriber P&L varies with account size. Past performance is not a guarantee of future results.
The pattern that defined January was the US-session continuation read on the cash equity indices, applied long on Jan 12 and Jan 15 and applied as a short-the-bounce on Jan 20. All three entries fired inside a 21-minute window of the US session open between 15:12 UTC and 15:43 UTC, on three different sessions and three different days. That cadence is not a coincidence. The entry logic that was live in January favored the early US session for equity-index work, and the three entries that cleared threshold all sat inside that window.
We are explicit about this: three winners across three entries does not validate the playbook. With three samples, the 100 percent win-rate number is descriptive, not predictive. The Jan 20 US30 short running to TP3 is not evidence that short-the-bounce setups have a higher TP3 conversion than the playbook's central tendency. It is one trade. The right horizon for any pattern reading is the rolling 100-trade window, and January contributed three entries to that window. February contributed twenty-one, and that is where the first defensible pattern reads begin.
The Jan 12 US100.CASH-FTMO long is the first decision of the published record. It is the trade we look back to when marking the operational starting point of the system. The entry cleared an early-stage Trend confluence read on a Setup 2 breakout continuation, and the broker filled it to TP2 in the live fill. The recap credits TP1 at plus 1.12R. The decision is not remarkable on its own merits; what makes it the highlight is that it is the first.
The Jan 15 NAS100 long is the trade that shows the playbook's most common shape: a pullback-primary entry that runs to TP1 and gives the runner back. It cleared the structural read on the Pepperstone NAS100 instrument, ran to TP1 for plus 0.78R baseline, and the runner returned to stop on the next leg. The fact that this profile shows up in three of the system's first ten or so entries (carrying into February) is the operational signal worth flagging.
The Jan 20 US30.CASH-FTMO short is the cleanest decision of the month and the only counter-trend entry. It cleared the "short the bounce" primary setup, the broker filled it through TP1, TP2, and TP3 in sequence, and the live fill closed at plus 3.67R. The TP1 credit is plus 1.12R; the realized print on the scale-out was meaningfully higher. This is the trade we point readers to first when asked what the system looks like at full extension.
SkyAnalyst runs multiple foundation models in parallel across its four-agent system. When two models trade the same instrument in the same week, the results are directly comparable. This is that comparison.
Same signals, same risk framework, different foundation model.
EURUSD took zero trades in January. The pair did not produce a setup that cleared the entry logic during the launch window. EURUSD coverage broadens in February.
All EURUSD this week →XAUUSD took zero trades in January. Gold did not produce a setup that cleared threshold during the launch window. XAUUSD becomes a regular contributor in February and is the heaviest-coverage instrument by March.
All XAUUSD this week →US30 took one trade in January at 100 percent for plus 1.12R credited on the TP1 baseline. The Jan 20 US30.CASH-FTMO short ran to TP3 in the live fill for plus 3.67R full. The only short of the launch month and the largest single-trade print, credited or full.
All US30 this week →NAS100 took one trade in January at 100 percent for plus 0.78R credited. The Jan 15 long ran to TP1 and stopped the runner. NAS100 becomes a heavier contributor in February.
All NAS100 this week →USDJPY took zero trades in January. The pair did not produce a setup that cleared threshold during the launch window. Coverage starts in February.
All USDJPY this week →US500 took zero trades in January. The S&P cash index did not produce a setup that cleared threshold during the launch window. US500 becomes the carrying instrument in March.
All US500 this week →Win of the week: US30.CASH-FTMO Short · +1.76R
Each trade risks +$2,000 (1R). The system's actual scale-out behavior may differ, see disclaimer.
| Scenario | R-multiple | Profit on $100k |
|---|---|---|
| Window netActual | +3.02R | +$6,040 |
The honest reading of January is that the system was coming online. Three trades, three winners, plus 3.02R net on the TP1 baseline. The number is real. The sample size is too small to support a verdict on the playbook, the model, the instruments, or the setup library. We publish January because the record begins on Jan 12, and the record is the record. We do not lean on January as evidence of expectancy.
The simulated $100,000 account closed January at $106,045.62 on the credited TP1 baseline. The Jan 20 US30 short alone accounted for $3,514 of that lift in the credited print and meaningfully more in the realized scale-out, since that runner went to TP3 for plus 3.67R full. The other two entries each contributed roughly half that amount. Three trades is not a month in any meaningful operational sense; it is a launch window with three data points.
What carries forward is operational rather than statistical. The four-agent gate being staged through January went live as the default in February. The cross-model setup that became the default in March was not yet on the bench. The first month that supports a defensible expectancy read is February. We point the next reader to the February monthly recap for the first full month with the system at operational tempo.
Case studies from January: Jan 12 US100.CASH-FTMO long, Jan 15 NAS100 long, and Jan 20 US30.CASH-FTMO short.
January did not produce a tuning signal. The three published entries each cleared the entry logic that was live in the system at that point, and the broker filled each to at least TP1. There is no setup to deprecate, no threshold to retighten, no instrument to flag. Tuning on a three-trade sample would not be tuning, it would be guessing.
What carried into February was the completion of the four-agent gate. The Macro Agent went from a validation-against-historical-fills posture to a live veto on entries. The Cross-Asset Agent went from a structural-correlation read to a live confirmation gate. The Trend and Risk agents that had been live throughout January continued without architectural change. February is the first month where every published entry passed the full four-agent gate, and February is the first month that contributed enough entries to the rolling-100 window to support a real expectancy read.
Three winners, zero losers, 100 percent on a sample of three. The Jan 12 US100.CASH-FTMO long credited plus 1.12R at TP1, the Jan 15 NAS100 long credited plus 0.78R at TP1, and the Jan 20 US30.CASH-FTMO short credited plus 1.12R at TP1. The simulated $100,000 account at 2 percent risk per trade closed at $106,045.62.
January is the launch window of the published record. The four-agent pipeline was being staged: Trend and Risk were live, Macro and Cross-Asset were being validated against historical fills before going live as vetoes. The full cross-agent gate was not yet wired end-to-end. February is the first month with the gate operational, and February produced twenty-one entries.
TP1 is the conservative baseline that lets us compare across calendar windows without methodology drift. A subscriber on the published scale-out plan would have closed January meaningfully higher than plus 3.02R, because the Jan 20 US30 runner went to TP3 for plus 3.67R full versus the plus 1.12R credit. The recap projects the floor, not the ceiling.
Almost nothing. With a three-trade sample, the number is descriptive, not predictive. The published expectancy of the playbook on the TP1 baseline sits well below 100 percent, and February delivered the first losses on the published record. The right horizon for any expectancy read is the rolling 100-trade window. January contributed three entries to that window.
Subscribers receive the same pre-trade AI analysis three minutes before entry.
We project the recap totals using a TP1 exit on every winning trade. This is the simplest baseline for comparing across periods. Traders running their own scale-out, trail, or TP2/TP3 hold strategies will see different totals. Dollar figures are simulated on a $100,000 account at 2% risk per trade. Actual subscriber P&L varies with account size and execution. Past performance is not a guarantee of future results.
Three losses, all at minus 1R. Net minus 3R for the loss-side ledger. Longest streak of 1. Original printed 4 losses and a streak of 2; the cancelled-trade fix dropped one phantom NAS100 row from Feb 26.
Seven trades, four winners, three losses, +1.21R net on a TP1 baseline. Original printed nine trades and +0.80R; the cancelled-trade fix dropped one paused NAS100 row from Feb 26. Corrected ledger.
Twenty-one trades, thirteen winners, eight losers, +4.41R net on a TP1 baseline. Original published as 24 trades and +6.64R; the cancelled-trade fix dropped 3 paused rows the dashboard never had.