Nine Trades, Six Winners, +5.1R: Feb 9-15 Weekly Recap

Restated: Gold (XAUUSD) was part of SkyAnalyst's coverage from inception (Jan 12, 2026) through May 2026. We've since narrowed coverage to six instruments — EURUSD, GBPUSD, USDJPY, US30, NAS100, US500 — and these numbers are restated for the current lineup. The original publish date is preserved; cumulative figures have been recomputed.

Nine trades. Six winners. Three losses. Net +5.10R on a TP1-baseline scoreboard, a 66.7 percent win rate, and the strongest week of February for the simulated account. The shape of the week ran on two clocks: three steady US30 longs across Monday, Tuesday, and Wednesday, then a quiet Thursday, then a 95-minute compression on Friday afternoon that fired five trades across US30, NAS100, and US500. Through Feb 16, 2026, the system has banked +7.12R YTD across 13 trades from Jan 12 inception. The simulated $100,000 account at 2 percent risk per trade sits at $114,249.25 on a static basis and $114,794.55 on the compounded line. By Wednesday's close the running net for the week was +5.24R, with the simulated account at $110,488. Thursday produced no entries; Friday's first 14 hours produced none either. Inside a 95-minute window between 15:02 and 16:37 UTC on Friday afternoon, the system fired five trades. Three winners, two losers, the week closing at +5.10R net. For the matching variance view see the [Weekly Drawdown Report](/blog/weekly-drawdown-report-2026-02-09); for the rolling February context see the [Monthly Recap](/blog/monthly-recap-2026-02). A note on methodology. Every winner in this scoreboard is exited at TP1 as the conservative baseline; two winners ran past TP1 to TP2 before close. Subscribers running scale-out (one-third at each of TP1, TP2, TP3) saw a higher number than +5.10R for the week.

Act 1: Mon-Wed, three US30 longs into rising VWAP

Feb 9 at 15:02 UTC the Trend Agent flagged a US30 long, pullback to support, confluence cleared the threshold, the position triggered. The trade ran to TP2 for +1.31R on the TP1-baseline accounting. The macro regime had rated lean-bullish-equity on the morning print and the structural read on the 5-minute chart was a clean retest of the rising VWAP.

Feb 10 at 16:31 UTC the same setup printed again. Pullback into VWAP and the 61.8 percent retracement, confluence cleared, the system entered. The position ran to TP1 and closed at +2.30R on the TP1-baseline. Thirty minutes earlier at 16:01 UTC the system had also entered a US30 long on a similar pullback that did not develop; that trade stopped at -1R within the hour. Two US30 longs Tuesday, one winner, one stop.

Feb 11 at 16:02 UTC the third US30 long of the week. Responsive entry into intraday support, confluence above threshold, the trade ran clean to TP1 for +2.64R on the TP1-baseline, the largest winner of the week. By Wednesday's close the running net was +5.24R and the simulated account sat at $110,488.

Act 2: Thursday, the gate closed

Feb 12 produced no entries. The macro regime softened in the morning and the Macro Agent's gate went neutral by the New York open. The Trend Agent ran every cycle and entered nothing. Friday's overnight session through 15:00 UTC also produced no qualifying setup. For 36 hours the system sat.

Act 3: Friday afternoon, 95 minutes, five trades

Feb 13 at 15:02 UTC the regime gate re-rated, this time to lean-bearish-equity on a softer-than-expected morning print. Within ninety-five minutes the system fired five trades. At 15:02 UTC a US30 short for +0.57R. At 15:24 UTC a NAS100 short that stopped at -1R after the cross-asset confirmation faded inside the same minute the entry triggered. At 15:42 UTC a US500 short, fade into resistance, for +0.74R. At 15:44 UTC a US30 short, fade into resistance, that stopped at -1R when the equity tape briefly bid back. At 16:37 UTC a US500 long on a momentum continuation read as the tape balanced, for +0.55R.

Five trades inside 95 minutes is sparse compared with a normal active session and dense compared with the Thursday silence. Three winners, two losers, net +1.86R for the window, the week closing at +5.10R.

Pattern of the week

The pattern of the week was the bullish-equity continuation long on US30. Three of the nine trades were US30 longs into the rising session VWAP, all entered between 15:02 and 16:31 UTC, all on the same macro read of lean-bullish-equity. Two of the three ran to TP1 or TP2 and one stopped. The setup is the same one documented in earlier case studies: a pullback into the rising VWAP with confluence above the 55 percent threshold.

Why the regime mattered more than the setup

A pullback-to-VWAP long and a pullback-to-support long are mechanically related but not identical. What unified the Mon-Wed sequence was the macro regime. With the morning prints supporting equity longs and the cross-asset reads aligned, every clean structural retest of intraday support cleared the Trend Agent's confluence math.

The compression on Friday

Friday's 95-minute window is the inverse pattern. The regime re-rated mid-session, the structural reads printed in tight succession across three instruments, and the system entered five trades. Two stopped on cross-asset divergences that softened within the same minute as entry, three ran to TP1 or TP2. The five-trade compression is unusual; the system does not normally cluster that many entries that tightly. The compression itself is the artifact of a regime change inside a single afternoon, not a system tilt toward higher activity.

Decision highlights

The decision to take three US30 longs across Monday, Tuesday, and Wednesday on the same setup was a confluence call, not a pattern habit. Each entry cleared the 55 percent threshold independently on its own evaluation cycle. The Trend Agent did not treat Tuesday's entry as a sequel to Monday's; it treated each as a separate calculation. The fact that the same pattern printed three days running is a feature of the macro regime, not of the system's memory.

The decision to enter the second Tuesday US30 long thirty minutes after the first one stopped is the discipline beat of the week. The 16:01 UTC entry took -1R; thirty minutes later at 16:31 UTC the same instrument printed another clean pullback-to-VWAP setup and the system entered. A discretionary trader who had just been stopped on the same instrument might have hesitated; the system did not, because hesitation is not in the threshold logic. The second entry ran to TP1 for +2.30R.

The decision to take all five Friday afternoon trades inside 95 minutes is the most aggressive activity the system has shown in February. Each entry cleared confluence on its own merits during a regime change that produced multiple structural reads in tight succession across three instruments. The Risk Agent sized each independently and did not apply any pace governor; the system has no per-window trade cap. Three winners, two losers, net +1.86R for the window.

Loss worth learning from

What the system saw that was right

The Friday NAS100 short triggered at 15:24 UTC. The macro regime had re-rated to lean-bearish-equity inside the previous half hour, the dollar was firming, and the structural read was a continuation short on a sell-the-rip pattern. Confluence cleared the 55 percent threshold and the Trend Agent's confluence math returned a clean entry.

What the system got wrong

Nothing in the entry logic. What killed the trade was that the cross-asset confirmation, the read on the parallel US500 chart, softened inside the same minute the entry triggered. By the time the system would have re-evaluated, the position was already underwater. The architecture does not re-evaluate open positions; the only exit paths are the stop, the trailing logic, and the take-profit targets. The stop triggered for -1R within thirty minutes of entry.

What we'd do the same

The entry. A 55-plus percent confluence trade on a fresh regime re-rate is what the system is built to take. The Friday 15:24 UTC entry was a clean confluence call. The fact that the cross-asset read softened inside the same minute is the kind of intraminute timing variance that the threshold does not protect against. Removing this trade from the playbook would tighten the win rate but lower the rolling expectancy.

From the desk

Through Feb 16, 2026, the cumulative ledger reads +7.12R YTD across 13 trades from Jan 12 inception. The simulated $100,000 account at 2 percent risk per trade sits at $114,249.25 on the static line and $114,794.55 on the compounded line. The spread between static and compounded is small at this point in the season because the equity curve has been mostly monotonic; the compounding kicks in more visibly once the system has banked enough R-multiples to materially shift the per-trade dollar risk.

The honest framing of the week is that the system did its job. Six winners, three losses, 66.7 percent win rate, net +5.10R on the conservative TP1 baseline. The structural fact that earned the result is that the macro regime supported the same pullback-to-VWAP long pattern three sessions running on US30, and the system took each clean structural read as a fresh confluence calculation rather than as a continuation of a streak.

Two readings of this week are worth keeping separate. The first is the reading that any subscriber would naturally make: nine trades, six winners, net positive, good week. The second is the reading the architecture invites: each entry was its own confluence math, the three US30 longs were not a single bet repeated, and the Friday compression was a regime-driven cluster, not a tilt toward over-activity. The system does not "warm up" or "go on a run." It runs the same threshold logic every cycle, and the visible clustering of activity is a function of the tape, not the system's internal state.

For readers comparing this to discretionary trading services: a 66.7 percent win rate on nine trades inside one week is well above the long-run distribution that the system is calibrated against. We expect this number to mean-revert. The rolling 100-trade win rate is closer to 50 percent, the rolling expectancy is the same +0.5 to +0.7R per trade on the TP1-baseline, and the variance week to week can be large. A 66.7 percent week is welcome; it is not a forecast.

The next case study will be the [Wednesday US30 long](/blog/us30-long-responsive-intraday-support-02-11-2026), the largest single winner of the week. The full February context lives in the [Monthly Recap](/blog/monthly-recap-2026-02); the matching variance lens lives in the [Weekly Drawdown Report](/blog/weekly-drawdown-report-2026-02-09).

What we're tuning

There is no actionable tuning signal in the Friday NAS100 stop. The trade entered cleanly, the macro context held through the entry, and the intraminute cross-asset softening is below the resolution at which the system makes decisions. Tightening the cross-asset window to sub-minute resolution would introduce significantly more noise than signal and would compromise entries on instruments where the confirmation read genuinely lags by a few seconds.

What is worth tracking for the running record: the Friday 95-minute compression produced five entries, the most the system has fired in a single window during February. That activity level is a function of the regime re-rate inside a single afternoon, not a tilt toward higher activity. We will continue to monitor the per-window entry count in the [February monthly recap](/blog/monthly-recap-2026-02) for any signs of a structural shift in entry pace.