Month-Bridge Week: Six Trades, +1.68R Net, and Two TP3 Winners Close March
Six trades, four winners, two losses, +1.68R net on a TP1 baseline at a 66.7 percent win rate. The week straddled the Mar/Apr boundary and closed March with two
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.
Before the trade, meet the system.
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.
Six trades, four winners, two losses, +1.68R net on a TP1 baseline at a 66.7 percent win rate. That is the scorecard for the week of March 30 to April 5, 2026, and it sits exactly where a recap should sit one week after a bumper. Cumulative equity traveled from $100,000 down to $98,000 by Monday's close, back to $103,340 by Wednesday afternoon, down to $101,340 after the Wednesday stop, and settled at $103,360 on Thursday's NAS100 long. Through Apr 6, 2026, the system has banked +7.79R YTD across 63 trades from Jan 12 inception. The simulated $100,000 account at 2 percent risk per trade sits at $115,576.23 on a static basis. The week's structural feature is the calendar. It bridges March and April, with Tuesday March 31 producing both of the week's TP3 winners (a USDJPY short and a EURUSD long inside 46 minutes) and closing March on a high note. The companion [weekly drawdown report](/blog/weekly-drawdown-report-2026-03-30) covers the gate-firing mechanics. Last week's bumper sits at [the previous recap](/blog/weekly-recap-2026-03-23); longer-window context lives in [February's monthly recap](/blog/monthly-recap-2026-02).
Act 1: Monday's stop, Tuesday's two TP3 winners close March
Monday March 30 produced one trade: a USDJPY short at 15:04 UTC on a bearish-continuation read that stopped at -1R as the pair held its structural shelf. Equity pulled to $98,000.
Tuesday March 31 was the inflection. A USDJPY short at 14:53 UTC on a pullback into the 159.20 to 159.30 zone ran the full ladder to TP3 for +1.20R. Forty-six minutes later, a EURUSD long at 15:39 UTC on a bullish pullback read also ran to TP3 for +0.75R. Tuesday closed +0.95R cumulative, with equity at $101,900. March closed there.
Act 2: Wednesday's mixed session, NAS100 wins and USDJPY stops
Wednesday April 1 produced two trades inside ten minutes. A NAS100 long at 14:37 UTC on a pullback into 5-minute dynamic support ran to TP2 for +0.72R. Ten minutes later, a USDJPY short on a pullback-rejection read stopped at -1R as the pair recaptured the rejection level. Wednesday closed +0.67R cumulative.
Act 3: Thursday's recovery, the week settles +1.68R
Thursday April 2 produced one trade. Equity had pulled to $101,340 after the Wednesday stop, and the Risk Agent's drawdown gate fired before the afternoon session opened.
A NAS100 long at 15:57 UTC on a 78.6 percent Fib pullback into structural support ran to TP2 for +1.01R. Equity settled at $103,360, leaving the week +1.68R cumulative. Friday April 3 produced no qualifying setups.
Every trade the system took.
| Date | Time | Instrument | Dir | Model | Setup | Grade | R | $ Sim | Result | Details |
|---|---|---|---|---|---|---|---|---|---|---|
| Mar 30 | 15:04 UTC | USDJPY | Short | Claude Opus 4.6 | USDJPY Short Bearish Continuation | C+ | -0.25R(SL) | -$500(SL) | Stop hit | Read case → |
| Mar 31 | 14:53 UTC | USDJPY | Short | Claude Opus 4.6 | USDJPY short on pullback to 159.20-159.30 | C+ | +0.30R(TP1) | +$600(TP1) | TP3 hit | Read case → |
| Mar 31 | 15:39 UTC | EURUSD | Long | Claude Opus 4.6 | Bullish Pullback Long | C+ | +0.75R(TP1) | +$1,500(TP1) | TP3 hit | Read case → |
| Apr 1 | 14:37 UTC | NAS100 | Long | Claude Opus 4.6 | NAS100 LONG — Pullback to 5m Dynamic Support | C+ | +0.72R(TP1) | +$1,440(TP1) | TP2 hit | Read case → |
| Apr 1 | 14:47 UTC | USDJPY | Short | Claude Opus 4.6 | SHORT USDJPY pullback rejection | C+ | -0.25R(SL) | -$500(SL) | Stop hit | Read case → |
| Apr 2 | 15:57 UTC | NAS100 | Long | Claude Opus 4.6 | NAS100 Pullback Long at 78.6% Fib / Structural Support | C+ | +1.01R(TP1) | +$2,020(TP1) | TP2 hit · ★ Trade of the week | Read case → |
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.
Pattern of the week
The week's recurring pattern was mean reversion of the recap signal itself. Last week was the bumper at +4.19R on fourteen trades. This week reverted to a 4-2 mix and a small positive net on six trades. The published win rate sits in the high 50s; a 66.7 percent week is above the median but well inside the variance band.
Inside the six trades, NAS100 carried the winners on pullback longs while USDJPY split across three different reads. NAS100 ran twice on the same setup family and netted +1.73R. USDJPY ran three times on three distinct reads and netted -0.80R.
How Tuesday produced two TP3 winners on opposite-direction trades
The USDJPY short at 14:53 UTC and the EURUSD long at 15:39 UTC were taken on different macro reads, not contradictory ones. The Cross-Asset Agent cleared both pairs as independently tradeable on the local rate-differential print. Both ran the ladder.
Decision highlights
The Tuesday decision to take a EURUSD long 46 minutes after the USDJPY short had triggered is the cleanest dollar-pairs judgment of the week. The two sat on opposite sides of the dollar but cleared confluence on independent macro reads. The Cross-Asset Agent confirmed the pairs as decoupled before the EURUSD entry was sized. Both ran to TP3.
The Wednesday decision to take a USDJPY short ten minutes after the NAS100 long is the harder one to read. Both cleared the same threshold; the USDJPY short stopped on a structural recapture inside the trade lifecycle. The regime shift that invalidated the short formed after the trigger.
The Risk Agent's drawdown gate fired on Thursday morning after equity pulled back from Wednesday's $103,340 peak. The gate held confluence steady and prevented the Trend Agent from loosening threshold on the afternoon's NAS100 setup. The recovery trade triggered on the same floor the morning had used.
Claude vs GPT: who led the week.
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.
- NAS100+1.7R · 2 trades
- EURUSD+0.8R · 1 trade
- USDJPY-0.2R · 3 trades
Same signals, same risk framework, different foundation model.
Six instruments, six stories.
EURUSD took one trade for a 100 percent win rate and +0.75R net. Tuesday's bullish pullback long at 15:39 UTC ran to TP3 and was one of the week's two month-closing winners.
All EURUSD this week →US30 was inactive. No setup cleared confluence; the index held a tight intraday range across all four sessions.
All US30 this week →NAS100 was the volume leader for winners with two trades, both pullback longs into structural support, both to TP2, netting +1.73R. The pair carried the week's positive net on its own.
All NAS100 this week →USDJPY took three trades for a 33.3 percent win rate and -0.80R net. The Tuesday pullback short to TP3 was the week's largest single winner at +1.20R; the Monday continuation short and the Wednesday pullback-rejection short both stopped at -1R.
All USDJPY this week →US500 was inactive. The index consolidated with no patterns scoring above the confluence floor.
All US500 this week →Win of the week: NAS100 Long · +1.01R
Loss worth learning from
What the system saw that was right
Both losses cleared the published confluence threshold at trigger. Monday's continuation short scored on structural shelf and macro lean. Wednesday's pullback-rejection short scored on the recapture failing in real time.
What the system got wrong
Nothing in the entries themselves. Both share a regime-shift sensitivity the exit logic does not address: the macro context repriced inside the trade lifecycle, and the stop was the only exit. The system does not re-evaluate in-position trades dynamically. That is a known cost of the architecture, not a tuning signal.
On a $100k account at 2.0% risk per trade.
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 | +2.28R | +$4,560 |
All six agents combined.
From the desk
Through Apr 6, 2026, the cumulative ledger reads +7.79R YTD across 63 trades from Jan 12 inception. The same $100,000 account at 2 percent risk per trade sits at $115,576.23 on the static line and $115,259.39 on the compounded line. The spread is the cost (or benefit) of compounding through a positive-expectancy edge as winners cluster around losses.
The honest reading is that the system reverted toward expectancy after last week's bumper, exactly as the rolling record predicts it should. Last week was +4.19R on fourteen trades. This week was +1.68R on six. It is published expectancy doing what published expectancy does.
The architecture point is the Tuesday trade pair. A USDJPY short and a EURUSD long, taken 46 minutes apart on the same dollar print, cleared their own confluence thresholds and ran independently to TP3. The Cross-Asset Agent did not throttle the second entry because the local rate-differential read had already cleared the pairs as decoupled. A discretionary trader would have hesitated on the second entry; the system did not.
The Risk Agent's gate fired on Thursday morning and held threshold steady through the afternoon. The NAS100 long that closed the week's positive net was taken on the same confluence threshold the morning had used. That is the gate working as designed.
What we're tuning
There is no entry-side tuning signal in either loss individually, and none in the cluster. Both cleared the published threshold; both stopped on regime shifts inside the trade lifecycle. The 4-2 mix is a fine outcome at the published win rate. The Tuesday dispersion (USDJPY and EURUSD running to TP3 inside 46 minutes on opposite sides of the dollar) is the cross-asset architecture working as designed.
The NAS100 pair (two pullback longs, both to TP2) sits inside the same setup family. We will track whether pullback-into-structural-support is producing TP2 stops at a higher rate than TP3 across the rolling sample.
At a Glance
Week at a glance
How did the week close +1.68R net after Monday's stop and Wednesday's USDJPY loss?
Tuesday's two TP3 winners (USDJPY at +1.20R and EURUSD at +0.75R) recovered Monday's stop with room to spare. Wednesday's NAS100 long added +0.72R against a USDJPY -1R loss. Thursday's NAS100 long added another +1.01R. The arithmetic settles at +1.68R net on six trades.
Why did only Claude Opus 4.6 trade this week?
Single-model windows occur naturally when one family's confluence math clears threshold and the other does not on the same setups. Six trades is too small a sample to read model dispersion. The longer-window head-to-head lives in February's monthly recap.
What does it mean for the system to revert toward expectancy after a bumper week?
The published win rate sits in the high 50s with average R per trade in the small positive. A bumper week is a positive outlier; the median weekly outcome is a small positive net on a 4-2 or 5-3 mix. This week landed above the median but well inside the variance band.
When does the drawdown gate fire, and what did it do this week?
The gate fires when cumulative equity drops a configured percentage below a recent peak. Once fired, it prevents the Trend Agent from lowering confluence threshold. The Thursday gate held through the afternoon; the recovery NAS100 long triggered on the same threshold the morning had used.
What is the difference between the +1.20R baseline and a case-study figure for the Tuesday USDJPY short?
Recap R-multiples use a TP1-baseline projection on every winner. The Tuesday USDJPY short ran the full ladder, so the recap baseline reads +1.20R while a standalone case study would document the full TP3 ladder. Both describe the same trade on different exit assumptions.
Get next week’s trades before they print.
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.
From the SkyAnalyst Journal
EURUSD Short on May 15: Six Waits, One Enter, a Clean TP1 Fill
A risk-off Euro short where the system scored six consecutive waits in the low 40s, then flipped to enter at 62 percent, and the position closed TP1 for +2.00R (TP1) with zero recorded drawdown.
Three stops, one regime: the week the tape refused to confirm
Three losses, 2.25R given back against a year that still reads +20.43R. The honest portfolio view: what every stop taught us, and what the drawdown curve says about a week that drew down 2.4 percent and recovered.
May 11-17, 2026: Seven-Three Week, +5.96R Net Across a Model Upgrade
Ten canonical trades, seven winners, three losers, +5.96R net at the TP1 baseline. Tuesday and Wednesday ran on Claude Opus 4.6, Friday switched to Opus 4.7, and GBPUSD came online as a new instrument and won both its trades.