Twelve Trades, Five Winners: A Chop-Regime Week That Paid Both Sides
Twelve filled trades, five winners, seven losses, -2.42R net on a TP1 baseline. A Monday +1.5R opener, a Wednesday EURUSD short paid twice, and a Friday USDJPY
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.
Twelve trades, five winners, seven losses, -2.42R net on a TP1 baseline. That is the scorecard for the week of March 16 to March 22, 2026. Equity descended from $100,000 to $95,167.32, a 4.8 percent drawdown across five sessions. Through Mar 23, 2026, the system has banked +2.39R YTD across 46 trades from Jan 12 inception. The simulated $100,000 account at 2 percent risk per trade sits at $104,777.62 on the static line and $103,769.83 on the compounded line. The week opened with a clean read. Monday's US500 long pullback at 14:59 UTC paid +1.5R on TP1, the largest baseline winner of the window. From there the tape repriced into chop. Tuesday produced three stop-outs inside 26 minutes across US500, NAS100, and US30 longs. Wednesday cleared two EURUSD shorts to TP3 inside 70 minutes on the same pair. Friday closed with a thirty-second sequence in which a US30 short stopped at -1R and a USDJPY long ran to TP3. Two trades documented as standalone case studies anchor the week, the US500 long pullback on March 16 and the EURUSD short on March 18. The companion weekly drawdown report covers the seven-loss arithmetic. The prior week's recap sets up the regime context for this one.
Act 1: Monday and Tuesday, the opener and the three-stop sequence
Monday March 16 produced one trade and one winner. At 14:59 UTC a US500 long pullback into prior breakout support ran to TP1 for +1.5R, the largest baseline winner of the week. Monday closed +1.5R cumulative.
Tuesday March 17 produced three trades inside 26 minutes, all stops. A US500 long at 14:10 UTC, a NAS100 long at 14:31 UTC, and a US30 long at 14:36 UTC all triggered on a tape that cleared lean-bullish at the open and rolled inside the trade lifecycle. Equity fell to $97,000 by end of Tuesday.
Act 2: Wednesday, the doubled EURUSD short pair
Wednesday March 18 produced three trades. A USDJPY pullback long at 14:15 UTC stopped at -1R when the retest did not hold. At 14:41 UTC a EURUSD short ran to TP3 for +0.37R. At 15:50 UTC a second EURUSD short, a rally fade into VWAP and resistance, also ran to TP3 for +0.75R. The doubled result on the same pair inside 70 minutes is documented in the EURUSD rally-fade case study. Wednesday closed -1.38R cumulative.
Act 3: Thursday and Friday, dispersion in both directions
Thursday March 19 produced two trades inside 20 minutes. A NAS100 short at 14:50 UTC stopped at -1R. At 15:10 UTC a US500 short paid +1.18R as a failed bounce into VWAP and prior-day-low resistance resolved lower. Thursday closed -1.20R cumulative.
Friday March 20 produced three trades inside 15 minutes. A US30 pullback-failure short at 15:13 UTC stopped at -1R. Thirty seconds later a USDJPY long ran to TP3 for +0.78R. At 15:28 UTC a EURUSD short into resistance stopped at -1R. The week closed at $95,167.32.
Every trade the system took.
| Date | Time | Instrument | Dir | Model | Setup | Grade | R | $ Sim | Result | Details |
|---|---|---|---|---|---|---|---|---|---|---|
| Mar 16 | 14:59 UTC | US500 | Long | GPT-5 | LONG pullback (buy-the-dip) | B | +1.50R(TP1) | +$3,000(TP1) | TP1 hit · ★ Trade of the week | Read case → |
| Mar 17 | 14:10 UTC | US500 | Long | GPT-5.4 | US500 LONG — Pullback buy into prior breakout support | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | Read case → |
| Mar 17 | 14:31 UTC | NAS100 | Long | GPT-5.4 | NAS100 LONG | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | Read case → |
| Mar 17 | 14:36 UTC | US30 | Long | GPT-5.4 | US30 LONG | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | Read case → |
| Mar 18 | 14:15 UTC | USDJPY | Long | GPT-5.4 | USDJPY pullback long retest-and-hold | C+ | -0.25R(SL) | -$500(SL) | Stop hit | Read case → |
| Mar 18 | 14:41 UTC | EURUSD | Short | GPT-5.4 | EURUSD SHORT | C+ | +0.37R(TP1) | +$742(TP1) | TP3 hit | Read case → |
| Mar 18 | 15:50 UTC | EURUSD | Short | GPT-5.4 | EURUSD SHORT rally fade into VWAP/resistance | C+ | +0.75R(TP1) | +$1,506(TP1) | TP3 hit | Read case → |
| Mar 19 | 14:50 UTC | NAS100 | Short | GPT-5.4 | NAS100 SHORT | B | -1.0R(SL) | -$2,000(SL) | Stop hit | Read case → |
| Mar 19 | 15:10 UTC | US500 | Short | GPT-5.4 | US500 SHORT - Failed bounce into VWAP / prior-day-low resistance | B | +1.18R(TP1) | +$2,360(TP1) | TP1 hit | Read case → |
| Mar 20 | 15:13 UTC | US30 | Short | GPT-5.4 | US30 SHORT (pullback failure into resistance) | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | Read case → |
| Mar 20 | 15:18 UTC | USDJPY | Long | GPT-5.4 | USDJPY LONG | C+ | +0.19R(TP1) | +$390(TP1) | TP3 hit | Read case → |
| Mar 20 | 15:28 UTC | EURUSD | Short | GPT-5.4 | EURUSD SHORT retracement into resistance | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | 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 the chop-regime split: a tape that paid both sides on the same day, on the same instrument, sometimes inside the same minute. The Wednesday EURUSD pair (two shorts on the same pair, both running to TP3) is the cleanest illustration. The Friday US30-then-USDJPY pair, thirty seconds apart with a stop and a TP3 winner, is the other.
Why a chop regime produces both winners and losers at threshold
A chop regime crosses the lean-bull or lean-bear threshold without committing. Some setups paid; some stopped when the tape repriced inside the trade lifecycle. Twelve filled trades is what a confluence-gated system produces when the macro keeps flipping at threshold and structure keeps offering setups on both sides of the same level.
Decision highlights
The Wednesday decision to take a second EURUSD short at 15:50 UTC, on the same pair the system had shorted to TP3 70 minutes earlier, is the clearest discipline beat. The confluence math did not register the prior winner; the second setup was scored on its own merits and ran to TP3 on the case-study ladder. The Macro Agent did not flip the regime read between the two entries.
The Friday USDJPY long, triggered thirty seconds after the US30 short stopped at -1R, is the cleanest demonstration of the chop regime. The two trades read the same macro state in opposite directions: a pullback-failure short on US30 and a separate dollar-strength long on USDJPY. The US30 stopped immediately; the USDJPY ran to TP3 on the same minute.
The Tuesday three-stop sequence (US500, NAS100, US30 longs inside 26 minutes, all stopping at -1R) is the hardest sequence in retrospect. Every entry triggered on a tape that cleared lean-bullish at the open and softened inside the trade lifecycle. We do not classify any as a system error; the threshold was met at every entry on the same confluence math the record runs on. The companion drawdown report covers the gate-firing mechanics.
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.
- US500+1.7R · 3 trades
- EURUSD+0.1R · 3 trades
- USDJPY-0.1R · 2 trades
- NAS100-2.0R · 2 trades
- US30-2.0R · 2 trades
Same signals, same risk framework, different foundation model.
Six instruments, six stories.
EURUSD took three trades, 66.7 percent win rate, +0.12R net. Wednesday's two short fades both ran to TP3 inside 70 minutes; Friday's short into resistance stopped at -1R.
All EURUSD this week →GBPUSD: no trades this week. The pair sat outside our setup criteria across all five sessions.
All GBPUSD this week →US30 took two trades, 0 percent win rate, -2.0R net. Tuesday's continuation long and Friday's pullback-failure short both stopped at -1R inside the same chop regime.
All US30 this week →NAS100 took two trades, 0 percent win rate, -2.0R net. Tuesday's morning long and Thursday's afternoon short both stopped at -1R on opposite reads of the same tape.
All NAS100 this week →USDJPY took two trades, 50 percent win rate, -0.22R net. Wednesday's pullback long stopped at -1R; Friday's long ran to TP3 thirty seconds after the US30 short stopped on the opposite read.
All USDJPY this week →US500 took three trades, 66.7 percent win rate, +1.68R net. Monday's pullback long paid +1.5R on TP1, the largest baseline winner of the week; Thursday's failed-bounce short paid +1.18R; Tuesday's pullback long stopped at -1R.
All US500 this week →Win of the week: US500 Long · +1.5R
Loss worth learning from
Loss worth learning: Tuesday's US30 long continuation
The hardest single loss to swallow is the Tuesday March 17 US30 long continuation at 14:36 UTC, the third stop-out inside 26 minutes after the US500 and NAS100 longs had already stopped on the same lean-bullish open.
What the system saw that was right
The macro tape had cleared lean-bullish at the New York open. DXY was soft on the 5-minute, yields neutral, and equities had printed an early-session push. The Trend Agent scored the setup at C+, a tradeable confluence inside the regime. The premise was a continuation of the morning push.
What the system got wrong
Nothing in the entry. The lean-bullish read had been confirmed at the open, and the structure offered a continuation level inside the trade lifecycle. What we got wrong, in retrospect, is the speed at which the tape repriced. The same regime that authorized three long entries inside 26 minutes flipped to chop before any of them reached TP1. The exit logic does not re-evaluate in-position trades; once the trigger fires, the only exit path is the stop. An exit-logic limit, not a confluence error.
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 | -2.25R | −$4,500 |
All six agents combined.
From the desk
Through Mar 23, 2026, the cumulative ledger reads +2.39R YTD across 46 trades from Jan 12 inception. The same $100,000 account at 2 percent risk per trade sits at $104,777.62 on the static line and $103,769.83 on the compounded line. The spread between static and compounded is small at this stage of the season, and that spread is what disciplined sizing produces: small, consistent edge plus 2 percent risk compounds into a different dollar figure than the same R does on a flat-stake basis.
The honest reading is that the system did almost everything right and still went into a 4.8 percent drawdown by Friday's close. A +1.5R baseline winner Monday, two EURUSD shorts to TP3 Wednesday, a US500 short to TP1 Thursday, a USDJPY long to TP3 Friday, and -2.42R net across twelve filled trades. To a subscriber watching equity descend from $100,000 to $95,167.32, the week looks like a regime mismatch or a sizing failure. It is neither.
The point is the architecture. Twelve trades is the trade-count a chop regime produces when the Macro Agent keeps not committing. Seven stop-outs and five winners (three of them TP3 winners on the case-study ladder) inside the same week are the same architecture producing different shapes. The Macro Agent did not flip its read mid-week, the Trend Agent did not loosen confluence after Tuesday's three-stop sequence, and the Risk Agent did not widen sizing after Wednesday's drawdown. The product is the dynamism, and the dynamism includes weeks where the math points to twelve setups and five pay.
The SkyAnalyst Team
What we're tuning
Nothing on the entries. The seven-loss cluster sits inside the variance envelope of a chop regime that produced twelve qualifying setups, and every loss was a regime-shift inside the trade lifecycle, not a confluence misread.
The twelve-trade count is itself the signal worth naming. A chop regime that lets macro reads cross threshold without committing produces this count by design, and the system traded all twelve at the same threshold used across the record. The per-trade model attribution gap surfaced by the head-to-head panel is the one operational item we are fixing forward: the executor will tag the family on every row so the Claude-versus-GPT view runs cleanly on the next window.
At a Glance
Week at a glance
How does a 41.7 percent win rate produce a -2.42R week across twelve trades?
At this hit rate the variance envelope includes weeks where losses outnumber winners. Five winners averaging slightly under 1R against seven losers at -1R each produced -2.42R net. The rolling 100-trade record is the right window for dispersion; a single week of twelve trades is one sample inside a much longer distribution.
Why did the system take twelve trades in a single week?
The Macro Agent flagged a chop regime that crossed lean-bull and lean-bear thresholds repeatedly without committing. The Trend Agent took every setup that cleared the confluence floor at the same threshold used across the record. Twelve filled trades is the count a chop tape produces by design, not by widening criteria.
What is the difference between the +1.5R TP1 baseline on Monday's US500 long and a full-TP case study figure?
Recap R-multiples use a TP1-baseline projection where every winner closes at the first take-profit. Standalone case studies document the full TP ladder. Both numbers describe the same trade on different exit assumptions; subscribers running their own scale-out or trail logic will see different totals on the same trade.
When the system takes opposite-direction trades thirty seconds apart, are those coordinated?
No. Each evaluation cycle is independent and reads the macro state, structure, and cross-asset confluence on its own. The Friday US30 short triggered on a pullback-failure read; the USDJPY long on a separate dollar-strength read. The two stacks scored independently on the same macro state and landed on opposite directions inside the same chop regime.
How should a subscriber read a twelve-trade losing week against the record?
As a single sample inside a longer distribution. The system is +2.39R YTD across 46 trades from Jan 12 inception, with the $100,000 simulated account sitting at $104,777.62 on a static basis. The right windows for evaluating performance are the monthly recap and the rolling 100-trade record, not any single week.
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.