Jun 1-7, 2026: a Monday wipeout, a Friday recovery
Twelve trades. Four losses inside thirty-eight minutes on Monday. One bounce-rejection short on NAS100 that paid the week. A 58.3% win rate that owes everything
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 losses. Seven wins. Net +3.11R for the week, and the bulk of it landed on Friday afternoon when NAS100 broke down through its rejection level and ran. Through Jun 8, 2026, the system has banked +23.11R YTD; a $100,000 simulated account at 2% risk per trade sits at $146,231.82 on the static path. That is the headline. The detail is less comfortable. The first day of the window cost us 3.5R inside thirty-eight minutes. Four losses, four instruments, one shared cause: a dollar bid that arrived right when our short-DXY and long-yen setups were ready to size. We will get into the calendar arc, the model split, and the loss that taught us the most, but the frame for the whole week is here. The week did not improve because the setups improved. The week improved because we kept the risk small, kept reading the tape, and Friday's setup arrived clean.
Monday: the dollar bid that broke four setups
Monday opened with four trades inside thirty-eight minutes between 15:06 and 15:44 UTC. GBPUSD short, USDJPY long, US30 short, EURUSD short. Each had its own structural premise. The Trend Agent had a fade on cable into a Fibonacci confluence; the trade plan on USDJPY was a pullback long into VWAP; US30 was a short into underside resistance; EURUSD was a rebound short into the same. The Macro Agent had not flagged a problem ahead of the prints. Then the Dollar Index bid, and every one of those setups failed in the same direction inside the same window. We took the full stops on all four. The window's drawdown peaked at -3.5R before lunch.
Wednesday: the second chance after ISM
Tuesday gave us one trade: a US30 long continuation on the breakout-hold of 51262 that paid +0.76R at TP1 and rolled there for the day. Wednesday opened with a GBPUSD short loss at 14:33, which put us deeper into drawdown. Then the post-ISM tape resolved into cleaner structure. The Trade Agent took three setups inside thirty minutes: a US30 short, an EURUSD short, and a USDJPY long, each on a confirmed second leg with volume. All three printed TP1, two of them eventually ran to TP3. Wednesday closed the week's drawdown from -3.74R back to -2.12R. The recovery was not heroic; it was patient.
Friday: the print that paid the week
Thursday's lone trade, a US30 long pullback reclaim at 14:56 UTC, GPT-routed, closed +0.95R at TP1 with a TP3 trail, which put the week within a single trade of break-even. Friday delivered two short setups inside twenty-two minutes. US500 broke down to a pullback into broken support; we shorted it for +1.04R. Then NAS100 rejected its bounce at 14:36 UTC. The bounce-rejection short cleared TP1 inside the hour and TP3 by 16:31 UTC for +3.26R, the week's largest print. Net for the week closed at +3.11R. That single Claude-routed NAS100 trade carried the window from a Monday hole into positive territory.
Every trade the system took.
| Date | Time | Instrument | Dir | Model | Setup | Grade | R | $ Sim | Result | Details |
|---|---|---|---|---|---|---|---|---|---|---|
| Jun 1 | 15:06 UTC | GBPUSD | Short | Claude Opus 4.7 | GBPUSD Post-ISM Pullback Short into VWAP/Fibonacci Resistance | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
| Jun 1 | 15:25 UTC | USDJPY | Long | Claude Opus 4.7 | USDJPY Pullback Long to VWAP/Support Continuation | C+ | -0.50R(SL) | -$1,000(SL) | Stop hit | - |
| Jun 1 | 15:26 UTC | US30 | Short | GPT-5.5 | Pullback Short into Underside Resistance | B | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
| Jun 1 | 15:44 UTC | EURUSD | Short | GPT-5.5 | Short the rebound into VWAP / Fib resistance | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
| Jun 2 | 15:51 UTC | US30 | Long | GPT-5.5 | Long continuation on breakout-hold of 51262 | B | +0.76R(TP1) | +$1,524(TP1) | TP1 hit | - |
| Jun 3 | 14:33 UTC | GBPUSD | Short | Claude Opus 4.7 | Post-Data Second-Chance Short | B | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
| Jun 3 | 14:56 UTC | US30 | Short | Claude Opus 4.7 | US30 Post-ISM Second-Chance / Opening Range Break Continuation Short | C+ | +0.77R(TP1) | +$1,535(TP1) | TP3 hit | Read case → |
| Jun 3 | 14:56 UTC | EURUSD | Short | Claude Opus 4.7 | EURUSD Short — Continuation after ISM Beat | B | +0.49R(TP1) | +$983(TP1) | TP1 hit | - |
| Jun 3 | 15:03 UTC | USDJPY | Long | GPT-5.5 | Pullback Long Into Intraday Support | C+ | +0.36R(TP1) | +$712(TP1) | TP3 hit | Read case → |
| Jun 4 | 14:56 UTC | US30 | Long | GPT-5.5 | US30 Pullback Reclaim into Trend Support | B | +0.95R(TP1) | +$1,896(TP1) | TP3 hit | Read case → |
| Jun 5 | 14:14 UTC | US500 | Short | Claude Opus 4.7 | Short on pullback to broken support | C+ | +1.04R(TP1) | +$2,074(TP1) | TP3 hit | Read case → |
| Jun 5 | 14:36 UTC | NAS100 | Short | Claude Opus 4.7 | NAS100 Short — Bounce Rejection | B | +3.26R(TP1) | +$6,513(TP1) | TP3 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
There was no clean recurring pattern this week. The setups were a mix: pullback longs into VWAP confluence on Tuesday and Thursday, post-data second-chance shorts on Wednesday, a bounce-rejection short on Friday. What did repeat was a meta-pattern, and it is worth naming honestly. Every winning trade entered after a confirmed second leg with volume, on instruments where the Macro Agent had not flagged a regime risk. Every Monday loss entered before a macro print or coincident with one. We were not wrong to take the Monday setups in isolation, but the regime under them shifted faster than the pattern math could update.
The honest read is that pattern-of-the-week framing rewards weeks with three or four of the same setup. This week did not have that. It had a mix, and the only thread we can pull is the discipline of waiting for confirmation. When we got it (Wednesday afternoon, Friday afternoon), the trades worked. When we tried to front-run a setup into a moving regime (Monday's four losses), they failed in the same direction at the same time.
Decision highlights
The Monday risk discipline. Four stops inside thirty-eight minutes is a textbook bad open. The decision we want to underline is that we did not double down after the third loss. The Risk Agent enforced the per-trade R budget on the EURUSD short at 15:44 UTC the same way it had on the first three; nothing about the drawdown was allowed to inflate position size on the next setup. By the time the dollar bid faded, the budget was intact for the rest of the week.
The Wednesday second-chance read. After Monday's wipeout and the GBPUSD short loss at 14:33 UTC on Wednesday, taking three more shorts inside thirty minutes (US30 at 14:56, EURUSD at 14:56, USDJPY long at 15:03) was a real judgment call. The Macro Agent's regime read had cleared after the ISM print; the Trade Agent's setups were confirmed on volume; the Risk Agent was already inside its tighter mid-week budget. We took the trades anyway. All three paid. That is the kind of read that does not show up in the equity curve but does show up in the year.
The Friday hold-through-TP1 on NAS100. The 14:36 UTC entry cleared TP1 inside the hour. The system held the trail through the next ninety minutes to TP3 at 16:31 UTC, lifting the trade from a +0.95R partial close to a +3.26R full close. That decision converted what would have been a +0.80R week into a +3.11R week. The Risk Agent's trail logic is the kind of background mechanic that does not get a headline most weeks; this week it printed the headline.
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+3.3R · 1 trade
- US500+1.0R · 1 trade
- US30+0.8R · 1 trade
- US30+0.7R · 3 trades
- USDJPY+0.4R · 1 trade
- EURUSD-1.0R · 1 trade
Same signals, same risk framework, different foundation model.
Six instruments, six stories.
EURUSD: two trades, net -0.51R. The first was a Monday short into the rebound that stopped out in the dollar bid; the second was a Wednesday post-ISM short that paid +0.49R at TP1. The pair traded clean both sides of the macro print; only the timing of the first entry was wrong.
All EURUSD this week →GBPUSD: two trades, both losses, net -2.0R. The Monday short into VWAP resistance and the Wednesday post-data second-chance short. Cable shorts into resistance in a dollar-bid week were the recurring tax of the window. We will look hard at what flagged the second entry on Wednesday given how the Monday tape closed.
All GBPUSD this week →US30: four trades, net +1.48R, 75% win rate. The Monday short was the loss; the Tuesday breakout-hold long, the Wednesday opening-range continuation short, and the Thursday pullback reclaim long all printed at TP1, with the Wednesday short trailing to TP3. US30 was the workhorse of the week.
All US30 this week →NAS100: one trade, net +3.26R. The Friday bounce-rejection short was the week's largest print and a TP3 fill. One trade is a small sample by definition; the trade was textbook and the outcome was the outlier, not the entry logic.
All NAS100 this week →USDJPY: two trades, net -0.14R. The Monday pullback long stopped out (-0.5R, half size on a C+ grade), and the Wednesday pullback long paid +0.36R at TP1 and trailed to TP3. The pair's two reads sat on opposite sides of the dollar move, which is the honest reading of why the net is near zero.
All USDJPY this week →US500: one trade, net +1.04R. The Friday short on a pullback to broken support paid +1.04R at TP1. The instrument only fit our setup criteria once this window.
All US500 this week →Win of the week: NAS100 Short · +3.26R
Loss worth learning from
The Monday wipeout is the loss worth learning, and we wrote it up in detail in [Five Stops, Thirty-Eight Minutes: the Jun 1 Monday Open](/en/blog/weekly-losses-2026-06-01), the same window's losses-only piece, published earlier today. The short version: GBPUSD short at 15:06, USDJPY long at 15:25, US30 short at 15:26, EURUSD short at 15:44. Four trades, four stops, -3.5R inside thirty-eight minutes. None of the entries was a bad setup in isolation. The shared failure was that the Dollar Index bid during the same window the trades were sized, and four positions that each had different individual logic ended up correlated through the same macro move.
The teardown question we keep coming back to is whether the Macro Agent should have gated the cluster. The honest answer is that no single regime signal had broken at 15:06 UTC; the bid arrived inside the trades, not before them. What we are tuning, then, is not the gate itself but the correlation read. Four short-dollar or long-yen setups inside the same thirty-eight minutes is a portfolio-level exposure that the system was treating as four independent reads. It was not. That is the lesson we want to carry into next week.
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 | +3.11R | +$6,220 |
All six agents combined.
From the desk
Twelve trades, +3.11R net, 58.3% win rate. The headline reads cleaner than the experience. A $100,000 simulated account at 2% risk per trade is at $146,231.82 on the static path through Jun 8, 2026, and at $154,125.13 on the compounded path. The compounded number is roughly $7,893 higher than the static number for the same R sequence. That difference is not the result of a bigger edge or a better setup grade; it is the result of letting the account size each new trade off the larger balance the previous wins produced. Disciplined sizing compounds. It is not a feature we sell; it is a property of fixed-percentage risk applied to a positive-expectancy strategy. The numbers are evidence of the discipline, not the discipline itself.
The week's other lesson is one we already published in the [Monday losses teardown](/en/blog/weekly-losses-2026-06-01): correlated entries inside a thirty-eight-minute window are a single portfolio bet, not four independent ones. We are tuning the cluster read for next week. The case studies on the individual trades that worked, including the [Wednesday post-ISM US30 short](/en/blog/us30-short-post-ism-second-chance-opening-range-break-continuation-06-03-2026), the [Wednesday USDJPY pullback long](/en/blog/usdjpy-long-pullback-into-intraday-support-06-03-2026), the [Thursday US30 pullback reclaim](/en/blog/us30-long-pullback-reclaim-trend-support-06-04-2026), the [Friday US500 short](/en/blog/us500-short-pullback-to-broken-support-06-05-2026), and the [Friday NAS100 bounce-rejection short](/en/blog/nas100-short-bounce-rejection-06-05-2026), are on the site if you want to read the decision logs trade by trade. The week is a story, and the trades are the sentences. Next Monday opens a new one.
What we're tuning
We are tightening the correlation read on cluster entries. Four trades inside thirty-eight minutes, all on the same side of the dollar, should trip a portfolio-level alert before the fourth gets sized. That is a Risk Agent change, not a Macro Agent change; the regime read at 15:06 UTC was not wrong, the position concentration was. We are also reviewing the Wednesday GBPUSD short for what flagged the entry after the Monday cable failure; the trade may have been correct on its own merits, but the second-chance bias on a pair that just took a stop deserves a harder look at the setup-grade math.
The longer tuning question is whether the pattern-of-the-week framing rewards us when there is no recurring pattern. This week we had a mix, and the honest read is that the only repeated edge was waiting for confirmation. We will not over-correct on one window; the YTD curve is the signal, and the year is still tracking +23.11R from inception.
At a Glance
Week at a glance
Why is the net only +3.11R when the NAS100 short paid +3.26R alone?
Because the rest of the week's wins and losses roughly canceled out. Seven wins summed to about +6.6R; five losses summed to about -3.5R. Subtract and you get +3.11R net, where the NAS100 print does most of the surplus. The honest read is that one outlier carried the week. The other wins kept the drawdown from compounding while we waited.
How does the system handle a Monday like Jun 1, where four trades stop out inside thirty-eight minutes?
It enforces the per-trade R budget without exception. The fourth stop loss is sized the same as the first. The Risk Agent does not let drawdown inflate the next entry, and the Macro Agent's regime gate stays in place for the next setup regardless of how the prior trades closed. The discipline is in the policy, not in the read.
Is the +23.11R YTD figure on TP1 exits or full-TP exits?
TP1-baseline. Every winning trade in the YTD snapshot is credited at the TP1 R distance; every loser is credited at -1R. The individual case studies publish full-TP figures for the same trades, which produce different numbers. The YTD snapshot uses TP1 baseline so the periods are comparable across the year.
Why does the compounded balance only edge slightly higher than the static balance?
Because the YTD R total is +23.11R and the risk-per-trade is 2% of equity. At those scales, compounding produces a roughly $7,893 surplus on a $100,000 starting account through five months. The longer the system runs at positive expectancy, the wider the gap. The static path is what a fixed-dollar risk strategy would produce; the compounded path is what fixed-percentage risk produces. We publish both because the difference is the math behind the sizing discipline, and we think it is worth seeing.
Did the Macro Agent miss the Monday dollar bid?
It did not flag a regime change ahead of the 15:06 UTC entry; the DXY bid arrived inside the trade window, not before it. The lesson we took is that the correlation read at the portfolio level (four short-dollar or long-yen setups inside thirty-eight minutes) is the gap, not the regime gate itself. We are tuning the cluster read for next 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
Weekly Losses: the Monday afternoon that cost us four R
Five losses, four of them in thirty-eight minutes on the same Monday session. The system gave back 4.5R against +23.11R YTD, then walked the curve back to a fresh peak by Friday close.
The NAS100 short the macro stack pre-cleared in one read
Yields at five-day highs, DXY firm, VIX rising, all six confluence factors clean on the first scan. A single evaluation took NAS100 short at 29,876.3, and the move ran 386 points.
US500 short: six of six confluences, four reads inside four minutes
NFP had cratered breadth and lit VIX. Six of six factors said short before the first evaluation ran. The interesting question wasn't whether. It was where.