The week only the euro paid, and every loss stayed small
Five of our seven trades stopped out. Only EURUSD found follow-through. The week cost 3.52R, and the reason it did not cost more is the whole point: not one los
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.
Some weeks the job is to catch the move. This one, the job was to lose well. Five of our seven trades stopped out between August 3 and 9, and the desk finished the week down 3.52R. We are not going to dress that up. What we will do is show you exactly how those losses were shaped, because a losing week where every stop held at minus 1R is a very different thing from a losing week where one trade runs away. Since the January 12 inception, the system has banked +31.43R. A simulated $100,000 account risking 2% per trade sits at $162,869 on a static basis, so a 3.52R week is a real dent but a shallow one against that base. The year is not built on weeks like this going well. It is built on weeks like this staying contained. The one instrument that worked was EURUSD. Both euro longs cleared their setups and booked gains, the pullback-to-breakout-retest on Tuesday and the soft-dollar continuation on Wednesday, while everything else, a US30 long, a NAS100 long, and two GBPUSD longs, ran into stops. That is the honest shape of the week: a single instrument carrying, the rest paying tuition, and a risk model that made sure the tuition was the fixed price we agreed to and not a dollar more.
Monday opened with a stop
The week started the way it would mostly continue. Monday's US30 long entered in the New York morning and was stopped inside two hours, a clean minus 1R. There was no drama in it, no gap, no violent reversal, just a setup that did not follow through and a stop that did its job. It set the tone: this was going to be a week where the market did not hand us continuation, and the discipline would be in accepting that quickly.
Midweek, the euro carried
Tuesday and Wednesday were the bright spots. The EURUSD long on Tuesday caught a pullback to a breakout retest and ran cleanly, our strongest read of the week. Wednesday's euro long added a second win on a soft-dollar continuation. For 48 hours it looked like the week might tilt green. But around those two euro wins, a NAS100 long on Wednesday stopped out, and the pattern of the week, EURUSD working while the indices and Cable did not, was already setting in.
The back half gave it back
Thursday and Friday erased the midweek gains and then some. A GBPUSD long stopped Thursday. Then Thursday's second entry, a EURUSD long, would go on to become the week's longest-running position, held open for five days before finally stopping this week. A second GBPUSD long also stopped. By Friday's close the two euro wins were buried under five losses, and the week settled at minus 3.52R. The back half is where a worse-run book turns a bad week into a disaster. Ours just turned it into a bad week.
Every trade the system took.
| Date | Time | Instrument | Dir | Model | Setup | Grade | R | $ Sim | Result | Details |
|---|---|---|---|---|---|---|---|---|---|---|
| Aug 3 | 14:45 UTC | US30 | Long | GPT-5.5 | US30 Pullback Reclaim Long | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
| Aug 4 | 14:43 UTC | EURUSD | Long | Claude Opus 4.7 | EURUSD LONG — Pullback to Breakout Retest | C+ | +0.80R(TP1) | +$1,592(TP1) | TP3 hit · ★ Trade of the week | Read case → |
| Aug 5 | 14:20 UTC | NAS100 | Long | Claude Opus 4.7 | NAS100 Long — Pullback to Fibonacci 61.8% / VWAP Reversion | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
| Aug 5 | 15:06 UTC | EURUSD | Long | Claude Opus 4.7 | EURUSD Pullback Buy | B | +0.68R(TP1) | +$1,364(TP1) | TP1 hit | Read case → |
| Aug 6 | 14:26 UTC | GBPUSD | Long | GPT-5.5 | Buy pullback into support | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
| Aug 7 | 14:29 UTC | EURUSD | Long | Claude Opus 4.7 | EURUSD Pullback Buy at Session Support | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
| Aug 7 | 14:29 UTC | GBPUSD | Long | GPT-5.5 | GBPUSD LONG (preferred pullback continuation) | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
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
One instrument working is still information
The cleanest pattern of the week was not a setup we caught repeatedly. It was a divergence: EURUSD found follow-through and nothing else did. Both euro longs cleared their targets while a US30 long, a NAS100 long, and two GBPUSD longs all failed at the same task, holding a breakout or a pullback long enough to travel.
That is worth sitting with rather than explaining away. When one instrument works and the rest do not, the temptation is to force the losing instruments into the winning template, to keep buying Cable because the euro is paying. We did not. Each instrument was traded on its own read, and when GBPUSD failed twice, the second failure was not a reason to abandon the pair, nor a reason to press it. It was one more data point in a week where sterling and the indices simply were not offering what our setups require. The pattern was thin follow-through everywhere but the euro, and the correct response to thin follow-through is smaller, fewer, and patient, which is exactly what the minus 1R stops enforced.
Decision highlights
Every loss closed at exactly minus 1R. Across five stopped-out trades there was not a single instance of slippage past the stop, a widened stop, or an averaged-down position. That is the decision that defined the week: the risk model was allowed to do its job on every trade, and the discipline of accepting five clean stops is what kept a bad week from becoming a costly one.
The Thursday EURUSD long became the week's longest-running position, held open for five days before it finally stopped this week. Rather than cut it early or move the stop to breakeven prematurely, the system let the trade breathe on its original parameters until the market resolved it. It resolved against us, but the decision to hold on plan, not on emotion, is the same discipline that lets winners run when the resolution goes the other way.
We did not chase the euro's success into the losing instruments. With EURUSD paying twice and GBPUSD failing twice, the easy error is to treat one pair's strength as license to force the other. Each instrument stayed on its own read, and GBPUSD's second stop did not trigger a revenge entry. Refusing to let a winning instrument justify a losing one is a decision you make before the week, and honor during it.
Six instruments, six stories.
EURUSD: The week's only green instrument. Three trades, two wins, +0.48R net. Both euro longs cleared their setups while the rest of the desk stalled, making the pair the sole source of follow-through.
All EURUSD this week →GBPUSD: The heaviest drag at minus 2R across two long attempts, both stopped. Neither breakout held. Sterling was not offering the clean continuation our setups need, and the second stop was a signal to step back, not to press.
All GBPUSD this week →US30: One long, stopped Monday inside two hours for minus 1R. A textbook clean stop: no follow-through, no hesitation, damage capped at the agreed risk.
All US30 this week →NAS100: One long, stopped Wednesday for minus 1R. The index would not hold its breakout, and the position was closed at its stop without argument.
All NAS100 this week →USDJPY: No trades this week. The yen sat outside our setup criteria across the window, and we do not manufacture entries to fill a grid cell.
All USDJPY this week →US500: No trades this week. The S&P book found no qualifying setup in the window and stayed flat, which on a week like this was its own small contribution.
All US500 this week →Win of the week: EURUSD Long · +0.8R
Loss worth learning from
Five stops, one shape
The five losses this week share a single anatomy, and it is a boring one, which is the point. Each trade entered on a qualifying setup, failed to find follow-through, and closed at its predefined stop for minus 1R. None of them were bad entries in the sense of ignoring the process; they were correct-process trades that the market simply did not reward. The US30 and NAS100 longs failed to hold breakouts. The two GBPUSD longs failed the same way on a pair that was not trending. The EURUSD Thursday long held longest and still resolved against us.
The teardown lesson is not about any single trade. It is about the absence of a sixth kind of loss, the one that does the real damage: the trade that runs past its stop, the position that gets doubled to "average in," the stop that gets walked lower to avoid the pain of being wrong. We had none of those. A week can be wrong on direction five times and still be survivable if it is never wrong on risk. This one was wrong on direction and right on risk, and that combination costs 3.52R, not 10.
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 | -3.52R | −$7,040 |
All six agents combined.
From the desk
From the desk
A losing week is the honest test of a track record, because anyone can publish the green ones. So here is ours, in full: minus 3.52R, five stops, two euro wins, no disasters. If you are evaluating this system, this is the week you should read most closely, because it shows you what happens when the reads are mostly wrong. What happens is that the losses stay the size we said they would.
The year-to-date numbers hold the context. Since inception the desk has banked +31.43R, and a $100,000 account at 2% risk sits at $162,869 on a static basis. Compound those same returns, letting each 2% risk grow with the account, and the figure is $180,579 instead. That gap between static and compounded, better than $17,000, is not from bigger wins. It is from disciplined, consistent sizing applied over many trades, the same discipline that made this week's five losses cost exactly five R and not a dollar more. The weeks that stay small are what make the compounding possible.
What we're tuning
There is nothing structural to tune out of this week, and saying so honestly matters more than inventing a fix. Five clean stops on qualifying setups is variance, not a broken process. If the losses had shown a common flaw, entering too early, stops too tight, chasing extended moves, that would be a signal to adjust. They did not. They showed a market that was not offering follow-through outside the euro, and a system that took its small, fixed loss each time and moved on.
What we watch going forward is instrument breadth. A week where only one of six instruments works is a narrow week, and if that narrowness persists we would rather trade less than force the quieter instruments into setups they are not offering. The tuning here is not to a parameter. It is to patience: fewer entries when only one book is paying.
At a Glance
Week at a glance
Why publish a recap of a losing week?
Because a track record that only shows winning weeks is not a track record, it is marketing. Losing weeks are where a trading system's risk discipline is actually visible. This week the desk lost 3.52R across five stopped-out trades, and every one of those stops held at its predefined level. Showing that is more useful to anyone evaluating the system than another green week would be.
How did every loss end at exactly minus 1R?
Each trade is entered with a predefined stop that represents 1R, one unit of risk. When the market hits that stop, the position closes, full stop. We do not widen stops to avoid being taken out, we do not average down into a losing position, and we do not hold past the stop hoping for a reversal. That discipline is what makes a five-loss week cost 3.52R rather than a multiple of it.
Why did only EURUSD make money this week?
EURUSD was the one instrument offering clean follow-through in the window. Both euro longs cleared their setups, a pullback-to-breakout-retest and a soft-dollar continuation, while the indices and GBPUSD failed to hold their moves. We trade each instrument on its own read rather than forcing a single directional view across the board, so when only the euro was working, only the euro was traded to a profit.
Does a losing week change the year-to-date numbers?
This week's result is part of the ongoing record, but the year-to-date figure we report, +31.43R, reflects the system through the last closed month. The early-August drawdown will show in the numbers as the month completes and is recorded. We freeze the year-to-date to closed months so the headline figure never bounces around on an open, in-progress week. The transparency is in publishing the down weeks in full, which is what this recap does.
What are you changing after a week like this?
Nothing structural, and that is a deliberate answer. Five clean stops on qualifying setups is variance, not a flaw. If the losses had shared a common error we would adjust, but they did not. The one thing we are watching is instrument breadth: when only one of six instruments is working, we would rather trade less than force the quiet instruments into setups they are not offering.
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
Eight losses, five on one pair, and why that is normal trading
Most of this week's losses came from a single instrument that would not cooperate. GBPUSD alone accounted for five of eight stops. Concentrated losing streaks in one instrument are a normal feature of professional trading, not a fault.
US30 went three for three, and the week still finished red
Six winning case studies and a perfect week from US30 were not enough. GBPUSD gave back more than the rest of the desk earned, and the week closed down 3.04R, the third losing week in a row.
NAS100 Short: Selling the Failed Bounce When Yields Turned Hostile
A ~223 point gap down, the 10Y at 4.698%, and a dead-cat bounce into Fibonacci resistance. One evaluation at 75% confidence, one entry, and a clean +0.69R (TP1) into the session low.