Eight Winners, Two Losses, and a Week That Graded C+
Ten trades, eight winners, +5.43R and an 80 percent strike rate, the best week of the month. The setup grades tell a quieter story: eight of the ten scored C+ b
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.
Ten trades this week, eight winners, +5.43R. That is an 80 percent strike rate against a year-to-date figure of 56.82, with a maximum drawdown of 1.91 percent and a longest losing streak of one. By every number on the scoreboard it is the best week the desk has had this month. There is one number that argues for restraint, and it is on the same scoreboard. The week's average setup grade is C+. Grades are assigned when the entry is scored, before anyone knows how it resolves, and eight of these ten trades graded C+ going in. Two graded B or better. Through Sep 14, 2026, the system has banked +28.75R since the Jan 12 inception. A $100,000 simulated account risking 2% per trade sits at $157,519 on static sizing. A week like this one is how that number grows. It is not evidence that it will keep growing at this rate.
Act I: One idea, repeated
There was a single trade on the board this week and the desk took it ten times.
Yields added ground in every session, running 4.772 to 4.784 to 4.792 to 4.841 and finishing above 4.92. VIX climbed from the mid fifteens to eighteen. Brent went from the high nineties through 101 and past 105. Breadth deteriorated from -500 on Tuesday to -1,244 by Thursday. Every one of those moved in the same direction all week.
Ten trades, ten shorts, across five instruments. A week shaped like that does not reward creativity, and the honest description of the desk's performance is that it identified one condition early and kept executing against it rather than that it found ten separate insights.
Act II: Thursday afternoon
Four of the ten positions opened inside forty-eight minutes on Thursday: Cable at 14:20, the Nasdaq at 14:26, the S&P at 14:27, the Dow at 15:08.
All four were scored independently against their own structure and given their own stops. All four won, returning +0.81R, +1.28R, +0.67R and +0.90R on the recap baseline. It is also, honestly stated, four positions leaning on one rate shock, which is 8 percent of the account exposed to a single macro reversal.
The spread of outcomes is the argument for having spread the risk. The best of the four returned nearly twice the worst, and nothing in the four setups predicted which would be which.
Act III: The two that did not work
A USDCAD short on Tuesday and an S&P short on Wednesday both stopped out, each for exactly 1R.
Neither loss clustered with the other and neither repeated a mistake. They were also both graded C+, which matters because it means the week's two failures came from the same quality band as most of its successes. The grade did not separate the winners from the losers at all this week, and that is the most useful observation the ten trades produced.
Every trade the system took.
| Date | Time | Instrument | Dir | Model | Setup | Grade | R | $ Sim | Result | Details |
|---|---|---|---|---|---|---|---|---|---|---|
| Sep 8 | 15:00 UTC | US30 | Short | US30 SHORT | B+ | +0.89R(TP1) | +$1,783(TP1) | TP3 hit | Read case → | |
| Sep 8 | 15:12 UTC | US500 | Short | US500 SHORT, VWAP Rejection / Bearish Continuation | C+ | +1.36R(TP1) | +$2,714(TP1) | TP3 hit · ★ Trade of the week | Read case → | |
| Sep 8 | 15:50 UTC | USDCAD | Short | USDCAD SHORT (Bearish Continuation Pullback) | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | - | |
| Sep 9 | 15:00 UTC | US30 | Short | US30 SHORT | B | +0.79R(TP1) | +$1,583(TP1) | TP3 hit | Read case → | |
| Sep 9 | 15:40 UTC | NAS100 | Short | NAS100 SHORT, VWAP Rejection Continuation | C+ | +0.74R(TP1) | +$1,485(TP1) | TP3 hit | Read case → | |
| Sep 9 | 15:50 UTC | US500 | Short | US500 SHORT, Bear Flag Breakdown Continuation (Pullback Entry) | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | - | |
| Sep 10 | 14:20 UTC | GBPUSD | Short | GBPUSD retracement short into resistance | C+ | +0.81R(TP1) | +$1,626(TP1) | TP3 hit | Read case → | |
| Sep 10 | 14:26 UTC | NAS100 | Short | NAS100 SHORT, Fibonacci/VWAP Rejection Fade | C+ | +1.28R(TP1) | +$2,557(TP1) | TP2 hit | Read case → | |
| Sep 10 | 14:27 UTC | US500 | Short | US500 SHORT, Pullback to Opening Range High / Fib Resistance | C+ | +0.67R(TP1) | +$1,333(TP1) | TP1 hit | Read case → | |
| Sep 10 | 15:08 UTC | US30 | Short | US30 SHORT, failed rebound into 5m fib resistance | C+ | +0.90R(TP1) | +$1,791(TP1) | TP2 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 pattern of the week is that setup grade and outcome came apart completely.
Of the ten trades, two graded B or better and eight graded C+. Both of the losses were C+. Six of the eight winners were also C+. If you had ranked the week's entries by grade before any of them resolved, the ranking would have told you close to nothing about which would pay.
That is not an indictment of the grading. Over a large sample, higher-graded setups should and do resolve better, which is why the losses report normally selects a B-or-better loss to tear down and why it could not do so this week. Over ten trades, a grade distribution this narrow has no discriminating power, and reading one into it would be exactly the mistake this recap exists to avoid.
The practical conclusion is unglamorous. The desk executed a correct directional read repeatedly in a tape that kept paying for it, and the quality of the individual entries was ordinary. Both halves of that sentence are true and only one of them is flattering.
Decision highlights
We opened four positions inside forty-eight minutes on Thursday rather than concentrating the same risk into the best-looking one. That is 8 percent of the account on a single macro thesis, stated plainly, and it is a deliberate exchange of concentration risk for selection risk. The four returned +0.81R, +1.28R, +0.67R and +0.90R, a spread wide enough that picking among them would have been a guess.
We published a +0.67R winner with the arithmetic showing that its geometry was marginal. The S&P short on Thursday had a single target 14 points out against a 21 point stop, a shape that needs a 60 percent strike rate to break even against a year-to-date 56.82. Letting a green number stand without that context is how a record becomes a highlight reel.
We changed nothing in response to either loss. Both were C+ setups that stopped out at exactly 1R, neither repeated the other, and USDCAD is two September trades into a sample. The threshold for acting is a pattern, and two unrelated stop-outs in a week with eight winners is not one.
Six instruments, six stories.
EURUSD did not trade this week. No setup cleared the confluence threshold and the book sat out rather than manufacture an entry into a tape that was already crowded with shorts.
All EURUSD this week →GBPUSD took one short on Thursday for +0.81R on the recap baseline. It needed nine refusals across thirteen minutes before the tenth evaluation cleared, which is the most this desk has logged on a single setup.
All GBPUSD this week →US30 was the book of the week: three trades, three winners, +2.58R. One of them reached its third target ten minutes after entry, the fastest resolution the desk recorded this month.
All US30 this week →NAS100 took two shorts for +2.02R combined. The pair is the clearest A/B test the week produced, one day apart with the second carrying half the stop width of the first and returning a third more.
All NAS100 this week →USDJPY did not trade this week. The book has taken six trades all year and remains deliberately on a short leash.
All USDJPY this week →US500 took three trades for +1.02R at a 66.7 percent hit rate, the widest spread of any book: one of the week's two losses, the smallest winner at +0.67R, and a clean +1.36R that was the best single result on the recap baseline.
All US500 this week →USDCAD took one short on Tuesday and lost it for -1R. It is the book's second September loss, which is worth watching rather than acting on at two trades.
All USDCAD this week →Win of the week: US500 Short · +1.36R
Loss worth learning from
Two losses this week, both at exactly 1R, and both graded C+ before they were taken.
That second fact is the one worth sitting with. The weekly losses report normally selects the highest-graded loss for a full teardown, on the reasoning that a well-formed trade that failed teaches more than a weak one that did. This week the selection gate refused to run: it requires a loss graded B or better and there was not one.
The reason is not that the losses were unusually bad. It is that almost nothing this week graded above C+, winners included. The gate was written to find the instructive failure in a normal distribution of setup quality, and a week with eight C+ entries does not have that distribution.
Both losses get a full teardown in this week's separate losses report, which also covers what a 1.91 percent drawdown and a longest losing streak of one mean against a 56.82 percent strike rate over 220 trades.
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 | +5.43R | +$10,860 |
All six agents combined.
From the desk
From the desk
Eight winners from ten trades is the best strike rate the desk has posted this month, and the number we would rather you take away is the C+ on the scoreboard.
Through Sep 14, 2026, the system stands at +28.75R across 220 trades since the January 12 inception, 125 winners against 95 losers, a 56.82 percent win rate. On a $100,000 account at 2% risk, static sizing puts that at $157,519. Compounding the identical sequence, where each position is sized off the balance the previous one produced, puts it at $169,638. The $12,119 gap is not a better trade. It is the same R, sized with discipline, allowed to work on itself.
An 80 percent week sits 23 points above the year-to-date rate. Weeks like that happen, they are not evidence of a new baseline, and the fastest way to lose the gain is to start sizing as though they are. The desk sized every one of these ten trades from the same 2 percent rule it used in August, the month it lost money.
One change shipped with this article. The setup grade tile on the weekly scoreboard used to print the same value every week no matter what had been traded. It now averages the real per-trade grades, which is why this week says C+. We would rather publish an uncomfortable number that is computed than a comfortable one that is not.
What we're tuning
Nothing changes on the back of this week, and the reason is the inverse of the usual one. A recap that recommends an adjustment after a bad week is usually reacting to noise. A recap that recommends nothing after a good week is resisting the same temptation from the other direction, which is harder to notice and at least as expensive.
The one thing now visible that was not before is the setup grade on the scoreboard. That tile previously carried a fixed value on every weekly recap regardless of what the desk had traded, which made it decorative rather than informative. It now shows the mean of the window's actual per-trade grades, which is why this week reads C+ rather than something more comfortable. Where a number is published as a metric it should be computed like one.
At a Glance
Week at a glance
An 80 percent win rate sounds excellent. Is it sustainable?
No, and nothing in this week suggests it is. The year-to-date strike rate is 56.82 percent across 220 trades, and an 80 percent week sits 23 points above that. Streaks in both directions are expected at this sample size. The week's average setup grade of C+ is the more informative number, because grades are assigned before outcomes are known.
What does the setup grade actually measure?
How much confluence an entry had when it was scored, before the outcome was known. Eight of this week's ten trades graded C+, including six of the eight winners and both of the losses. Over a large sample better grades resolve better; over ten trades a distribution this narrow separates nothing, which is itself the week's most useful finding.
Every trade was a short. Is that a house view?
No, it is what the tape paid for. Yields rose in every session and finished above 4.92, VIX ran to eighteen, Brent went past 105, and breadth fell to -1,244. No long setup cleared the confluence threshold all week, and the desk did not manufacture one to look balanced.
How is R calculated in this recap?
On a TP1 baseline. Winners are credited the R distance from entry to the first take-profit and losers are counted at exactly -1R. It is deliberately more conservative than the full-potential R the individual case studies report, so totals stay comparable across periods. The Nasdaq short on Thursday appears here at +1.28R and in its own case study at +2.32R for that reason.
Are these real trades?
Yes. Every trade in the table above was placed by the live system with its entry, stop and targets recorded before the outcome was known. The dollar figures are a simulation of what a $100,000 account risking 2% per trade would have produced from that exact sequence.
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
The Week Our Own Selection Gate Refused to Pick a Loss
This report normally tears down the best-graded loss of the week. This week it could not: neither loss graded above C+, because almost nothing did. Two stop-outs, minus 2R, and a 1.91 percent drawdown.
Four Shorts in Forty-Eight Minutes, on One Reading
Between 14:20 and 15:08 the desk sold Cable, the Nasdaq, the S&P and the Dow. One macro read, four instruments, four stops, four winners. This is the Dow leg, and the question is what that correlation costs.
The Winner That Paid Less Than a Loss Would Have Cost
This trade was right and returned +0.67R. The target sat 14 points away and the stop sat 21, so being correct paid two thirds of what being wrong would have cost. It is the least flattering trade of the week.