A +1.53R week ended on two straight stops at exactly 1R each: an Orion gold short sold into an exhausted liquidation day, and an Atlas Dow short that ran 128 po
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.
The week finished positive and still ended on two stops in a row. Each lost exactly the 1R it was sized for, and together they gave back 2.0R, a simulated 4,000 dollars on a 100,000 dollar account at 2% risk. The weekly recap covers the +1.53R the whole window produced. This report covers the two trades that did not work: Thursday's Orion gold short and Friday's Atlas Dow short. The split by line is even this week. Orion finished the window at +0.43R across two trades, one win and one loss. Atlas finished at +1.11R across three trades, two wins and one loss. Each line paid early in the week and gave one trade back at the end of it. Against the year to date ledger the week is small. The system has banked +28.63R YTD from Jan 12 inception, through the October close, which puts the same simulated account at $157,278.64. The desk gave back 2R in losing trades this week against +28.63R YTD. To read approvals like these as they happen, see SkyAnalyst run your markets.
The Orion Nasdaq long filled at 27,728.4 at 15:35 UTC on Monday and paid +1.43R, the best trade of the week. The Atlas Dow long followed at 16:09 UTC from 52,206.6 and reached its first target for +0.93R. On Tuesday the Atlas S&P 500 short filled at 7,523.2 at 15:35 UTC and paid +1.18R. That put the simulated account at 107,073.34 dollars, the high of the window.
Wednesday brought no trades. On Thursday gold was in a liquidation day: the Orion note, written at 11:31 ET, counted a 148 point range, about 175% of the prior day's, with the dollar and yields rising. The Orion gold short sold the bounce into a 3,880 to 3,886 shelf and filled at 3,880.22 at 16:51 UTC. It reached 3,866.23, about 14 dollars our way against roughly 26 to its 3,854.5 first target, then turned and stopped overnight at 01:21 UTC Friday.
The Atlas Dow short filled at 52,233.7 at 15:41 UTC, after breadth had rolled over from an intraday high back into negative territory. It ran to 52,105.9, about 128 points our way and roughly 16 points from its 52,090 first target. It never printed that target, and it stopped at 52,368 at 20:51 UTC, a little over five hours after entry. That stop set the window's deepest drawdown.
| Date | Time | Instrument | Dir | Model | Setup | Grade | R | $ Sim | Result | Details |
|---|---|---|---|---|---|---|---|---|---|---|
| Nov 12 | 16:51 UTC | XAUUSD | Short | XAUUSD SHORT Pullback | D | -1.0R(SL) | -$2,000(SL) | Stop hit | - | |
| Nov 13 | 15:41 UTC | US30 | Short | US30 NY AM failed-reclaim short | 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.
Both losses this week were plans that carried their doubts in writing. The difference is where the doubt was handled.
The Orion gold note named its own case against the trade before any order existed: the day's range was already 175% spent, the 60-minute RSI sat near 19, it was the lunch hour with London gone, and the first target sat at a level that had already held once. It scored the trade 5.6 out of 10 and wrote it anyway, as the prompt asks it to. Orion books enter when price reaches the planned zone and the news and risk gates have cleared, so that 5.6 was the whole judgment, made in advance.
The Atlas Dow short handled its doubt at the trigger. The plan asked for a 5-minute rejection followed by a 5-minute close back below 52,232. The first evaluation at 15:37 UTC declined because the last candle had closed at 52,253. Three minutes later the approval at confidence 63 entered, noting that the close below 52,232 was only being approximated rather than confirmed.
Two trades cannot make a rule. Last week's report found the same thing on three Atlas losses, though: the approval names the reason to wait, then enters anyway.
Friday's Atlas Dow short is the decision we would most like back. The plan asked for a 5-minute close back below 52,232, the gate declined at 15:37 UTC because that close had not printed, and three minutes later the approval at confidence 63 entered while calling the confirmation approximated. The trade then came within 16 points of paying, so the entry was not far wrong, but it was not the entry the plan wrote.
The Dow short's stop deserves its own look. The plan placed it at 52,368, two points under the 52,370 level the Trend Agent named as resistance and invalidation, with VWAP above at 52,400.5, while the prompt that wrote the plan asks for a buffer beyond the structural level. A stop at the invalidation level rather than beyond it can be taken out by a test of the level, which is not the same as the idea being proved wrong.
Thursday's Orion gold short is the judgment worth watching rather than changing. The note wrote down exhaustion, lunch-hour timing and a first target at tested support, and scored the trade 5.6 out of 10, a probability that will lose a large share of the time by design. The system's threshold turned it into an order, and the trade lost the way the note said it might, which is the grading working rather than a fault in it.
No trades this week. EURUSD sat outside our setup criteria and nothing filled.
All EURUSD this week →No trades this week. The pound produced no fill on either line.
All GBPUSD this week →One loss, Friday's Atlas short at 52,233.7, which came within about 16 points of its first target before it stopped at 52,368. Monday's Atlas long paid and is not counted in this card.
All US30 this week →No losses this week. The book's one trade, Monday's Orion long, paid +1.43R at its first target.
All NAS100 this week →No losses this week. Tuesday's Atlas short paid +1.18R at its first target and is not counted here.
All US500 this week →No trades this week. USDCAD sat outside our setup criteria and nothing filled.
All USDCAD this week →One loss, Thursday's Orion short at 3,880.22, sold into a liquidation day that was already 175% of the prior day's range. It stopped overnight at 01:21 UTC Friday.
All XAUUSD this week →Loss of the week: US30 Short · -1R
The loss with the clearest read on both sides.
The direction. Breadth had rolled from an intraday high of +1,012 back to -237, below its 5-day EMA, VIX sat above its 5-day EMA, the Trend Agent was bearish at 72%, and the 60-minute and 15-minute structure were both bearish. The trade ran to 52,105.9, about 128 points our way against roughly 144 to the first target at 52,090.
The trigger and the stop. The plan asked for a 5-minute close back below 52,232, and the approval at confidence 63 entered at 52,233.7 with that close only approximated, three minutes after the gate had declined for the same reason. The stop at 52,368 sat two points inside the 52,370 invalidation the plan itself named, with no buffer beyond it. The Macro Agent leaned bullish at 60% on a softer PPI print.
The size. It stopped at 52,368 a little over five hours after entry, for exactly 1R.
The regime read. The dollar and yields were rising, the 60-minute and 15-minute charts were stacked bearish, and that morning's pullbacks after breaks had been about 50% retracements that failed. The plan sold the 3,880 to 3,886 shelf at the 50% retracement and the broken 3,882.83 swing low, and filled at 3,880.22.
The timing. The note said the day's range was already 175% spent, the 60-minute RSI sat near 19 and London had gone, and it scored the trade 5.6 out of 10. Price reached 3,866.23, about 14 dollars our way against roughly 26 to the 3,854.5 first target, and stopped overnight.
The stop. It sat beyond the 61.8% retracement, the 15-minute EMA20 and round 3,900, where the idea was wrong, for 1R.
Each trade risks +$2,000 (1R). The system's actual scale-out behavior may differ, see disclaimer.
| Scenario | R-multiple | Profit on $100k |
|---|---|---|
| Window drawdownActual | -2R | −$4,000 |
The two losses cost 2.0R, or 4,000 dollars on the simulated 100,000 dollar account at 2% risk. Three winners at their first targets brought the window to +1.53R, about 3,070 dollars. Atlas made +1.11R and Orion +0.43R, and each line gave one trade back in the last two days of the week.
The year to date ledger reads +28.63R across 294 trades at a 55.1% win rate from Jan 12 inception, through the October close. On a static 100,000 dollar account at 2% risk that is $157,278.64, and this week's 4,000 dollars of losing trades is about 2.5% of that figure. Compounding the same trades, where each position is sized off the balance it actually had, gives $166,650.60. The gap, about 9,370 dollars, is the same R run through fixed fractional sizing and adds no edge of its own. It is what disciplined sizing does to a positive record over time, and it cuts the other way in a losing stretch.
November is not in that figure yet. This week's trades, the two losses included, enter the ledger when the month closes.
We keep looking at how Atlas treats the trigger its own plan wrote. When a plan asks for a specific 5-minute close and the gate has just declined because that close did not print, an approval three minutes later that calls the confirmation approximated should have to clear a higher bar, not a lower one. This is the second week running in which an Atlas approval named its reason to wait and entered anyway.
We are also looking at stop placement relative to the invalidation level a plan names. A stop placed at that level, rather than a buffer beyond it, turns a test of resistance into a loss. On the Orion side we change nothing yet: a 5.6 out of 10 trade that loses is the score doing its job, and two gold stops in two weeks are too few to judge the book.
Two losses in five trades is a 60% win rate for the window, against 55.1% across 294 trades from Jan 12 inception. At 55.1%, roughly 45% of trades lose, so a five-trade window should expect two or three losses. This week produced two, and they came last, back to back, which is why the longest streak in the panel reads 2.
Van Tharp's framework in Trade Your Way to Financial Freedom treats a system as a distribution of outcomes measured in units of risk. The 1.18R target in this panel is the average of the week's three first-target winners, not a fixed system parameter. At 1.18R per winner against a 1R loser, breakeven sits near a 46% win rate. The useful part of the framework is that it keeps the loss side fixed: a stopped trade is -1R whatever the instrument, so two stops in a row cost the same 2R whether they come from gold or the Dow.
Jack Schwager's Market Wizards interviews make a related point. The traders who last are not the ones who avoid losses, they are the ones whose losses stay uniform and bounded, and who expect streaks. With about 45% of trades losing, any given pair of consecutive trades will both lose roughly one time in five, so a two-trade streak is ordinary over a record of 294 trades. A stop is still an order and not a promise: the gold short's exit came at 3,904.65, about 2.85 dollars past its 3,901.8 stop on a trade that risked about 21.6. The ledger books every stopped trade at -1R by convention, so that slippage appears in no R figure here.
The caveat is sample size. The 3.74% drawdown in the panel comes from this window's five-trade equity walk alone, in entry order, measured from the week's own high. It is not the system's drawdown from its all-time high, which has been larger in harder months. The 294-trade record and its 55.1% win rate are the numbers with weight behind them, and this week is a thin slice of that record.
They measure different things. This report covers only the losing trades, so it totals 2.0R across two losses. The recap covers all five trades and nets those losses against three winners. Both come from the same ledger records for the same window.
They are the two trading lines subscribers run in the app, and every trade in this report belongs to one of them. This week Orion made +0.43R across two trades with one loss, and Atlas made +1.11R across three trades with one loss.
Because the report follows the losses and not the result. Any week with at least one loss gets one, so every stopped trade is named and examined whatever the week's total came to. Drawdown reporting is standard practice in professional fund management, and a record that only appears in bad weeks would tell readers less.
The ledger adds a month once its window closes. November is still open, so the figure still reads +28.63R through the October close, and this week's trades enter it when November ends. We publish it this way so a number printed in one month still means the same thing later in the year.
The 30 minutes is the life of the order, the window in which price can reach the zone and fill. Once it filled at 16:51 UTC Thursday, the position stayed open with its stop and first target until one of them printed, and the stop printed at 01:21 UTC Friday.
Subscribers receive every signal, winners and losers alike, three minutes before entry, with full reasoning.
Dollar figures are simulated on a $100,000 account at 2% risk per trade. Drawdown trajectories shown reflect a small window sample size and are not projections of forward performance. Past performance, including losses, is not a guarantee of future results. Actual subscriber P&L varies with account size and execution. YTD context: +28.63R YTD across 294 trades, see stats strip.
The trades in this article are taken by the SkyAnalyst agents in live demo accounts at a regulated broker, in a live market environment, so the demo account reflects every transaction SkyAnalyst makes. The app logs each trade and confirms it against the live price feed of the broker; the capital is simulated and no real capital is at risk. The demo account is linked to MyFxbook, which publishes its results publicly. We publish these results to study how the agents trade and reason, for education and trade analysis, not as a recommendation. Trade at your own risk.
A +5.88R week still opened with three straight stops. Four losses at exactly 1R each, three of them Nasdaq shorts sold into a bounce that had not finished, and one Atlas euro short approved before its trigger fired.
Thirteen trades, all of them shorts, nine wins and +5.88R. Tuesday opened with three stops in a row, then seven straight winners from Thursday morning through Friday turned a 3R hole into the week's result.

The Atlas GBPUSD trader sold Cable at 1.3112 only after the London high was rejected. Eleven minutes later the market tested that rejection again, and it held. TP2 printed at 2:26 PM ET.