Five Trades, Four Winners, Two Clean TP3 Shorts: Feb 16-22
A short, sharp week. Five entries, four winners, two clean US30 shorts that rode all the way to TP3. Net +1.73R on the simulated account, the best win rate of F
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.
Five trades. Four winners. One loss. Net +1.73R and an eighty percent win rate, the highest of any week in February. By the standards of a reader looking for activity, the week was thin. By the standards of a reader looking for quality, it was the best the month had produced. The shape was unusual. Nothing on Monday, paired NAS100 / US30 longs Tuesday that both hit TP1 before pulling back to stop, silence Wednesday, then two short windows on Thursday and Friday that produced the week's cleanest runners. Both Thursday's US30 VWAP-EMA fade short and Friday's pullback-to-supply short are documented separately. The fifth trade, a NAS100 short thirty minutes before the Thursday US30 entry, was the week's only stop.
Act 1: Tuesday, paired longs, TP1-and-out
Feb 17 re-rated to lean-bullish-equity by the New York morning. At 16:35 UTC the Trend Agent flagged a NAS100 long, and within the same evaluation cycle a parallel US30 long on the same regime read. Both triggered together, both hit TP1, both rolled back to stop. NAS100 closed at +0.62R, US30 at +0.43R, banked under TP1-baseline accounting.
Act 2: Thursday, one stop, then the cleanest short of the month
Feb 19 produced the week's only loss and its biggest winner inside thirty minutes of each other. At 15:02 UTC the system entered a NAS100 short on a continuation read; the trade stopped at -1R. Thirty minutes later the same regime read produced a US30 short, the system entered, and the position ran clean through TP3 for +1.08R on the TP1-baseline scoring. Same direction, same hour, same macro regime. One instrument did not have the structural confirmation, the other did. The Trend Agent treated them as separate setups. The Thursday US30 short is covered in its own VWAP-EMA fade case study.
Act 3: Friday, second clean TP3 in twenty-four hours
Feb 20 at 17:02 UTC the system extended the streak with a US30 short on a pullback into a freshly minted supply zone from the morning's earlier breakdown. Same regime as Thursday, different structural read. The position ran through TP1, TP2, and closed at TP3 for +0.61R on the TP1-baseline scoring. Two clean TP3 shorts in twenty-four hours on the same instrument is not coincidence; it is the macro regime giving the same setup twice and the structural reads being clean enough both times to clear confluence. The Friday short publishes as its own case study.
Related reading: February 2026 monthly recap · Feb 19 case study · Feb 20 case study.
Every trade the system took.
| Date | Time | Instrument | Dir | Model | Setup | Grade | R | $ Sim | Result | Details |
|---|---|---|---|---|---|---|---|---|---|---|
| Feb 17 | 16:35 UTC | NAS100 | Long | Claude Opus 4.6 | Setup #1 · NAS100 LONG (pullback-to-go) | C+ | +0.62R | +$1,248 | TP1 hit | - |
| Feb 17 | 16:35 UTC | US30 | Long | Claude Opus 4.6 | Setup #1 · US30 LONG (pro-trend intraday) | C+ | +0.43R | +$855 | TP1 hit | - |
| Feb 19 | 15:02 UTC | NAS100 | Short | Claude Opus 4.6 | Setup #1 · NAS100 Short (Continuation) | C+ | -1.0R | -$2,000 | Stop hit | - |
| Feb 19 | 15:32 UTC | US30 | Short | Claude Opus 4.6 | SHORT — Sell the VWAP/EMA Fade | C+ | +1.08R | +$2,151 | TP3 hit · ★ Trade of the week | Read case → |
| Feb 20 | 17:02 UTC | US30 | Short | Claude Opus 4.6 | Setup #1 · US30 SHORT (pullback-to-supply) | C+ | +0.61R | +$1,213 | TP3 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 was the bearish-equity continuation short on US30. Two of the three US30 trades were short, both ran through TP3, and both shared the same macro regime: dollar-strength, yields up, equity weakness. The setup names differ, Thursday's was a VWAP-EMA fade and Friday's was a pullback into supply, but the regime read is identical.
Why the regime mattered more than the setup
A pullback-to-supply short and a VWAP-EMA fade short are mechanically different trades. What unified them was the macro regime. With the dollar bid and yields rising, every equity rally became a fade candidate, and the system's job was to find the cleanest structural reads inside that regime.
The discipline beat
On Wednesday Feb 18 the system ran every cycle and entered nothing. The regime gate had gone neutral after Tuesday's softening, and no setup scored above the 55-percent threshold. Wednesday is the day that does not show up in the trade index but is, in our view, half the reason the week's record looks the way it does.
Decision highlights
The decision to enter both Tuesday legs simultaneously was a confluence call, not a hedging move. The NAS100 long and the US30 long triggered on the same regime read, the same evaluation cycle, the same minute. The Risk Agent sized each independently. Both cleared TP1 and both rolled back to stop.
The decision to enter the Thursday US30 short thirty minutes after the NAS100 short stopped is the discipline beat of the week. The NAS100 stop did not contaminate the US30 read. The Trend Agent treated them as separate confluence calculations, and the US30 read cleared on its own merits. A discretionary trader who had just been stopped might have hesitated; the system did not, because hesitation is not in the threshold logic.
Friday's US30 short twenty-four hours after Thursday's is the cleanest demonstration of the week. Same regime, different structural read, same clean execution. The system did not require novelty to take the trade; it required confluence above threshold and clean structural confirmation. Both were present, the position ran.
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.
Same signals, same risk framework, different foundation model.
Six instruments, six stories.
EURUSD was inactive. The pair held a tight range on a neutral euro tape and no setup scored above the threshold all week.
All EURUSD this week →XAUUSD was inactive. Gold consolidated between resistance and support with no clean directional read.
All XAUUSD this week →US30 was the workhorse. Three trades, two winners, one half-runner; net +2.11R on the TP1-baseline. Both Thursday and Friday shorts ran clean to TP3 and produced the week's largest contributors.
All US30 this week →NAS100 was mixed. Two trades, one TP1 winner Tuesday and one stop-out Thursday; net -0.38R on the TP1-baseline. The instrument tracked US30 directionally but the structural reads were thinner.
All NAS100 this week →USDJPY was inactive. The dollar-yen pair held an overnight range and never produced a structural pattern that cleared the threshold.
All USDJPY this week →US500 was inactive. The S&P moved with the broader equity tape but no S&P-specific setup scored separately above threshold.
All US500 this week →Win of the week: US30 Short · +1.08R
Loss worth learning from
What the system saw that was right
The Thursday NAS100 short triggered at 15:02 UTC. The macro regime had rated bearish-equity, the dollar was bid, yields were rising, and the structural read was a rejection at a prior breakdown level. Confluence cleared the 55-percent threshold.
What the system got wrong
Nothing in the entry logic. The structural read on NAS100 was thinner than the parallel read on US30 thirty minutes later, but it was above threshold. What killed the trade was that NAS100 found a buyer twenty points below the entry where the equivalent level on US30 found sellers. The instruments are correlated but not identical, and the NAS100 supply zone was not as well-defended.
What we'd do the same
The entry. A 55-plus percent confluence trade in a regime-supportive macro tape is what the system is built to take. Removing it from the playbook would tighten the win rate but lower the rolling expectancy; the average winner at this confluence range is +2.1R against a 1R stop.
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 | +1.73R | +$3,460 |
All six agents combined.
From the desk
Five trades is a sparse week. Four winners is not. The system was selective, the macro tape gave it two clean continuation windows on Thursday and Friday, and the Tuesday paired longs delivered TP1 before rolling back. By win rate this was the best week of February. By trade count it was one of the quietest. Both are true.
The two case studies we published, the VWAP-EMA fade short on Thursday and the pullback-to-supply short on Friday, are the cleanest pair of structural reads we have published in February. Read together they show how the system finds two different setups inside the same regime. For the running monthly view see the February monthly recap. No drawdown report this week; the gate never opened on a single loss.
What we're tuning
There is no tuning signal in the Thursday NAS100 stop. The trade entered cleanly, the macro context held, and the cousin instrument cleared at the same moment. The variance was in the instrument selection, not in the decision logic.
The per-instrument dispersion is worth tracking. US30 produced both of the week's largest winners; NAS100 took its TP1 winner on the same regime but stopped on a read that US30 cleared. We will continue to monitor this in the February monthly recap.
At a Glance
Week at a glance
How did the system produce a 4W/1L week after the prior week's losing run?
The macro tape changed. Feb 16-22 produced two clean dollar-strength windows on Thursday and Friday, both of which supported continuation shorts on US30. The system did not adjust its threshold or change its setup library; it entered when confluence cleared, which was four of five times this week. The architecture was identical across both weeks.
Why did the Tuesday paired longs both stop after hitting TP1?
Both entries cleared TP1 within the first leg and gave up the runner when the macro tape softened by mid-afternoon. The system books winners at TP1 under the published TP1-baseline accounting, regardless of where the runner finishes. A trader using a different exit strategy would record different numbers.
What made the Thursday and Friday US30 shorts run all the way to TP3?
The Thursday and Friday regimes had more durable macro support. The dollar stayed bid, yields stayed up, and the equity tape stayed weak across both sessions. When the regime that produced the entry stays in place, the position has room to run through multiple take-profit levels. The system does not predict regime persistence; it enters when confluence clears and exits at the stop, the trail, or the take-profit targets.
Why is the win rate this week so much higher than the rolling-window average?
Single-week win rates are dominated by variance. The system's rolling 100-trade win rate is closer to 35 percent and the average winner runs above 2R against a 1R stop. An eighty-percent week is well within the variance distribution; so was the prior week's losing stretch. Subscribers should evaluate the system on the rolling window, not on 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.
Del diario de SkyAnalyst
Mar 2-8, 2026: Four Losses, Three Offsetting Winners
Four US500 losses, -4.00R given back, a 2-trade losing streak Thu into Fri. Three winners in the same five sessions covered most of the draw. The companion recap nets -0.62R.
US30 Short on March 2: Eleven Waits, One Enter, +1.2R at TP1
March opens with a sell-the-rally on the Dow. Twelve evaluations across fourteen minutes, eleven of them wait. The twelfth fired short at 48842 and banked TP1 at 48700.
NAS100 Long on March 4: TP1 Banked, Runner Stopped, +0.93R
A breakout continuation on the Nasdaq 100 cleared TP1 inside the New York session, then the runner reversed and tagged the original stop. Reported result reflects the TP1-baseline R.