SkyAnalyst/Journal/Recaps/Apr 20-26, 2026
SkyAnalyst Journal · Weekly RecapApr 20-26, 2026

Apr 20-26, 2026: One TP3 Winner, Three Stops, and a Corrected -1.33R Republish

Four filled trades, one winner, three losses, -1.33R net at TP1 baseline. The Apr 23 EURUSD short ran clean to TP3. The other three stopped. Republish after a m

Net result
−1.3R
4 trades · 25.0% win rate · Apr 20-26, 2026
SA
The SkyAnalyst Team
AI Research & Trading Desk
May 3, 2026·7 min read·Weekly Recap · Short
Instrument
Multi · Weekly Recap
Direction · Session
Short · Apr 20-26, 2026
Duration
Outcome
-1.33R
4 trades · 25.0% win rate
Section 00 · The system

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.

Trend
Reads 5m / 15m / 60m charts, scores structure, triggers entries when confluence clears the threshold.
Macro
Gates regime before any pattern. Reads yields, DXY, VIX, oil — the tape behind the tape.
Cross-Asset
Checks correlated markets. Vetoes false breaks, confirms real ones.
Risk
Sizes positions, sets stops, enforces portfolio exposure.

Four filled trades, one winner, three losses, -1.33R net on the TP1 baseline. That is the corrected scorecard for Apr 20-26, 2026. This article is a republish. The first version of this recap counted seven trades and -4.33R because the prior methodology silently treated three unfilled conditional orders as -1R losses. The fix dropped those phantom rows. The honest window reads 25 percent win rate and -1.33R, and that is what we publish here. The single winner was a EURUSD short Thursday afternoon that ran the full TP1, TP2, TP3 ladder. Under our recap baseline we book it as +1.67R from entry to TP1. The broker fill at TP3 ran +4.17R. The full anatomy lives in the case study at the Apr 23 EURUSD short writeup. The three losses are walked in detail in the companion drawdown report. Last week's recap sits at the Apr 13 recap; March's monthly recap covers the longer window.

Act 1: Mon-Wed produced no fills

Mon Apr 20 through Wed Apr 22 were zero-fill sessions. Conditional orders were posted. Price did not interact with any trigger level inside the active window. The first half of the week was a clean abstain.

Act 2: Thursday produced the winner, then the first loss

Thu Apr 23 opened the entry column. EURUSD Short at 14:58 UTC on a resistance-rejection setup graded B cleared confluence. Macro gated bearish-EUR on a firming DXY. Cross-Asset confirmed with US yields bid into the cash open. The trade ran entry through TP1 to TP2 and out at TP3. TP1-baseline credit: +1.67R. Equity printed $103,333 at the peak.

Fifty-three minutes later, NAS100 Long at 15:51 UTC on a conditional pullback to a VWAP/structure zone graded C+ cleared. The index rolled through the zone instead of holding it. Stop at -1R inside the hour. Thursday closed +0.67R cumulative.

Act 3: Friday gave back the rest

Fri Apr 24 brought two filled entries, both losses. US500 Short at 14:05 UTC on a VWAP rejection setup graded C+ stopped at -1R as the index reclaimed VWAP through the cash open. Forty-four minutes later, XAUUSD Long at 14:49 UTC on a pullback to the Trend Agent invalidation zone graded C+ stopped at -1R. Equity closed Friday at $97,333. Window net: -1.33R.

Key insight
“Four filled trades, one TP3 winner, three SL hits. The honest read for the window is 25 percent and -1.33R, not the 14.3 percent and -4.33R the original build showed.”
SkyAnalyst Risk Agent · Weekly review
Section 03 · The audit trail

Every trade the system took.

1 winners3 losers·Winners link to full case study
|
DateTimeInstrumentDirModelSetupGradeR$ SimResultDetails
Apr 2314:58 UTCEURUSDShortClaude Opus 4.6Conditional Short EURUSD at Resistance RejectionB+1.67R+$3,333TP3 hit · ★ Trade of the weekRead case →
Apr 2315:51 UTCNAS100LongClaude Opus 4.6Conditional Pullback Long at VWAP/Structure ZoneC+-1.0R-$2,000Stop hit-
Apr 2414:05 UTCUS500ShortClaude Opus 4.6VWAP Rejection / Opening Range Breakdown ShortC+-1.0R-$2,000Stop hit-
Apr 2414:49 UTCXAUUSDLongClaude Opus 4.6Bullish Pullback to Trend Agent Invalidation / Support ZoneC+-1.0R-$2,000Stop hit-
EURUSD · Short
Apr 23 · 14:58 UTC
Claude Opus 4.6TP3 hit · ★ Trade of the week
Setup
Conditional Short EURUSD at Resistance Rejection
Grade
B
R
+1.67R
$ Sim
+$3,333
Read case →
NAS100 · Long
Apr 23 · 15:51 UTC
Claude Opus 4.6Stop hit
Setup
Conditional Pullback Long at VWAP/Structure Zone
Grade
C+
R
-1.0R
$ Sim
-$2,000
US500 · Short
Apr 24 · 14:05 UTC
Claude Opus 4.6Stop hit
Setup
VWAP Rejection / Opening Range Breakdown Short
Grade
C+
R
-1.0R
$ Sim
-$2,000
XAUUSD · Long
Apr 24 · 14:49 UTC
Claude Opus 4.6Stop hit
Setup
Bullish Pullback to Trend Agent Invalidation / Support Zone
Grade
C+
R
-1.0R
$ Sim
-$2,000

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 week's pattern was one clean directional read on EURUSD, then three lower-confluence setups that did not survive the cash open. The single B-grade entry won and ran the full ladder. The three C+ entries each stopped at -1R inside the first hour after fill.

Inside the four filled trades, the EURUSD short was the only setup that scored above the C+ floor band. The three losses clustered at the floor of the actionable range, where a roughly 40 percent failure rate is the design assumption.

How one TP3 winner carried the recap

The Apr 23 EURUSD short combined a B grade with a regime read the cash open confirmed. Under TP1 baseline it credits +1.67R. The broker fill at TP3 ran +4.17R. Subscribers running scale-out at TP1, TP2, and TP3 booked closer to the broker figure. The recap baseline holds the projection at TP1 across every winner so the period comparisons stay consistent.

Decision highlights

The Thursday EURUSD short at 14:58 UTC at a B grade is the single highest-quality entry of the window. Macro gated bearish-EUR on a firming DXY, Cross-Asset confirmed with bid US yields, the structure read held, the trade ran the full ladder. TP1-baseline credit +1.67R; the case study documents the full +4.17R run.

The Friday XAUUSD long at 14:49 UTC, forty-four minutes after the US500 short stopped, defined the loss side. The Risk Agent did not widen the floor or pause the engine. Confluence cleared at the same threshold the morning's loser had used. The setup stopped at -1R.

The methodology decision sits behind every number in this republish. The prior build counted three unfilled conditional orders as -1R losses. The fix drops rows that never produced a fill price. Four filled trades, 25 percent win rate, -1.33R is the honest window.

Key insight
“Thursday's EURUSD short at 14:58 UTC ran the full ladder to TP3. Under the TP1 baseline we book it as +1.67R. The broker fill at TP3 ran +4.17R.”
SkyAnalyst Trend Agent · 14:58 UTC
Section 04 · Head-to-head

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.

C
Claude
Opus 4.6
-1.3R
Trades
4
Win rate
25%
Avg R
-0.33
Led this week on
  • EURUSD+1.7R · 1 trade
  • XAUUSD-1.0R · 1 trade
  • US500-1.0R · 1 trade
Notable trade
EURUSD Short · Apr 23 · +1.67R
G
GPT
-
No GPT trades this window.

Same signals, same risk framework, different foundation model.

Section 07 · Instrument deep dive

Six instruments, six stories.

EURUSD
+1.7R
1 trade · 100% WR

EURUSD took one trade for 100 percent win rate and +1.67R net. The Apr 23 short at 14:58 UTC ran from resistance rejection to TP3. The TP1-baseline credit is +1.67R; the broker fill ran +4.17R.

All EURUSD this week →
XAUUSD
-1.0R
1 trade · 0% WR

XAUUSD took one trade for 0 percent win rate and -1R net. The Apr 24 long stopped at -1R when the support zone failed inside the first hour after fill.

All XAUUSD this week →
US30
-
0 trades

US30 was inactive. No setup cleared the confluence floor in this window.

All US30 this week →
NAS100
-1.0R
1 trade · 0% WR

NAS100 took one trade for 0 percent win rate and -1R net. The Apr 23 conditional pullback long stopped as the VWAP/structure zone gave way through the cash open.

All NAS100 this week →
USDJPY
-
0 trades

USDJPY was inactive. No trigger level interacted with price during the active window.

All USDJPY this week →
US500
-1.0R
1 trade · 0% WR

US500 took one trade for 0 percent win rate and -1R net. The Apr 24 VWAP rejection short stopped as the index reclaimed VWAP through the New York open.

All US500 this week →
Final Outcome
+1.7R
TP3 HIT
Dollar figures calibrated to a $100k account at 2% risk appear below in Simulated Returns.

Win of the week: EURUSD Short · +1.67R

Loss worth learning from

What the system saw that was right

All three losses cleared the published confluence threshold at trigger. The NAS100 long had a clean VWAP/structure zone read with macro risk-tolerant. The US500 short was a textbook VWAP rejection with bond bid into the open. The XAUUSD long had Trend Agent invalidation alignment with cross-asset confirmation. None of the three were structurally indefensible at trigger time.

What the system got wrong

All three were C+ grade entries at the floor of the actionable confluence range. A C+ trade has, by design, a roughly 40 percent failure rate. Three C+ entries clustered inside two sessions, the failure rate compounded, and all three resolved at the stop. The entries were not the error. The cash open repriced the local tape inside the trade lifecycle, and the stop was the only exit on each.

Simulated Returns

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.

Max potential captured
−$2,660
-1.33R · Window net
ScenarioR-multipleProfit on $100k
Window netActual-1.33R−$2,660
Simulated equity · $100,000 baseline · 2% risk per trade
Thu 23Fri 24$97,333$100,000
System Performance · Year to date

All six agents combined.

Net R
-7.2R
Trades
13
Win rate
15%
US30
-2.2R
8 trades
25%
NAS100
-3R
3 trades
0.0%
US500
-2R
2 trades
0.0%
Updated 25 minutes ago
View live stats →
Key insight
“The three losses all resolved at -1R: NAS100 long Apr 23, US500 short Apr 24, XAUUSD long Apr 24. Confluence cleared on each, the local tape did not pay on any.”
SkyAnalyst Macro Agent · Decision log

From the desk

The honest reading: four fills, one TP3 winner, three SL hits, -1.33R net. A 25 percent week at our published expectancy is the kind the system was designed to absorb on the way to the rolling-window math. The single B-grade entry ran clean. The three C+ entries clustered at the floor of the range and the failure rate landed where the design says it should.

The architecture point is the methodology fix. The prior version reported seven trades and -4.33R because the build counted three unfilled conditional orders as -1R losses. They were never filled. The fix drops those rows from the ledger and republishes the honest count. We would rather correct a published number than leave a wrong one in the record.

The TP1-baseline reading of -1.33R undercounts the EURUSD short's broker fill at TP3. Subscribers running scale-out at TP1, TP2, and TP3 booked closer to +4.17R on that single trade. The recap holds the baseline across windows so comparisons stay clean. From the SkyAnalyst Team.

What we're tuning

The three losses do not surface a single tunable artifact. All three were C+ entries at the floor of the actionable range. The design accepts a roughly 40 percent failure rate at that grade in exchange for the volume of opportunities the floor produces. Removing the C+ band would lower expected value over the rolling 100-trade window and would have skipped real winners in prior weeks. The four-trade window is too small to recalibrate the floor in either direction.

The methodology fix is the operational item out of this republish. The benchmark pipeline now drops conditional orders that never produced a fill from the trade ledger. The companion drawdown report applies the same fix on the loss side.

The Short Version

At a Glance

Week Setup Grade
A-
Decisive Trades
4
Best R
+1.67R
Win Rate
25.0%
What subscribers actually see
Three things that hit your phone or inbox this session.
Full subscriber tour →
01 · Signal Alert
SkyAnalyst · now
Enter signal · US30 long
71% confidence
Push notification the moment an agent issues an Enter. Mobile + desktop.
02 · Live Dashboard
US30 +1.5R
SPX idle
NDX −0.4R
EUR live
XAU idle
OIL +0.8R
All six markets at once. Status, open P&L, and every agent reasoning live.
03 · Morning Briefing
Daily briefing
Macro: lean-bull · DXY soft. Trend agents watching US30 micro-support and EURUSD range break.
Rolling aggregate updates each publish
What the agents are watching, delivered at 08:00 local.
0 traders joined

Week at a glance

Why is this recap a republish?

+

The first version counted seven trades and -4.33R. The benchmark pipeline previously treated three unfilled conditional orders as -1R losses. A methodology fix dropped phantom rows from the ledger. The corrected window reads four filled trades, one winner, three losses, 25 percent win rate, -1.33R net.

How did the EURUSD short read +1.67R in the recap and +4.17R in the case study?

+

Recap R-multiples use a TP1-baseline projection on every winner so period comparisons stay consistent. The Apr 23 short ran the full ladder, so the baseline reads +1.67R while the case study documents the full +4.17R TP3 run. Scale-out subscribers booked closer to the case-study figure.

Why did only Claude Opus 4.6 trade this week?

+

Single-model windows occur when one family's confluence math clears threshold and the other does not on the same setups. Four trades is too small a sample to read model dispersion. The longer-window head-to-head lives in March's monthly recap.

What is the system tuning in response to the three losses?

+

Nothing instrument-specific. All three were C+ entries at the floor of the actionable range, where the design accepts a roughly 40 percent failure rate. The operational item out of this republish is the methodology fix that drops unfilled conditional orders from the ledger.

Get next week’s trades before they print.

Subscribers receive the same pre-trade AI analysis three minutes before entry.

Start 7-day free trialWatch a 2-min demo
$79/mo after trial · Cancel anytime

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.

Key insight
“Republishing a thin window with corrected math is the discipline. The three phantom rows the prior build counted as losses were unfilled conditional orders. They never had a fill price.”
From the desk · April 27, 2026
Keep reading

From the SkyAnalyst Journal

All case studies →
trade-analysis

Republished: Feb 23 to Mar 1 Drawdown Report, 3 Losses for -3R, Streak of 1

Three losses, all at minus 1R. Net minus 3R for the loss-side ledger. Longest streak of 1. Original printed 4 losses and a streak of 2; the cancelled-trade fix dropped one phantom NAS100 row from Feb 26.

9 min read
trade-analysis

Republished: Feb 23 to Mar 1 Closes +1.21R After the Cancelled-Trade Fix

Seven trades, four winners, three losses, +1.21R net on a TP1 baseline. Original printed nine trades and +0.80R; the cancelled-trade fix dropped one paused NAS100 row from Feb 26. Corrected ledger.

8 min read
trade-analysis

February 2026 Monthly Recap (Republished): 21 Trades, 61.9 Percent, +4.41R Net

Twenty-one trades, thirteen winners, eight losers, +4.41R net on a TP1 baseline. Original published as 24 trades and +6.64R; the cancelled-trade fix dropped 3 paused rows the dashboard never had.

10 min read