How SkyAnalyst Waited Through a Falling Confidence Print on US30
SkyAnalyst AI journal entry: US30 Short on Jun 3, 2026 closed +1.83R on TP3. Full workspace view, decision log, and AI reasoning, unedited. SkyAnalyst AI journa
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. That’s what makes the system auditable — and it’s what this case study will show, step by step, on a specific setup the trend agent almost passed on.
The tape that opened the door
US30 had already done most of its work by the time the New York open landed. Yesterday's close at 51,391 had been pierced at the bell, and the gap filled near 51,406 on a thin first print. By the time the 10:00 ET ISM Services beat hit the wire at 54.5 against a 53.7 forecast, price was already cycling down. The 10:00 candle dropped from 51,070 to a spike low near 50,907. Headlines called it counterintuitive. The cross-asset readings did not.
The Macro Agent ingest at 10:11 ET carried a lean_bull US30 tag at 58% confidence, sub-threshold for the confluence count and 31 minutes old by the time entry came. That stale tag mattered less than what the live cross-asset stream said. The 10Y yield was at 4.477%, above its 5-day EMA and yesterday's high. The dollar index was at 99.488, also above both. The VIX had printed 16.63 intraday against a 5-day high of 16.29. The NYAD breadth reading was at -974, deep into negative territory and well below its 5-day EMA. Each of those reads is a small piece of evidence on its own. Stacked together they reshape what "good news" from ISM actually means for the Dow.
The Trend Agent was BEARISH at 70% on a TRENDING regime (the inverse posture from our recent US30 long), with invalidation pinned at 51,258 and the daily pivot at 51,162 as the first overhead. The setup name in the workspace right rail was the literal Post-ISM Second-Chance / Opening Range Break Continuation Short. The opening range from 13:30 to 14:00 UTC ran 51,236 high to 51,001 low. The 10:00 candle broke the OR low. The 14:30 UTC candle had not reclaimed it. That is the structure that made the second-chance read possible. Not the ISM print. Not the macro tag. The price action after both.
What professional traders call the Post-ISM Second-Chance / Opening Range Break Continuation Short is one of the cleaner intraday setups in index futures, and one of the easiest to mis-execute. The pattern asks three questions in order. First: did a high-impact data print break the opening range. Second: did price retrace into the range without reclaiming the broken level. Third: is the cross-asset regime supportive of the original break direction. Two yeses and a third "weak yes" is the conviction window. Three full yeses are rare.
The first leg: the data-driven range break
The opening range itself is a piece of information. From 13:30 to 14:00 UTC, US30 traded a 235-point range, high 51,236, low 51,001. The 10:00 ET data print closed the 10:00 candle decisively below the OR low. That break alone is not a trade. A range break on data is often a fade setup as the initial reaction unwinds. What makes the break tradeable is whether the broken level holds as resistance on the retrace, and whether the cross-asset regime kept up with the move or moved against it.
The second leg: the second-chance retrace
The retrace is where the setup gets honest. Price has to come back. If it does not, the move is too fast and chasing is a mistake. If it comes back too far, the regime read was wrong and the retrace will become a reclaim. The sweet zone is the 50 to 62 percent retrace of the data candle. On June 3, that zone was 50,988 to 51,008 against an OR low at 51,001. The retrace held there. The 5m candle stalled. We watched the 5m candles leading into the 14:42 UTC entry for an upper-wick rejection at the OR low. The retest high printed at 51,065 and rejected.
The third leg: the cross-asset regime cosign
Confluence is what separates a setup from a guess. Five of seven confluences scored on this short. The two that did not score are worth naming. The Macro Agent's discrete US30 tag was a stale lean_bull at 58%, generated 31 minutes before entry and pre-dating the full sell-off. The 60-minute EMA stack had not yet completed a clean bearish cross with the fast above the slow; price was below both but the cross was still pending. The remaining five all confirmed. Breadth was strongly negative, the VIX was elevated and rising, the Trend Agent was bearish above the 60% threshold, the price was at the second-chance retrace zone with a visible 5m reaction, and no further high-impact data was inside the 30-minute window.
Why this is one pattern, not a doctrine
This is not the only intraday pattern we trade. It is one of about a dozen the Trend Agent maintains in its library. A Post-NFP fade. An Asian-range reclaim. A VWAP reject in trend. A 60m higher-low continuation, of the kind we ran in a NAS100 breakout-retest. A liquidity sweep at session change. We do not lead with any one of them. We lead with the tape and the regime, and let the pattern surface from the data the four agents are reading.
Setup grade C+ is the agent's own grade, and C+ is the bottom edge of the conviction window. The Trend Agent assigns a grade based on how cleanly each leg lined up. A C+ means the setup is takeable but not exemplary. On this trade, the C+ came from two things: the stale Macro snapshot and the incomplete 60m EMA cross. The win came anyway, because price discovery at the OR-low retest is a higher-probability event than the grade alone implies when breadth and volatility are this risk-off.
That is the product angle worth naming. The system doesn't favor any single strategy. It reads the tape first and asks which pattern the cross-asset regime is currently rewarding. Dynamic, not dogmatic. The agents are paid to disagree with each other in the open before agreeing on a number. That is the framework readers should hold in mind as they read the rest of this case study.
US30 NY AM Session Analysis — June 3, 2026
1. Breadth Assessment (Primary Driver)
| NYAD Metric | Value | Assessment |
|---|---|---|
| Current | -974 | Deeply negative |
| 5-Day EMA | -200.8 | Current well below EMA |
| Today's Low | -1,267 | Near 5-day low territory |
| Yesterday Close | +216 | Massive 1-day reversal (from +216 → -974) |
| 5-Day Trajectory | +376 → -485 → -158 → +216 → -974 | Deteriorating sharply |
| Range Position | Below yesterday's low | Breaking down |
Breadth Verdict: STRONGLY NEGATIVE. NYAD at -974 is well below its 5-day EMA (-200.8), crashing from yesterday's +216 close. This is approaching 5-day low territory (yesterday's low was -418, and today has already printed -1,267 intraday). The breadth picture is unambiguously negative — the broad market is declining aggressively. This sets the default directional bias to SHORT.
Breadth-Extreme Veto Check: NYAD is near/at its 5-day low with price declining → No longs permitted.
| VIX Metric | Value | Assessment |
|---|---|---|
| Current | 15.98 (last snapshot showed 16.57) | Slightly above EMA |
| 5-Day EMA | 15.97 | At/slightly above |
| Today's High | 16.63 | Above yesterday's high (16.29) |
| Yesterday Close | 15.76 | VIX rising today |
| 5-Day High | 16.29 (yesterday) | Today's 16.63 exceeds it |
VIX Verdict: VIX is above its 5-day EMA and has spiked above yesterday's high to 16.63 intraday. This supports wider stops, mean-reversion approaches, and favors short setups. VIX is not at panic levels (still sub-17), but the directional move is clearly supportive of downside.
Risk-Off Compound Check: VIX spiking above 5-day high (16.63 > 16.29) AND NYAD strongly negative (-974) → Only shorts qualify. ✅
2. Macro Regime & Cross-Asset Confirmation
Macro Agent Summary
| Factor | Value |
|---|---|
| Group Bias | lean_bear (58%) |
| US30 Bias | lean_bull (score 35, confidence 58%) |
| Tradeability | High (72/100) |
| Key Factors | Price near 5d EMA, cyclical tilt favorable, oil rising |
| Catalyst | ISM Services PMI 54.5 (beat; already released at 10:00 ET) |
Important Nuance: The Macro Agent carries a lean_bull US30 bias, but this was generated 31 minutes ago before the full sell-off materialized. The group bias is lean_bear. With confidence at 58% (below the 60% threshold for confluence), this factor does not count as aligned for shorts, nor does it confirm longs.
Cross-Asset Confirmation
| Asset | Current | vs 5d EMA | vs Yesterday | Signal |
|---|---|---|---|---|
| 10Y Yield | 4.477% | Above (4.469) | Above yest high (4.463) | Sharp yield spike — headwind for rate-sensitives |
| DXY | 99.488 | Above (99.176) | Above yest high (99.345) | Surging — pressures multinational Dow components |
| Oil | 97.55 | Above (95.96) | Above yest high (96.28) | Rising — mixed for Dow (energy +, consumer -) |
| Gold | 4,448 | Below (4,491) | Below yest low (4,462) | Risk-off rotation out of gold into USD |
Cross-Asset Verdict: The combination of sharply rising yields, surging DXY, and VIX above its 5-day high creates a clear headwind for US30. Rising yields pressure financials (JPM, GS — heavy Dow weighting). A surging DXY pressures multinationals (MSFT, AAPL, JNJ, etc.). The Trend Agent explicitly flags this: "VIX, DXY, and 10Y yields all above yesterday's highs — a near-term headwind."
Regime Classification: RISK-OFF
- NYAD: Strongly negative ✅
- VIX: Elevated and rising above 5-day levels ✅
- Yields + DXY: Both surging above ranges ✅
- All three confirm downside → Maximum short conviction
3. Trend Structure & Key Levels
Trend Agent Summary
| Parameter | Value |
|---|---|
| Direction | BEARISH |
| Confidence | 70% |
| Strength | Moderate |
| Regime | TRENDING |
| Invalidation | 51,258 |
| Key Resistance | 51,162 (daily pivot) |
| Key Support | 50,829 (prior-day low) |
| VWAP | ~51,248–51,273 (declining through session) |
| Macro Assessment | HEADWIND |
60-Minute EMA Analysis
| Candle Time (ET) | Close | vs Fast EMA | vs Slow EMA | RSI | MACD Hist |
|---|---|---|---|---|---|
| 5:00 AM | 51,166 | Below (51,239) | Above (51,106) | 45.9 | -28.8 |
| 6:00 AM | 51,203 | Below (51,235) | Above (51,109) | 49.7 | -30.3 |
| 7:00 AM | 51,227 | Below (51,235) | Above (51,113) | 52.1 | -28.8 |
| 8:00 AM | 51,148 | Below (51,226) | Above (51,114) | 44.6 | -32.1 |
| 9:00 AM | 51,054 | Below (51,210) | Below (51,111) | 37.6 | -39.0 |
| 10:00 AM | 50,986 | Below (51,189) | Below (51,106) | 33.6 | -46.1 |
60m EMA Stack: Price is now below both fast and slow EMAs. The fast EMA (51,189) has rolled over above the slow EMA (51,106) but is converging — a bearish cross is imminent. This is NOT a clean bearish stack yet (fast still > slow), but price action is decidedly bearish with accelerating MACD histogram and RSI at 33.6 (approaching oversold).
Key structural takeaway: The 60m slow EMA (~51,106) acted as support early in the session but was decisively broken at 9 AM ET. Price is now trading ~120 points below it with no visible bounce attempt holding.
Gap Analysis
- Yesterday's close: 51,391
- Today's high: 51,390 (gap filled essentially at the open)
- Current: ~50,986
- Move from high: ~404 points
- Daily ATR context: 60m ATR is ~52 points, implying a daily ATR around 200-250 points. This 404-point decline represents ~1.6-2x daily ATR — an expanded-range day.
Key Levels Map
| Level | Price | Significance |
|---|---|---|
| Yesterday High / Today High | 51,390–51,406 | Major resistance, session ceiling |
| Trend Invalidation | 51,258 | Above = bearish thesis dead |
| VWAP (declining) | ~51,244 | Institutional anchor, declining |
| Daily Pivot | 51,162 | First overhead resistance |
| 60m Fib 38.2% | 51,049 | Just broken |
| 60m Fib 23.6% | 50,965 | Immediate downside level |
| Today's Low | 50,899 | Today's session low — key target |
| Yesterday's Low | 50,807–50,829 | Major structural support / prior-day magnet |
| Support Cluster | 50,795–50,837 | Multiple 60m support levels |
4. Lower-Timeframe Entry Analysis
15-Minute Assessment
The 15m chart is in full bearish alignment:
- EMA: Price (50,986) far below fast EMA (51,126) and slow EMA (51,193). Fast < Slow = bearish cross confirmed.
- RSI: 26.3 — Oversold. Has been oscillating between 26-31 for the last 4 candles.
- MACD: Line at -62.8, histogram at -18.2, strong bearish momentum. Histogram has been expanding.
- VWAP: Price at lower 2 standard deviation band — deeply extended.
- Volume: Spike detected at 10:00 AM bar (176 ticks vs 73 avg), confirming the ISM-driven sell-off.
5-Minute Assessment
The 5m chart confirms:
- EMA: Price (50,986) far below fast EMA (51,059) and slow EMA (51,127). Full bearish alignment.
- RSI: 36 — bearish but not oversold on this timeframe. Recently bounced from 28.8.
- MACD: Line at -53.0, histogram at -9.7 — histogram is contracting from the -15.1 peak. This suggests the immediate momentum impulse is decelerating.
- VWAP: Lower 2SD band. Deeply extended.
- Volume: Normalizing after the 10:00 AM spike.
Post-Data Second-Chance Assessment
The ISM Services PMI at 10:00 AM (54.5 vs 53.7 forecast — a beat) produced a counterintuitive reaction: the good news is bad news paradigm where strong economic data reinforces the "higher for longer" rate narrative, aligning with the yield spike and DXY surge.
- ISM spike low: ~50,907 (at 10:15-10:20 AM ET)
- Current price: ~50,986
- Retracement from the spike move: The 10 AM candle opened at 51,070, dropped to 50,907 — a 163-point move. Current price at 50,986 represents a ~48% retracement.
- 50-61.8% retrace zone of the 10 AM spike move: 50,988–51,008
Price is currently sitting in the post-data second-chance zone, consolidating between 50,979-50,996 on declining volume. This is the textbook setup: the initial ISM spike has retraced ~50% back toward the 5m EMA9 area, and the question is whether sellers re-engage here.
Opening Range Context
- NY Open range (9:30-10:00 ET / 13:30-14:00 UTC): High ~51,236, Low ~51,001
- OR was broken to the downside at ~10:00 AM ET (14:00 UTC candle)
- The OR break held — price continued to 50,907 and has not reclaimed the OR low (51,001)
- OR low at 51,001 is now resistance for any bounce attempt
5. Confluence Gate — SHORT Setup Evaluation
Setup: Short on Post-Data Retracement / OR Break Continuation
| # | Confluence Factor | Status | Detail |
|---|---|---|---|
| (i) | NYAD direction matches SHORT | ✅ YES | NYAD at -974, deeply negative, below 5d EMA |
| (ii) | VIX supports SHORT | ✅ YES | VIX at 15.98-16.57, above 5d EMA, spiked above 5d high |
| (iii) | Macro Agent aligns ≥60% | ⚠️ PARTIAL | Group bias lean_bear (58%), US30 bias lean_bull but stale; confidence 58% < 60% threshold |
| (iv) | Trend Agent aligns ≥60% | ✅ YES | BEARISH at 70% confidence, TRENDING regime |
| (v) | 60m EMA stack supports direction | ⚠️ PARTIAL | Price below both EMAs, but fast EMA still > slow (no completed bearish cross). Effectively bearish. |
| (vi) | Price at key level with 5m reaction | ✅ YES | Price at 50% retrace of ISM spike (50,988-51,008 zone), at OR low (~51,001), visible stalling on 5m with declining momentum |
| (vii) | No high-impact event within 30 min | ✅ YES | ISM and ADP already released; Bessent speech ongoing but priced in |
Score: 5/7 = Medium-High (6.5–7.5)
Factors (i), (ii), (iv), (vi), (vii) confirmed. Factors (iii) and (v) are partial — Macro confidence just under threshold and 60m EMA hasn't completed the bearish cross despite price being well below both.
Additional checks:
- Breadth-extreme veto: NYAD at 5-day low + price declining → No longs ✅ (not applicable to our short)
- Risk-off compound: VIX above 5d high + NYAD negative → Only shorts ✅
- FOMC/Fed: Bessent speaking but not Fed policy — standard threshold applies (4/7 min) ✅
- US30 vs NYAD divergence: No divergence — both declining ✅
6. US30-Calibrated Risk Management
Volatility Assessment
| Metric | Value |
|---|---|
| 60m ATR | 51.7 points |
| 15m ATR | 51.6 points |
| 5m ATR | 55.5 points |
| VIX Level | 15.98-16.57 (expanding) |
| Day Type | Expanded range — >400 pts from high to low |
This is an expanding volatility day — stops must be wider to survive noise, but R:R must still clear 1.5:1 minimum.
Stop Loss Placement
- Structural level: OR Low at 51,001 + the 50% retracement zone ceiling at ~51,008
- Buffer for execution slippage: +12 points
- Stop Loss: 51,020
- Stop distance from entry zone (~50,975): ~45 points
- This exceeds 1x 5m ATR (55.5) — marginally tight. Adjusting to structural: the 5m candle at 14:05 bounced to 51,065 from the OR low area, so a more conservative stop at 51,070 (above the post-OR-break retest high) gives ~95 points of risk.
- Check vs Trend Agent invalidation (51,258): 51,070 is well below → ✅
Preferred stop: 51,070 — above the post-OR-break reaction high, beyond 1x 60m ATR (51.7), structural.
Target Structure
From entry ~50,975:
| Target | Price | Distance | R Multiple | Structure |
|---|---|---|---|---|
| TP1 | 50,899 | ~76 pts | 0.8R | Today's session low |
| TP2 | 50,829 | ~146 pts | 1.54R | Yesterday's low / 60m pivot low |
| TP3 | 50,795 | ~180 pts | 1.89R | 60m support cluster |
TP1 Problem: At 0.8R, the nearest structural target (today's low at 50,899) is under 1R. However, this is the first structural magnet — yesterday's low at 50,829 is the high-probability magnet on a day where the prior-day low acts as a gravity point, especially given the breadth washout. Per the framework: "a close TP1 with a strong TP2 at 2R+ is valid" — while TP2 at 1.54R doesn't reach 2R, the structural path to yesterday's low is clean with no intervening support.
Adjusted approach: Use 50,899 as a partial profit level (reduce risk) and 50,829 as the primary target.
R:R Assessment
- Stop: 95 points (entry ~50,975 to stop 51,070)
- TP1 (partial): 76 points (0.8R) at 50,899 — today's low
- TP2 (primary): 146 points (1.54R) at 50,829 — yesterday's low
- TP3 (extended): 180 points (1.89R) at 50,795 — support cluster
The minimum 1.5:1 R:R is met at TP2. On an expanded-volatility day with this level of breadth deterioration and cross-asset confirmation, targeting yesterday's low is high-probability.
PRIMARY SETUP
SHORT — Post-ISM Second-Chance / Opening Range Break Continuation
| Parameter | Detail |
|---|---|
| Directional Bias | SHORT |
| Entry Zone | 50,975 – 51,010 |
| Entry Trigger | 5m candle closes below 50,975 after stalling in the 50,980-51,010 retracement zone, OR a rejection candle (upper wick) at/near 51,001 (OR low acting as resistance). Price must remain below 5m EMA9 (~51,059). |
| Stop Loss | 51,075 (above the post-OR-break retest high at 51,065 + 10pt buffer) |
| TP1 | 50,900 (~76 pts / 0.8R) — Today's session low. Partial profit: reduce position by 40%. |
| TP2 | 50,830 (~146 pts / 1.54R) — Yesterday's low / 60m pivot low. Structural magnet. Take additional 40%. |
| TP3 | 50,795 (~180 pts / 1.89R) — 60m support cluster. Runner with trailing stop. |
| Confidence Score | 7.0/10 (Medium-High) — 5/7 confluences |
| Invalidation | Price reclaims 51,258 (Trend Agent invalidation) on a 15m close → bearish thesis fully invalidated. Tactical invalidation at 51,075 (stop). |
Confluences Confirmed
- ✅ NYAD deeply negative (-974), below 5d EMA — strongest short signal
- ✅ VIX above 5d EMA and spiking above 5d high — risk-off confirmed
- ✅ Trend Agent BEARISH at 70%, TRENDING regime
- ✅ Price at post-ISM 50% retrace zone + OR low (51,001) with visible 5m stalling
- ✅ No high-impact events remaining within 30 minutes
Risk Factors
- ⚠️ 15m RSI oversold (26.3): A mechanical bounce is possible. This is the primary counterargument. However, on expanded-range days with breadth washouts, 15m RSI can stay oversold for extended periods.
- ⚠️ 5m MACD histogram contracting: Immediate momentum is decelerating. Need sellers to re-engage.
- ⚠️ Macro Agent lean_bull on US30 (stale, pre-selloff): Creates ambiguity, though group bias is lean_bear and cross-assets have deteriorated.
- ⚠️ Deep below VWAP (lower 2SD): A VWAP mean-reversion squeeze is possible if any positive catalyst emerges. This argues for disciplined stop placement, not for avoiding the trade.
Trade Management Notes
- Position sizing: On this expanded-volatility day with a 95-point stop, reduce standard position size to maintain consistent dollar risk (~1% of equity). The wider stop should be compensated by smaller lot size, not by tightening the stop.
- If price breaks below 50,899 (today's low) without pausing: Trail stop to entry (breakeven) and hold for TP2/TP3.
- If price bounces sharply above 51,001 within 3 candles of entry: Consider early exit — the OR low rejection thesis has failed.
- Time consideration: It is approximately 10:30 AM ET. The NY AM session has ~2 hours of active flow remaining before the midday lull. Yesterday's low at 50,829 is 146 points away — achievable within this window given today's pace (~400 points traversed in 1 hour).
SECONDARY CONSIDERATION (Not Active — Monitoring)
Potential Short Add on VWAP Rejection
If price bounces to the 51,100–51,160 zone (daily pivot area / 60m Fib 38.2% at 51,049), this would offer a higher-entry short with a tighter stop at 51,258 (Trend invalidation). This setup would score at least 5/7 and offer superior R:R (~2.5:1+ to yesterday's low). Monitor but do not pre-position — this requires price to first bounce significantly.
Summary
| Item | Assessment |
|---|---|
| Regime | Risk-Off (NYAD negative + VIX spiking + yields/DXY surging) |
| Breadth | Strongly bearish — NYAD -974, near 5d low, massive 1-day deterioration |
| Trend | Bearish, 70% confidence, trending regime |
| Active Setup | SHORT @ 50,975–51,010 → Stop 51,075 → Targets 50,900 / 50,830 / 50,795 |
| Score | 7.0/10 (Medium-High, 5/7 confluences) |
| Primary Risk | 15m oversold conditions could produce a mechanical bounce before continuation |
| No-Trade Conditions | Longs are vetoed by breadth-extreme rule + risk-off compound rule |
Decision log
At 14:37 UTC, the first eval went out with confidence at 48% on a WAIT. The setup was forming but the 5m candle had not yet stalled. Price was inside the 50-62 percent retrace zone but the wick read was incomplete. The framework rejects a confluence threshold below the 60% line. The Trend Agent did exactly what the threshold tells it to do. It published the 48% number, tagged the decision WAIT, and left the field open for the next print.
At 14:39 UTC the eval came back at 40%. Eight points lower. A momentum-style system would treat that as the setup deteriorating. The framework reads it differently. The 14:39 candle was a 2-minute candle inside the larger 5m structure. The 5m had not closed. The MACD histogram on the 5m was contracting from -15.1 toward -9.7, which is the immediate impulse decelerating, not the structure failing. The 15m RSI was at 26.3, oversold. None of those reads cleared the gate. The Trend Agent published 40%, tagged WAIT again, and re-armed for the next 5m close. The number got worse before it got better. That is allowed.
At 14:42 UTC the print came back at 68% on ENTER. The post-break retest high on the 5m had stalled at 51,065 and price had traded back down toward the OR low. The 5m EMA9 was at 51,059 and price was below it. The cross-asset stream still confirmed the regime. The OR low was holding as resistance. The Trend Agent moved the decision to ENTER short at 50,976, stop at 51,075, with TP1 at 50,900, TP2 at 50,830, and TP3 at 50,795. The Risk Agent sized the position against a 99-point stop. The Cross-Asset Agent did not veto. The Macro Agent's stale lean_bull tag was no longer load-bearing because four other reads had carried the gate.
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 |
|---|---|---|
| Stop hit (invalidated) | -1R | −$2,000 |
| TP1 hitActual | +0.77R | +$1,540 |
| TP2 hit | +1.47R | +$2,940 |
| TP3 hit (max potential) | +1.83R | +$3,660 |
All six agents combined.
What a +1.83R (TP3) actually paid for
The hero number on this case study is +1.83R (TP3). The realized number on the same trade is +0.77R (TP1). Both are correct. The broker closes the full position at the first take-profit; the hero R is the path the market actually traveled before the move exhausted. The simulated-returns panel above shows both at 2% account risk on a $100,000 example: $1,540 realized at TP1, $3,660 at the TP3 max-potential line. The track-record line, the one we carry forward into the monthly tally, is the TP1 line at +0.77R (TP1). Showing the hero alongside the realized is what lets readers compare the path with the ledger.
The 0-drawdown story is real, but it is one trade
The result reveal block reports max drawdown against this position at 0. That is accurate for this trade. Entry at 50,976, never touched a level above the entry print. The position spent five hours trading away from the entry, not toward it. We surface this because it is rare and because the surrounding journal needs the context: out of seven trades MTD the desk is net -1.76R, on the back of the three-stop week we published last Monday. This was the only deep TP3 ticket of the month so far. We are not selling the 0 drawdown as a generalization.
The patience number nobody else publishes
Three evaluations across five minutes. Two waits. One enter. The middle wait printed lower confidence than the first. The framework still held the wait posture, because the gate is not based on the confidence trajectory; it is based on whether the live read clears the threshold. Most readers will recognize this as the opposite of how a discretionary trader behaves at a screen. A discretionary trader who sees their own number drop almost always backs off. The system did not back off. It re-evaluated. That is what we mean when we say the agents are auditable: the 14:39 UTC print is on the record, the 14:42 print is on the record, and the reasoning between them is published, not concealed.
The setup was C+. The framework took it because five of seven confluences cleared, not because anyone in the room had a strong opinion on US30 today.
How we read this in the running track record
The MTD ledger sits at 7 trades, -1.76R, 42.9% win rate. The QTD and YTD lines match each other at 31 trades and -6.55R (see the full May recap for the rolling picture) because the year began with the same set of trades the quarter holds. A +1.83R (TP3) hero with a +0.77R (TP1) realized contribution lifts the MTD line by 0.77R when the broker closes everything at TP1. The hero number affects the journal narrative. The realized number affects the books. We separate them on purpose.
What we will be watching this week
The Macro snapshot freshness window is the single most important item on the audit log. A 31-minute-old lean_bull tag survived the gate today because four other reads carried it. That is the design intent. A lean_bull tag at 58% confidence is sub-threshold and the framework de-weights it. But the freshness window is a different problem. The Macro Agent re-evaluates on a fixed cadence and on price-action triggers. We are pushing on whether the cadence should compress on data-print days, because the 31 minutes that elapsed between the macro print and entry covered the full move from 51,391 to 50,907. The macro print was right when it was written, and it was less useful by the time the trade triggered.
What this means for readers running the platform
We do not believe the lesson here is "trust the system through a falling confidence number." We believe the lesson is that the gate is a threshold check, not a momentum check on the confidence series itself. If you want to see SkyAnalyst run your markets, the practical implication is that you can read a falling confidence print as a "still gathering evidence" signal, not as a "step back" signal, as long as the next print is inside the window the framework expects.
This is one trade. It is not a proof. The journal continues.
At a Glance
What this teaches about AI-driven trading
How does SkyAnalyst decide to enter when confidence falls between evaluations?
The entry gate is a threshold, not a trajectory check. Each evaluation publishes a confidence number alongside a wait, enter, skip, or exit decision. The threshold for ENTER on most setups is 60% with at least five of seven confluences scoring. A falling number from print one to print two does not flip the gate; only a sub-threshold print on the eligible candle does. The system waits for the next eligible candle and re-evaluates without adjusting the prior posture. That separates evidence gathering from sentiment.
Why does the Macro Agent's read sometimes disagree with the Trend Agent's read?
The two agents read different time horizons and different data streams. The Macro Agent looks at multi-asset regime data and runs on a re-evaluation cadence that can lag intraday price moves by tens of minutes. The Trend Agent looks at 5m, 15m, and 60m chart structure and updates every eligible candle. When they disagree, the framework checks the freshness of each read and weights the live one higher when the gate is otherwise clearing. A stale macro tag does not block a trade. A live cross-asset veto does.
What is a Post-ISM Second-Chance setup, and why does it work after a data print?
A second-chance setup forms when a data release breaks the opening range and the broken level then holds as resistance during the retrace. The pattern exists because the initial reaction often overextends, and the second test of the broken level is where institutional flow either re-engages with the trend or reclaims the range. If the retrace stalls at the 50 to 62 percent zone and the 5m candle prints a rejection, the path of least resistance is continuation. The pattern requires cross-asset confirmation to score.
When does SkyAnalyst grade a setup as C+ instead of A or B?
Setup grades reflect how cleanly the legs of the pattern lined up at the moment of evaluation. A grades require every confluence to score with margin, a fully aligned cross-asset regime, and no stale agent reads. B grades allow one minor degradation. C+ grades show two minor degradations and at least five of seven confluences scoring. The grade is takeable but conservative. C+ trades win at a lower rate than A trades, and we publish the grade so readers can size their own confidence against the system's own.
How does the realized R differ from the hero R reported on a winning trade?
The hero R is the path the market actually traveled before the move exhausted or invalidated. If the market reaches TP3, the hero R equals TP3's R. The realized R is the R the broker actually booked on the position, and because the broker closes the full position at TP1, the realized R on every winning trade equals TP1's R. Both numbers are honest. The hero R describes the move. The realized R describes the ledger.
Run your markets with SkyAnalyst
Seven-day free trial. No credit card. Full access to the Trend Agent, Macro Agent, and six-factor confluence scoring.
Trading involves substantial risk of loss. Past performance is not indicative of future results. The analysis shown was produced by an AI model operating on SkyAnalyst’s live trading infrastructure; it is shared for educational and research purposes only and is not financial advice. About reported results. Every AI Trader publishes three take-profit targets (TP1, TP2, TP3) per trade. The broker closes 100% of the position at TP1, so two distinct R-multiples appear in this article. The hero R-multiple is the full-potential R: where the market actually traveled (the highest take-profit hit, or the stop loss) before the setup was invalidated or exhausted. The realized R, shown on the TP1 row of the simulated returns panel, is TP1’s R (or -1R on a stop out). The realized R is what we log to our running track record. Both numbers are honest. Showing both is what lets readers see the full arc of the move and the conservative ledger entry it produced. Simulated returns in this article are calculated against a hypothetical $100,000 account at 2% risk per trade (1R = $2,000). These are educational reference figures and do not reflect any specific account or broker execution. Your actual result depends on your position size, your risk parameters, and live market conditions.
From the SkyAnalyst Journal
Three Stops, One Tape: A 5.77% Drawdown on a Thin Week
Three stops, three R given back, a peak-to-trough drawdown of 5.77 percent, and a longest losing streak of three across two trading sessions. What each stop taught us, and what the curve says about a week the runner never showed up.
May 25-31, 2026: A Three-and-Three Week Closed -1.01R Net
Six trades, three winners, three losers, -1.01R net at the TP1 baseline. Tuesday opened with a 3-for-3 morning printing +1.99R inside thirty minutes; Wednesday and Thursday gave it back through two US30 stops and an EURUSD stop.
May 2026: Forty-Seven Trades, +8.33R Net, the Month Claude Pulled Ahead
Forty-seven trades, twenty-seven winners, twenty losers, +8.33R net at the TP1 baseline. Claude ran +6.89R on 26 trades at 61.5 percent; GPT printed +1.45R on 21 trades at 52.4 percent. The two-model lineup stabilized this month.