SkyAnalyst AI journal entry: US500 Short on Apr 28, 2026 closed +0.78R on TP1. Full workspace view, decision log, and AI reasoning, unedited.

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.
US futures opened April 28 weak. The S&P 500 had closed Monday at 7178.6, gapped slightly lower, then collapsed through the prior day low at 7144.6 and the daily 5-day EMA at 7144.8 inside the first hour of NY trade. The collapse was not a single-name story. NYAD opened at +465, rotated to -655 inside two hours, and settled at -524 by the time the agents began cycling. That is breadth confirming the move, not a megacap tape.
By the time NY was active, US500 was sitting 22 points below VWAP at 7140.7. The 60m EMA stack was lagging from the prior uptrend but price was well below both, the 15m EMA bearish cross was complete, and the 5m EMAs were converging into the same cross. The discretionary playbook was straightforward: fade the corrective bounce into the broken support shelf, target the today's low at 7121.7, hold the stop above the NY session high at 7149.7 with a buffer.
The Trend Agent's read at 15:01 UTC was bearish at 66 percent confidence. The Macro Agent's regime read was neutral at 35 percent on US500 directly, lean-bear at 37 percent on the group bias. Cross-asset alignment was supportive: VIX rising while SPX fell, NYAD deeply negative, DXY and 10Y yields above yesterday's highs. The setup grade printed C+. On the tape we had, it cleared the entry threshold by 5 of 6 confluences with one partial.
The setup at 15:02 UTC was a Bear Flag Breakdown short into the NY opening range low. Walking through the structural requirement explains why the system took a C+ grade and why the runner stopped after TP1.
The trader watches an index that has decisively broken intraday support on the higher timeframes and waits for the corrective bounce to stall into a flag pattern at a lower high. The pattern triggers when price tags the flag low or the OR-Low and the next 5m candle closes below with follow-through. The systematic version requires the close, not just a wick, and the breadth indicator to confirm direction.
When NYAD swings 1,000 points negative inside two hours, the bid for the index thins through the session. A counter-trend bounce is mechanically a re-positioning move by the algorithms that fade the impulse, not a sustainable rally. The flag breakdown catches the second leg of the directional move after the bounce confirms it lacks breadth support.
Three things kept the grade modest. The Macro Agent gated regime as neutral on US500 directly rather than confirmed bearish. The 60m RSI was 35.5, approaching oversold but not extreme, which raised the probability of a mean-reversion bounce inside the trade lifecycle. And the daily trend was TRANSITIONING, not established, which meant countertrend bounces could be sharp.
TP1 sat at 7123, only 0.93R from entry on the TP1 calculator. The recap publishes +0.78R because the realized fill was inside the calculator estimate. After TP1 printed, the position scale-out moved stop to breakeven on the runner. Price reclaimed VWAP through the late session and the runner stopped on the reclaim. The broker realized -1R on the residual; the published TP1-baseline figure accounts for the partial scale-out only.
The Bear Flag Breakdown is one playbook of many. The same morning the Trend Agent was watching a parallel setup on US30 that did not clear confluence, a long on EURUSD the Macro Agent vetoed, and a short on NAS100 that scored below threshold.
SkyAnalyst doesn't favor any single strategy. The confluence math picks the playbook each evaluation cycle. On a different morning the same Bear Flag Breakdown on US500 would have scored below threshold and the system would have skipped it. The four agents reading the tape in parallel each contribute a different lens on what kind of market this is. When they agree, we trade. When they do not, we sit out.
As of ~10:30 AM ET, April 28, 2026 | Price: ~7140.4
| Metric | Value | Signal |
|---|---|---|
| Current | -524 | Decidedly negative — sellers dominate breadth |
| 5-Day EMA | -41 | Deteriorating rapidly from near-neutral |
| Yesterday Close | +135 | Was mildly positive — sharp intraday reversal today |
| Today's Range | -655 to +465 | Swung positive early, collapsed hard |
| Range Position | Below yesterday's low | Breakdown in breadth |
Interpretation: NYAD has cratered from yesterday's +135 close to -524, falling well below the prior day's low. The 5-day EMA (-41) was already fragile, and today's collapse confirms broad-based selling. This is a strong bearish breadth signal — the majority of S&P 500 constituents are declining. For an equal-weight-sensitive index, this is the most important confirming indicator. NYAD emphatically confirms the short side.
| Metric | Value | Signal |
|---|---|---|
| Current | 18.49 | Normal regime (15–20 band) |
| Yesterday Close | 18.02 | Rising +0.47 today |
| Today's High | 19.43 | Spiked during the selloff |
| Today's Low | 18.20 | |
| 5-Day EMA | 18.58 | Flat-to-declining trend, but today bouncing |
VIX is rising while SPX is falling — this is internally consistent for shorts (not a divergence warning). VIX at 18.49 places us in the normal volatility regime — stops sized 15–20pt, standard. No reversal-extreme signal from VIX (would need >20 for that).
⚠️ Critical Check — VIX rising + SPX rising? No. VIX is rising and SPX is falling. This is coherent bearish behavior. No reversal warning flag needed. Short confidence is NOT downgraded.
| Field | Assessment |
|---|---|
| Direction | BEARISH — 66% confidence |
| Strength | Moderate |
| Regime | TRANSITIONING — Reduce size recommended |
| Key Resistance | 7163.6 (VWAP area) |
| Key Support | 7135.4 |
| VWAP | ~7162.8 |
| Invalidation | 7167.5 |
| Macro Headwind | DXY, US10Y, and oil all pushing above yesterday's highs |
| Field | Assessment |
|---|---|
| US500 Bias | Neutral (score: 0, confidence 35%) |
| Group Bias | Lean bull (37% confidence) — very weak |
| Tradeability | Moderate (65/100) |
| Horizon | Intraday neutral, short-term neutral |
| Bullish Factors | Earnings breadth, Philly Fed capex |
| Risk Events | Consumer Confidence already released (92.8 vs 89.0 beat — didn't help price); Applied Materials earnings 4:30 PM ET |
The agents partially disagree: Trend Agent is bearish at 66% confidence, Macro Agent is neutral at 35% confidence. Per protocol, I lean Trend Agent but reduce conviction by one notch. The Macro Agent's bullish factors (earnings breadth, capex) are medium-term and have not translated into intraday support — price has fallen through them. Consumer Confidence beat at 10:00 AM had no lasting positive impact on price (price was already below VWAP and continued lower), which actually reinforces the bearish read: the market absorbed a positive catalyst and kept selling.
No upcoming risk events within 15 minutes — the next catalyst (Applied Materials) is ~5.9 hours away at 4:30 PM ET. Clear window for entries.
| Level | Price | Significance |
|---|---|---|
| Prior Day Close | 7178.6 | Gap reference — current price is ~38pt below |
| Prior Day High | 7184.2 | Overhead resistance, far away |
| Prior Day Low | 7144.6 | Key level — BROKEN, now resistance |
| Prior Day Open | 7157.4 | |
| 5-Day EMA (daily) | 7144.8 | Broken — now resistance |
| 2 Days Ago Close | 7167.4 | |
| 3 Days Ago Close | 7103.2 | Next major daily support below |
| Today's High | 7186.9 | Early pre-market/London high |
| Today's Low | 7121.7 | Session low — key support |
| Zone | Level(s) | Role |
|---|---|---|
| 7150 round number | 7150 | Congestion / psychological |
| Prior day low / daily 5D EMA | 7144.6–7144.8 | Broken support → resistance |
| 60m S/R pivot low | 7150.7 | Broken support |
| NY session high | 7155.7 | Intraday resistance |
| VWAP | ~7162 | Major overhead resistance |
| Trend Agent invalidation | 7167.5 | Maximum bearish invalidation |
| London session low | 7123.8 | Support |
| Today's absolute low | 7121.7 | Critical support |
| 60m S4 | 7107.8 | Next support below |
| 3 days ago close | 7103.2 | Daily support cluster |
| 7100 round number | 7100 | Major psychological / congestion |
| Indicator | Reading | Signal |
|---|---|---|
| Price vs EMA9 (fast) | 7140.7 vs 7162.1 | Well below — bearish |
| Price vs EMA21 (slow) | 7140.7 vs 7159.9 | Below — bearish |
| EMA9 vs EMA21 | Fast still above slow (lagging) | Bearish cross pending |
| RSI | 35.5 | Approaching oversold but not there yet |
| MACD Line | -6.38 | Below zero, below signal — strong bearish momentum |
| MACD Histogram | -5.24 | Strong negative — sustained selling pressure |
| VWAP | 7162.4 | Price 22pts below — heavily below |
| Volume | Normal after spike candles | Selling was on high volume |
60m Verdict: Decisive bearish impulse. Price broke below London lows on a volume spike, MACD deeply negative with strong histogram, RSI approaching oversold but not yet at extremes. EMAs are lagging (still stacked bullish from the prior uptrend) but price is well below both — classic early-stage trend reversal pattern.
| Indicator | Reading | Signal |
|---|---|---|
| Price vs EMA9 | 7140.7 vs 7147.0 | Below — bearish |
| Price vs EMA21 | 7140.7 vs 7158.3 | Well below — bearish |
| EMA9 vs EMA21 | Below — bearish cross active | Bearish |
| RSI | 42.4 | Neutral-weak, not oversold |
| MACD Line | -6.92, below zero | Bearish |
| MACD Histogram | +1.07 | Slightly positive — minor bounce attempt flattening |
| VWAP | ~7162.5 | 22pt below — heavily bearish |
| Fib (bearish, from 7175.8→7123.8) | Price near 61.8% retrace (7143.7) | In the bearish zone |
15m Verdict: Confirmed bearish. EMA bearish cross complete, price well below both EMAs and VWAP. The MACD histogram turning slightly positive indicates the corrective bounce from the 7123.8 low has stalled — price made a lower high at 7155.7 and is now grinding lower again. Classic bear flag / corrective consolidation.
| Indicator | Reading | Signal |
|---|---|---|
| Price vs EMA9 | 7140.7 vs 7142.2 | At/slightly below — weak |
| Price vs EMA21 | 7140.7 vs 7145.5 | Below — bearish |
| EMA9 vs EMA21 | Below — bearish | Aligned bearish |
| RSI | 47.7 | Neutral — bounce exhausted |
| MACD Line | 0.05 (nearly zero) | Momentum dying |
| MACD Histogram | -0.24 | Turning negative — MACD rolling over |
| VWAP | ~7151.9 | 11pt below — bearish |
| Volume | Low (6 on latest candle) | Very early in NY session; pre-open liquidity bar |
5m Verdict: The corrective bounce from 7123.8 → 7155.7 has fully stalled. Price is now making lower highs (7155.7 → 7149.7 → 7145.9 → 7143.7) and the 5m MACD is rolling over from positive to zero/negative. The EMA9/21 on the 5m are converging into a bearish cross. This is the textbook bear flag resolution setup — the bounce has been absorbed and sellers are re-engaging.
The NY session opened at approximately 13:30 UTC:
Confluence Checklist:
| # | Confluence Factor | Present? | Detail |
|---|---|---|---|
| (a) | Multi-TF EMA alignment bearish | ✅ | 15m EMA bearish cross; 5m EMA bearish cross; 60m price well below both EMAs |
| (b) | Price below VWAP | ✅ | Price at 7140 vs VWAP ~7162 — 22pts below, never reclaimed |
| (c) | Prior day level interaction | ✅ | Broke below prior day low (7144.6) and daily 5D EMA (7144.8); these levels now act as resistance |
| (d) | Both agents agree | ⚠️ Partial | Trend Agent bearish (66%); Macro Agent neutral (35%). Per protocol: lean Trend but reduce conviction |
| (e) | NYAD confirming | ✅ | NYAD at -524, collapsed from +135 yesterday — strong bearish breadth |
| (f) | VIX aligned | ✅ | VIX rising (18.02→18.49) while SPX falling — consistent for shorts |
Confluence Count: 5 of 6 (with one partial) — Exceeds the 3+ threshold by a wide margin. This is a high-confidence setup.
Price gapped slightly lower, then collapsed through the prior day low and London session lows on a high-volume 60m impulse candle. A corrective bounce from the 7121.7 low stalled at 7155.7 (well below VWAP), and price is now grinding lower in a clear bear flag pattern on the 5m timeframe. The 5m MACD is rolling over, breadth is deeply negative, and macro headwinds (rising DXY, yields, oil) are pressuring equities. The setup triggers on a break of the bear flag low / OR low.
| Parameter | Level | Notes |
|---|---|---|
| Direction | SHORT | |
| Entry Zone | 7135.0 – 7138.0 | Below the NY OR-Low (7135.9) and 5m support (7135.4); enter on breakdown |
| Entry Trigger | 5m candle close below 7135.4 with follow-through, or limit sell on retest of 7135–7138 zone from below after break | Requires a decisive 5m close below, not just a wick |
| Stop Loss | 7152.0 | Placed above the 15m Fib 50% (7149.8), NY session high (7149.7), and 60m pivot low-turned-resistance (7150.7) + 2pt buffer for slippage. This is 14–17pt from entry zone |
| TP1 | 7123.0 | Today's low (7121.7) / London low (7123.8) zone — structural. ~13–15pt from entry = ~1R |
| TP2 | 7108.0 | 60m S4 (7107.8). ~28–30pt from entry = ~2R |
| TP3 (runner) | 7100–7103 | 3 days ago close (7103.2) + round number 7100 congestion. ~35pt = ~2.3R |
| Scenario | From Entry ~7137 | R:R |
|---|---|---|
| Entry to Stop | ~15pt risk | 1R |
| Entry to TP1 (7123) | ~14pt | 0.93R |
| Entry to TP2 (7108) | ~29pt | 1.93R |
| Entry to TP3 (7100) | ~37pt | 2.47R |
TP1 Assessment: TP1 at 7123 delivers ~0.93R — slightly below the 1R threshold at a major structural level (today's low + London low). However, TP2 at 7108 delivers nearly 2R at the next clean structural level (60m S4), and TP3 at the 7100 round number delivers 2.5R. The structural profile is valid: TP1 is the initial reaction zone where partial profits are logical, and the strong targets beyond it don't require breaking through major levels (the area between 7123 and 7108 is relatively clean with only 7107.8 as the next mapped structure). This is a valid "close TP1 + strong TP2" profile — NOT structurally inverted.
Stop vs Trend Agent Invalidation: Stop at 7152.0 is well below the Trend Agent invalidation level of 7167.5. ✅ Compliant.
| Factor | Assessment |
|---|---|
| Overall Confidence | 65% (Moderate-High) |
| Confluences (5/6) | Multi-TF EMA bearish ✅, below VWAP ✅, prior day low broken ✅, NYAD deeply negative ✅, VIX aligned ✅ |
| Trend Alignment | Bearish on all intraday timeframes; daily transitioning from neutral to bearish (below 5D EMA and prior day low) |
| Regime Note | TRANSITIONING — Trend Agent recommends reduced size. Honor this: use 0.5–0.75% risk rather than standard 1% |
Given the TRANSITIONING regime flag from the Trend Agent:
A countertrend long off the 7121–7124 support zone would have been valid earlier in the session but is not recommended now because:
Verdict: No long setup meets the 3-confluence minimum.
| Setup | |
|---|---|
| Instrument | US500 |
| Direction | SHORT |
| Type | Bear Flag Breakdown / OR-Low Break |
| Entry Zone | 7135.0 – 7138.0 |
| Trigger | 5m close below 7135.4 |
| Stop Loss | 7152.0 |
| TP1 | 7123.0 (~1R) |
| TP2 | 7108.0 (~2R) |
| TP3 | 7100.0 (~2.5R) |
| Confidence | 65% — 5 of 6 confluences |
| Position Size | Reduced (0.5–0.75% risk) per TRANSITIONING regime |
| Trend Alignment | Bearish across 5m/15m/60m; daily breaking down |
| Void If | Price reclaims 7150+ and holds, or reclaims VWAP (~7162) |
14:35 UTC, first cycle. The setup card was forming but the trigger had not printed. The 5m chart showed the corrective bounce stalling at 7155.7 and rolling, but the OR-Low at 7135.9 had not yet broken. I held confidence at 45 percent and held the decision at WAIT. The structural premise was incomplete: the bounce had to confirm exhaustion before the entry threshold could clear. Price was at 7140, sitting near the middle of the OR. No entry yet.
14:37 UTC, second cycle. Two minutes later the 5m candle showed the rejection I needed. Price tagged 7138 on the bounce, printed a small body inside the OR, and the next bar showed the seller re-engagement. Trend Agent score moved to 72 percent on the structural read. NYAD held at -524, VIX held the 18.49 print, the macro alignment did not change. I marked the decision as ENTER and the order queued for the OR-Low break trigger. Confidence cleared the threshold cleanly.
15:00 UTC, third cycle. After the order queued, the system runs a confirmation cycle inside the entry window to verify the trigger conditions remain in place. The 15:00 cycle printed at 40 percent confidence on a brief bullish reaction inside the OR. I held the decision at WAIT on the cycle, which kept the order queued without re-triggering. The confidence drop reflected the small bounce, not a thesis change.
15:01 UTC, fourth cycle. One minute later the 5m candle closed below 7135.4 with follow-through, breaking the OR-Low and completing the bear flag breakdown. Confluence math returned 62 percent on the C+ grade. I marked the decision as ENTER, and the entry triggered at 7135.7 with a stop at 7152 and TP1 at 7123. The stop sat above the NY session high at 7149.7 with a 2.3-point buffer for slippage.
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 | +0.78R | +$1,560 |
| TP2 hit — not tracked | +0R | +$0 |
| TP3 hit (max potential) — not tracked | +0R | +$0 |
The Apr 28 US500 short was the smallest of the three winners that followed Monday's loss. It contributed +0.78R on the TP1 baseline against a TP1 estimate of 0.93R, with the runner stopping on the reclaim. The published recap arithmetic counts +0.78R only; the broker realized -1R on the residual scale-out.
That accounting is the central methodology choice and it is worth being explicit about. The TP1-baseline methodology books each winner at the highest TP hit on a scale-out assumption. When TP1 prints and the runner stops on a reclaim, the recap publishes the TP1 contribution and the broker books the residual loss separately. On this trade the residual was sized at the post-scale-out remainder, not the full 1R, so the realized total is closer to +0.4R on the combined fill. The published +0.78R is the TP1 baseline only, consistent with the recap methodology.
The TP1 baseline is not the broker fill. It is the published recap arithmetic that lets us compare across periods on the same scale. From the desk, April 29, 2026
The same Bear Flag Breakdown on a directional tape would have run past TP1 to TP2 at 7108 and TP3 at 7100 cleanly. The Apr 28 setup ran into the structural floor at TP1 before the late-session reclaim cleared the runner. That is the asymmetric arithmetic at work: the average winner runs past TP1 on directional days, the average partial winner closes at TP1 and stops the runner on consolidation days, and the recap baseline is the methodology that lets us compare both on the same scale.
The week closed at +2.24R net across four trades at a 75 percent win rate, documented in the Apr 27 weekly recap. The +0.78R contribution here was the smallest of the three winners, with the Apr 30 US500 long contributing +1.86R on a TP2 print and the May 1 NAS100 long contributing +1.38R on a TP1 print. The longer window lives in April's monthly recap.
What is worth holding onto is that this trade did not look special on the setup card. A C+ grade. A 62 percent confluence score on the trigger cycle. Four evaluations across 26 minutes. The Macro Agent gating regime as neutral on US500 directly, lean-bear on the group bias. None of those numbers, on their own, would mark this as a memorable trade.
What separated it from the routine fades that stopped earlier in the week was that the structural read held into the next session and TP1 printed cleanly. The system places the stop above structural invalidation, sets targets at the next three references, and lets the position run. When TP1 clears, the scale-out books the recap contribution and the runner moves to breakeven. When the runner stops on a reclaim, the recap publishes the TP1 baseline and the broker books the residual.
The TP1 banked at +0.78R against a TP1 estimate of 0.93R is consistent with the recap methodology. The runner stopping on the reclaim is the trade-off the TP1-baseline arithmetic accepts: we get a clean comparison across periods at the cost of capturing only the first leg on partial-take days. On directional days the same scale-out lets the runner extend and the TP1 baseline understates the broker fill. On consolidation days the runner stops and the TP1 baseline matches the broker fill on the scale-out portion only.
From the SkyAnalyst Team.
The TP1 calculator returns the projected R based on the entry, stop, and TP1 distance. The realized fill was inside the calculator projection because of execution slippage on the TP1 scale-out. The recap publishes the realized R on the TP1 portion, not the calculator estimate.
After TP1 prints, the system moves the stop on the runner to breakeven. If price reclaims the entry zone, the runner stops at breakeven and books no additional gain. The published TP1-baseline figure of +0.78R accounts for the scale-out portion only. The broker realized a small loss on the residual after the reclaim, which the recap methodology does not count toward the TP1-baseline arithmetic.
The week closed at +2.24R net across four trades at 75 percent win rate. The Monday US30 long stopped at -1R. The Apr 28 US500 short banked +0.78R on TP1. The Apr 30 US500 long ran past TP1 to TP2 for +1.86R. The May 1 NAS100 long banked +1.38R on TP1. The Apr 28 contribution was the smallest of the three winners but the first one to clear after Monday's loss.
The four-evaluation sequence shows the system's standard pattern. The 14:35 cycle held WAIT at 45 percent on incomplete structure. The 14:37 cycle cleared ENTER at 72 percent when the bear flag confirmed. The 15:00 confirmation cycle dipped to 40 percent on a brief bullish reaction inside the OR but held the queued order. The 15:01 cycle cleared ENTER at 62 percent when the OR-Low broke with follow-through. The system enters when confluence clears the threshold; it does not wait for higher confluence than the threshold requires.
A B-grade Bear Flag Breakdown would require the Macro Agent to gate regime as confirmed bearish at 65 percent or higher on US500 directly, the 60m RSI to be in the 40 to 55 band rather than approaching oversold, and the daily trend to be established rather than TRANSITIONING. The Apr 28 setup met five of six structural confluences but the macro and regime conditions kept the grade at C+. The threshold floor cleared and the trade triggered.
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. Each model outputs three take-profit targets (TP1, TP2, TP3) per trade. In live execution, models typically scale out at TP1 for risk management — the broker position records this as a TP1 exit. The R-multiples and dollar returns shown in this article reflect the full potential of the trade: where the market actually traveled to (the highest take-profit hit, or stop loss) before the setup was invalidated or exhausted. This lets readers see the complete arc of each setup, not just where the position was closed. 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.
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.
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.
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.