US500 Short on March 25: The First TP3 of a Seven-Trade Streak
SkyAnalyst AI journal entry: US500 Short on Mar 25, 2026 closed +3.23R on TP3. Full workspace view, decision log, and AI reasoning, unedited.
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 morning the rally ran out of breadth
US500 opened March 25 with the auction trying one more push. Price probed 6636.8, marginally above the prior-session high of 6630.8, and was rejected back inside the prior range. The reversal was a sequence: the move into 6636.8, a 5-minute close back under VWAP at 6611, a slide into 6595 by the time the workspace opened. The Macro Agent had written `regime = transitioning, lean = bear` to shared state at 13:55 UTC, with confidence flagged low at 15 percent.
The breadth read confirmed the rejection was structural. NYAD printed +447 against yesterday's +1653 close, below the prior session's low and inside a five-day range from -1920 to +1653. The Trend Agent flagged 60-minute as transitioning: price below the fast EMA at 6599, above the slow EMA at 6589, RSI rolled to 48, MACD negative. The 15-minute had committed: bearish EMA cross, MACD below zero, support cluster at 6589.8 and 6587.7.
VIX was the gate. At 26.09 with a 5-day EMA of 26.02, the volatility regime was elevated, the kind of tape where reversal trades work and trend-continuation trades get trapped. The Risk Agent widened the stop to 15.5 points, above the Trend Agent's invalidation at 6618.9. Setup grade printed C+: structural read clean, macro lean correct but low confidence, every floor cleared and nothing more.
The setup at 14:04 UTC was a Short VWAP / Prior Close Rejection. Walking the structural requirement explains why the system held seven waits before firing the eighth.
What the pattern is
Price probes above prior-session resistance, fails, and rotates back under VWAP. The pattern triggers when a counter-trend bounce returns to the VWAP zone, prints a rejection candle, and the next bar fails to retake. The systematic version requires the rejection to close, not just wick.
How professionals use it
A first failure at prior-session resistance holds roughly forty percent of the time. The second test, after the rotation back under VWAP, holds closer to sixty-five. The tell is volume on the rejection: a quiet retest means thin participation, a loud rejection on a 5-minute close means real offers are stepping in.
Why this works on a high-VIX tape
Elevated volatility punishes trend-continuation trades because each push gets faded, and rewards reversal trades because each extension finds liquidity. The VWAP zone is the magnet of the day's auction. When breadth has weakened and price has failed at session highs, the bounce back to VWAP is mechanically a re-positioning move.
Why this graded C+ rather than B
The Macro Agent's confidence was 15 percent, low enough to read as "lean correct, conviction not high." The 60-minute had not committed to a bearish regime. Cross-Asset returned neutral rather than supportive. C+ is the system's notation for tradeable, every floor clears, conviction not high enough for B.
How the agents reach quorum
Four agents read the same shared state. Macro gates regime. Trend scores structure. Cross-Asset checks correlated markets. Risk sizes against the regime. Each writes structured fields the others read on every cycle. Coordination produces the entry.
How the system reads this, dynamically not dogmatically
The Short VWAP / Prior Close Rejection is one playbook of many. The same morning the Trend Agent scored a parallel breakdown on NAS100 (which ran to TP3) and a fade-to-VWAP on USDJPY (also TP3). SkyAnalyst doesn't favor any single strategy. The confluence math picks the playbook each cycle. On a different morning the same VWAP rejection would have scored below threshold and been skipped. The dynamism is the product, and the math reads the tape first.
US500 NY AM Session Analysis
Current context: US500 ~6595 at/just after NY open, after failing 6636.8 and rotating back under VWAP 6611.
Bottom line: Bearish intraday bias, but this is a high-volatility transition regime. Best setup is sell failed rallies, not chase lows into support.
1) Breadth & Volatility Regime
NYAD / Breadth
From market context:
- Current NYAD: +447
- Yesterday close: +1653
- 2 days ago: -1920
- Today range: -1629 to +614
- Range position: below yesterday’s low
Read
- Breadth is weaker than yesterday and highly unstable.
- The 5-day trend cannot be measured perfectly because one value is missing, but the sequence suggests deteriorating / inconsistent breadth, not broad risk-on expansion.
- NYAD being only +447 after yesterday’s +1653 while price has tested higher and failed is not strong confirmation for sustained upside.
- This is neutral-to-bearish breadth confirmation, especially after the rejection from session highs.
VIX Regime
- VIX: 26.09
- 5-day EMA: 26.02
- Today high: 26.47 / low: 24.82
- VIX is elevated >20, which means:
- Favor reversals at extremes
- Use wider stops: 20–25 pts
- Avoid assuming smooth continuation
VIX/SPX interaction
- SPX rose earlier into 6636.8 while VIX remained elevated and rebounded intraday.
- That is a reversal warning for longs.
- Long confidence must be downgraded.
Regime conclusion:
- Breadth: weak/unstable, not confirming aggressive longs
- Volatility: high fear regime
- Implication: prefer fade failed rallies and be cautious pressing shorts into major support
2) Agent Synthesis
Macro Agent
- Bias: lean_bear
- Confidence: 15% only
- Tradeability: moderate
- Drivers:
- Soft services PMI
- Breadth deterioration masked by mega-cap concentration
- Risk events:
- No high-impact USD event within the next 15 minutes
- Trump speech later, unemployment claims tomorrow
Macro take
Macro is not strongly directional, but it does lean bearish and matches the broader “fragile risk appetite” story.
Trend Agent
- Direction: BEARISH
- Confidence: 63%
- Regime: TRANSITIONING
- Recommendation: REDUCE_SIZE
- Key levels:
- VWAP: 6611.1
- Resistance: 6611
- Support: 6593
- Invalidation: 6618.9
Synthesis
- Agents do not fully agree in confidence, but they do agree directionally bearish
- Since Trend Agent is clearer and more actionable, we lean Trend Agent
- Conviction is moderate, not max, because:
- Macro confidence is low
- 60m is still transitioning, not fully bearish
- VIX > 20 increases reversal risk
Signal quality status:
Best-quality signals are short-biased, especially on VWAP rejection / failed retest.
3) Gap & Daily Structure
Daily reference
- Current: 6595
- Prior close: 6599.6
- Prior high: 6630.8
- Prior low: 6525.7
- 5-day EMA: 6586.68
Gap analysis
- Open was effectively near prior close; gap is well under 0.3%
- Small gaps often fill / mean revert
- Today’s auction pushed above prior high marginally to 6636.8, then failed back below
Key daily/intraday structure
- Prior close: 6599.6
- Prior high zone: 6630.8–6636.8
- Pivot / support zone: 6588.4–6593
- 5d EMA: 6586.7
- Next major downside support: 6525.7
- Round numbers:
- 6600 = congestion / magnet
- 6650 = upper congestion if squeeze
- 6550 = lower magnet if breakdown extends
Structure conclusion
- Market is back below prior close and testing the 6593 / 6588 support pocket
- This area is important:
- Hold + reclaim = possible mean reversion bounce
- Break + accept below = opens room lower
4) Multi-Timeframe Technicals
60-Minute Bias
Using the latest 5 hourly candles:
- Price recently failed at 6636.8
- Latest 60m:
- price below fast EMA (~6599)
- still slightly above slow EMA (~6589)
- RSI ~48 neutral-weak
- MACD histogram negative
- price below VWAP
- This is transitioning from bullish to bearish, not a fully established higher-timeframe trend breakdown yet.
60m takeaway
- Bias is bearish rotation
- But because slow EMA/support are just below, chasing shorts into 6588/6587 is lower quality
15-Minute Confirmation
Latest 15m:
- price below EMA fast
- still above slow EMA, but momentum has rolled over
- RSI ~39.5
- MACD below zero and bearish
- price below VWAP
- support cluster: 6589.8 / 6587.7
- session impulse from 6636.8 -> 6591
15m takeaway
- Confirms bearish intraday momentum
- But current price is already near first support pocket
5-Minute Execution
Latest 10x 5m candles show:
- Failed push into 6636.8
- Sharp downside impulse
- Bearish EMA turn on 5m
- MACD strongly negative
- Price below VWAP 6612.8
- Current price pressing 6593 / 6587.3
- Opening range high effectively became 6636.8
- Opening range low now around 6587.3
5m takeaway
Best precision entries are:
- Short on retest failure of 6600–6611/VWAP
- Short on breakdown only if 6587 fails and retest rejects
Not ideal:
- fresh short directly into 6593/6587 support without a retest
5) Setup Filtering
Required: at least 3 of 6 confluences
Confluence checklist:
- (a) multi-timeframe EMA alignment
- (b) price on correct side of VWAP
- (c) prior day level / daily S/R interaction
- (d) both agents agree
- (e) NYAD confirms
- (f) VIX aligned
We have for shorts:
- (a) partial-to-good: 5m and 15m bearish, 60m transitioning bearish
- (b) yes: price below VWAP
- (c) yes: interacting with prior close / intraday support / failed prior high
- (d) yes: both agents lean bearish
- (e) yes enough: NYAD weak vs yesterday, unstable breadth
- (f) mixed but acceptable: VIX elevated and fear regime supports downside/reversal behavior
So short setups qualify.
For longs:
- Longs lack clean breadth and VIX confirmation.
- Any long would be counter-regime scalp only, not high-probability by your filter.
High-Probability Setups
Setup 1 — Short VWAP / Prior Close Rejection
Preferred setup
| Item | Level |
|---|---|
| Direction | Short |
| Entry zone | 6602 to 6611 |
| Entry trigger | 5m rally into prior close / VWAP zone, then bearish rejection candle or 5m close back below 6603/6600 after testing 6608–6611 |
| Stop loss | 6620.5 |
| TP1 | 6590–6588 |
| TP2 | 6578–6575 |
| Stretch TP3 | 6555–6550 only if 6587 breaks and breadth worsens |
Why it qualifies
Confluences:
- 5m + 15m bearish, 60m transitioning bearish
- Below VWAP, with VWAP at ~6611 acting as resistance
- Prior close / intraday structure in the entry zone
- Macro + Trend agents agree bearish
- NYAD not confirming upside
- Elevated VIX supports reversal/fade behavior
Risk logic
- Stop is placed above Trend Agent invalidation 6618.9 with a small slippage buffer.
- That makes 6620.5 the practical automated stop.
- This is a wide stop regime, but still valid because entry is near resistance.
R:R
Example from 6607 entry:
- Risk: ~13.5 pts to 6620.5
- TP1 at 6590 = ~17 pts, about 1.25R
- TP2 at 6578 = ~29 pts, about 2.1R
Risks
- Because price is already near support, the market may bounce before giving the retest.
- If price tags 6611 and reclaims/holds above it, short thesis weakens quickly.
- If price moves straight down without retest, do not chase.
Trend alignment
- Aligned with Trend Agent bearish view
Confidence
- Moderate-High (7/10)
- Best available NY AM setup
Setup 2 — Short Breakdown Retest Below 6587
Only if support fails cleanly
| Item | Level |
|---|---|
| Direction | Short |
| Entry zone | 6585 to 6588 after breakdown |
| Entry trigger | 5m close below 6587.3, then a retest of 6587–6588 fails with rejection |
| Stop loss | 6601.5 |
| TP1 | 6570–6568 |
| TP2 | 6552–6550 |
Why it qualifies
Confluences:
- 5m/15m bearish momentum
- Price below VWAP
- Break of key support 6587.7 / 6587.3
- Agents aligned bearish
- Weak breadth backdrop
Risk logic
- Breakdown trades in high VIX can be trap-prone, so this needs:
- actual break
- retest failure
- Stop above reclaimed structure and round-number area.
- Keeps stop below the Trend Agent invalidation issue, since this is a different local structural setup and doesn’t require holding above 6618.9, but from a practical standpoint the setup should fail well before that if breakdown is real.
R:R
Example from 6586 entry:
- Risk to 6601.5 = 15.5 pts
- TP1 6569 = 17 pts, just over 1R
- TP2 6552 = 34 pts, over 2R
Risks
- Selling a breakdown after a fast impulse can get trapped if this becomes an opening flush reversal.
- VIX > 20 means breakdowns can reverse hard from extremes.
- If NYAD stabilizes and price reclaims 6593/6600, stand down.
Trend alignment
- Aligned with Trend Agent bearish view
Confidence
- Moderate (6/10)
- Valid only with a clean retest failure; otherwise no trade
What I Would Avoid
Avoid Longs Right Now
Reasons:
- VIX is too high for confident trend-continuation longs
- NYAD is not strong enough
- Price is below VWAP
- Trend Agent is bearish
- Earlier rally failed at/above prior high
A long would need:
- reclaim of 6600, then 6611/VWAP
- VIX softening intraday
- NYAD improving materially
- ideally a 5m/15m hold above VWAP
Until then, no high-probability long setup.
Trade Plan Summary
| Setup | Direction | Entry Zone | Trigger | Stop | TP1 | TP2 | Confidence |
|---|---|---|---|---|---|---|---|
| 1 | Short | 6602–6611 | 5m rejection at VWAP/prior close | 6620.5 | 6590–6588 | 6578–6575 | Moderate-High |
| 2 | Short | 6585–6588 | Break below 6587.3, then failed retest | 6601.5 | 6570–6568 | 6552–6550 | Moderate |
Final Call
Best NY AM bias: Sell rallies, not lows
- Primary setup: short a 6602–6611 retest failure
- Secondary setup: short a 6587 breakdown retest
- No high-probability long setup is available right now
Risk guidance
- Because this is a high-VIX fear regime, use reduced size
- Think ~0.5% to 1% equity risk per trade, not full normal size
- Do not widen stops beyond structure just to stay in
- If price action becomes too expanded and structural R:R drops below 1.5:1, it becomes No Trade
If you want, I can also convert this into a strict execution checklist for the next 30–60 minutes with exact “if/then” conditions for automation.
Decision log
First evaluation, 14:04 UTC, confidence 86 percent. Price tagged 6602 and printed a rejection wick. A wick is not a closed body. Confluence returned high on six of six factors, but the trigger bar was open. Declining.
Second evaluation, 14:06 UTC, confidence 84 percent. The rejection bar closed bearish, but volume sat below the 60-period average. The system requires confirmation volume on the trigger bar. Declining.
Third evaluation, 14:08 UTC, confidence 88 percent. Highest score of the cycle. A 5-minute closed under VWAP with RSI rolling 52 to 44 and MACD extending negative. The close came on an open candle that subsequently formed an upper wick filling the rejection. Declining.
Fourth evaluation, 14:09 UTC, confidence 86 percent. Follow-through bar opened lower, then probed back toward VWAP at 6611. The retest was happening but the response candle had not printed. Declining.
Fifth evaluation, 14:11 UTC, confidence 82 percent. Bounce firmed back toward 6608. A 5-minute closed marginally below the prior bar's high, but the structure of the rejection had softened. Declining.
Sixth evaluation, 14:12 UTC, confidence 82 percent. Building bar inside the entry zone, neither a clean lower high nor invalidation. Cross-Asset flagged DXY firming, supportive but not the trigger. Declining.
Seventh evaluation, 14:13 UTC, confidence 61 percent. Confidence dropped as the bounce probed 6610. Price tagged 6610.5 and printed an upper wick, but the threshold was 63 percent. Below the floor, no entry. Declining.
Eighth evaluation, 14:14 UTC, confidence 63 percent. The most recent 5-minute closed as a printed bearish reversal at the VWAP zone, volume above the 60-period average, RSI through 48, MACD decisively negative, next print opening lower with no upper wick. All four agents aligned on a C+ above every floor. Entering short at 6605, stop 6620.5, TP1 6590, TP2 6578, TP3 6555.
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 hit | +0.97R | +$1,940 |
| TP2 hit | +1.74R | +$3,480 |
| TP3 hit (max potential)Actual | +3.23R | +$6,460 |
All six agents combined.
What this trade teaches
The first TP3 of the week did not come from the highest-conviction setup. The Trend Agent scored 88 percent on the third evaluation and 86 percent on the first and fourth. The system entered at 63 percent on the eighth. The score crossed threshold when the mechanics aligned, not when the math peaked.
The score on an open bar can finish anywhere. The 88 percent print came from a structural read that subsequently filled with an upper wick. The 63 percent print came from a closed bar with confirmation volume, RSI rolled to 48, MACD decisively negative, and the next print already opening lower.
The score crossed threshold once at 88, twice at 86, then dropped to 61 before the system entered at 63. The number was never the trigger. The closed bar with confirmation was. - From the desk - March 26, 2026
The trade ran fifty points to TP3 over eighteen hours and fifteen minutes, spanning the overnight into the European open. Zero recorded drawdown. The next leg of the streak ran the same kind of fade on a different premise in the March 26 counter-trend squeeze fade.
From the desk
A discretionary trader watching this tape at 14:08 UTC, when score touched 88 percent, would have shorted the close. They would have been short at a 3-point worse price than the system entered six minutes later, then sat through the 14:09 to 14:12 retest into 6610.5 with their stop two ticks above and likely been picked off on the wick at the seventh evaluation.
The system entered at 63 percent because the threshold cleared when a closed bar printed with confirmation volume. The number was descriptive, not prescriptive. What changed across ten minutes was not the agents' read. It was the bar.
The shape we want is not perfect win rates. It is large winners on directional tapes the system does not predict, paired with small losers on tapes it filters out. The +3.23R on March 25 was the first leg of the bumper week.
From the SkyAnalyst Team.
At a Glance
What this teaches about AI-driven trading
How does the system decide that an 88 percent confidence score is not enough to enter?
Score is computed continuously on whatever the trigger bar is showing, and an open candle can finish anywhere. The 88 percent score on the third evaluation came from a 5-minute that subsequently formed an upper wick and partially filled the rejection. The system requires a closed trigger bar with confirmation volume and follow-through opening lower.
Why did the system widen the stop to 15.5 points instead of the calm-regime default?
The Risk Agent reads VIX on every evaluation. At 26.09, volatility was elevated, which means each push extends further before reversing. A calm-regime stop of 8 points would have been picked off by routine noise. The 15.5-point stop sat above the Trend Agent's invalidation at 6618.9, anchored to structure.
What does it mean that the trade ran 18 hours with zero recorded drawdown?
The position never traded above the entry price after the system entered. From 14:14 UTC on March 25 to 08:29 UTC on March 26, the broker's high-water-mark remained at 6605. TP1 at 6590 banked inside the first hour, TP2 at 6578 in the late NY session, TP3 at 6555 on the European open.
How does this single trade fit into the week and the published quarter?
This was the first of seven consecutive TP3 winners across March 25 through 27. The week of March 23 through 29 closed at 10W / 4L, 71.4 percent, +4.19R net, the best week of the published record. Month-to-date through March 25 sat at +4.41R across 37 trades. Quarter-to-date sat at -1.92R across 55 trades.
What would have invalidated the setup before the eighth evaluation?
A 5-minute close above 6611 with acceptance over 6618.9 would have cleared VWAP and the invalidation level, inverting the pattern. A NYAD print improving toward yesterday's +1653 close would have re-opened the long thesis. A VIX print collapsing under 24 with risk-on alignment would have pulled the Macro Agent off the lean-bear read.
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. 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.
From the SkyAnalyst Journal
March 2026 Monthly Recap: 42 Trades, Variance Both Directions, Break-Even Net
Forty-two trades. Twenty-two winners, twenty losers, 52.4 percent win rate. Net minus 0.13R, essentially flat on a TP1 baseline. The month produced both the deepest published drawdown and the bumper week of the record.
USDJPY Short on March 31 - Closing the Month on a TP3 Winner
A pullback short on USDJPY entered at 159.23 ran to TP3 at 158.75 in 2h 32m, closing at +3.20R. The closing-day winner of a March that finished -0.13R / 22W-20L on the TP1-baseline tally.
EURUSD Long on March 31 - Second of Two Same-Day TP3 Winners
A Bullish Pullback long on EURUSD entered at 1.1520 ran to TP3 at 1.1558 over four hours and seventeen minutes, closing at +1.58R. The second of two TP3 winners on the closing day of a near-flat March.