NAS100 Short: A Second Setup, a Disciplined Wait, and a Give-Back Exit
SkyAnalyst AI journal entry: NAS100 Short on May 12, 2026 closed +1.26R on TP2. 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 macro that set the table
The morning's tape was decided long before NAS100 printed its second setup. Core CPI cleared the consensus by a tenth of a point, ten-year yields broke decisively above the prior session's high and printed fresh five-day highs at 4.457%, VIX advanced above its five-day EMA, and the dollar followed both higher. The Macro Agent flagged the group as lean-bear at 62% conviction. Every cross-asset signal that mattered to the index was pointed in the same direction.
Inside that regime, NAS100 had already given its first answer of the session. The earlier trade, a short into the same resistance complex around the VWAP rejection, had run cleanly to TP3 inside an hour. By the time the second setup began to form near 15:35 UTC, the index was consolidating off the session low at 28,776, and the Trend Agent was watching the bounce for evidence of either a continuation rally or a corrective retrace into resistance.
The decision the system faced was not whether the trend was intact. The decision was whether the second entry was a confirmed rejection at the 28,840 to 28,870 zone, or a chase into a relief bounce off oversold conditions. That distinction is the entire content of the discipline section below.
We teach our subscribers a single framing for professional traders: a pullback to resistance zone short is a conditional trade, not an outright bearish bet. The pattern requires structure to test a defined level and then visibly fail at that level. The Trend Agent's job is to wait for the failure, not to anticipate it.
What the pattern requires
The full setup has four components. Price has to retrace from session lows back into a marked resistance zone. The reaction at that zone has to be a bearish candle, ideally an engulfing or pin bar. The 5m RSI has to roll back below 40 from the relief bounce. And the higher-timeframe structure (60m EMAs, MACD) has to remain bearish so the rejection has a thesis to lean on.
Why the second look is harder than the first
The first setup of the morning had cross-asset confirmation, a fresh VWAP rejection, and a clean rejection candle. The second setup had the same macro backdrop but ambiguous price action. The 5m RSI was sitting in the low thirties from sustained selling rather than rolling back down from a bounce.
Three consecutive 5m candles were printing higher lows toward the zone, which read as recovery, not retest. The structure was right. The trigger was not, until it was. That is the entire reason the system waited four evaluations rather than firing on the macro alone.
The dynamic, not dogmatic, product angle
SkyAnalyst does not favor any single strategy. The system reads the tape first, then asks which pattern, if any, applies to the regime in front of it. On 12 May, the regime selected was bearish-continuation on rates-driven selling. The pattern that fit was a pullback short, but the system would have rejected a long with equal precision had the tape pointed up.
We ship a tool that adapts to what the market is doing, not one that hunts for a favored setup. The trend is read dynamically, and the playbook follows it. The system doesn't favor bullish patterns over bearish ones, or breakouts over pullbacks. It reads structure, gates by macro, and waits for confluence.
NAS100 NY AM Session Analysis — May 12, 2026
Step 1: 10Y Yield Assessment (Primary Driver)
| Metric | Value |
|---|---|
| Current 10Y Yield | 4.451% |
| 5-Day EMA | 4.409% |
| Position | Above EMA by +4.2 bps |
| Yesterday High | 4.416% |
| Today's High | 4.457% |
| 5-Day High Status | Making NEW 5-day highs |
Verdict: Maximum bearish signal for NAS100. The 10Y yield is not just above its 5-day EMA — it has broken decisively above yesterday's high (4.416) and is printing fresh 5-day highs at 4.457. This is directly driven by today's hotter-than-expected CPI print (Core CPI 0.4% vs 0.3% forecast, CPI y/y 3.8% vs 3.7% forecast). Rising yields repricing rate-cut expectations are the single most powerful headwind for the Nasdaq 100. Default directional bias: BEARISH. Longs are blocked by this condition.
Step 2: Macro Regime & Cross-Asset Confirmation
Triple Headwind Check
| Cross-Asset | Current | 5d EMA | Position | Signal |
|---|---|---|---|---|
| US10Y | 4.451 | 4.409 | Above, new 5d high | 🔴 BEARISH |
| VIX | 19.02 | 18.13 | Above EMA, above yesterday's high (18.47) | 🔴 BEARISH |
| DXY | 98.43 | 98.09 | Above EMA, above yesterday's high (98.14) | 🔴 BEARISH |
All three headwinds confirm simultaneously — this is maximum cross-asset conviction for NAS100 downside.
Macro Agent Assessment
- Group Bias: lean_bear (62% confidence)
- NAS100 Bias: lean_bull (58%) — but this is stale (87-minute-old data) and contradicted by the live cross-asset picture
- Key Factors: Technical divergence (price below yesterday's low despite prior momentum), equipment/semi weakness as leading indicator for NVIDIA capex
- Catalyst: Core CPI beat already released at 8:30 AM ET — this is the driver of today's selloff
- Risk Event: Fed Chair Nomination Vote ~5:00 PM ET (5.4h away — not imminent for AM session), Core PPI tomorrow 8:30 AM
Note on Macro Agent NAS100 lean_bull score: The agent's NAS100-specific lean_bull (58%) appears to be lagging — the group bias is lean_bear (62%), and the live data shows price 500+ points below yesterday's close with all cross-assets confirming bearish. I weight the group bias and live cross-asset signals over the stale NAS100 sub-score. The macro factors cited (rate sensitivity, technical divergence) support shorts.
Verdict: Maximum cross-asset bearish conviction. Yields spiking + VIX elevated + DXY surging = the worst possible macro cocktail for NAS100.
Step 3: Trend Structure & Key Levels
Trend Agent Reading
| Field | Value |
|---|---|
| Direction | BEARISH |
| Confidence | 78% |
| Strength | MODERATE |
| Regime | TRENDING |
| Recommendation | REDUCE_SIZE |
| Invalidation | 29,217.3 |
| Key Resistance | 28,851.2 |
| Key Support | 28,796 |
| VWAP | 29,154.45 |
60-Minute Structure Analysis
EMA Stack: Price (28,800) is far below EMA fast (29,121) and EMA slow (29,106) — both EMAs remain above price. The fast EMA is still technically above slow (29,121 > 29,106) but this is a lagging residual from the prior range; price has blown through both and left them behind. This is a bearish displacement — price has accelerated away from moving averages, not a mean-reversion setup but a trend-continuation pattern.
RSI: 25.1 — oversold territory. This confirms extreme selling pressure but also warns that immediate continuation shorts at current levels carry mean-reversion risk.
MACD: Line at -70.28, histogram at -43.22 (medium strength, expanding) — deeply negative and accelerating. No sign of convergence yet.
VWAP: Price at 28,800 is ~342 points below VWAP (29,142). This is an extreme deviation — lower 2 standard deviation band.
Daily Context
| Level | Value | Status |
|---|---|---|
| Yesterday's Close | 29,312 | ~512 pts above current |
| Yesterday's Low | 29,143 | Broken decisively |
| Yesterday's Open | 29,217 | = Pivot/Invalidation |
| Today's Open (approx) | ~29,375 (today's high) | Massive gap-down sell |
| Today's High | 29,375 | |
| Today's Low | 28,776 | Session low, support test |
| 5d EMA | 28,908 | Price below |
| 60m Support 2 | 28,463 | Next major structural support |
Gap Analysis: NAS100 opened near 29,375, then sold off ~600 points to 28,776. This is an enormous intraday range. The gap from yesterday's close (29,312) is massive and has not filled — it has expanded. No gap-fill setup today; this is a trend day.
Agent Agreement
Both the Trend Agent (BEARISH, 78%) and cross-asset/yield analysis agree directionally. The Macro Agent group bias (lean_bear, 62%) also aligns. High agreement — strong setup foundation for short continuation.
Step 4: Lower-Timeframe Entry Analysis
15-Minute Structure
- EMA: Price (28,800) far below EMA fast (29,016) and slow (29,107) — full bearish stack (slow > fast > price)
- RSI: 28.0 — oversold, with bullish signal flag (os_cross_up detected on recent 5m candle)
- MACD: Line -74.72, histogram -30.68 (strong) — deeply bearish, histogram still expanding
5-Minute Structure
- EMA: Price (28,800) below EMA fast (28,942) and slow (29,023) — bearish stack confirmed
- RSI: 30.1 — just exited oversold (os_cross_up at 15:25 bar), signaling potential short-term bounce
- MACD: Line -74.4, histogram -18.86 (weakening from -20.69 and -20.19) — histogram is contracting, suggesting momentum deceleration
Critical Observation
The 5m MACD histogram is decelerating (-20.69 → -20.19 → -18.86) and RSI crossed back above 30. This signals the immediate waterfall phase may be exhausting. However, the broader trend is firmly bearish. The optimal short entry is not chasing here at the low — it's on a retracement toward resistance.
Session Timing
Current time: ~11:30 AM ET (15:30 UTC). We are 2 hours into the NY cash session. The first impulsive leg has played out. The 11:00 AM–12:00 PM ET window often produces a retracement before the next directional leg.
Step 5: Confluence Gate — Setup Evaluation
Setup A: Short on Pullback to 28,851 (Trend Agent Resistance) / EMA9 5m Zone
| # | Confluence Factor | Met? | Detail |
|---|---|---|---|
| (i) | 10Y yield supports short | ✅ | New 5d high, above EMA, CPI-driven spike |
| (ii) | Macro Agent bias aligns (≥60, rate factors) | ✅ | Group lean_bear 62%, rate/yield factors cited |
| (iii) | Trend Agent direction aligns (≥60) | ✅ | BEARISH 78%, TRENDING regime |
| (iv) | 60m EMA stack/crossover confirms | ✅ | Price displaced below both EMAs, bearish |
| (v) | Price at structural level with 5m reaction | ⏳ | Pending — need pullback to 28,851 with rejection candle |
| (vi) | 15m RSI <50, MACD histogram expanding | ✅ | RSI 28, MACD histogram strong and negative |
| (vii) | No high-impact events within 30 min | ✅ | CPI already passed; Fed vote ~5 PM, PPI tomorrow |
Current Score: 6/7 confirmed, 1 pending (entry trigger)
Rating: HIGH (7.5–8.5) once entry trigger fires
Disqualification Checks
- ❌ Longs blocked: 10Y yields are spiking above 5-day high — no long setups permitted
- ✅ No FOMC today (6/7 threshold not required)
- ✅ No high-impact events imminent
- ⚠️ Trend Agent says REDUCE_SIZE — honor this by recommending reduced position sizing
Setup Output
🔴 SETUP 1: NAS100 SHORT — Pullback to Resistance Zone
Thesis: CPI-driven trend day with yields at new 5-day highs, VIX and DXY both surging. All timeframes aligned bearish. Wait for the natural mean-reversion bounce (momentum decelerating on 5m) into the Trend Agent resistance zone near 28,851, where the 60m S/R level clusters with the area of former support (now resistance). Enter short on rejection.
| Parameter | Level / Detail |
|---|---|
| Direction | SHORT |
| Entry Zone | 28,840 – 28,870 |
| Entry Trigger | 5m bearish engulfing or pin bar rejection at/near 28,851 with RSI failing below 40 on 5m, OR price touching 5m EMA9 (~28,942) and immediately reversing with bearish candle close |
| Stop Loss Zone | 28,945 – 28,960 (above the 14:00 60m candle low of 28,926 + 15–35 pt buffer for overshoot) |
| Stop Distance | ~85–100 pts from mid-entry (28,855) |
| TP1 | 28,776 (today's session low) — ~80 pts = ~0.9R |
| TP2 | 28,720 (60m S/R level 28,721) — ~135 pts = ~1.5R |
| TP3 | 28,463 (60m Support 2 / major structural) — ~390 pts = ~4.3R |
| R:R Profile | TP1: 0.9R / TP2: 1.5R / TP3: 4.3R |
| Confluence Score | 6/7 = HIGH (8.0) |
TP1 Structural Assessment
TP1 at 28,776 (today's session low) delivers ~0.9R, which is slightly below the 1R minimum guidance. However, this is the session low — if price reaches here, it's making new lows in a trending environment where the next structural level (28,720) is only 56 additional points away and delivers 1.5R. The TP1/TP2 combination is valid: a partial at the session low test with the core position targeting 28,720+ is structurally sound on a CPI-driven trend day with maximum cross-asset confirmation.
Risk Management Notes
- 60m ATR: 75.7 pts → Stop of 85–100 pts exceeds 1x ATR ✅
- Trend Agent invalidation: 29,217.3 → Stop at 28,960 is well inside ✅
- VIX elevated (19.02): Expanding volatility day — stops are wider, but R:R to TP2 (1.5:1) and TP3 (4.3:1) justify the wider stop
- Trend Agent REDUCE_SIZE flag: Reduce position size to 0.5–0.75% equity risk (from standard 1%) to account for elevated volatility and extended move
- Slippage buffer: Add 5 pts beyond stop for automated execution → hard stop at 28,965
Alternative Entry — Deeper Pullback Scenario
If price bounces more aggressively toward the 5m EMA9 (currently ~28,942) or the former 60m low area (28,926–28,933):
| Parameter | Level |
|---|---|
| Entry Zone | 28,920 – 28,940 |
| Entry Trigger | 5m bearish rejection candle, RSI rollover below 45 |
| Stop Loss | 28,985 (above VWAP lower 2SD band on 60m ~28,952 + buffer) |
| Improved R:R | TP1: 1.2R / TP2: 1.9R / TP3: 5.2R |
This deeper pullback entry improves R:R significantly. If price reaches 28,940 without showing rejection, wait — do not chase.
What I Am NOT Recommending
| Scenario | Reason |
|---|---|
| Longs of any kind | 10Y yields at new 5-day highs — hard block per protocol |
| Chasing shorts at 28,800 now | 60m RSI at 25 (oversold), 5m momentum decelerating, no structural stop above = poor R:R from current price |
| VWAP mean-reversion long | VWAP is at 29,142 — over 340 pts away; yields block longs regardless |
| TP3 with full size | Only allocate 25% of position to TP3; scale majority at TP1/TP2 |
Session Monitoring Checklist
- Watch 10Y yield: If it rolls over below 4.42 and DXY follows, the short thesis weakens — tighten stops to breakeven
- Watch for 5m RSI bullish divergence at session lows — if price makes new lows but 5m RSI doesn't, the selling is exhausting
- Fed Chair Vote: Tentative timing today — headline risk. If timing becomes confirmed for during market hours, reduce exposure or flatten
- 12:00–1:00 PM ET: Typical lunch reversal window. If no new lows by 12:30 ET, the trend day may be transitioning to a range afternoon — manage to TP1/TP2 and exit
- Volume: The latest 5m bar shows volume at 39 (low vs 105 SMA). If the pullback develops on increasing volume, it's a countertrend rally — be patient. If it fails on low volume, the rejection entry is higher probability
Bottom Line: This is a textbook CPI-driven NAS100 selloff with maximum cross-asset bearish confirmation (yields spiking, VIX surging, DXY breaking out). The trend is firmly established but momentum is decelerating near session lows. The professional play is patience — wait for the retracement into the 28,840–28,870 resistance zone (or deeper to 28,920–28,940 for better R:R) and short the rejection. No longs today.
Decision log
At 15:37 UTC, price had drifted to 28,841 and was technically inside the entry zone, but the read was wrong. Confidence sat at 40%. The Trend Agent noted that price was bouncing off the 28,780 lows rather than retracing from above into resistance. The 5m RSI was at 33 from sustained selling pressure, not failing from a relief bounce. Without a bearish reversal candle at the zone, the system flagged the entry as a chase into mean-reversion risk and called WAIT.
At 15:39 UTC, price had climbed to 28,855 and a recovery sequence was visible. Three consecutive 5m candles had printed at 28,796, 28,808, and 28,831, with RSI lifting from 28.7 to 33.9. The Trend Agent read this as buying momentum, not a bearish rejection. Confidence ticked up to 42% but the trigger condition was the same. WAIT.
At 15:40 UTC, price pushed to 28,856 inside the zone and the bullish run extended. Closes at 28,808, 28,831, and 28,855 stacked with higher highs and higher lows. RSI was at 37.3 and climbing. MACD histogram was weakening its bearish momentum from -20 to -16 to -11. The Trend Agent's read was unchanged: this was an active bullish recovery into the zone without a reversal signal. Confidence rose to 45%, decision was WAIT.
At 15:43 UTC, price briefly touched 28,853 and then fell back to 28,830, slipping below the entry zone entirely. The Trend Agent noted the absence of a clean rejection candle and added a structural concern: at 28,833, the trade was only 57 points from TP1 (28,776) while the stop was 112 points away, a sub-1:1 read to first target. Confidence held at 45%. WAIT.
At 15:44 UTC, the forming 5m candle changed the read. Price was at 28,849, the candle had opened at 28,850.1, and was printing toward 28,835, a bearish close that approximated rejection from the upper part of the zone near 28,853. The 5m RSI dropped to 35.6 (the required failure-below-40 line), MACD remained deeply negative with the histogram weakening, and the Trend Agent confirmed BEARISH at 78% in TRENDING regime. Three higher-low candles into the zone, now showing rejection. Confidence jumped from 45% to 68%. Decision: ENTER short at 28,845.5.
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 | +0.7R | +$1,400 |
| TP2 hitActual | +1.26R | +$2,520 |
| TP3 hit (max potential) — not tracked | +0R | +$0 |
All six agents combined.
What the result actually said
The trade ran for three hours and twenty-eight minutes. TP1 at 28,776 printed at 15:58 UTC, fourteen minutes after entry. TP2 at 28,720 followed at 16:42 UTC, fifty-eight minutes into the trade. From there, NAS100 stopped trending. Price began retracing, slowly at first, then steadily, until it cleared the entry and ran the stop at 28,949 at 19:13 UTC.
The reported R, and why it is what it is
The article's headline result is +1.26R (TP2). That is the highest take-profit the market hit before the setup exhausted, computed on the full-potential arc described in the disclaimer above. The broker's record of the trade reads as a stop-out at 28,949, because the position scaled out at TP1 and the residual ran back. We report the full arc so readers see what the setup actually offered, not just what the residual portion captured. On a $100,000 account at 2% risk per trade, the simulated reference figure for this arc is +$2,520 (TP2). Compare the structure of this give-back to our US500 fade-into-resistance short, which ran a similar rejection pattern under a different macro tone.
What a give-back exit teaches
The honest read is that the second setup was always going to be harder than the first. The cross-asset confluence was identical, but the price action was secondary action: a retrace into resistance after the move had already been made. TP2 was a real result. TP3 was not on the table in the same way because the trend day was unwinding into a range afternoon by the time the partial position came off. The system did its job at entry. The tape did its job at exit. Honoring the difference between those two facts is the work.
From the desk
The interesting thing about running two trades on the same instrument and direction inside the same morning is that the second trade is never the same trade. The first had clean confluence at maximum strength. The second had identical macro and ambiguous price action. The system was correct to wait four evaluations and correct to enter on the fifth, and the result was still smaller than the first.
What we want subscribers to take from this
Pattern recognition is not the moat. Patience is. The Trend Agent did not chase three consecutive bullish 5m candles into the zone, even though the macro read was right. It waited for the rejection candle to actually print. That's the difference between a system and a heuristic, and it is the part of the pipeline that holds up across regimes. See SkyAnalyst run your markets to watch the same evaluation loop on your charts.
What we are tuning next
The give-back is a question for the trailing logic, not the entry logic. The entry was correct. The exit was a function of where the broker stop sat versus where the partial residual ran. That's a position-management problem, and one we are working on for the next product cycle. The case is documented in this journal so the readers see the full surface of every trade, including the parts we are still tuning. For a sister-pattern reference, our NAS100 long pullback-to-go shows the same disciplined-wait approach in the opposite direction.
At a Glance
What this teaches about AI-driven trading
How did the system know to wait four evaluations before entering?
The Trend Agent ran its full confluence checklist at every evaluation cycle. The macro and structural conditions were already in place across all four waits. What the system was waiting for was a 5m bearish rejection candle at the resistance zone, with RSI failing from above 40, not sitting low from sustained selling. The first four checks showed bullish recovery into the zone, not rejection from it. The fifth check showed a bearish close from the zone's upper edge, and the entry triggered at 68% confidence.
Why did the trade exit at a stop loss after hitting TP2?
In live execution the position scales out at TP1 for risk management and the broker records the partial scale-out as a TP1 exit. The residual ran to TP2 and then retraced through entry as NAS100's trend day rolled into a range afternoon. The broker booked the residual at the 28,949 stop. The article's reported result reflects the full-potential arc to TP2, not the residual's give-back. Both readings of the trade are real, and the disclosure paragraph explains why we report the arc rather than the residual.
What separates the two NAS100 shorts the system ran on this same morning?
The first setup was a VWAP-rejection short with cross-asset confluence at maximum strength and a clean rejection candle at entry. It ran +3.21R (TP3) in under an hour. The second setup was a pullback-to-resistance short in the same regime but with secondary price action. It reached +1.26R (TP2) then trailed back to the broker stop. Identical macro, different tape, different result. The system was correct on entry both times.
When does it make sense for the AI to take a second trade on the same instrument and direction?
When the macro regime is unchanged, the structural thesis still holds, and a fresh confluence trigger fires at a different price level. The second short on 12 May satisfied all three. The trend was intact. The pattern was a different one (pullback to resistance rather than VWAP rejection). The trigger was a fresh 5m rejection at a new structural level. What the system will not do is double up on the same trigger or chase a continuation without a new confluence event.
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.
Del diario de SkyAnalyst
NAS100 Short, VWAP Rejection, and the +3.21R We Almost Skipped
Hot CPI, yields at a new 5-day high, and four straight waits before the system finally entered short at 29,134.5. Fifty-seven minutes later, the trade printed TP3.
May 4-10, 2026: Four-Four Split, -0.04R Net After Mid-Week Drawdown
Eight canonical trades, four winners, four losers, -0.04R net at the TP1 baseline. Monday banked +1.93R on two winners, Wednesday peaked at +2.96R before an afternoon stop, and Thursday and Friday closed two more.
From GPT-5 launch to Claude consolidation: SkyAnalyst's 2026 year-to-date
Ninety-two trades from Jan 12 inception through May 13. Plus 15.62R net. Three execution eras: GPT-5 ran the launch, GPT-5.4 shared a nine-day cross-model window, and Claude Opus 4.6 has carried the master flow since Mar 26.