The Nasdaq Short That Said Wait Eight Times Before It Said Enter
SkyAnalyst AI journal entry: NAS100 Short on Sep 1, 2026 closed +2.97R 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 Nasdaq short on September 1 was obvious for twenty-seven minutes before the system agreed to take it. From 15:37 UTC the gate evaluated nine times. The first eight returned WAIT, sitting between 38 and 42 percent confidence, while the macro backdrop was about as one-sided as this desk sees. The 10-year yield had pushed to 4.77 percent and was printing new 5-day highs. The Macro Agent read strong_bear at 87 percent confidence, its highest-conviction bearish configuration, citing an acute duration headwind against long-duration tech. At 16:04 UTC the ninth evaluation returned 68 percent and ENTER. The system sold 29,229.9 with a stop at 29,310, and price fell to 28,992 for +2.97R against the second target. Eight refusals, one entry. The refusals are the interesting part.
A hostile tape, and a trap inside it
The macro configuration on September 1 was close to maximum bearish for the Nasdaq, and every cross-asset input agreed.
Post-Jackson Hole repricing had pushed the 10-year to 4.77 percent, well above its 5-day EMA of 4.730 and printing new 5-day highs at 4.798 against a prior high of 4.768. September rate-hike odds sat at 60 to 66 percent. For an index whose valuation is more sensitive to the discount rate than any other, that is the single most bearish input available.
Cross-asset confirmation was near total. VIX at 15.39 above its 5-day EMA of 15.06 and rising. DXY at 99.637 above its EMA and pressing toward the prior day's high, a double headwind of rates and dollar together. Brent above $92.50, adding cost-push inflation to the rates narrative. And gold breaking below the prior day's low, which in a risk-off session is the tell that matters: gold sold alongside equities is liquidation and margin stress, not a safe-haven bid, and it is what a real-rate shock looks like from the inside.
The technical picture matched. NAS100 gapped down, failed its gap-fill attempt, and collapsed through the prior day's low at 29,374.9, the 5-day EMA at 29,354, and prior daily support at 29,222.3, bottoming at 28,992.1 across a 530-point range.
And that is precisely where the trap was. The Trend Agent confirmed BEARISH at 70 percent confidence but flagged the regime as TRANSITIONING, explicitly warning of a countertrend relief bounce from oversold 60-minute conditions. An index that has fallen 530 points is not a safe place to sell simply because it has fallen 530 points.
VWAP Rejection at Session-High Resistance. That is the pattern the ninth evaluation finally scored, and the eight refusals before it are the more instructive half of this trade. Direction was never in doubt. Timing was the only open question, and the system treated it as one.
Direction is not timing
A market can be unambiguously bearish and simultaneously be the worst possible short at that moment. Those two facts do not conflict. Macro answers which side to be on. Structure answers when to get there, and conflating the two is how a correct directional call still loses money.
When an oversold index bounces, it bounces hardest in exactly the tape where the fundamental case is strongest, because that is where positioning has become most crowded on one side. The Trend Agent had flagged that risk explicitly, calling the regime TRANSITIONING and warning of a countertrend relief bounce off oversold 60-minute conditions.
What the system was actually waiting for
It was not waiting for a better price in the abstract. It was waiting for a specific event: the relief rally running into resistance and failing there. That failure does two things at once, and both are necessary.
It confirms that sellers are still defending the level, which the macro case alone cannot tell you. And it supplies a stop location directly above a rejection the market has just demonstrated it will honour. Without the second of those there is no way to size the position sensibly, however strong the thesis.
Why patience changed the arithmetic
The result was an 80.1 point risk band on a 238 point move. Entered into the falling knife at 15:37, the same directional call would have required a materially wider stop, would have sat underwater through the bounce, and would have paid a fraction of the R for identical price action.
Professional traders: this is why SkyAnalyst does not favor any single strategy or a fixed entry trigger. It reads the tape first and lets structure decide the moment, which is why the Dow short that same morning cleared on its first evaluation while this one took nine. Same thesis, same desk, different tape, different answer.
NAS100 NY AM Session Analysis: September 1, 2026
Market Environment Summary
The NAS100 is trading in a hostile macro environment. Post-Jackson Hole yield repricing has pushed the 10-Year Treasury yield to 4.77%, well above its 5-day EMA (4.730%) and printing new 5-day highs (today's high 4.798% vs. yesterday's high 4.768%). This is the single most bearish signal for the Nasdaq-100. September rate-hike odds have jumped to 60-66%, directly pressuring long-duration tech valuations. The Macro Agent reads strong_bear at 87% confidence, explicitly citing "acute duration headwind" and yields at cycle highs, this is the highest-conviction bearish macro configuration available.
Cross-asset confirmation is near-maximum: VIX at 15.39 sits above its 5-day EMA (15.06) and is rising, confirming risk-off. DXY at 99.637 is above its 5-day EMA (99.577) and pushing toward yesterday's high, a double headwind for NAS100 (yields up + dollar up). Oil surging above yesterday's high (Brent 92.52) adds inflationary pressure reinforcing the rates narrative. Gold crashing below yesterday's low signals liquidation/margin stress, not a safe-haven bid, which is consistent with a real-rate shock environment.
On the technical side, NAS100 opened with a massive bearish gap: today's high of 29,522.6 vs. yesterday's close of 29,455.3 created an early gap-fill attempt, but the sell-off has been relentless. Price collapsed through yesterday's low (29,374.9), the 5-day EMA (29,354), and the prior daily support at 29,222.3, bottoming at 28,992.1: a 530-point intraday range. The Trend Agent confirms BEARISH at 70% confidence in a TRANSITIONING regime, flagging a countertrend relief bounce off oversold 60m conditions but maintaining the bearish call with a recommendation to REDUCE_SIZE.
Price currently sits at 29,183.5, roughly 100 points above today's low. The 5m chart shows a bullish EMA cross (fast above slow) with RSI 61 and MACD above zero, this is a countertrend relief rally within a bearish structure. The 15m shows price above its fast EMA but still well below the slow EMA and VWAP (~29,291). The 60m remains firmly bearish: price below all EMAs, VWAP (29,289), RSI at 40, MACD deeply negative with a weakening histogram.
ISM Manufacturing PMI (10:00am) came in at 54.6 vs. 55.2 expected, softer but not weak enough to shift the rates narrative. ISM Prices at 71.1 (in-line) and JOLTS at 7.27M (slightly below 7.33M forecast) are already absorbed. No remaining high-impact USD events today: the calendar is clear for the rest of the session.
Directional Bias: Bearish Volatility: Normal-to-Elevated (VIX 15.39, 60m ATR expanding to ~77 pts)
Confluence Assessment: Short Setup
| # | Confluence Factor | Status | Score |
|---|---|---|---|
| (i) | 10Y yield direction supports short | ✅ 4.77% above 5d EMA, new 5d highs | ✓ |
| (ii) | Macro Agent bearish ≥60% citing rates | ✅ Strong_bear 87%, explicit yield/rate factors | ✓ |
| (iii) | Trend Agent bearish ≥60% | ✅ Bearish 70%, moderate strength | ✓ |
| (iv) | 60m EMA stack/crossover bearish | ✅ Price below fast and slow EMA, bearish alignment | ✓ |
| (v) | Price at VWAP/Fib/session level showing reaction on 5m | ⏳ Price approaching resistance cluster 29,189-29,222 (NY session high, pivot), 5m showing deceleration | ✓ |
| (vi) | 15m RSI <50 with MACD histogram expanding | ⚠️ 15m RSI at 52.7 (slightly above 50), MACD histogram positive but flattening | ✗ |
| (vii) | No high-impact USD events within 30 min | ✅ All events past, calendar clear | ✓ |
Score: 6/7 = High (7.5-8.5 range)
The only miss is the 15m RSI sitting marginally above 50 due to the relief bounce, but this is expected in a countertrend move that is approaching resistance where we'd initiate the short. Once price rejects the resistance zone, the 15m RSI will roll back below 50, confirming the entry.
Setup Logic
The trade thesis is a VWAP rejection short: price has rallied ~190 points off the 28,992 low in a countertrend squeeze. This rally is now approaching a dense resistance cluster:
- NY Session High: 29,218.8
- Trend Agent Resistance / Invalidation: 29,209.5 / 29,222.3
- 60m Fibonacci 23.6% retrace: 29,355.9 (above, indicating the bounce hasn't even reclaimed the shallowest Fibonacci level from the higher-timeframe swing)
- 15m Fibonacci 23.6% (bearish): 29,158.2 (already exceeded)
- VWAP: ~29,289-29,299 (still ~100 points overhead, strong overhead magnet/resistance)
- Previous daily low / support-turned-resistance: 29,222.3
The key zone is 29,200-29,230: this is where the NY session high (29,218.8), the Trend Agent invalidation (29,222.3), and the resistance cluster converge. The 5m chart shows MACD histogram weakening (from 14.88 to 2.85 over the last 6 candles) with RSI at 61, momentum is decelerating into this resistance zone. This is a textbook exhaustion pattern into supply.
The more aggressive entry targets the 29,200-29,220 zone as price pushes into the session high resistance. A more conservative entry waits for VWAP at ~29,290 to be tested and rejected, but given the bearish weight of all higher-timeframe factors and the Trend Agent's REDUCE_SIZE flag on the relief bounce, the near-resistance entry offers superior R:R.
Why not long? 10Y yields are spiking above their 5-day highs, the explicit rule prohibits longs. Additionally, both agents are bearish, VIX and DXY confirm, and the 60m structure is firmly bearish. The relief bounce is a short-selling opportunity, not a reversal.
Setup #1: NAS100 SHORT, VWAP Rejection / Session High Resistance
- Entry Zone: 29,200-29,230 (cluster: NY session high 29,218.8 + Trend Agent R level 29,209.5 + daily prior support-turned-resistance 29,222.3)
- Entry Trigger: Bearish rejection candle on 5m (bearish engulfing, pin bar, or lower-high formation) within the entry zone, ideally with 5m RSI rolling below 60 and MACD histogram turning negative. Price must remain below 29,230 on the close of the trigger candle.
- Stop Loss: 29,310 (87 pts from mid-entry at 29,215; placed 15 pts above VWAP ~29,290-29,299 to account for NAS100 overshoot tendency; within 1.5x the 60m ATR of 77 pts = 115 pts max allowable, and well below the Trend Agent's structural framework)
- Targets:
- TP1: 29,110, London session low area / 60m support cluster (~105 pts = ~1.2R), structural level where the 11:00 UTC candle closed
- TP2: 28,992, Today's low / session floor (~223 pts = ~2.6R), clear structural level with price memory
- TP3: 28,900, Round number extension below today's low (~315 pts = ~3.6R), only if price breaks 28,992 with momentum and both agents maintain bearish posture
- R-Multiples: TP1 = 1.2R, TP2 = 2.6R, TP3 = 3.6R
- Position Management: Move stop to breakeven on TP1 fill. Trail stop 10 pts above the most recent 15m swing high after TP1.
- Quality Score: 7.5 / 10
- Confidence: High, 6/7 confluence factors met; yield spike is the primary driver with maximum cross-asset confirmation (VIX + DXY + Macro Agent all aligned); the only caution is the Trend Agent's TRANSITIONING regime and REDUCE_SIZE recommendation, which appropriately caps the score. The countertrend bounce is losing momentum into a well-defined resistance cluster, offering textbook mean-reversion entry within a dominant bearish trend.
Risk Considerations
- Countertrend squeeze risk: The relief bounce from 28,992 has been sharp (~190 pts). If short-covering accelerates through 29,230, it could run toward VWAP at 29,290-29,300. The stop at 29,310 accommodates this overshoot. If price closes a 5m candle above 29,230 without immediate rejection, do not chase, wait for the VWAP test instead.
- Trend Agent REDUCE_SIZE flag: The transitioning regime warrants smaller position sizing. Use 0.5-0.75% equity risk rather than a standard 1% on this setup to respect the regime signal.
- ADP tomorrow at 8:15am ET: A significantly weak ADP number could trigger a dovish yield reversal. This trade is an intraday setup, manage or exit before the close if carrying overnight risk is not intended.
- No sector rotation flag needed: Both Dow futures (-0.5%) and NAS100 (-1%) are declining, with NAS underperforming as expected in a rates-driven selloff. No divergence detected in the available data.
- Slippage buffer: The automated system should place the stop-loss order at 29,315 (5 pts buffer above the analytical stop of 29,310) to account for execution slippage in a fast market.
Decision log
15:37 UTC, 40 percent, WAIT. The macro gate was already emphatic and the Trend Agent already bearish, but the 60-minute read was oversold and the regime tagged TRANSITIONING. Selling the low of a 530-point collapse is the highest-variance version of a correct idea, and the confidence score reflects that: directionally right, structurally premature.
15:39 UTC, 40 percent, WAIT. No change of substance. Price was still extended below the 5-day EMA with no rejection structure to sell into, and no candidate stop location that did not require an uncomfortably wide band.
15:40 UTC, 42 percent, WAIT. A marginal improvement as price began to stabilise, but stabilisation is not rejection. A base forming near the low is as consistent with a bounce as with continuation, and the gate does not act on ambiguity.
15:41 UTC, 42 percent, WAIT. Holding. The relief bounce the Trend Agent had flagged was starting to develop, which is the correct moment to be least interested in selling, not most.
15:43 UTC, 38 percent, WAIT. The session's lowest confidence reading, and it arrived as the bounce gathered pace. This is the number behaving properly: as the countertrend move strengthened, the case for selling into it weakened.
15:44 UTC, 40 percent, WAIT. The bounce continued. Still no resistance test, so still nothing to sell against.
15:44 UTC, 40 percent, WAIT. A second evaluation inside the same minute, triggered by fresh price action rather than the clock. Same conclusion.
16:03 UTC, 42 percent, WAIT. Nineteen minutes later the bounce had carried into the session-high and VWAP zone. The setup was now forming, but forming is not confirmed, and the gate held for one more pass.
16:04 UTC, 68 percent, ENTER. The rejection printed. Price tested the session-high resistance confluence with VWAP, failed there, and the confidence score jumped 26 points in a single evaluation. The Risk Agent sized against a stop at 29,310, 80.1 points above the 29,229.9 entry, and the position went short. Everything the previous eight evaluations were waiting for arrived in one bar.
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 | +1.5R | +$3,000 |
| TP2 hit | +2.97R | +$5,940 |
| TP3 hit (max potential) (not tracked) | +0R | +$0 |
All six agents combined.
What this trade teaches
The most expensive mistake available on September 1 was not being wrong about direction. It was being right about direction and early about timing.
Eight evaluations returned WAIT while a retail read of the same screen would have said the case was already made. Yields at cycle highs, dollar bid, breadth negative, gold liquidating: every one of those was true at 15:37 and every one of them was still true at 16:04. What was not true at 15:37 was a structure worth risking money against.
The second lesson is about what a confidence score is for. It is not a measure of how bearish the market is. It is a measure of how good this specific entry is right now. Those two numbers move independently, and a system that conflates them will sell every low in every downtrend.
From the desk
This was one of three shorts the desk took on September 1, alongside US30 and US500, all on the same macro read and all winners. The US30 short cleared on its first evaluation. This one took nine. The difference was not conviction, it was that the Dow offered a clean retest immediately while the Nasdaq was mid-collapse and had to be allowed to bounce first.
The trade also opens the month after August closed at -2.67R, the system's first losing month since the January 12 inception. That loss was concentrated in one place: the GBPUSD long book took ten trades, won two and gave back 6.18R, while the desk's short books finished August up 3.52R.
On August 27 we split every instrument trader into independent long and short books precisely so that a one-sided failure cannot hide inside a pooled instrument average. This NAS100 short is now measured as its own book, separately from NAS100 long, which lost 2.13R across four trades in August while the short side made 0.62R across five. Same instrument, same agents, opposite outcomes, and until the split we were reporting the two of them as one number.
At a Glance
What this teaches about AI-driven trading
Why did the system evaluate nine times in twenty-seven minutes?
Evaluations are triggered by price action rather than by a fixed clock, which is why two of them fall inside the same minute at 15:44. Each pass re-scores the setup against current structure, and the gate keeps running until the inputs agree or the setup expires. A high evaluation count is not indecision, it is the system repeatedly declining an entry that has not yet earned its stop placement.
What is the difference between a 42 percent and a 68 percent score?
The score measures the quality of entering right now, not the strength of the directional case. The macro case did not change across the twenty-seven minutes: yields, breadth, the dollar and gold were all saying the same thing at 15:37 as at 16:04. What changed was that a rejection printed at resistance, supplying both confirmation that sellers were present and a defensible place to put the stop.
What does +2.97R mean in dollars?
R is the trade's risk unit, the distance from entry to stop. On a $100,000 account risking 2% per trade, 1R is $2,000, so +2.97R is roughly $5,940. Reporting in R keeps results comparable across instruments whose point values differ. The simulation panel in this article shows the same trade at several account sizes.
Would entering earlier have made more money?
No, and this is the most common misreading of a patient entry. An earlier fill into the falling knife would have required a wider stop to survive the relief bounce the Trend Agent had already flagged. Because R is measured against risk taken, the same 238-point move on a wider stop pays materially less, and that assumes the position survives the bounce at all rather than being stopped out first.
How do you count R on this trade?
This case study reports full potential, the R distance to the furthest target price reached, here the second target at 28,992. Our weekly, monthly and year-to-date recaps use a stricter TP1 baseline crediting only the first target, so the same trade appears smaller there. See the [August monthly recap](/blog/monthly-recap-2026-08) and the [2026 year-to-date review](/blog/ytd-2026) for the conservative totals.
Run your markets with SkyAnalyst
21-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
The S&P Short Where a 7.8 Point Stop Turned 32 Points Into 4R
The smallest move of the three shorts SkyAnalyst took on September 1, and by far the largest R. Price travelled 31.7 points. The risk band was 7.8. That ratio is the entire trade.
US30 Short, September 1: One Evaluation, One Entry, TP3
The Dow short that needed a single evaluation to clear. Breadth at -683, yields at 4.77%, and a retest that never gave back a tick. Entry to third target in 1 hour 17 minutes for +1.63R.
2026 So Far: +28.75R, and the Month That Changed How We Measure
Two hundred and twenty trades since the January 12 inception, 125 winners, +28.75R net. Seven green months and one red one, and the red one taught us more about the system than any of the seven that came before it.