Twelve of Thirteen: The Long Book Answers Back at TP3
SkyAnalyst AI journal entry: NAS100 Long on Sep 3, 2026 closed +2.84R 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.
SkyAnalyst has now won twelve of its last thirteen trades. The run started on August 24 with seven consecutive winners, broke on August 31 when a NAS100 short stopped out for -1R, and has since added five more. Today's is the fifth, and it is the one worth stopping on, because it is a long. At 14:36 UTC the gate cleared on its first evaluation and the system bought NAS100 at 29,261.9 with a stop at 29,183, a risk band of 78.9 points. Fifty-three minutes later price had run through all three targets to 29,486. Two hundred and twenty-four points, +2.84R, and no drawdown against the position at any point. It was taken over the objection of the desk's own Macro Agent.
The tape disagreed with the macro read
The setup for September 3 was a genuine internal disagreement, and the way it resolved is the whole trade.
The Macro Agent returned lean_bear at 71% confidence, citing NAS100's sensitivity as a discount-rate proxy with September hike probability still elevated at 60 to 70 percent. That is a structural position, and on its own terms it is not wrong.
The live rate tape said something different. The 10-year sat at 4.756%, marginally above its 5-day EMA of 4.755 but below the prior close of 4.780 and well below the 5-day high of 4.816. Yields had closed lower for two consecutive sessions, 4.796 to 4.780 to 4.756, and the day printed a low of 4.732, beneath the prior day's low. Yields were not spiking into the trade. They were rolling over from a five-day high.
Cross-asset agreed with the tape rather than with the agent. VIX at 14.97 was below its 5-day EMA of 15.19 and below the prior day's low, so fear was draining. DXY at 99.052 was well below its 5-day EMA of 99.507, removing the dollar headwind entirely. Gold was pushing above the prior day's high, which is what a weaker dollar and lower real rates look like from the other side of the trade.
The Trend Agent read BULLISH at 66% in a TRENDING regime, with invalidation at 29,157.4. Price held above VWAP at 29,153 and above the prior day's high of 29,167. The broader tape confirmed it was not a Nasdaq-specific move: US30 was trading above its own prior-day high and above its 5-day EMA.
Fibonacci Pullback Continuation. The pattern is the least exotic one the desk trades: an established intraday uptrend pulls back into a measured retracement, holds, and continues. What made it worth writing about is that the system had to overrule one of its own agents to take it.
How a disagreement gets resolved
When the Macro Agent and the Trend Agent disagree, the framework reduces confidence and narrows the permitted setups. It does not simply defer to the macro read, and it does not average the two into mush.
What it does is ask whether the macro bias is confirmed by the instruments that would have to move for it to be true. Here it was not. A rate-driven bearish case for the Nasdaq requires yields to be rising, and yields had fallen for three sessions. It requires a firm dollar, and DXY was below its 5-day EMA. It requires fear to be building, and VIX was below its own.
Why the divergence lowered the score without killing the trade
The disagreement is why this entry scored 62 rather than the high seventies. That number is not a hedge, it is an accurate discount for a real, unresolved tension: the ISM Services print had come in at 55.4 against a 54.2 forecast, a beat hot enough to push yields back up and vindicate the Macro Agent within hours.
So the position was taken at a size that respected the disagreement, against a stop at 29,183, 78.9 points below entry and just under the Trend Agent's stated invalidation at 29,157.4. Being wrong here was designed to be survivable, because the case against the trade was legitimate rather than imaginary.
What the desk does with a conflict
Professional traders: the instinct when two reliable signals conflict is to stand aside, and standing aside is genuinely the right answer when neither has an edge over the other. It is the wrong answer when one of them is describing a market that is not currently doing what it says.
SkyAnalyst does not favor any single strategy, and it does not rank its agents by seniority either. It reads the tape first and lets live cross-asset behaviour arbitrate between them. On September 3 the tape sided with the trend read, the trade was sized down to respect the doubt, and it went the full distance anyway.
NAS100 NY AM Session Analysis: Thu, Sep 3, 2026
Market Environment Summary
The NAS100 is trading at 29,275 in the early NY AM session, having gapped above yesterday's high (29,167) and rallied strongly from the London session low near 29,059 to a session high of 29,332 before consolidating. The macro backdrop presents a nuanced picture with conflicting signals across the rate/yield and cross-asset complex.
Yields are the key story today. The 10Y yield at 4.756% sits marginally above its 5-day EMA (4.755%) but is critically below yesterday's close (4.780%) and well below the 5-day high (4.816%). Yields are in a short-term downtrend, closing lower for two consecutive days (4.796 → 4.780 → current 4.756), with today's session printing a low of 4.732, which is below yesterday's low (4.768). This is constructive for NAS100: yields are not spiking, they are rolling over from a 5-day high. The near-flat position relative to the 5-day EMA prevents a full bullish yield signal, but the directional drift is supportive.
Cross-asset confirmation is mixed-to-bullish. VIX at 14.97 is below its 5-day EMA (15.19) and below yesterday's low (15.12), confirming declining fear, this is supportive. DXY at 99.052 is well below its 5-day EMA (99.507) and below yesterday's low (99.43), removing the dollar headwind entirely. Gold is surging above yesterday's high, confirming the weaker-dollar/lower-rate narrative. Oil at 96.72 (Brent) is elevated but not accelerating enough to reignite inflation fears today. The VIX + DXY combination is double supportive, not a headwind.
Macro Agent flags NAS100 as lean_bear (confidence 71%) primarily on rates/Fed pricing, September hike probability remains elevated at 60-70%, and the agent cites NAS100's sensitivity as a discount-rate proxy. However, the agent also acknowledges the improvement in price context (above yesterday's high) and notes the ISM/NFP catalysts. ISM Services PMI just printed at 55.4 vs 54.2 forecast: a beat that could reignite rate-hike concerns. This is a material development: a hot ISM reading could push yields back up, partially offsetting the supportive yield drift. The macro agent's bearish bias is driven by structural rate positioning, not by today's real-time yield action, which creates a divergence between the agent's bias and the actual yield tape.
Trend Agent reads BULLISH (66% confidence, MODERATE strength, TRENDING regime) with invalidation at 29,157.4. Price holds above VWAP (29,153), above yesterday's high (29,167), and 5m/15m EMAs are in bullish alignment. The 60m EMA structure is still transitioning (fast EMA 29,148 just below slow 29,151), but price has been above both for two candles with MACD turning positive. The daily picture shows NAS100 essentially at its 5-day EMA (29,277 vs 29,275 current), a potential inflection point.
Key divergence: Macro Agent (lean_bear 71%) vs Trend Agent (bullish 66%). Per the framework, this divergence reduces confidence by 2-3 points and limits setups to VWAP mean-reversion trades. However, the Macro Agent's bearish bias is not confirmed by the real-time yield tape (yields declining) or by VIX/DXY (both declining), which weakens the macro bearish case for the current session. The ISM beat introduces uncertainty, yields could reverse higher.
Broad market context: US30 at 53,410 is trading above yesterday's high (53,251) and above its 5-day EMA, showing broad market strength, not NAS-specific weakness. No sector rotation risk flagged; breadth (ADD at 173, below 5d EMA of -318) is modestly positive but not divergent.
Economic calendar: Both events today are complete, Unemployment Claims (8:30 ET, 206K) and ISM Services PMI (10:00 ET, 55.4). No further high-impact events today. NFP tomorrow at 8:30 ET is the next major catalyst. No Fed speakers imminent. The event window is clear for NY AM trading.
Directional Bias: Cautiously Bullish (yields declining, VIX declining, DXY declining, trend bullish, but Macro Agent and ISM beat create tension)
Volatility: Normal (VIX 14.97, 60m ATR 73 pts, normal regime, no expansion)
Step-by-Step Confluence Assessment
Step 1: 10Y Yield Direction
Yield at 4.756% is essentially flat to EMA (4.755%) but trending lower from 4.796 → 4.780 → 4.756 over the past three sessions. NOT making new 5-day highs, the 5-day high was 4.816, and today's high of 4.788 is well below. Today's low of 4.732 is a new 5-day low. Verdict: Marginally supportive for longs. Not a strong bullish signal given the near-EMA position, but the directional drift is clearly lower. The ISM beat (55.4 vs 54.2) could reverse this, monitoring required. Score: Supportive (lean bullish).
Step 2: Macro Regime & Cross-Asset
- Macro Agent: lean_bear, 71% confidence, citing rates/Fed pricing. This is a high-conviction bearish signal per the framework. However, the agent's own notes acknowledge price improvement and the signal is based on structural rate expectations, not today's yield action.
- VIX: 14.97, below 5d EMA (15.19), below yesterday's low. Bullish confirmation.
- DXY: 99.052, below 5d EMA (99.507), below yesterday's low. Bullish confirmation.
- Both VIX and DXY are confirming the lower-yield environment, which contradicts the Macro Agent's bearish bias for the intraday session. The Macro Agent's lean_bear is structurally valid (rate expectations) but is not supported by today's real-time cross-asset tape.
- Score: Mixed, VIX/DXY bullish, Macro Agent bearish. Net: slight headwind due to agent confidence >70 with rate factors.
Step 3: Trend Structure & Key Levels
- Trend Agent: Bullish, 66% confidence, TRENDING regime, invalidation 29,157.4.
- 60m EMA alignment: Fast (29,148) still slightly below Slow (29,151), NOT a clean bullish stack, but converging with price above both. Fresh bullish crossover is imminent but not confirmed. Partial bullish.
- 60m RSI: 63.2 (above 50, bullish zone, not overbought).
- 60m MACD: Line 25.6 above signal 14.5, histogram expanding at +11.2. Bullish momentum confirmed.
- Daily reference: Price at 29,275 is effectively at the 5-day EMA (29,277). Yesterday's close was 29,100; today opened with a gap up (>100 pts from close to today's high). The gap from close (29,100) to today's low (29,033) was actually a gap down initially, followed by a full reversal, this is a bullish V-reversal pattern from the Asian session.
- Key levels: R1 = 29,332 (today's high/session high). S1 = 29,198 (NY session low). S2 = 29,157 (Trend invalidation). VWAP = 29,153.
- Score: Bullish lean, moderate conviction. 60m EMA crossover not yet confirmed is the main weakness.
Step 4: Lower-Timeframe Entry Analysis
15m: EMA fast (29,197) above slow (29,162), bullish stack confirmed. RSI at 62.6 (above 50, supports longs). MACD histogram at +18.4, strong and expanding. Price above VWAP in upper 2SD band. All signals aligned bullish on 15m.
5m: EMA fast (29,245) above slow (29,194), strong bullish stack. RSI at 57.2 (above 50, pulled back from 70 overbought, healthy reset). MACD histogram fading from +14 to +0.8, momentum is waning on the 5m, suggesting the initial thrust from 29,060 to 29,332 has paused. Price at 29,276 is above EMA9 (29,245), using it as dynamic support. VWAP at 29,152 is ~123 points below, too far for an immediate mean-reversion entry.
Current 5m structure: After the 29,332 high, price has pulled back to 29,259-29,276 range, forming a consolidation/flag pattern. The 5m Fibonacci from the NY session pivot low (29,157 → 29,332 high) puts:
- 23.6% = 29,291
- 38.2% = 29,265 (current price is near this level)
- 50.0% = 29,245
- 61.8% = 29,224
Price is consolidating around the 38.2% retracement, a classic continuation zone in trending moves.
Entry type assessment:
- VWAP mean reversion: Not applicable, price is 123+ pts above VWAP, no reversion opportunity at current levels.
- Fibonacci pullback: Active: price is sitting at the 38.2% retracement of the NY leg, which aligns with yields declining and VIX/DXY supportive. A bounce from the 29,245-29,265 zone (38.2-50% fib) would be a textbook pullback entry.
- Opening range breakout: The first 15m candle (13:30 UTC / 9:30 ET) ranged 29,157-29,316. A breakout above 29,316/29,332 would confirm continuation. Not yet triggered.
- EMA9 pullback: 5m EMA9 at 29,245 is acting as dynamic support. A touch and hold would be an entry trigger.
Step 5: Confluence Gate Scoring
| # | Factor | Assessment | Score |
|---|---|---|---|
| i | 10Y yield direction supports long | Yields declining from 5-day high, below yesterday's close, new 5-day low printed | ✅ |
| ii | Macro Agent bias aligns (≥60, citing rates) | Agent is lean_bear 71% citing rates, opposes the long | ❌ |
| iii | Trend Agent direction aligns (≥60) | Bullish 66%, supports the long | ✅ |
| iv | 60m EMA stack or fresh crossover | Fast EMA converging on Slow but NOT yet crossed above, partial | ❌ |
| v | Price at VWAP/Fib/session level with 5m reaction | Price at 38.2% Fib retracement (29,265), holding above with 5m EMA9 (29,245) as support | ✅ |
| vi | 15m RSI >50 with MACD histogram expanding | RSI 62.6 (>50) ✅, MACD histogram 18.4 strong ✅ | ✅ |
| vii | No high-impact events within 30 min | ISM completed at 10:00 ET; next event is NFP tomorrow 8:30 ET | ✅ |
Score: 5/7 = Medium-High (6.5-7.5)
Note: The Macro Agent divergence is the primary headwind. The 60m EMA crossover is pending but not confirmed. Five of seven factors support a long. The ISM beat (55.4 vs 54.2) is a risk, could push yields back up, but yields have not reacted negatively as of the snapshot (4.756 still below yesterday's close).
Step 6: Risk Management Calibration
Structural stop level: The Trend Agent invalidation is 29,157.4. Below that is the NY session structure at 29,157 (the pre-market low that initiated the rally) and VWAP at 29,153. A structural stop below 29,150 with buffer = 29,135.
Distance from entry: If entry at 29,250 zone, stop at 29,135 = 115 points. The 60m ATR is 73 pts, a 115-point stop is 1.57x ATR, which is within the "normal-to-wider" range and appropriate for NAS100's overshoot tendency. However, R:R must be checked.
Alternative tighter stop: The 5m pivot low at 29,198.9 with buffer = 29,185. From entry 29,250, that's 65 pts, essentially 1x 60m ATR (73). This requires the pullback to hold without breaching the session's structural low. This is more aggressive but still valid if the 38.2% fib holds.
Target profile:
- TP1: 29,332 (today's high / session resistance) = 82 pts from 29,250 entry. With 65-pt stop = 1.26R. With 115-pt stop = 0.71R (fails minimum).
- TP2: 29,400-29,420 (next structural zone, 60m upper ATR band) = ~160 pts. With 65-pt stop = 2.46R. With 115-pt stop = 1.39R.
- TP3: 29,486-29,522 (60m S/R levels from historical data) = ~250 pts. With 65-pt stop = 3.85R.
R:R Assessment: With the tighter stop (29,185), TP1 at 29,332 delivers 1.26R at structure, acceptable. TP2 at 29,400+ delivers 2.46R. The wider structural stop (29,135) makes TP1 unworkable (<1R) and degrades the entire profile. The tighter stop is required for this setup to be valid, accepting the risk that a deeper pullback to the 50-61.8% fib zone would stop out before continuation.
Trade Setup
Setup #1: NAS100 LONG, Fibonacci Pullback Continuation
- Entry Zone: 29,240-29,265 (38.2-50% Fibonacci retracement of 29,157→29,332 leg, confluence with 5m EMA9 dynamic support at ~29,245)
- Entry Trigger: 5m candle closes above 29,265 after touching the 29,240-29,265 zone, OR price holds 5m EMA9 (currently ~29,245) and prints a higher low on 5m with RSI above 55
- Stop Loss: 29,183 (15 points below the NY session pivot low at 29,198.9, allows for NAS100 overshoot; 1x 60m ATR from the entry midpoint; well above the Trend Agent invalidation at 29,157)
- Targets:
- TP1 = 29,332 (today's session high / London session high, first structural resistance)
- TP2 = 29,400 (round number + 60m ATR upper band 29,385-29,399)
- TP3 = 29,486 (60m historical S/R level, only if price clears 29,332 cleanly with volume)
- R-Multiples: (using entry 29,252, stop 29,183 = 69 pts risk)
- TP1 = 29,332 → 80 pts = 1.16R
- TP2 = 29,400 → 148 pts = 2.14R
- TP3 = 29,486 → 234 pts = 3.39R
- Quality Score: 6.5/10
- Confidence: Medium, Five of seven confluence factors align. The Fibonacci pullback structure is textbook, yield/VIX/DXY cross-asset confirmation is solid, and the Trend Agent supports. The primary risks are: (1) Macro Agent lean_bear at 71% with rate concerns, structurally valid even if today's yield tape doesn't confirm yet; (2) the hot ISM print (55.4) could reignite yield pressure; (3) the 60m EMA bullish crossover is pending, not confirmed; (4) NFP tomorrow at 8:30 ET may cap upside ambition today. TP1 at 1.16R is slightly below ideal (1.25R) but sits at hard structure (session high). The trade is valid because TP2 at 2.14R is strong, a close TP1 with a high-probability TP2 at real structure satisfies the framework.
Management notes:
- If price breaks below 29,198 before entry triggers, the setup is void, wait for a re-test of VWAP (~29,153) for a secondary entry evaluation.
- Move stop to breakeven if TP1 (29,332) is hit. Trail remaining position with 5m EMA9.
- If yields begin spiking above 4.780 (yesterday's close), close the position regardless of price action, the rate-sensitive thesis would be invalidated.
- Reduce position size from standard 1% risk to 0.75% given the Macro Agent divergence and pre-NFP positioning risk.
No second setup qualifies. A short setup cannot be justified when yields are declining, VIX is falling, and DXY is breaking down, the macro tape does not support shorts despite the Macro Agent's structural lean_bear view. A breakout long above 29,332 would require waiting for confirmation and would offer poor R:R with stops still needing to be below 29,198. The pullback entry is the only actionable setup in this environment.
Decision log
14:36 UTC, 62 percent, ENTER. One evaluation, and it cleared on a score well below what this desk usually enters on, because the Macro Agent's lean_bear at 71% sat directly against the Trend Agent's bullish at 66%. The framework does not break that tie by rank. It checks whether the market confirms the macro claim, and it did not: the 10-year had closed lower for two straight sessions and printed a new low beneath the prior day's, VIX was under its 5-day EMA, and DXY was under its own by nearly half a point. Price was above VWAP at 29,153 and above the prior day's high at 29,167, with 5m and 15m EMAs in bullish alignment. The Risk Agent sized against a stop at 29,183, just under the Trend Agent's 29,157.4 invalidation, and took the long at 29,261.9 at reduced size. The 62 is the disagreement priced in, not confidence in the read.
On a $100k account at 2.0% risk per trade.
Each trade risks +$2,000 (1R). The system's actual scale-out behavior may differ, see disclaimer.
| Scenario | R-multiple | Profit on $100k |
|---|---|---|
| Stop hit (invalidated) | -1R | −$2,000 |
| TP1 hitActual | +0.89R | +$1,780 |
| TP2 hit | +1.75R | +$3,500 |
| TP3 hit (max potential) | +2.84R | +$5,680 |
All six agents combined.
What this trade teaches
A confidence score is not a forecast quality rating. It is a measure of how much the inputs agree, and those are different things.
This trade entered at 62, lower than most entries the desk takes, and returned +2.84R by running through every target. The GBPUSD short the day before entered at 62 as well, after four consecutive refusals at 79, 81, 82 and 84. Neither outcome was predictable from the number, because the number is describing internal agreement rather than the probability of the next 224 points.
The second lesson is about which signal wins when two conflict. The Macro Agent was making a structural argument about rate positioning that may well be right over weeks. The Trend Agent was describing what the tape was doing in the session. For an intraday position with a 78.9 point stop, the second is the one that determines whether the stop gets hit, and the framework knows the difference.
From the desk
Twelve of the last thirteen trades have been winners. Seven in a row from August 24, one NAS100 short that stopped out on August 31, and five in a row since, running from Tuesday's three shorts through yesterday's Cable short and today's long.
The direction of this one is the part that matters most. August closed at -2.67R, the system's first losing month since inception, and the damage was almost entirely on the long side: the GBPUSD long book took ten trades, won two and gave back 6.18R, while the short books collectively finished the month up 3.52R. On August 27 every instrument trader was split into independent long and short books so a one-sided failure could no longer hide inside a pooled average, and the underperforming long books were taken offline.
NAS100 long was not one of the books that came off. It lost 2.13R across four trades in August, a genuinely bad month for it, and it kept its allocation because four trades is not a sample and the losses came from ordinary variance rather than a structural mismatch. Today it returned +2.84R in 53 minutes.
That is what the split was for. Not to switch off every book having a bad month, but to be able to tell the difference between a book that is losing to variance and a book that is on the wrong side of a regime. The year stands at +28.75R across 220 trades, and it is a better number today than it was a week ago.
At a Glance
What this teaches about AI-driven trading
What does twelve of thirteen actually mean?
Of the last thirteen trades the system has taken, twelve closed as winners and one closed as a loss. That run began on August 24 and covers eight trading days through today. It is trade count rather than calendar days, and the single loss was a NAS100 short that stopped out for -1R on August 31.
Why enter at 62% confidence when the desk usually wants more?
The score measures how much the inputs agree, not how good the setup is. Here the Macro Agent read lean_bear at 71% while the Trend Agent read bullish at 66%, and that genuine disagreement discounts the score. The position was sized down to respect it. The trade still went the full distance to the third target, because the tape confirmed the trend read rather than the macro one.
What does +2.84R mean in dollars?
R is the trade's risk unit, the distance from entry to stop, here 78.9 points. On a $100,000 account risking 2% per trade, 1R is $2,000, so +2.84R is roughly $5,680. Reporting in R rather than dollars keeps results comparable across account sizes and across instruments whose point values differ.
Should the system have skipped a trade its own Macro Agent opposed?
Not when the market contradicts the macro read. The bearish case required rising yields, a firm dollar and building fear, and all three were moving the other way at the time of entry. Standing aside is correct when two signals genuinely conflict; it is not correct when one of them describes conditions that are not present in the tape.
How do you count R on a trade that reaches a third target?
This case study reports full potential, the R distance to the furthest target price reached, here the third target at 29,486. Our weekly, monthly and year-to-date recaps use a stricter TP1 baseline that credits only the first target, so the same trade enters those totals at +0.89R. The two numbers are deliberately different and we never mix them.
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 Cable Short That Entered on Its Lowest Confidence Reading
Four refusals at 79, 81, 82 and 84 percent. Then the system entered at 62, the weakest score of the sequence, and took TP1 for +1.02R. The number was never the trigger.
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 the best return per point by a distance. Price travelled 31.7 points. The risk band was 7.8. That ratio is the entire trade.
The Nasdaq Short That Said Wait Eight Times Before It Said Enter
Nine evaluations in twenty-seven minutes. Eight declined at 38 to 42 percent confidence while the macro case was already overwhelming. The ninth cleared at 68 percent, and price ran 330 points to the third target.