Waited at 88, Entered at 72: Why SkyAnalyst Sold the Dow on Its Lowest Score
Three reads scored 84, 88 and 82 percent and all said wait. The fourth scored 72 and sold US30 at 50,785. TP1 printed 16 minutes later: +0.87R (TP1).
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.
Eighty-eight percent was the highest score SkyAnalyst put on Thursday's Dow short, and it came attached to the word wait. Over five minutes of the New York morning the system read the US30 setup four times. The first three reads scored 84, 88 and 82 percent, and all three declined to enter. The fourth scored 72, the lowest of the four, and sold the Dow. The order filled at 50,785 at 10:45 ET, the bottom of the planned 50,785-50,800 zone, with the stop at 50,928, 143 points away. TP1 at 50,660 printed at 11:02 ET, 16 minutes after the fill, for +0.87R (TP1), or +$1,740 (TP1) on the simulated $100,000 account in the returns panel. The broker closes the whole position at TP1 and TP2 never printed, so the full-potential R and the realized R are the same number on this trade. This is a small winner, and we picked it for the order of those four scores and for what price did after 11:02 ET. Yesterday's Nasdaq long showed a score climbing into its entry. This one shows a score falling into its entry, and the explanation is the same in both cases. If you want an analysis that holds to its own trigger, see SkyAnalyst run your markets on a 21-day free trial.
A Dow already below yesterday's low
The Macro Agent's 10:02 ET refresh had the US index group at lean bear, 74 percent, and the Dow itself at lean bear with a bias score of -35 and 76 percent confidence. Its reasons were specific to the Dow's mix of industrials and financials: the 10-year yield at 5.321% and rising, the Dollar Index at 101.72, and Brent at $100.89. In its read, a softer August PCE print, 3.4% headline, did not offset them. The Dow sat near 50,873, about 0.9% below its 5-day EMA of 51,352 and under Wednesday's low of 50,935, after a cash close down about 0.86%. The same refresh read the Nasdaq as bullish at 73 percent on chip and AI strength, which made the Dow the weaker index of the two.
The Trend Agent, at 10:12 ET, called US30 bearish at 72 percent in a trending regime. It noted that the earlier rebound had stalled under the 60-minute slow EMA and reversed hard from the 51,220-51,240 zone. It placed key resistance and invalidation at 50,935.2, VWAP at 50,967.6 and key support at 50,661.9. The setup analysis added the breadth read: NYAD at -1052 against its 5-day EMA of -679.8, which it called deteriorating participation rather than a healthy dip, with VIX at 17.14 above its 5-day EMA of 16.30. On the 60-minute chart RSI was 37.7 and MACD sat below zero. On the 15-minute chart RSI was 36.1.
What the analysis flagged against the short
The analysis named its own risk: the 5-minute RSI had rebounded off oversold, so a squeeze toward 50,875 was possible before the trend resumed. It also told itself not to chase a fresh flush into the 50,660 support. One label deserves a note. The analysis described breadth and VIX together as a risk-off compound, while the Macro Agent's regime field, with VIX at 16.86, still read normal. We do not adopt that label in our own voice. Structure, breadth and the macro bias all pointed the same way.
Failed Reclaim in a Retracement Pocket. This is a continuation pattern for a market that is already falling. Instead of selling the low, the trader marks the zone where a bounce is likely to stall, waits for price to climb into it, and sells only when price tries to push above a nearby level and fails. The failure, not the arrival, is the signal.
The pocket
The analysis marked a retracement pocket at 50,780-50,820, built around the 78.6% Fibonacci retracement near 50,780. On the 5-minute chart price was still below its EMAs and below VWAP, so the bounce into that pocket read as a retracement, not a reversal. The entry zone was 50,785-50,800, inside the pocket.
The trigger
The trigger asked for a 5-minute rejection candle or a lower-high failure inside the pocket, while price stayed below VWAP and the 5-minute and 15-minute EMAs. The preferred version was specific: a failed reclaim of 50,820, followed by a move back under 50,780. Price touching the pocket did not qualify. Price had to try the top, lose it and come back through the bottom.
The stop and the targets
The stop went to 50,928-50,932, just inside the Trend Agent's invalidation at 50,935.2. The targets were 50,660, 50,520 and 50,400. TP1 sat 1.9 points under the Trend Agent's key support at 50,661.9, and the analysis warned against chasing any flush into that level.
Why there was no long
The analysis checked the other side and closed it: breadth was negative, VIX was elevated, and the plan for the New York morning was short only. It added one more rule for the short itself. If price lost 50,660 without a retest, skip the chase and wait for a clean breakdown and retest.
SkyAnalyst doesn't favor any single strategy. Yesterday it sold a failed retest on the Dow and bought a pullback on the Nasdaq. Today it waited for a failed reclaim inside a Fibonacci pocket. Each plan comes from the chart in front of it, with the trigger written down before price arrives.
NY breadth is the primary read and it is decisively bearish: NYAD -1052 vs 5-day EMA -679.8, below its EMA and below yesterday’s low, showing deteriorating participation rather than a healthy dip. VIX 17.14 vs 5-day EMA 16.30 is elevated and above yesterday’s high, so this is not a clean low-volatility breakout tape; it is a wider-stop, risk-off environment. That breadth/VIX combination creates a risk-off compound, so only shorts qualify.
Macro confirms the technical picture. The Macro Agent is lean_bear (76% confidence, tradeability 82), citing price below the 5-day EMA and below yesterday’s low, with US10Y 5.31% above its 5-day EMA and above yesterday’s high, DXY above its 5-day EMA and above yesterday’s high, and oil above yesterday’s high. That mix pressures cyclicals and multinationals. Regime classification is risk-off, with maximum directional conviction because both NYAD and VIX confirm the same side.
Trend structure also agrees. The Trend Agent is BEARISH (72%), TRENDING, with R=50935.2, S=50661.9, VWAP=50967.6, invalidation=50935.2. On 60m, price is below both EMAs, RSI is weak at 37.7, and MACD is below zero and below signal. On 15m, price is below both EMAs, RSI 36.1, MACD histogram negative. On 5m, price remains below EMAs and below VWAP; the current bounce is only a retracement into the 78.6% fib zone near 50780, not a bullish reversal. The 10:00 ET data/speaker window has passed, so entries are allowed again, but this is a sell-the-bounce / sell-the-failed-reclaim tape, not a buy-the-dip tape.
Directional Bias: Bearish
Volatility: High
Setup #1: US30 SHORT
- Directional Bias: Bearish continuation
- Entry: 50785-50800
- Entry Trigger: 5m rejection candle / lower-high failure in the 50780-50820 retracement pocket, while price stays below VWAP and below 5m/15m EMAs. Best trigger is a failed reclaim of 50820 followed by a move back under 50780.
- Stop Loss: 50928-50932
- Targets: TP1=50660, TP2=50520, TP3=50400
- R-Multiples: TP1≈1.0R, TP2≈2.0R, TP3≈2.8-3.0R
- Quality Score: 8.6/10
- Confidence: High, 7/7 confluences
- NYAD bearish and below 5-day EMA
- VIX above 5-day EMA and elevated
- Macro Agent lean_bear with confidence >60
- Trend Agent bearish with confidence >60
- 60m bearish EMA/MACD structure
- Entry zone sits at fib resistance with visible 5m reaction
- No high-impact USD event within 30 minutes
- Risks: 5m RSI rebounded off oversold, so a squeeze toward 50875 is possible before trend resumes. Do not chase a fresh flush into 50660 support.
- Invalidation Condition: Any sustained 5m reclaim/close above 50935.2; additionally, if NYAD sharply improves back toward/above its EMA while VIX fades, bearish conviction drops materially.
No long setup. Breadth is negative, VIX is elevated, and the risk-off compound veto keeps the NY AM bias short-only. If price loses 50660 without a retest, skip the chase and wait for a clean breakdown-retest sequence.
Decision log
14:37 UTC, 84 percent, WAIT. Price had entered the pocket seconds after monitoring began, but the 5-minute candle was still forming and had pushed back above 50,780 toward 50,815. The system saw no rejection candle and no failed reclaim followed by a move back under 50,780. The bearish context held, below VWAP and both EMAs with MACD under zero, but it called an entry here anticipatory rather than confirmed.
14:39 UTC, 88 percent, WAIT. The highest score of the session came with a refusal. The forming candle was pushing up into 50,819, near the top of the pocket, with RSI rebounding. The system judged the squeeze risk too high to short before confirmation and kept waiting for a completed rejection or a move back below 50,780.
14:40 UTC, 82 percent, WAIT. The latest 5-minute candle reached 50,820 and closed near 50,806. The system read it as a retracement candle rather than a clear rejection, and price had not moved back under 50,780. With the 5-minute RSI still rebounding, it named the squeeze risk again and asked for a visible rejection or a break below 50,780.
14:42 UTC, 72 percent, ENTER. Price had probed above 50,820 to 50,824.7, failed to hold the reclaim and rotated back under 50,780, still below VWAP and both EMAs with MACD under zero on both timeframes. That was the trigger as written. The system named two cautions, a candle still forming and price slightly below the ideal band, and entered anyway. The order filled at 50,785 at 14:45 UTC.
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 | +0.87R | +$1,740 |
| TP2 hit (not tracked) | +0R | +$0 |
| TP3 hit (max potential) (not tracked) | +0R | +$0 |
All six agents combined.
What this trade teaches
The score is not a countdown. On the Nasdaq yesterday the score rose as price came back to the plan. Here it fell on the entry read, from 82 to 72, at the moment the plan was met. The three waits scored higher, but none of them had the thing the setup asked for: a failed push above 50,820 and a return under 50,780. The entry read also listed two cautions of its own and entered anyway. The gate was the trigger. A high number without the trigger stayed a wait.
The rejection/failure pattern gives enough edge to allow entry now. SkyAnalyst entry evaluation, 14:42 UTC
The second lesson is in what happened after 11:02 ET. Price kept falling through the 50,661.9 support to 50,582.3 at 11:11 ET, 62.3 points short of TP2. Then it turned, and by 12:33 ET it had climbed 345.7 points back to the 50,928 stop. The position was not there for any of that. The broker had closed it in full at TP1. A position held for TP2 would have ended at the stop, because TP2 never printed. Tuesday's EURUSD short made the same trip from a further target. What drove the Dow back up is not in our data, and we will not supply a reason for it.
From the desk
The honest line on this trade is +0.87R (TP1). It is a small one, and the 16 minutes it lasted are not the interesting part. The interesting part is the five minutes before it, when the system held a high score and still said no.
This is the third Dow short we have written up since September 23. Yesterday's US30 short entered on its first read because its trigger had already printed before monitoring began. The short on the 23rd sold a failed reclaim with breadth on its side. This one waited three reads for its trigger to finish. One read or four, the question asked of the chart was the same, and one trade settles nothing on its own.
At a Glance
What this teaches about AI-driven trading
What is a failed reclaim short?
It is a setup in a falling market where price bounces, pushes briefly above a nearby resistance level, cannot hold above it, and drops back below the zone. Traders sell that failure with a stop above the resistance. The logic is that buyers had a clear chance to reclaim the level and lost it, which often hands control back to sellers and the earlier downtrend.
Why would a trading system wait when its confidence score is high?
A confidence score and an entry trigger answer different questions. The score reflects how the system reads the overall picture, while the trigger is a specific event the plan requires, such as a rejection candle closing back below a level. If the event has not happened yet, entering means guessing that it will. Waiting until it prints keeps the entry tied to the written plan.
What is a retracement pocket?
A retracement pocket is a price zone where a bounce inside a downtrend is expected to stall, often built around a Fibonacci level such as 61.8% or 78.6% of the prior move. Traders use it to locate a short entry with a nearby stop. Price reaching the pocket is not a signal by itself. Traders wait for a rejection or failure inside it before selling.
How does a full close at the first target change a trade's outcome?
Closing the whole position at the first target locks in a smaller gain and removes exposure to whatever price does next. If the market later reverses through the entry to the stop, the closed position is unaffected. The tradeoff is that a trade which would have run to the second or third target books only the first. Both outcomes follow directly from the exit rule chosen in advance.
When does a forming candle matter for an entry decision?
A candle that has not closed can still change shape. A bar that looks like a rejection with two minutes left can finish as a strong close in the other direction. Many traders require a closed candle before acting on a pattern. Others accept a forming candle when price has already done what the trigger asked, treating the open bar as a stated risk rather than a reason to wait.
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
2026 So Far: +38.35R, and What Nine Months Taught Us About Risk
Two hundred and fifty-seven trades since the January 12 inception, 148 winners, +38.35R net. Three threads ran through the year: how the edge was built, how we learned to see it, and what it is still exposed to.
September 2026: Thirty-Two Shorts, One Thesis, and Two Bad Sessions
Thirty-seven trades, 23 winners, +9.59R. Almost all of it came from one directional read, and so did almost all of the pain: when the short book was wrong, it was wrong everywhere at once.
Long Nasdaq, Short Dow: The US30 Short That Reached All Three Targets
Thirty-four minutes after buying the Nasdaq, SkyAnalyst sold the Dow at 51,422.2 on a failed retest under VWAP. TP3 printed at 3:48 PM ET for +3.40R (TP3). The record books +1.04R (TP1).