Yields at 5.054%, RSI at 13.8: Why SkyAnalyst Sold the Nasdaq Bounce
SkyAnalyst AI journal entry: NAS100 short on Sep 23, 2026 ran to +3.02R (TP3) full potential and closed +0.81R (TP1) realized. 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 Nasdaq short the system took on its first evaluation.
Oversold does not mean buy. SkyAnalyst's Nasdaq analysis wrote that line on Wednesday with the 15-minute RSI between 13.8 and 19.6, the index about 400 points under the prior close of 30,741, and the 10-year Treasury yield at 5.054 percent, a fresh 5-day high. For the most rate-sensitive US index, a spiking yield vetoes longs outright. The analysis also refused the obvious alternative. Do not chase shorts at 30,400, it wrote, because price was sitting on the prior daily support with every momentum gauge stretched. What it planned instead was a pullback short: let the selling exhaust, let price bounce back toward the level it broke from, and sell there. The zone was 30,550 to 30,595. Twenty-two minutes later price rallied into the bottom edge of it, the single evaluation entered at 68 percent confidence, and the order filled at 30,550.5 with a stop at 30,650. The trade never went more than 2.7 points against the entry. TP1 at 30,470 was hit 49 minutes after the fill, and TP3 at 30,250 printed at 07:03 UTC on Thursday, for a full-potential move of +3.02R (TP3). The broker closes the whole position at TP1, so the result logged to our track record is +0.81R (TP1). Tuesday's Nasdaq pullback long bought the same index on a retracement, and the same session's Dow short sold a sibling index into a broken low. By Wednesday the Nasdaq had turned lower, and the entry rule had not changed: let price come back to the level, then trade it. If you want a morning read that tells you when oversold is a warning and not an invitation, see SkyAnalyst run your markets with a 21-day free trial.
A rates shock the Nasdaq could not ignore
The 10-year yield opened the story. At 5.054 percent it sat 6 basis points above its 5-day EMA of 4.994 percent, above the prior day's high of 4.984 percent, and through the 5.000 percent high from three days earlier. The analysis treats the 10-year as the primary driver for NAS100, and it called this reading maximum bearish. The dollar agreed: DXY at 101.052 was above its 5-day EMA of 100.58 and above the prior day's high of 100.685. Gold was down $75 on the day at 4,284, consistent with a rates and dollar move.
Breadth ruled out the easy explanation. The advance-decline line (ADD) read -1,354 against a 5-day EMA of -280.6, and the analysis wrote that this was broad selling pressure, not a rotation out of tech. VIX was the one soft confirmation. At 14.79 it was up about 4 percent from the prior close of 14.22 but still under its 5-day EMA of 15.02, so volatility was reacting without panicking.
The Macro Agent disagreed with all of it. Its NAS100 read was strong bull at 82 percent, built before the yield spike and 87 minutes old when the analysis ran. The Trend Agent flagged that read as a headwind, and the analysis overrode it on the live yield and dollar data, at the cost of one confluence. The Trend Agent itself was bearish at 82 percent in a strong-trend regime, with the 60-minute RSI at 26.7 and the 60-minute MACD histogram at -43.2. Price was 285 points under VWAP on the 15-minute chart, which is exactly why the analysis would not sell it there.
Pullback Short to a Breakdown Zone. The pattern sells a market that has broken hard to the downside, on the bounce back toward the level it broke from. The breakdown shows who is in control; the bounce gives a price where a stop above structure is reasonable and the targets below are far enough away to pay for it. Professional traders use it because a falling market that is already stretched tends to snap back before it continues, and selling the snapback beats selling the hole.
The zone was drawn from four levels
The entry zone was 30,550 to 30,595. It stacked the breakdown low of the 13:00 UTC candle at 30,559.8, the open of the New York 60-minute candle at 30,594.1, the Trend Agent invalidation cited in the analysis at 30,596.5, and a Fibonacci retracement of the last leg. From the 15-minute swing high near 30,738 to the session low of 30,369, the 50 percent level sat at 30,554 and the 61.8 percent at 30,597.
Three ways in, and three ways out
The analysis wrote three triggers: a 5-minute rejection candle that wicked above and closed back below 30,560, a failure to reclaim the 5-minute EMA9 inside the zone, or a 5-minute lower high at that level. It also wrote the exits before any entry existed. A drop in the 10-year back under 5.00 percent would cancel the idea. A 5-minute RSI above 50 with the 15-minute histogram turned positive would mean the structure had broken. A straight break of 30,369 with no bounce would mean no trade at all.
That last rule matters most. A pullback short that never gets its pullback is not a trade to chase later.
The stop above structure, the targets at the next shelves
The stop went to 30,650, 13.4 points above the 30,636.6 level the Trend Agent marked as key resistance and invalidation at entry time. On the 30,550.5 fill that was 99.5 points of risk. TP1 at 30,470 was a cluster of 15-minute lows from the 14:00 UTC candle. TP2 at 30,370 sat at the session low, just under the prior daily support at 30,393.9. TP3 at 30,250 was an extension below that support, conditional on yields staying high.
Where the plan read better than it measured
The analysis priced its targets from a mid-zone entry near 30,570 and wrote about 1.1R, 2.1R and 3.4R. The fill came at the bottom of the zone, the edge closest to the targets, so the risk grew and every target shrank. Measured from the real fill the targets sat at 0.81R, 1.81R and 3.02R. The record carries the measured numbers.
SkyAnalyst doesn't favor any single strategy. On Tuesday it bought a Nasdaq breakout on the pullback, and on Wednesday it sold a Nasdaq breakdown on the bounce. It reads the tape first and takes whichever pattern the session offers.
NAS100 NY AM Session Analysis: Wed, Sep 23, 2026
Market Environment Summary
The NAS100 is experiencing a sharp intraday selloff driven by a violent spike in the 10-Year Treasury yield, which has surged to 5.054%: well above its 5-day EMA (4.994%) and decisively above yesterday's high of 4.984%, making a fresh new 5-day high. This is the single most important signal for NAS100: yields are spiking, and as the most rate-sensitive US equity index, the Nasdaq is responding accordingly, having fallen approximately 400 points from yesterday's close of 30,741 to the current area near 30,400.
Cross-asset confirmation is at maximum conviction:
- DXY has surged to 101.052, well above its 5-day EMA (100.58) and above yesterday's high of 100.685, a fresh breakout representing a powerful dollar headwind for tech/growth.
- VIX at 14.79, slightly below its 5-day EMA (15.02), but rising sharply from yesterday's close of 14.22 (+4%), partial bearish confirmation. VIX is not yet screaming panic but is clearly responding to the move.
- Gold down sharply (-$75 from yesterday's close to 4,284), consistent with a rates/dollar shock.
- Advance/Decline (ADD) at -1,354, deeply negative and well below its 5-day EMA (-280.6), confirming broad market weakness, this is not NAS100 rotation, it is broad selling pressure.
The Macro Agent reads "strong_bull" with 82% confidence, but this is stale (87 minutes old) and critically, the Trend Agent correctly flags it as a HEADWIND: "Macro agent bias is bullish for NAS100, but current tape is dominated by a strong-dollar/high-yield backdrop with DXY and US10Y above yesterday's highs." The macro read was anchored to the pre-selloff breakout structure near 30,678 and has not yet updated for today's yield spike and price collapse. I am overriding the Macro Agent's bullish bias based on the real-time yield/DXY data, this is a bearish rates-driven environment.
Trend Agent is BEARISH at 82% confidence, regime STRONG_TREND, invalidation at 30,596.5. Price at ~30,401-30,434 is deeply below VWAP (30,728), below yesterday's low (30,730), below the 5-day daily EMA (30,407-30,412), and making new session lows. The 60m RSI is oversold at 26.7, and 15m RSI is deeply oversold at 13.8-19.6. MACD histograms are strongly negative across all timeframes.
No high-impact USD events remain today (all Euro PMIs released pre-market). Tomorrow's Unemployment Claims at 8:30am ET is the next catalyst. The calendar is clear for trading.
Directional Bias: Bearish Volatility: High (expanding, 60m ATR rising to 47.9 pts, 5m ATR at 40.9 pts, 15m ATR high at 41.1 pts)
Step-by-Step Confluence Analysis
1. 10Y Yield Assessment
- Current: 5.054%, above 5-day EMA (4.994%) by 6 bps
- New 5-day high: Yes (prior high was 5.000% from 3 days ago; today's high is 5.058%)
- Signal: Maximum bearish for NAS100. Yields are spiking, not just drifting. No longs can be proposed.
2. Cross-Asset Confirmation
| Indicator | Current | 5d EMA | Signal |
|---|---|---|---|
| US 10Y | 5.054% | 4.994% | ⬆ BEARISH, new 5d high |
| DXY | 101.05 | 100.58 | ⬆ BEARISH, above yesterday's high |
| VIX | 14.79 | 15.02 | ⚠ Mixed, rising but below EMA |
| ADD | -1,354 | -280.6 | ⬇ BEARISH, broad selling |
| Gold | 4,284 | 4,341 | ⬇ Consistent w/ rates shock |
Verdict: 10Y + DXY = double headwind confirmed. VIX partial. ADD confirms broad selling, not rotation. This is near-maximum bearish conviction for NAS100.
3. Trend Structure & Key Levels
- Trend Agent: BEARISH, 82% confidence, STRONG_TREND regime
- Invalidation: 30,596.5
- Key Levels: R=30,636.6 | S=30,393.9 | VWAP=30,727.8
60m EMA alignment: Price (30,434) is well below EMA fast (30,670) which remains above EMA slow (30,513), the EMAs haven't crossed yet but price is deeply below both, characteristic of a waterfall move that leads the moving averages. The bearish crossover is imminent. RSI at 26.7 (oversold), MACD histogram at -43.2 (strong negative, accelerating).
Daily context: Price has gapped down ~100 pts from yesterday's close (30,741) and has extended the gap further. Price is now below the 5-day daily EMA (~30,407-30,412) and below the prior day's low by over 300 points. The daily pivot point sits at 30,636.6 with the previous daily support at 30,393.9, price is now testing this exact level (today's low: 30,369).
Pre-market gap: Opened near 30,820 (today's high), fell steadily through London session. This is not a gap-fill scenario, it's a sustained directional sell.
4. Lower-Timeframe Entry Analysis
15m structure: All recent candles show persistent lower highs/lower lows. EMA9 well below EMA21 (bearish alignment). RSI at 19.6 (deeply oversold). MACD at -67.95 with histogram at -28.76 (strong negative). Price is 285 points below VWAP.
5m structure: Ten consecutive candles all below EMA9, below VWAP, with RSI ranging 15-25 (deeply oversold). MACD histogram has been expanding negatively but the latest candle shows histogram contracting slightly (-20.32 vs -25.54 prior), suggesting the most intense selling pressure may be easing into a potential minor consolidation/bounce, but this does NOT indicate reversal.
Critical observation: While the trend is clearly bearish and yields confirm shorts, chasing at deeply oversold levels (RSI 15-20 on 5m, 13-19 on 15m, 27 on 60m) into the prior daily support at 30,393.9 carries significant adverse-excursion risk. The setup here is a pullback short, not a chase short.
5. Confluence Gate: Short Setup
| # | Factor | Status | ✓/✗ |
|---|---|---|---|
| i | 10Y yield supports short | 5.054% > 5d EMA, new 5d high | ✓ |
| ii | Macro Agent aligns (≥60, rate factors) | Agent says bullish (STALE), but real-time yields/DXY bearish, override | ✗ |
| iii | Trend Agent aligns (≥60) | BEARISH, 82% confidence | ✓ |
| iv | 60m EMA stack/crossover confirms | Price below both EMAs, crossover imminent | ✓ |
| v | Price at VWAP/Fib/session level w/ reaction | Price at prior daily support 30,393.9, needs pullback to structure for entry | Conditional ✓ |
| vi | 15m RSI <50, MACD expanding | RSI 19.6, MACD histogram -28.76 strong | ✓ |
| vii | No high-impact events ±30 min | Calendar clear | ✓ |
Score: 5-6/7 (depending on entry location) → Medium-High to High
The Macro Agent miss (stale bullish) costs a point, but real-time yield data more than compensates. On a pullback entry with price reacting at structure, this reaches 6/7.
Trade Setup
The setup is a pullback short: waiting for price to retrace toward the 30,559-30,596 breakdown zone before entering, rather than chasing into deeply oversold conditions at the 30,369-30,394 daily support. This zone represents:
- The 13:00 UTC candle low breakdown (30,559.8)
- The Trend Agent invalidation area (30,596.5)
- The 60m hourly candle open of the NY session candle (30,594.1)
- The 15m Fibonacci retracement zone from the last lower high
The logic: NAS100 routinely produces 38.2%-50% retracements in strong trend moves before continuing. From the 30,738 area (last 15m swing high at 13:15 UTC) to today's low of 30,369, the retracement levels are:
- 38.2%: 30,510
- 50.0%: 30,554
- 61.8%: 30,597
The 30,555-30,597 zone clusters with the breakdown level, the 5m EMA9 catch-up zone, and the Trend Agent invalidation. This is the ideal short-entry zone.
Setup #1: NAS100 SHORT, Pullback to Breakdown Zone
- Entry Zone: 30,550-30,595 (wait for price to retrace into this zone)
- Entry Trigger: Bearish 5m candle rejection (wick above with close back below 30,560), or failure to reclaim the 5m EMA9 within the zone, or a lower-high formation on 5m at this level
- Stop Loss: 30,650 (15 pts above the Trend Agent invalidation of 30,596.5, beyond the 61.8% Fib at 30,597, and above the 60m structural resistance at 30,636.6, buffers for NAS100 overshoot)
- Targets:
- TP1 = 30,470 (~1.0R-1.25R, today's 14:00 UTC 15m candle low cluster / round number support)
- TP2 = 30,370 (~2.0R, today's session low / prior daily support 30,393.9 with 20pt overshoot)
- TP3 = 30,250 (~3.0R+, extension target below daily support, only if yields remain elevated and no reversal signals appear)
- R-Multiples: TP1 ≈ 1.1R, TP2 ≈ 2.1R, TP3 ≈ 3.4R (based on ~85pt stop from 30,570 midpoint entry to 30,650 stop)
- Quality Score: 7.5/10
- Confidence: Medium-High, Strong yield/DXY/trend confluence, but deeply oversold conditions create bounce risk that could stop the setup from triggering if the pullback never reaches the entry zone. The stale Macro Agent bullish read introduces a small uncertainty discount.
Risk Management Notes
- 60m ATR: 47.9 pts, the 80pt stop exceeds 1x ATR (minimum requirement satisfied at 1.67x ATR), appropriate for expanding volatility.
- Position Sizing: With high volatility (5m ATR at 40.9 pts), reduce from standard 1% risk to 0.5%-0.75% equity risk per this trade. The wide stop accommodates NAS100's overshoot tendency but demands smaller position size.
- Management: Move stop to breakeven after TP1 hit. Trail stop by 5m EMA9 after TP2.
- No-Trade Condition: If price reaches the entry zone but RSI on 5m has already climbed above 50 and MACD histogram has flipped positive on 15m, the bearish structure is broken, skip the trade. If price never pulls back to 30,550+ and simply breaks below 30,369 directly, do not chase.
- If 10Y yield reverses sharply below 5.00% before entry triggers, the bearish thesis weakens materially, cancel the setup.
What NOT to Do
- Do not go long. 10Y yields are spiking above their 5-day high. Per the framework, this is an absolute prohibition on longs regardless of oversold technicals. Oversold does not mean "buy", it means "wait for a pullback to short."
- Do not chase shorts at 30,400. You are 300+ points below VWAP, RSI is 15-20 on 5m, and you're sitting on prior daily support. The risk of a snapback squeeze is elevated.
If the pullback short does not trigger within the next 60-90 minutes (by approximately 12:00 PM ET), the setup quality degrades as we move past the optimal NY AM window. Re-evaluate fresh at that time.
Decision log
14:56 UTC, 68 percent, ENTER. Price touched the zone at 14:55:48 UTC and the evaluation ran about 15 seconds later with price at 30,550.2, the zone floor. The last closed 5-minute candle had rallied from 30,442 to 30,539 and price had just crossed above the 5-minute EMA9 near 30,524, so the EMA9 trigger was not cleanly met. The system weighed that against a 15-minute RSI bouncing from 24 to 37 inside a strong downtrend, called the entry acceptable but not perfect, and trimmed its confidence for the missing rejection wick. The order filled at 30,550.5 at 14:59 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 | +0.81R | +$1,620 |
| TP2 hit | +1.81R | +$3,620 |
| TP3 hit (max potential) | +3.02R | +$6,040 |
All six agents combined.
What this trade teaches
Oversold readings answered one question and only one. RSI at 13.8 on the 15-minute chart said the selling was stretched and that a bounce was likely. It did not say the trend had turned, and with the 10-year at a 5-day high the analysis would not consider a long at any price. The trade used oversold as timing information: wait for the snapback, then sell into it at the level the market had broken.
The second lesson is about where inside a zone the fill lands. On a short, the bottom of the entry zone is the worst price in it, because it sits closest to the targets. The analysis planned from about 30,570 and got 30,550.5, and TP1 went from roughly 1.1R on paper to 0.81R measured. That matters because TP1 is what the ledger books. The trade was right and the logged result was still under 1R.
Entry is acceptable here but not perfect. The bounce is still active and no clear rejection wick has formed yet. SkyAnalyst entry evaluation, 14:56 UTC
The caveats belong beside the win. The evaluation entered without the rejection candle the first trigger described, and it said so. The Macro Agent's read was bullish and stale, and the trade overrode it rather than waited for it to update. The setup carried a C+ grade on the workspace. The execution itself was clean: the worst price after the fill was 30,553.2, 2.7 points against the entry and 96.8 points short of the stop.
The full arc took most of a day. TP1 filled at 15:49 UTC on Wednesday, TP2 at 30,370 printed at 05:57 UTC on Thursday, and TP3 at 30,250 at 07:03 UTC, 16 hours after the entry. The +3.02R (TP3) measures how far the market ran in the trade's favor. The account closed at TP1, and the ledger holds +0.81R (TP1).
From the desk
We picked this trade because it refuses two trades before it takes one. A falling Nasdaq with RSI in the teens tempts buyers who see a bargain and sellers who see momentum. The analysis turned both down in writing, named the price it wanted, and then took it 22 minutes later when the bounce arrived.
This is one trade, and one trade proves very little. It sits next to our September 10 Nasdaq Fibonacci short and the August 20 Nasdaq short into a failed bounce as the same idea in a different tape: sell the rebound inside a downtrend, not the low. What this one adds is a rates shock doing the directional work, and an honest look at what a fill at the zone floor costs the ledger.
At a Glance
What this teaches about AI-driven trading
What is a pullback short to a breakdown zone?
It is a trade that sells a falling market after price bounces back toward the level it broke through. The broken support tends to act as resistance on the retest. The trader waits for the rebound, sells near that level, and places the stop above nearby structure, so the risk is defined by a price the market has already rejected once.
Why is an oversold RSI not a reason to buy in a downtrend?
RSI measures how stretched recent moves are, not which way the trend points. In a strong downtrend it can stay oversold for a long time while price keeps falling. An oversold reading mostly warns that a bounce is likely soon. Trend traders often use it as a signal to stop selling at the lows and wait for that bounce to offer a better entry.
How do rising Treasury yields affect the Nasdaq 100?
Much of the Nasdaq 100's value sits in growth companies whose earnings are expected further in the future. Higher yields reduce what those future earnings are worth today, so the index tends to be the most rate-sensitive of the major US benchmarks. When the 10-year yield breaks to new short-term highs, many traders treat that as a strong bearish input for the index.
What does it cost when a short fills at the bottom of its entry zone?
For a short, the lowest price in the entry zone is the least favorable fill, because it sits closest to the profit targets. The stop stays where it is, so the risk per trade grows and every target is worth fewer R. On this trade a planned mid-zone fill near 30,570 became 30,550.5, and TP1 fell from about 1.1R to 0.81R.
When should a pullback setup be cancelled instead of chased?
A pullback setup should be cancelled when price never retraces to the planned zone and simply breaks further, when the conditions behind the direction reverse, or when time runs out. Chasing a move that skipped the pullback means a worse price, a wider stop, and usually a weaker reward. Writing those cancel rules down before the entry keeps the decision mechanical.
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
Take TP1 Aggressively: The GBPUSD Short That Ran to TP2 Anyway
Cable had fallen for three hours when SkyAnalyst sold the bounce at 1.3263 and told itself to bank TP1 fast. TP2 printed 1 hour 43 minutes later for +1.63R (TP2); the record books +0.74R (TP1).
Bull Tape, Bear Dow: Why SkyAnalyst Shorted US30 Against the Group Read
Our index macro read leaned bull at 68 percent. NYSE breadth and the Dow's own read said sell. SkyAnalyst shorted 51,827.4 on its first evaluation, and TP1 filled 28 minutes later for +1.02R (TP1).
No Chasing at 30,705: The NAS100 Long SkyAnalyst Bought on the Pullback
The morning read called the Nasdaq breakout too stretched to buy. Eleven minutes later SkyAnalyst filled 69.7 points lower at 30,635.3, and TP2 printed overnight for +2.45R (TP2).