Four Tenths of a Pip Short: Why SkyAnalyst Sold Cable Anyway
SkyAnalyst AI journal entry: GBPUSD Short on Oct 5, 2026 closed +0.72R on TP1. 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 plan for Monday's GBPUSD short named its trigger to the fifth decimal: a 5-minute close back below 1.32190. Over thirteen minutes of the New York morning SkyAnalyst read the setup eight times and declined to sell every time, because price kept closing above that line. On the ninth read the last completed candle faded to 1.32194, four tenths of a pip short of the trigger, and the system sold anyway. The order filled at 1.3219 at 11:24 ET, the bottom of the planned 1.32190-1.32235 zone, with the stop at 1.3237, 18 pips away. TP1 at 1.3206 printed at 12:11 ET, 47 minutes after the fill, for +0.72R (TP1), or +$1,440 (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 ninth read: how a system that had held its trigger to the decimal for eight reads decided a near miss was close enough, and what it said that would cost. Thursday's Dow short entered once price had done everything its trigger asked. This one entered a fraction short of that. If you want that reasoning written out on your own charts, see SkyAnalyst run your markets on a 21-day free trial.
A London selloff that had already found its low
London set the tone before New York arrived. Cable printed its London high at 1.32393, almost exactly the prior session's high of 1.32395, then rotated lower, broke below the prior session's low at 1.32290 and sold to 1.31904. The setup analysis called that structure trend-down rather than a balanced range, and drew its conclusion from where the selloff had stopped. The move had already reached a session extreme, so New York should not chase the lows. It should sell retracements back into the 60-minute VWAP and the underside of broken support.
The dollar side agreed. The setup analysis had the Dollar Index at 102.182, above its 5-day EMA of 102.026 and above the prior session's high, with the US 10-year yield also above its prior-session high. The Macro Agent's 10:04 ET refresh had the FX group at lean bear, 72 percent, and GBPUSD at lean bear with a bias score of -35 and 70 percent confidence. Its read on sterling was blunt: "largely a passenger on global rates and risk flows." With VIX at 15.53, under its 5-day EMA of 15.71, the analysis framed this as a dollar-trend short, not a risk-off one.
A bounce the Trend Agent was watching
By the time the plan was written, Cable was bouncing. The Trend Agent's 10:50 ET read had GBPUSD bearish at 68 percent in a trending regime. Its 11:00 ET refresh cut that to 62 percent, relabeled the regime as transitioning, and noted that the latest 5-minute and 15-minute candles had reclaimed VWAP. Shorts still had the edge, it said, but should be sized down until price either rejected the 1.3227-1.3239 resistance band or broke back below 1.3209/1.3190. The analysis took the hint: "the only qualifying NY AM idea is a conditional short on rejection, not a market sell."
Conditional Short on Rejection. This is a pattern professional traders use when the trend and the tape disagree. The higher timeframes point down, but the short-term chart is bouncing. Instead of selling into the bounce, the trader writes down the level where it should fail and the exact candle that would prove it failed, then sells only if that candle prints. Until it does, there is no trade.
The level
The level was the 60-minute VWAP at 1.3219, which sat under the support London had broken. The entry zone ran from 1.32190 to 1.32235, from VWAP to just above it, so a short would be sold into the retest rather than after price had already left.
The trigger
The analysis allowed two versions: a 5-minute wick rejection from 1.3220-1.3223, or a 5-minute close back below 1.32190 after price probed higher. Both asked for proof that buyers had tried the zone and lost. It also wrote the reason for the rule into its risks: 15-minute RSI had recovered above 50 and 5-minute momentum had improved, "so do not short blindly into strength."
The stop, the targets and the clock
The stop went to 1.32370-1.32388, inside the Trend Agent's hard invalidation at 1.32393. The targets were 1.32060, 1.31905 and 1.31820, the second sitting on the London low. The analysis added a time stop, no new entry after 11:30 ET, and a note on the first target: after the multi-hour selloff, TP1 "should be treated as a forced take-profit due to Cable's V-reversal tendency."
Why there was no long
The analysis checked the other side and closed it in one line. Longs failed the London bias, the dollar trend and both agent biases.
SkyAnalyst doesn't favor any single strategy. On Thursday it sold a failed reclaim inside a Fibonacci pocket on the Dow. On Monday it wrote a Cable short that would not exist unless one specific candle printed. The plan is built from the chart that morning, and the trigger is written down before price gets there.
GBPUSD NY AM Environment (8:00 AM-12:00 PM ET)
London set the day’s tone as bearish after a failed test of resistance. Cable printed a London high at 1.32393, almost exactly at yesterday’s high 1.32395, then rotated lower, broke below yesterday’s low 1.32290, and sold to 1.31904/1.31901. That makes London’s structure trend-down, not balanced range, but because the selloff already hit a session extreme, the preferred NY tactic is not chasing lows: it is selling retracements back into 60m VWAP / underside resistance.
Macro/regime also leans bearish:
- DXY 102.182 > 5-day EMA 102.026, and DXY is trading above yesterday’s high → USD trend supports GBPUSD downside.
- US 10Y above yesterday’s high reinforces the USD headwind.
- VIX 15.53 < 5-day EMA 15.71, so this is not a full risk-off panic, but it does not contradict shorts.
- Macro Agent: lean_bear, 70% confidence, tradeability high.
- Trend Agent: BEARISH, 68% confidence, trending regime, reduce size, with VWAP 1.3219 and invalidation 1.32393.
Lower timeframes are bouncing, so the only qualifying NY AM idea is a conditional short on rejection, not a market sell. Also, no new setup after 11:30 ET.
Directional Bias: Bearish
Volatility: Normal
Setup #1: GBPUSD SHORT
- Entry: 1.32190-1.32235
- Entry Trigger: 5m bearish rejection at/just above 60m VWAP 1.3219 and underside of broken support, such as:
- a 5m wick rejection from 1.3220-1.3223, or
- a 5m close back below 1.32190 after probing higher
- Stop Loss: 1.32370-1.32388
- keeps minimum 15 pip structural risk
- stays inside Trend Agent invalidation 1.32393
- Targets: TP1=1.32060, TP2=1.31905, TP3=1.31820
- R-Multiples: TP1≈1.0R, TP2≈2.0R, TP3≈2.5R
- Quality Score: 7.2/10
- Confidence: Medium-High - London bearish bias, DXY supports downside, Macro Agent bearish, Trend Agent bearish, 60m structure remains below broader resistance/VWAP context, and there is no high-impact event within 30 minutes. Main risk is that 15m/5m momentum is still in a corrective bounce, so wait for rejection confirmation and use reduced size.
Confluences (5/7 now, 6/7 on trigger):
- London session bias alignment: Yes
- DXY supports trade: Yes
- Macro Agent bias aligns ≥6/10: Yes
- Trend Agent aligns at moderate+ confidence: Yes
- 60m EMA/momentum supports direction: Yes
- 5m defined-level reaction visible: Only after trigger
- No high-impact event within 30 minutes: Yes
Risks:
- 15m RSI has recovered above 50 and 5m momentum improved, so do not short blindly into strength
- VIX is not risk-off, so this is a USD-trend short, not a panic short
- After the multi-hour selloff, TP1 should be treated as a forced take-profit due to Cable’s V-reversal tendency
Invalidation Condition:
- Any 5m acceptance above 1.3230-1.32324 weakens the idea materially
- Hard invalidation: 1.32393
- Time invalidation: no new entry after 11:30 ET
No qualifying long setup. Longs fail London bias, DXY trend support, and both agent biases.
Decision log
15:01 UTC, 90 percent, WAIT. Monitoring had started twelve seconds earlier with price already inside the sell zone. The last completed 5-minute candle had closed bullish at 1.32227, above the 1.3219 trigger, with 5-minute RSI near 66 and MACD above zero. Selling here, the system said, would be a blind fade rather than a confirmed rejection.
15:02 UTC, 89 percent, WAIT. The forming candle held above the zone with 5-minute RSI and MACD still rising. The system read that as acceptance above resistance, not rejection. There was no close back below 1.32190 and no clear upper wick from 1.3220-1.3223, so neither version of the trigger existed.
15:04 UTC, 87 percent, WAIT. Another 5-minute candle closed strong above VWAP, with MACD expanding to the upside. The system called it continuation strength rather than failure at resistance, and restated the trigger in full: a wick rejection from 1.3220-1.3223, or a 5-minute close back below 1.32190 after a probe higher.
15:09 UTC, 89 percent, WAIT. The last completed candles had pushed above 1.3219 and closed strong, and 5-minute RSI was rising into the high 60s. The system granted that the broader intraday bias was still mildly bearish, then quoted the plan back to itself: this was exactly the kind of strength it had been told not to short blindly into.
15:11 UTC, 61 percent, WAIT. Something changed. Price probed up to 1.3229 and the latest 5-minute candle rejected from above, closing bearish near 1.32223. The system called the trigger only partially satisfied, because the close was still above 1.32190, and with RSI above 50 and MACD above zero it chose to wait for a cleaner rejection or a decisive close under the level.
15:12 UTC, 79 percent, WAIT. The upper wick was there, but price was still above VWAP and the fast EMA, with 5-minute RSI around 64 and MACD positive. The system read the candle as a pause after a push higher rather than a clean bearish rejection, and kept waiting for a decisive 5-minute close.
15:13 UTC, 78 percent, WAIT. The same picture. The wick after the probe to 1.3229 counted for something, but the candle had not closed back below 1.32190, and the forming candle was still holding above it. The system said a completed close under 1.3219 would give it better edge.
15:14 UTC, 82 percent, WAIT. The system named the 15:05 candle's upper-wick rejection after the push to 1.3229, then set it aside. Five-minute momentum was still too firm, price was above the fast EMA and VWAP, and RSI and MACD on the 5-minute and 15-minute charts were not yet aligned bearishly. Its summary: close, but not confirmed.
15:16 UTC, 61 percent, ENTER. The last completed 5-minute candle had faded from 1.3229 to 1.32194, and live price was 1.3218, back under the 60-minute VWAP. The system said plainly that the close had missed the 1.32190 trigger by a fraction, and entered anyway, calling it a usable signal in a context that still leaned bearish. It named its caution, improving 5-minute and 15-minute momentum, and set the rule: a decent but not perfect entry, to be managed tightly toward TP1. The order filled at 1.3219 at 15:24 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 | +0.72R | +$1,440 |
| TP2 hit (not tracked) | +0R | +$0 |
| TP3 hit (max potential) (not tracked) | +0R | +$0 |
All six agents combined.
What this trade teaches
A trigger is a line, and lines get missed by small amounts. For eight reads the system treated 1.32190 as exact. A close at 1.32227, or at 1.32223 after a rejection wick, was not a close below 1.32190, and every one of those reads said wait. On the ninth the close was 1.32194. What changed was not the candle, which still missed, but live price, which by then was trading at 1.3218, under the level the close was supposed to reclaim. The system did not pretend the trigger had printed. It said the close had missed by a fraction, judged the rejection from 1.3229 good enough, and wrote down what that judgment cost: a decent but not perfect entry, to be managed tightly toward TP1.
That gives a usable short signal even though the last closed 5m candle missed the 1.32190 close trigger by a fraction, and the broader intraday context still leans bearish. SkyAnalyst entry evaluation, 15:16 UTC
The second lesson is arithmetic. The plan described TP1 as roughly 1.0R. With the fill at the bottom of the zone, 1.3219, and the stop at the near edge of its range, 1.3237, the risk was 18 pips and TP1 sat 13 pips away, which is 0.72R. The plan's round number and the filled trade are different numbers, and the record keeps the filled one.
The third is what happened after 12:11 ET. Price later reached 1.32014 at 2:15 ET on Tuesday, 4.6 pips past TP1 and 10.9 pips short of TP2. Then it turned, and by 3:56 ET it was back at the 1.3237 stop, 35.6 pips higher. The position was not there for any of it. The broker had closed it in full at TP1, which is how the analysis had told itself to treat that level. Our September 23 Cable short carried the same instruction and ran to TP2 anyway. This one did not. What drove Cable back up overnight is not in our data, and we will not supply a reason for it.
From the desk
The number on this trade is +0.72R (TP1), on an entry the system itself called decent but not perfect. The result is the smaller half of this story. A system that only enters on a textbook trigger is easy to describe. This one entered on a near miss, said so in writing, and attached a management rule to the admission.
Cable has handed us a run of judgment calls lately. On September 22 two reads at the same score gave two different answers, and on September 21 a short came 1.1 pips from its stop before it paid. One trade settles nothing on its own. The record is the place to judge, and this one goes into it at +0.72R (TP1).
At a Glance
What this teaches about AI-driven trading
What is a conditional short on rejection?
It is a short that exists only if a specific event happens. The trader marks a resistance level, such as VWAP or broken support, and names the candle that would prove buyers failed there, for example a wick rejection or a close back below the level. If that candle never prints, there is no trade. The condition keeps the trader from selling into a bounce that may still be gaining strength.
Why would a trader enter when a trigger is missed by a small amount?
Triggers are drawn as exact prices, but markets rarely respect them to the decimal. If a closing candle misses the level by a fraction while live price has already moved through it, some traders treat the intent of the trigger as met. The cost is a lower-quality entry, so the trade usually gets tighter management, such as taking the first target rather than holding for more.
How does the fill price change a trade's R-multiple?
R-multiple is reward divided by risk, both measured from the fill. A short filled at the bottom of its entry zone sits closer to the target and farther from the stop than one filled at the top. With the stop and target fixed, reward shrinks and risk grows at the same time. A target described as about 1R from the middle of the zone can land well below 1R at the edge.
What does VWAP tell a trader during an intraday bounce?
VWAP is the average price of the session weighted by volume. In a falling market, a bounce that reclaims VWAP shows buyers paying above the average traded price, which weakens the downtrend case. A bounce that fails at VWAP and closes back below it suggests sellers still control the average. That is why many short setups use VWAP as both the entry level and the confirmation line.
When should a first target be treated as a forced take-profit?
Traders often bank the first target in full when a move has already run a long way before the entry, or when the instrument tends to reverse sharply after extended moves. In those conditions the extra reward from later targets may not justify the risk of giving back the gain. Taking the first target caps the upside, and it also removes exposure to a reversal back through the entry.
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.
Past performance is not indicative of future results
The trades in this article are taken by the SkyAnalyst agents in live demo accounts at a regulated broker, in a live market environment, so the demo account reflects every transaction SkyAnalyst makes. The app logs each trade and confirms it against the live price feed of the broker; the capital is simulated and no real capital is at risk. The demo account is linked to MyFxbook, which publishes its results publicly. We publish these results to study how the agents trade and reason, for education and trade analysis, not as a recommendation. Trade at your own risk.
From the SkyAnalyst Journal
Three Losses, One Habit: Approvals That Ran Ahead of the Candle
Three stops in under a day, exactly 3R given back from a +4.08R Thursday peak. Each approval wrote down, in its own words, that part of its confirmation was still missing.
Up 4R by Thursday, Then Three Entries That Did Not Wait
Seven trades, +1.08R. The desk banked four winners from Tuesday to Thursday morning, then gave back 3R on three entries approved on less than full confirmation, two of them on a payrolls Friday.
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 after the fill for +0.87R (TP1).