Right Direction, Wrong Price: The Cable Short SkyAnalyst Refused to Chase
SkyAnalyst AI journal entry: GBPUSD Short on Oct 7, 2026 reached TP2 for +1.3R full potential, booked +1.02R (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.
At 10:44 ET on Wednesday, SkyAnalyst had the rejection it was waiting for and still would not sell. The read scored 88 percent, the highest of the morning, and it agreed with everything about the GBPUSD short except the price. Cable had already slipped below the 1.32130-1.32155 sell zone, and selling there, it wrote, "would be chasing after the bounce-sell reaction rather than selling the rally." Two minutes later an 85 percent read said the same thing more bluntly: the trigger had fired, but the reward to TP1 was "badly compressed." The fifth read, at 10:48 ET, scored 72 and entered, noting that price was now only "slightly below the zone." The order filled at 1.32134, inside the zone, at 10:51 ET, with the stop at 1.32310, 17.6 pips away. Then Cable took its time. TP1 printed at 2:02 AM ET Thursday for +1.02R (TP1), or +$2,040 (TP1) on the simulated $100,000 account in the returns panel, and the broker closed the whole position there, as it does on every trade. An hour later price reached TP2 at 1.31905, a full-potential move of +1.30R (TP2). We picked this trade for the two waits. Our October 1 Dow short also entered on the lowest score of its morning. This piece is about why the higher scores said no, and the answer is location, not direction. If you want a desk that checks where it is selling as well as which way, see SkyAnalyst run your markets on a 21-day free trial.
London did the selling before New York arrived
By 10:31 ET, when the setup analysis was written, London had done the heavy lifting. GBPUSD had sold off from the London high area to a London low at 1.31929, and it was trading below Tuesday's low of 1.32578, below its 5-day EMA at 1.3240 and below 60-minute VWAP near 1.3233. The analysis called London "a trend session, not a range session." It also noticed where the selling had stopped: in a daily support and prior-low zone at 1.32014 and 1.31904, with the 1.3200 round number in between. That is why the New York plan was to sell a relief rally, not to chase fresh lows.
The dollar side agreed. The Macro Agent's 10:05 ET forex refresh had the Dollar Index at 102.326, above its 5-day EMA of 102.038 and above Tuesday's high of 102.279, and the US 10-year yield at 5.324 percent, above its own 5-day EMA. It read GBPUSD lean bear with 71 percent confidence and moderate tradeability, bearish on the intraday horizon, and listed the break below the 5-day EMA and Tuesday's low as a primary factor. Equity breadth on the NYSE advance-decline line had swung from +475 at Tuesday's close to -1,550. VIX at 15.72 kept the regime classified as normal, not risk-off.
What could undo it
The same refresh named the risks. The FOMC minutes were due at 2:00 PM ET, Bank of England Governor Bailey was scheduled to speak at 8:15 AM ET Thursday, and the agent's strongest counter-argument was that a "dovish Fed interpretation combined with supportive Bailey remarks could challenge USD-positive setup." Across the major pairs it also warned that four of five implied a strong dollar, which makes them one correlated idea rather than five independent ones.
The Trend Agent called the bounce relief
At 9:50 ET the Trend Agent read GBPUSD bearish at 65 percent. At 10:00 ET it was bearish at 72 percent, with the 60-minute fast EMA now crossed below the slow. By 10:25 ET, as Cable bounced off the 1.3193 area, it held bearish at 70 percent in a trending regime and changed its sizing advice to reduce. Its read on the bounce: "this looks like relief rather than a confirmed reversal," with the upcoming FOMC minutes arguing for smaller size.
Rally Sell Continuation. This is a pattern professional traders use when a market has already trended hard and then bounces. Instead of selling the low, where the trend has just run into support, the trader waits for a relief rally into a defined level and sells the rejection there. The bet is that the bounce is a pause in the trend, and the entry location is what keeps the risk small.
The zone
On Wednesday the level was 1.32130-1.32155, bracketing the New York high at 1.32152 and well under VWAP near 1.3233. The analysis was explicit that this was "a sell-the-bounce setup only," because London had already pushed into the 1.3200 and 1.3190 support area.
Location is most of the edge in this pattern. Sell lower and the same stop sits further away while the first target gets closer. The trade still points the right way, but its arithmetic gets worse with every pip.
The trigger
The plan asked for a 5-minute rejection inside the zone: a wick or failure at or just under 1.32152, or a 5-minute close back below 1.32130, ideally with 5-minute RSI rolling back under 55 and the MACD histogram fading. It also wrote down the case for no trade at all: if price reached the zone and accepted above 1.32155 without rejection, there was no short.
Stop, targets and expiry
The stop went to 1.32310, beyond the 5-minute swing zone and below the Trend Agent's invalidation at 1.3239. TP1 sat at 1.31955, TP2 at 1.31905 and TP3 at 1.31820, and the setup expired if it had not triggered by 11:30 ET. The plan estimated the targets at roughly 1.0R to 1.1R, 2.0R or more, and 2.8R to 3.0R. From the actual fill the same levels measured 1.02R, 1.30R and 1.78R. The first estimate held. The other two were too generous, and we would rather say so here than let the setup description oversell the trade.
SkyAnalyst doesn't favor any single strategy. On Monday it sold Cable on a conditional rejection, on Thursday morning it sold a coil under VWAP on the Nasdaq, and on Wednesday it sold a relief rally that stopped well short of VWAP. The pattern comes from that morning's chart, and the conditions are written down before price gets there.
London set a clear bearish tone into the NY overlap: GBPUSD sold off from the London high area and printed a London low at 1.31929, with price still trading below yesterday’s low (1.32578), below the 5-day EMA (1.3240), and below 60m VWAP (~1.3233). That makes London a trend session, not a range session. However, London also hit a meaningful downside extreme near daily support / prior low zone (1.32014 / 1.31904) and the 1.3200 round figure, so the higher-probability NY play is not chasing fresh lows: it is selling a relief rally.
Macro and cross-asset context reinforce that view. DXY is above its 5-day EMA and above yesterday’s high, near its recent upside extreme, while VIX is above its 5-day EMA, signaling mildly risk-off conditions that favor Cable shorts over longs. The inverse GBPUSD/DXY relationship is intact today, so there is no divergence-gate problem. Both agents agree: Trend Agent = bearish (70%, trending, reduce size) and Macro Agent = lean_bear (71%, tradeability moderate). Because Macro confidence is above 70 and DXY aligns with GBPUSD downside, the hard rule applies: do not take GBPUSD longs against this backdrop.
Directional Bias: Bearish
Volatility: Normal
Setup #1: GBPUSD SHORT
- Directional bias: Bearish continuation via rally-sell, not breakout-chase
- Entry: 1.32130-1.32155
- Entry trigger: Wait for a 5min rejection in that zone:
- wick/failure at or just under NY high 1.32152
- or a 5min close back below 1.32130
- ideally with 5min RSI rolling back under 55 and MACD histogram fading after the bounce
- Stop Loss: 1.32310 hard stop
(meets 15-pip minimum, sits beyond the immediate 5min swing zone, and remains below Trend Agent invalidation 1.3239) - Targets:
- TP1 = 1.31955
- TP2 = 1.31905
- TP3 = 1.31820
- R-Multiples:
- TP1 ≈ 1.0R-1.1R
- TP2 ≈ 2.0R+
- TP3 ≈ 2.8R-3.0R
- Quality Score: 8.1/10
- Confidence: High - London bearish bias, DXY support, Macro/Trend agreement, bearish 60m structure, defined 5m resistance entry, and no high-impact event inside 30 minutes. Main risk is that London already tagged support, so NY can squeeze higher before rolling back.
Confluences (6/7 pre-trigger, 7/7 after trigger prints):
- London bias alignment: Yes, London trended down decisively
- DXY supports trade: Yes, above 5-day EMA and strong
- Macro bias aligns ≥6/10: Yes, lean_bear, 71%
- Trend Agent aligns moderate+: Yes, bearish, 70%
- 60m EMA structure supports: Yes, bearish cross / price below EMAs and VWAP
- 5min defined-level entry with reaction: Yes only after rejection trigger prints
- No high-impact event within 30 min: Yes
Risks / caveats:
- London already extended into 1.3200 / 1.3190 support, so this is a sell-the-bounce setup only.
- Trend Agent explicitly says reduce size.
- Treat TP1 as a forced partial / primary take-profit because Cable is vulnerable to sharp V-reversals after multi-hour directional pushes.
- Setup expires if not triggered by 11:30 ET.
Invalidation condition:
- No short if price reaches the zone and accepts above 1.32155 without rejection.
- Broader bearish setup is weakened on a sustained push above 1.32320 and fully invalidated above 1.3239.
No long setup qualifies.
Reasons: hard macro rule blocks longs (Macro confidence 71% + DXY aligned bullish USD), VIX is mildly risk-off, and DXY is near upside extremes, which materially lowers long-side reliability.
Decision log
14:40 UTC, 10:40 in New York, 78 percent, WAIT. Price had entered the sell zone 14 seconds earlier and was sitting at its bottom edge near 1.32130. The direction was not in question, the trigger was: no decisive rejection from the 1.32152 area yet, 5-minute RSI still above 55, and the MACD histogram not yet rolled over. The read asked for a fresh 5-minute failure back below 1.32130 or a clearer rejection wick.
14:43 UTC, 77 percent, WAIT. The 14:40 candle was rejecting lower from the zone, but it had not closed. The last completed 5-minute candle had closed at 1.32137, above the 1.32130 line, and 5-minute RSI was 55.8 with MACD still positive. In the read's own words, "the fade is only developing."
14:44 UTC, 88 percent, WAIT. The highest score of the five. The rejection now looked like the one the plan wanted, with RSI back under 55 and the histogram fading. But the candle was still incomplete and price was "already well below the entry zone." Selling there would have meant chasing the reaction instead of selling the rally, at a materially worse reward-to-risk against the 1.32310 stop and TP1.
14:46 UTC, 85 percent, WAIT. The trigger had fired on the last completed candle, which traded up to 1.32135-1.32137 and closed back below 1.32130 at 1.32073, with RSI at 51.5 and the histogram fading from 0.00017 to 0.00013. Price was around 1.32071. The read called it "no longer a fresh rally-sell entry" with the reward to TP1 "badly compressed," and passed.
14:48 UTC, 72 percent, ENTER. The trigger had fired: a top at 1.32137, a completed 5-minute close at 1.32079, RSI rolling from 59.6 to 52.2 and the histogram fading from 0.00017 to 0.00014. Price was now "slightly below the zone rather than inside it." The read said the entry was no longer perfect but the short was still valid. The order filled at 1.32134, inside the zone, at 14:51 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 | +1.02R | +$2,040 |
| TP2 hit | +1.3R | +$2,600 |
| TP3 hit (max potential) (not tracked) | +0R | +$0 |
All six agents combined.
What this trade teaches
Direction and location are separate questions. Between 10:44 and 10:46 ET the system was more confident than at any other point that morning, and both reads declined to sell. The arithmetic shows why. At about 1.32071, where price sat during the 85 percent read, the 1.32310 stop was 23.9 pips away and TP1 only 11.6 pips, roughly 0.49R to the first target. From the 1.32134 fill it was 17.6 pips of risk for 17.9 pips to TP1, or 1.02R. Same trade, same direction, about twice the payoff for each pip of risk.
However, right now price is already around 1.32071, well below the intended sell zone, so this is no longer a fresh rally-sell entry and the reward to TP1 is badly compressed. SkyAnalyst entry evaluation, 14:46 UTC
The second lesson is about which number we keep. TP1 printed at 2:02 AM ET Thursday and the broker closed the whole position there, so the ledger records +1.02R (TP1). The market went on to TP2 at 3:02 AM ET for a full-potential +1.30R (TP2), and reached 1.31876 at 3:07 AM ET, 5.6 pips short of TP3. Then it turned. The tracking record shows price back through the 1.32310 stop at 10:25 ET Thursday, more than eight hours after the position was already flat. The plan had warned that Cable is "vulnerable to sharp V-reversals after multi-hour directional pushes" and asked for TP1 to be treated as the primary take-profit. The reversal took about seven hours rather than minutes, but it went all the way back.
Set it beside our September 23 Cable short, which carried a similar instruction about TP1 and ran on to TP2 without coming back. This one reached TP2 and then gave the whole move back. What turned GBPUSD on Thursday morning is not in our data, and we will not supply a reason for it.
From the desk
The headline number on this trade is the market's, +1.30R (TP2). The number we keep is +1.02R (TP1), on a read that waited for price to come back to the zone instead of selling a move it had just watched happen. Being right about direction is the common part. Caring where you sell is the part we want readers to check.
Cable has given us a run of these shorts lately. On Monday the system sold a trigger it had missed by four tenths of a pip, and on September 22 two reads at the same score gave two different answers. One trade settles nothing on its own. The record is the place to judge, and this one goes into it at +1.02R (TP1).
At a Glance
What this teaches about AI-driven trading
What is a rally sell continuation setup?
It is a short taken inside an established downtrend after price bounces into a defined resistance level. Rather than selling at the low, where the trend has just met support, the trader waits for the relief rally to stall and sells the rejection. The stop goes above the bounce, so a genuine reversal takes the trade out quickly while the trend's next leg does the work.
Why does entry location change a trade's reward-to-risk?
The stop and the targets are fixed price levels, so the entry price decides the distances between them. Selling lower in a short moves the stop further away and the first target closer, which shrinks the reward for each unit of risk. Two traders with the same view and the same levels can end up with very different payoffs simply because one sold the bounce and the other sold after it.
How can a trade reach its second target and still show a stop in its record?
Some execution setups close the entire position at the first take-profit. The trade record keeps tracking price afterward, so the market can travel on to later targets and then reverse through the original stop while the position is already flat. The account books the first target's R-multiple, while the furthest target reached describes the setup's full potential.
When should a trader skip a trigger that has already fired?
A trigger confirms the idea, but the entry still has to make sense at the current price. If price has already moved well past the planned zone, the reward to the first target may be too small for the risk. Many traders treat that as a missed entry and wait for price to return to the level, accepting that sometimes it never does.
Why do scheduled central bank events matter for an open forex trade?
Events such as policy minutes or a governor's speech can shift rate expectations quickly, and currencies reprice on rate expectations. A trade opened before the event carries that risk whether or not the event is the reason it was taken. Traders often reduce size, tighten management, or avoid new entries near the release so a surprise does not dominate the outcome.
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
The Floor That Held Once: SkyAnalyst's Nasdaq Short Through 30,895
The plan named the 30,895 floor as its risk and tied TP3 to breaking it. One read, a fill at 30,996.3, 57 quiet minutes, then TP1 to TP3 in 13. The broker booked +0.84R (TP1).
Four Tenths of a Pip Short: Why SkyAnalyst Sold Cable Anyway
Eight reads waited for a 5-minute close under 1.32190. The ninth got 1.32194, sold GBPUSD at 1.3219 anyway, and TP1 printed 47 minutes after the fill for +0.72R (TP1).
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.