Five stops in a row: what a losing week actually costs
Five trades stopped in a row for a combined 5R, and every one of them was a long. None was a mistake in isolation. This is what a normal losing streak looks lik
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.
Every trading system that has ever made money has also produced losing streaks, and last week ours produced one. Between July 14 and July 15, five consecutive trades stopped out, each for exactly 1R, for a combined loss of 5R. All five were longs. We are publishing this the way a fund publishes its drawdowns, because a track record that only shows you the winning weeks is not a track record. It is a highlight reel. The honest framing starts with proportion. Through Jul 20, 2026, the system has banked plus 26.24R since the January 12 inception. On a 100,000 dollar account risking two percent a trade, that is 152,498 dollars on a static basis. This week gave back roughly 10,000 dollars of simulated equity across the five stops, a drawdown of about 9.72 percent from the intra-week peak. Set the give-back beside the year and the shape is clear: a 5R week against a plus 26.24R record is a slice of the ledger, not a change in its direction. Nothing broke this week. The setups were graded, the stops were placed where the plan said, and the losses arrived at their planned size. What follows is the anatomy of five ordinary losing trades and the statistics that explain why a week like this is not only survivable but expected.
Act I — the peak before the slide
The week's high-water mark was set early, on Monday, and it matters here only as the reference point the drawdown is measured from. Simulated equity reached about 102,916 dollars before the losing sequence began. From that peak, everything the report covers is downhill, which is exactly what a drawdown curve is: the distance from the best the account had been to the worst it got.
Act II — five stops in two sessions
The slide itself was compact. Tuesday brought three long entries, on US500, US30, and GBPUSD, and all three stopped at 1R. Wednesday brought two more longs, US500 and US30 again, and both stopped as well. Five losses across two afternoons walked simulated equity from its peak down to roughly 92,916 dollars, a peak-to-trough drawdown of 9.72 percent. The losses were evenly sized and evenly spaced. There was no single catastrophic trade, just five ordinary ones pointing the same way.
Act III — the floor and the turn
The drawdown found its floor on Wednesday afternoon and did not deepen. No trade was taken Thursday. By the end of the week two shorts, outside the scope of this losses report, had lifted equity back toward 95,519 dollars and trimmed the drawdown to about 7.19 percent. The floor held because the risk policy never let any single loss run past its 1R limit. A drawdown that stays linear is a drawdown you recover from.
Every trade the system took.
| Date | Time | Instrument | Dir | Model | Setup | Grade | R | $ Sim | Result | Details |
|---|---|---|---|---|---|---|---|---|---|---|
| Jul 14 | 14:37 UTC | US500 | Long | Claude Opus 4.7 | US500 LONG — Opening Range Breakout | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
| Jul 14 | 14:40 UTC | US30 | Long | GPT-5.5 | US30 NY AM Pullback/Second-Chance Long | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
| Jul 14 | 15:06 UTC | GBPUSD | Long | GPT-5.5 | GBPUSD London pullback buy | B | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
| Jul 15 | 14:37 UTC | US500 | Long | Claude Opus 4.7 | US500 Bullish VWAP/Structure Pullback Buy | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
| Jul 15 | 14:52 UTC | US30 | Long | GPT-5.5 | US30 pullback-long continuation | C+ | -1.0R(SL) | -$2,000(SL) | Stop hit | - |
Dollar figures are simulated on a $100,000 account at 2% risk per trade. Actual subscriber P&L varies with account size. Past performance is not a guarantee of future results.
Pattern of the week
The losses shared one trait above all others: they were all longs. Five entries bought strength or pullbacks into an index and currency tape that was, that week, fading exactly those bounces. The read at entry was defensible in each case, but the market was in a posture that punished dip-buying, and our long setups kept arriving into it. When every loss in a streak points the same direction, the useful question is not which entry was wrong. It is why the same directional bet kept clearing our filter into an unfavorable tape.
There is a second thread the numbers make plain. Four of the five losses came from the two US index books, US500 and US30, and twice those two fired long in the same session within minutes of each other. Two highly correlated instruments taken the same direction at the same time are, in risk terms, close to one doubled position. The streak was not five independent losses. It was closer to three independent decisions, two of which were doubled by correlation.
Decision highlights
The risk policy held flat through all five losses. Every trade in the streak risked the same fixed 1R, so five losses cost exactly 5R with no acceleration. A discretionary trader who widened stops or doubled size to recover the third or fourth loss would have booked a far worse week from the same five reads.
The system took no trade on Thursday. Sitting in a drawdown near the weekly low is precisely when the pull to force a recovery entry is strongest, and the system felt none of it because no setup cleared the filter. Doing nothing was the correct decision, and it is one a human in a losing streak often cannot make.
The correlated index longs were allowed to run together twice. On both Tuesday and Wednesday, a US500 long and a US30 long were open at the same time, effectively doubling a single directional bet. This is the one decision the week argues we should have caught, and we address it in what we are tuning below.
Six instruments, six stories.
EURUSD: No losses this window. The pair produced no trades at all and sat outside our setup criteria.
All EURUSD this week →GBPUSD: One loss, minus 1R. A Tuesday London pullback long that stopped cleanly at its planned risk.
All GBPUSD this week →US30: Two losses, minus 2.0R combined. Both were longs, both stopped, and both fired alongside a correlated US500 long in the same session.
All US30 this week →NAS100: No losses this window. Its only trade was a winning short, outside the scope of this report.
All NAS100 this week →USDJPY: No losses this window. The pair stayed outside our setup criteria the entire week.
All USDJPY this week →US500: Two losses, minus 2.0R combined. Both were longs stopped at 1R, the other half of the week's correlated index give-back.
All US500 this week →Loss of the week: US30 Long · -1R
Losses worth learning from
GBPUSD long, Tuesday's London pullback
The best-graded loss of the week, a B setup, and the most instructive because it was not sloppy. The trend agent flagged a pullback buy during the London session, the structural premise was intact, and the entry filled with a stop placed at a level the plan defined in advance. What was right was the process: a graded setup, a defined risk, a stop that did its job at exactly 1R. What was wrong was external, not internal. The pair was in a posture that did not reward the long, and the pullback that looked like a base was really a pause on the way lower. What we would do the same is all of it. A B-grade setup that stops at its planned risk is not an error to correct. It is the cost of participating in an edge that plays out over hundreds of trades, and pretending otherwise would be the actual mistake.
US30 long, Wednesday's continuation buy
A C+ setup, marginal but inside our filter, and the second of two US30 longs that stopped that week. The read was a pullback-long continuation into an intraday uptrend, leaning on prior structure for the stop. What was right was the discipline around size: it risked the same 1R as every other trade, no more for being a "continuation" of a trend we liked. What was wrong sits partly in correlation. This long was open alongside a US500 long the same afternoon, so the account was effectively twice as exposed to a single US-equity direction as the trade count suggests. What we would do differently is not the entry itself but the portfolio context around it. Taking two correlated index longs at once turned one wrong read into two stops, and that is a sizing question, not a setup question.
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 |
|---|---|---|
| Window drawdownActual | -5R | −$10,000 |
All six agents combined.
From the desk
A note, before we move on.
We publish losses reports because the alternative is dishonest. Any system can show you its green weeks. What tells you whether a track record is real is what the operator does with the red ones, and what we do is put them here, in the same detail and the same voice as the wins. Five stops, all longs, all 1R, no drama and no excuses.
Put the week in dollars and it stays in proportion. On a 100,000 dollar account at two percent risk, the static balance sits at 152,498 dollars, and this week's five stops removed roughly 10,000 dollars of that, the 9.72 percent drawdown the curve shows. The compounded version of the same record, where each trade risks two percent of the growing balance rather than a fixed amount, sits higher at 163,402 dollars. The gap between those two figures, near 11,000 dollars, is what disciplined, consistent sizing produces over 151 trades. A losing week narrows that gap for a few days. It does not close it, because the sizing that built it does not change when the week goes against us.
The reason we can write about a 5R week this calmly is that the risk policy makes the worst case knowable in advance. Five losses were always going to cost 5R, no more, and the day the streak ran we did nothing differently than the day it did not. That steadiness is the product. A losing week is where you find out whether a system has it, and this is us showing you that ours does.
— The SkyAnalyst Team
What we're tuning
There is nothing to fix in the individual losses. Each was a graded setup stopped at its planned 1R, which is the system operating exactly as designed. The flat risk policy did its job and held a five-loss streak to a linear 5R rather than letting it spiral, and no change to entry logic would have turned an unfavorable tape into a favorable one.
The one genuine tuning point is correlation. Twice this week a US500 long and a US30 long were open in the same session, and because those two instruments move together, the account carried close to a doubled directional bet without the trade count showing it. We do not currently enforce a portfolio-level correlation check across the two index books. This week is a clean argument for one. It would not have prevented the losing reads, but it would have kept a single wrong direction from costing us two stops at once, which is the difference between a 3R week and a 5R one.
What the numbers actually mean
The first thing to understand about a five-trade losing streak is that in a system winning 58.94 percent of its trades, it is not an anomaly. It is a scheduled event. When roughly 41 percent of trades lose, runs of consecutive losses are guaranteed to appear across a long record, and the longer the record, the longer the worst expected run. Over 151 trades, a streak of five is well inside the range of normal. Jack Schwager's interviews across the Market Wizards series return to this point repeatedly: even the traders with the best long-run records endure losing streaks that would shake anyone watching in real time. The streak is not evidence the edge is gone. It is evidence you are running a probabilistic strategy honestly.
The second thing is why we report in R at all. Van Tharp's work on R-multiples, in Trade Your Way to Financial Freedom, reframes every trade as a multiple of the amount risked. Thinking in R is what turns five losses from an emotional event into an arithmetic one. Five stops at 1R each is minus 5R, full stop, and because our risk per trade is fixed, that number was knowable before the week began. A system with a positive expectancy, where the average win in R times the win rate exceeds the average loss in R times the loss rate, still produces minus 5R weeks. Expectancy is a statement about the average of many trades, not a promise about any five of them.
The third thing is sample size, and it cuts against over-reading this report as much as it cuts against panic. A week is five to fifteen trades. This one was seven, five of them losses. That is far too small a sample to update your belief about a 151-trade edge in either direction. The 9.72 percent intra-week drawdown, measured on a 2 percent risk model, is well within the range a strategy like this generates in normal operation. A drawdown report is one slice of a much longer record, and the honest use of it is context, not conclusion.
At a Glance
Drawdown questions
Is a five-trade losing streak a sign something is broken?
No. In a system that wins about 59 percent of its trades, roughly 41 percent lose, and over a record of 151 trades runs of five consecutive losses are statistically expected rather than surprising. The streak arrived, cost a fixed 5R, and ended. Nothing in the entry logic, risk policy, or agent coordination changed because of it.
How much did this week actually cost in dollars?
On a simulated 100,000 dollar account risking two percent per trade, the five stops removed about 10,000 dollars, a drawdown of roughly 9.72 percent from the intra-week peak. That figure is the model's, not a live account's. It scales with account size and with whatever risk fraction a subscriber chooses to run.
Why report losses in R instead of just dollars?
R normalizes every trade to the amount risked, which is Van Tharp's framework from Trade Your Way to Financial Freedom. Thinking in R turns five losses into a knowable minus 5R rather than an emotional event, because risk per trade is fixed in advance. It also lets you compare this week to any other period on the same scale, regardless of account size.
Does a losing week change your expectancy?
No. Expectancy is a property of the strategy measured across many trades, the average win in R times the win rate against the average loss in R times the loss rate. A five-trade sample is far too small to move that number. A drawdown report is context on a long record, not a revision of the edge that record is built on.
What are you changing after this week?
Nothing in how individual trades are judged, because each loss was a graded setup stopped at its planned risk. The one thing under review is a portfolio-level correlation check, since twice this week a US500 long and a US30 long ran together and doubled a single directional bet. That is a sizing refinement, not a change to entry logic.
Trade with the system that publishes its drawdowns.
Subscribers receive every signal — winners and losers — three minutes before entry, with full reasoning.
Dollar figures are simulated on a $100,000 account at 2% risk per trade. Drawdown trajectories shown reflect a small window sample size and are not projections of forward performance. Past performance — including losses — is not a guarantee of future results. Actual subscriber P&L varies with account size and execution. YTD context: +26.24R YTD across 151 trades, see stats strip.
From the SkyAnalyst Journal
The short the system refused to chase, and why it paid
London sold Cable hard, and the tempting play was to hit the lows. The system did the opposite. It waited for price to retrace to VWAP, shorted the failure to reclaim, and never went a single pip underwater.
Five longs stopped, two shorts paid: the week in review
Seven trades, five of them losses, a net give-back of 2.24R. Yet the setups graded A- for the week and both winners were shorts. Here is what the tape actually taught us, trade by trade.
The first trade Claude Fable 5 ever took as a beta model.
Claude Fable 5 is a new beta trader running in the SkyAnalyst sandbox, outside the tracked Pro and Lite record. Its very first trade was an eight-of-eight-confluence EURUSD pullback long that ran clean to TP3.