September ended with every trade on the short side and a +3.57% month. October was another month of falling gold, and the system stayed short for most of it. Gold opened the month around $4,200, slid through the second half and closed below $4,000 on 23 October for the first time in weeks. The Fed held rates at 3.75–4.00% on 28 October, with two members voting for a hike, so the pressure on gold never really lifted.
The direction was right, and the month was still the worst this account has had since it went live. Here is what happened and why.
The month in numbers
- 28 trades (long-run average ~31 per month)
- 15 winners, 13 losers, a win rate of 54%
- Net result: −6.67% (−€291.30)
- Profit factor: 0.37
- 23 shorts, 5 longs
- Best trade: +0.98% (+€39.30), a short from 3,943.73 to 3,929.03 on 30 October
- Worst trade: −1.82% (−€73.13), a short from 4,066.68 stopped at 4,094.02 on 22 October
- Active on 6 of 22 trading days
That worst trade is now the largest single realised loss in the account's two-year history. The previous record was −€66.81 in August. I am stating that plainly because it matters more than the headline percentage.
Two days made the month
Day by day, the month looks like this:
- 2 October: −€73.83
- 6 October: −€197.52
- 7 October: −€0.74
- 9 October: +€17.24
- 22 October: −€147.41
- 30 October: +€110.96
The 6th and the 22nd together cost €344.93, more than the whole monthly loss. The other four active days were net positive.
On 6 October, gold pushed below a level and the system opened five shorts together around 4,109. Price came straight back up through the level, to 4,116–4,134, and stopped all five.
On 22 October, gold started the move that took it under $4,000. The system sold the breakdown around 4,067–4,075. Price snapped back to 4,073–4,094 before continuing lower, hit the stops, and the next day gold closed below $4,000. The trade idea was correct. The entry was early enough that the retest took us out before the move we were positioned for actually happened.
What October actually shows
I have written several times on this blog about the difference between a level that is broken and a level that is taken. A level is taken when the market moves through it and the retest from the other side fails. A level is merely broken when price pokes through and then comes back.
In August, shorts lost money because they fought an uptrend. In September, shorts made money because the levels they sold were genuinely taken. October is the uncomfortable third case: the trend was down and the system was positioned with it, but twice it acted on the break before the level had been taken. The retest came, the stops were sitting right where the retest went, and the move continued without us.
That is a timing problem, not a direction problem. It is also the reason a falling market does not automatically mean easy money for a short-biased month.
The honest part about losses
The system opens positions as a cluster. When the trigger is right, that structure pays well. 28 September and 30 October were both cluster days, and they produced the best results of their respective months. When the trigger is wrong, every piece of the cluster is stopped together. On 6 October that turned one bad decision into five losing trades.
That cluster behaviour is not new. It is visible throughout the history and it is part of why the good days are as good as they are. October is simply the month where its downside showed up twice in three weeks.
From the late-September peak to the low on 22 October, the account gave back about 9.2%. That is uncomfortable, but it is still inside the 15.32% maximum drawdown the account has already lived through and recovered from. I am not changing the rules in the middle of a drawdown. Changing a system after two bad days is how you end up optimised for last month instead of the next one.
The full track record (verified, live)
- +190.67% total gain since August 2024
- 847 trades, win rate 79%
- Profit factor 1.93 (down from 2.22 a month ago, October did real damage here)
- Average monthly return: 4.09%
- Maximum drawdown: 15.32%
- 21 of 27 closed months positive
- 2026 so far (January to October): +54.1%
The +190.67% covers the full live history on SignalStart, measured against deposits since August 2024. The MQL5 signal page only starts counting from February 2026, when the signal was published there, which is why it shows a lower growth figure (around +146%) for the same account.
Looking to November
Gold has started November back around $4,080, above the $4,000 level that decided most of October. The next scheduled Fed decision is on 8–9 December, and with a split committee, every inflation and labour release between now and then can move rate expectations.
That kind of environment produces a lot of breaks that are not yet takes. I am not forecasting direction, and I am not setting a target for the month. The system will trade when its levels are in play and stay flat when they are not.
You can follow AlphaEdge Smart with the full verified statistics on my profile. Questions about October, the drawdown or how the system handles clusters? Ask in the comments — I answer everyone.
Past performance is not indicative of future results. Trading leveraged products such as XAUUSD carries a substantial risk of loss and is not suitable for every investor. Results of copied trades may differ from the original account due to slippage, spreads and execution.
