Why every trade needs a stop loss: lessons from 11,439 of my own real trades
- I traded by hand for more than 7 years, 11,439 trades, mostly gold. The account ended down 742,241 baht (about $22,000).
- Yet the trades I closed myself made +459,617 baht with a 52.9% win rate. What wiped everything out were 782 forced stop-outs that cost 1,171,926 baht.
- 97.4% of my trades had no stop loss. That group was stopped out almost 4 times as often as the trades that had one.
- The problem was not reading the market. It was not controlling risk. That is why MindCore puts a stop loss on every trade.
Why I am publishing this
Most people selling EAs or signals only show their winners. I chose the opposite: I took my full MT4 statement from April 2019 to September 2026, analysed it trade by trade, and I am showing both what worked and what broke.
If you trade the way I used to, these numbers will help you more than another smooth profit chart.
The whole account
| Period | April 2019 – September 2026 (about 7.4 years) |
| Trades | 11,439 (about 128 per month) |
| Account result | −742,241 baht |
| Profit factor | 0.68 |
| Overall win rate | 38.1% |
On these numbers alone you would conclude I cannot trade. Split the trades into two groups and the picture changes completely.
Two groups: trades I closed, and trades closed for me
| Closed by me | Forced stop-out | |
|---|---|---|
| Trades | 7,654 | 3,594 (782 events) |
| Result | +459,617 baht | −1,171,926 baht |
| Win rate | 52.9% | 6.7% |
The trades I chose to close were genuinely profitable, winning more than half the time. But every time price ran hard against me with no exit planned, margin ran out and the broker closed several positions at once, at the worst prices of the move.
Working backwards, if the account had never been stopped out it would have finished about +429,685 baht. The gap between plus four hundred thousand and minus seven hundred thousand is the price of not managing risk, and nothing else.
How much a stop loss changed things, measured on my own trades
97.4% of all my trades had no stop loss. Grouping them:
| Trades without a stop loss | 32.0% ended in a stop-out |
| Trades with a stop loss | 8.8% ended in a stop-out |
Almost 4 times the difference. Trades with a stop loss were still caught sometimes, because other positions in the same account without one drained the shared margin and the whole account was closed together.
What a stop-out looks like
- On average −1,499 baht per event, closing about 4.6 trades each time
- The worst single event was −20,917 baht, nine positions closed together on 18 June 2022
- In 2022, trades I closed myself made +185,735 baht, while stop-outs cost −303,194 baht
- The account lost money every year from 2019 to 2026, even though my own closed trades were positive almost every year
Another factor made it worse: I traded several assets from the same margin, including BTC (−157,605 baht) and EURUSD (−97,224 baht). When one of them moved hard, it dragged down gold positions that were doing fine.
Why no stop loss breaks you, even when you read the market well
A 52.9% win rate on my own closed trades says I called direction right more often than wrong. But trading is not scored on how often you are right. It is scored on how big your wrong trades get.
- A trade with no stop loss can lose without limit until the margin is gone
- One loser left open can erase dozens of winners in a single night
- In a stop-out you do not choose the price; the broker closes you at the worst point of the move
A stop loss does not make you win more often. It makes the size of a wrong trade known in advance, and that is the only thing that keeps an account alive long enough for real skill to pay.
What the data says I did right
To be fair to myself, my own closed gold trades show several things worked:
- Two to three positions at a time was the sweet spot: +104 baht per trade, 57.4% win rate. At 7–10 positions it fell to −19 baht per trade.
- 0.16–0.35 lots gave the best results. Above 0.36 lots the average turned negative (−173 baht per trade).
- The money was in trades held 1–24 hours. Trades closed within 5 minutes lost on average.
- I was clearly better at selling gold than buying it: SELL +228,993 baht at 56.0%, BUY +42,516 baht at 50.7%.
Almost every item is about discipline — number of positions, position size and patience — not about finding better entries.
What I built into MindCore
MindCore comes directly from these numbers. I did not try to build something that reads the market better than I do. I built something that does what I could never do consistently:
- ✅ A stop loss on every trade, set when the trade opens. No trade can lose without limit.
- ✅ No averaging down, no grid, no martingale. It never adds positions to average out a wrong one.
- ✅ Position size from your balance, not from how confident it feels.
- ✅ Gold only. One market, one set of rules.
A system like this still has losing years. MindCore's 10-year test includes a maximum drawdown of about 26%, and every year is published on the results page. The difference is that each loss has a known size.
Checklist for anyone still trading by hand
- Set the stop loss before you open the trade, not after price has moved against you
- Decide your risk per trade as a percentage of your balance, then size the position from the stop distance (try the position size calculator)
- Limit how many positions you hold at once; my own data says two to three is best
- Do not trade several assets from one margin pool; if you must, use separate accounts
- Every month, split your results into "closed by me" and "stopped out". If the second group exists, your problem is risk control, not market reading
Notes
- All figures come from the author's real MT4 statement, analysed trade by trade, in the account currency (Thai baht).
- One trader's results do not represent everyone, but the pattern "profitable when I close, wiped out by stop-outs" is very common among traders who skip stop losses.
This article is for education only and is not investment advice. Trading gold and CFDs carries a high risk and you may lose all your capital. Past results do not guarantee future results.
