Ultra-Scalping System — Can Someone Please Help Me Figure This Out?

New comment

This is a screenshot of the trades. I’ve extracted the entire trading history, consisting of more than 2,307 trades, and they all look similar to what you see in the screenshot.

He is trading on a demo account as part of a trading contest. If needed, I can upload the full trading history in the comments later.

Does anyone recognize this technique? If so, please let me know anything you can about it. I honestly can’t figure out how this guy is scalping like this.






Luu Tuan Trung:

This is a screenshot of the trades. I’ve extracted the entire trading history, consisting of more than 2,307 trades, and they all look similar to what you see in the screenshot.

He is trading on a demo account as part of a trading contest. If needed, I can upload the full trading history in the comments later.

Does anyone recognize this technique? If so, please let me know anything you can about it. I honestly can’t figure out how this guy is scalping like this.

Sorting by "Open date" in the full trading history would be more conducive to reverse engineering.
Ryan L Johnson #:
Sorting by "Open date" in the full trading history would be more conducive to reverse engineering.
He scalps for very short periods, averaging just a few minutes per trade. That’s how he traded from start to finish. The only difference was that, at the beginning, he used smaller lot sizes because the leverage was only 1:100. As he kept winning, he gradually increased his lot size until reaching the maximum of 50 lots.
Luu Tuan Trung:

Does anyone recognize this technique?

Telegram users were discussing this kind of trading. Here is an automatic translation of that.

Hope you don't mind the spam, friends 🤓


I've been racking my brain over this all day...


Gold.

It uses a simple algorithm. Every minute, it places two pending orders equidistant from the current price (40–60 pips away). When one triggers, the other is deleted, and a stop-loss is set for the active order—around 80–130 pips out—with a matching trailing stop...


But how does it determine the values ​​for these ranges? It clearly involves some simple ATR calculation that affects both ranges...


In case anyone else is curious to dig into this... I can share my MQL5 version if needed...


Algorithm.


Two pending stop orders in opposite directions are placed every minute.

As soon as one is triggered, the other is deleted.

The position is closed via a trailing stop.


Result.

Expected value: 20 pips. Commission: -6 pips. Fixed spread: 6 pips. No slippage.


Live trading.

Commission: -16 pips; slippage is exclusively negative. Spread is several times higher.


Conclusion.

If this trading strategy were replicated perfectly, the expected value would be negative even without slippage.

fxsaber #:
It uses a simple algorithm. Every minute, it places two pending orders equidistant from the current price (40–60 pips away). When one triggers, the other is deleted, and a stop-loss is set for the active order—around 80–130 pips out—with a matching trailing stop...

No, that’s not correct. Scalping like that would gradually lead to a steady decline in capital. As for that guy, he trades both buy and sell positions simultaneously. I’m not completely sure, but he doesn’t hold onto either winning or losing positions for too long. There are always both buy and sell positions open, and the buy/sell ratio is very well balanced. Both sides have a profit rate of approximately 52%, and his account equity grows very steadily.

Luu Tuan Trung #:

No, that’s not correct.

I have no idea how you perceive the result of the translation of the original message.


Open a demo account with this broker and evaluate how well the trading conditions match reality.


Use your critical thinking.

Luu Tuan Trung #:

No, that’s not correct. Scalping like that would gradually lead to a steady decline in capital. As for that guy, he trades both buy and sell positions simultaneously. I’m not completely sure, but he doesn’t hold onto either winning or losing positions for too long. There are always both buy and sell positions open, and the buy/sell ratio is very well balanced. Both sides have a profit rate of approximately 52%, and his account equity grows very steadily.

Without seconds nor milliseconds listed in the "Open time" and "Close time" columns, it's difficult to guesstimate whether the strategy implements OCO (one-cancels-the-other) orders. For example, the first 2 trades listed up top appear to have had hold times that overlapped by 1 minute but if one trade closed at second #0 and the other trade closed at the preceding second #59, that could merely be a second/millisecond-based execution latency in closing one leg of the OCO bracket─masquerading as a full minute.
Ryan L Johnson #:
Without seconds nor milliseconds listed in the "Open time" and "Close time" columns
Unfortunately, there aren't any milliseconds , all the orders are just like the ones in that photo. I'm really confused—could someone suggest a few trading styles similar to that.
Luu Tuan Trung #:
Unfortunately, there aren't any milliseconds , all the orders are just like the ones in that photo. I'm really confused—could someone suggest a few trading styles similar to that.

Please do that, and I'll have a look at it.

For now, I extracted the text from the image, pasted it into AI, and got this:

OCO orders (one-cancels-other): NO evidence

    • SL (Pips) and TP (Pips) columns are - on all 20 rows — no stop/target was attached to any position, so there was no SL–TP OCO pair to cancel.
    • The close prices are irregular market prices (4,369.76, 4,363.61, 4,357.11…), not round-number take-profit fills. Exits look like direct market closes issued by the EA when its logic flips.
    • The report contains only market Buy/Sell deals — no pending-order rows (buy_limit/buy_stop etc.) would appear here even if they existed, but there is nothing in the trade-derived data suggesting them.

So: the closest probable architecture is a hedging EA that opens opposite legs and closes each side directly with market orders, not an OCO/SL-TP-based system.

Risk note (from the data)

    • None of the 20 trades carried SL or TP — during gaps between offset legs the account held full naked 50.00-lot exposure (≈ $21.8M notional). That's exactly when the two biggest drawdowns occurred: −$21,300 (−2.17%) at 15:33 while the profitable Buy leg was still open, and −$16,500 (−1.74%) at 15:14–15:17.
    • The "hedge" protects only when both legs are open; the moment one leg is closed, the remaining leg is fully exposed. With no SL, a fast adverse move on the remaining leg can be catastrophic.
Extracted text:
Luu Tuan Trung #:
No, that’s not correct.

The expected value is 29 points. For Gold this is extremely thin, to the extend that it is close to the average spread for a CFD feed.

The edge comes from the demo account. Most likely little to no slippage and no commission.

Saber is right, think critically and open the demo account and verify if the trading conditions bear any relation to reality. Only then, after answering this question, worry about the actual strategy.

Imp there is nothing here, just as tester-grails exists so do demo-grails. Heck there are even fake real account grails, some of which you can find here in the signal service as wel with 10000+ percentages of growth. It is not real.

From what I can see in the data
Lot 50 every trade. No SL or TP anywhere. Average loss bigger than average win. He wins on frequency not on RR. Also it is demo contest, so blowing up costs nothing. Different incentive.
New comment