AI Thomas

v1.2 is a good version

Do not be fooled by the historical backtests of AI Thomas. It is a classic textbook example of an over-optimized, curve-fitted strategy that completely fails in real market conditions.Consistent Live Losses: While the history shows perfect profits, the actual live performance is a steady, painful drain on capital. It simply cannot adapt to live markets.Terrible Risk-to-Reward: The EA uses an excessively far stop-loss. This "small wins, huge losses" logic means a single negative trade consistently wipes out weeks of minor gains.The Verdict: The strategy is fundamentally flawed and highly overfitted. It won't blow your account instantly, but it will bleed your live capital dry through continuous losses. Avoid.

I bought this based on the claim that it was a single-position strategy with SL and TP set for every position, but after the update, it had been automatically changed to a grid system. With a $1,000 account, 0.01 lots should be plenty, but before I knew it, five additional positions had been added via the grid—completely ignoring risk management. This is terrible. If there had been an explanation of the grid system along with the update notice, I would have stopped using it right away, but because I didn’t know, I suffered massive losses. Even when used as a single-position strategy, the stop-loss is set at 950 pips—about four times the take-profit level—which is extremely high. The actual results were a far cry from the developer’s signals, and it was a huge disappointment. I do not recommend it.

The dev has a great track record - saw a live signal on his profile page displaying over 2000% profit with under 15% max drawdown a year ago if I'm remembering correctly. It’s been deleted but I recall it didn’t seem like the cause was due to an account blowup. Perhaps a broker issue or something else. Whatever the case he has enough experience to make a robust algorithm.

The one star reviews are pretty funny - first of all two weeks isn’t enough time to quantify if an algorithm performs in a live account like in the backtest. Second, the backtest performance statistics like sharp and return to drawdown ratio are similar to other sellers gold EAs live statistics like gold reaper. The claim of over optimization is rich and is literally backed up with nothing; it's merely a longwinded rant by a beginner.

The EA has roughly a 1:1 risk to reward ratio when not using the break even or trailing stop which is ideal. Even if used it’s still roughly 1:1. It’s not a grid system obviously, there’s a grid option however it can be disabled. Don’t recommend using it.

Vitali - please don't be discouraged keep making this type of algorithm.

Accurate on trend

Update:The SL is wide so the trade can remain for longer period. If it recovers its fine but an SL hit can be twice the TP