Trend following EA: what a trend following forex robot looks like on a verified account

What a trend following EA actually does

A trend following forex robot waits for price to start moving in one direction, joins it, and stays in until the move shows signs of ending. It does not predict. It reacts late, on purpose, and accepts that it will be wrong most of the time. The profit comes from the few trades that catch a 300-pip move and are allowed to run, not from being right often.

Most trend EAs on MT4 and MT5 use some combination of moving-average alignment, channel breaks or momentum to decide a trend exists, place a stop-loss behind the entry, and then trail that stop as the move develops. Some pyramid into a winning position. The holding time is days to weeks, and the trading bot is usually flat more often than it is in the market.

This is the category with the longest real history. Managed futures funds have run trend following for decades, which is why it is also the one with the fewest vendors: it is hard to sell a product that loses six trades in a row as a matter of routine.

The statistics you will see

On a verified account page, a genuine trend EA shows a win rate between roughly 30% and 45%. Profit factor on a long record sits around 1.3 to 1.8. The average win is two to four times the average loss, so the payoff ratio does the work that win rate cannot. Expectancy per trade is decent in currency terms but arrives in lumps.

Average trade length is measured in days. In the Activity distributions histogram, most trades land in the 1–7 days and over-7-days buckets, with a cluster of quick losers in 4–24 hours where the initial stop was hit. Trades per week is low, often under five per symbol, so a trend account needs a year or more to reach the 200 trades that earn full longevity points in the Score.

The equity curve is a staircase with long flat or sagging stretches between steps. Drawdowns are moderate in depth but long in duration: 15–25% falls that take three to six months to recover are normal. The Z-score tends to be negative, meaning wins and losses cluster more than chance, because trends come in regimes. The long-versus-short split in Detailed stats is usually balanced over time but lopsided over any given quarter, since one direction is trending. Open positions are few, typically one or two, and they carry a stop-loss that has been moved into profit.

A worked example

Say the EA wins 35% of the time with an average win of 2.8R, where R is the risk per trade. Expectancy is 0.35 × 2.8 − 0.65 × 1 = 0.33R per trade. At 1% risk that is a third of a percent per trade, which compounds nicely over 150 trades a year.

Now the uncomfortable side. With a 65% loss rate, the chance of eight consecutive losses is 0.65⁸, about 3%. Over 300 trades you will see an eight-loss streak several times and a ten-loss streak at least once. At 1% risk per trade, a ten-loss streak is a 10% drawdown before any winner arrives. Nothing is wrong with the EA when this happens. It is the price of admission. The risk of ruin calculator will tell you what risk per trade keeps that streak survivable.

Advantages

• Losses are capped by a real stop on every trade, so the worst case is a long grind rather than a single event.
• The strategy benefits from volatility and crisis moves, which is exactly when most other EA types die.
• It transfers between brokers well: a 200-pip winner does not care about a 0.3-pip spread difference.
• Long holding times mean open positions do not leak the strategy, and reveal delays matter less.

Disadvantages and failure modes

Ranging markets kill it slowly. A trend EA in a six-month sideways market will be stopped out repeatedly with small losses, and the account can bleed 20% with nothing to show. Swap costs bite when positions are held for weeks against the interest differential; check the Swap column in Trade records if the owner exposes it.

The psychological failure mode is the real one. Buyers switch it off after the fifth losing trade, then watch the sixth catch the move. Another failure mode is over-optimisation: an EA tuned to the last two years of trending gold will have a perfect backtest and no forward edge. The monthly returns calendar shows this as a cluster of green months followed by steady red.

What the AI review and Score tend to flag

The review classifies an average hold over 48 hours as Trend / Swing. High-severity flags are rare for a real trend EA because stop-loss usage is near 100% and average loss is smaller than average win. What does get flagged is concentration: many trend EAs trade one pair or only gold, and above 85% in one symbol the review raises a watch flag. A drawdown watch flag appears above 40%, which a well-sized trend EA should never reach.

The Score usually lands in the B range for a good one. Profitability is capped by a modest profit factor, risk control is decent, consistency loses points because a third of months are red, and longevity only fills in after a year.

Red flags when buying a trend following EA

• Win rate above 60% or average loss bigger than average win: the EA is cutting winners, or it is a hidden grid.
• Stop-loss on fewer than 15% of trades: the review will flag it, and a trend EA without stops is a swing martingale in waiting.
• A record that contains only one trending period. Use the calendar to see whether it survived a range.
• Maximum drawdown under 5% on a two-year record: trend following does not work that smoothly, so something is being hidden or it is a demo.

How to check on an account page

In Trade statistics confirm the low win rate, the payoff ratio above 2 and the multi-day average trade length. In the Monthly returns calendar count the red months and look for the long flat stretches; they should exist. In Detailed stats and long vs short, check that both directions contribute over the full record. On deeper plans, Performance depth's deepest drawdowns table shows depth, duration and recovery for the five worst spells; expect durations in months. Risk Lab's chance of ruin at the −25% rung should be low if position sizing is sane, and the Kelly note should not say positions are oversized. If the account runs several robots, By magic number separates the trend EA from anything else. Compare against the opposite style in mean reversion EA, and read win rate is lying before you judge the 35%.