In July 2026, GBPUSD started climbing and didn't stop for six weeks. I watched the first 200 pips from the sidelines, told myself "it's overextended," and nearly missed the entire move. DarkWolf MT5 didn't have my hesitation. Its trend-following algorithm entered on the pullback at 1.2680, added on the second pullback at 1.2840, and rode the pair to 1.3480. One position sequence, +$6,200 on a $10,000 account, and I did absolutely nothing except not interfere.
That trade rewired how I think about the darkwolf mt5 trend following algorithm. Trend following isn't about predicting direction — it's about having a mechanical answer to three questions: when to enter, when to add, and when the trend is over. Humans answer those questions with feelings. The algorithm answers with scores. This is the full breakdown of how those scores work and how I configure them.
The calculator fields behind every trend trade I take
Before I let the trend follower loose on any pair, I sanity-check the setup on the DarkWolf MT5 Signal Calculator. If the calculator says the risk math is ugly, I skip the trade no matter how pretty the trend looks. These are the fields I fill in every single time.
| Input Field | What it means | Sample value |
|---|---|---|
| Trend candle High / Low | The impulse candle that confirmed the trend — anchors the whole setup | 1.0921 / 1.0865 |
| Account Balance | Live equity the EA is actually trading with | $25,000 |
| Risk Per Trade | Fixed loss budget — trend trades need room, but the cap never moves | 1.5% |
| Stop Loss (pips) | Below the last swing low; wide enough to survive pullbacks | 60 pips |
Letting winners run: the hardest setting to trust
The entry is the easy part. The hard part is the exit — specifically, not exiting. My journal from the GBPUSD run shows I wanted to close the trade four separate times: at +$1,800 ("bank it"), at +$3,100 ("it's going to reverse"), at +$4,400 ("definitely overextended now"), and at +$5,600 ("okay seriously, close it"). Each time, the algorithm's trailing logic said the trend structure was intact. Each time, it was right.
In the darkwolf mt5 trend following algorithm, the setting that makes this possible is the trend-exit mode: instead of a fixed take-profit, the EA trails the stop below the last swing low (for longs) and only exits when structure breaks. My darkwolf mt5 trend following algorithm configuration uses a 1.5x ATR trail — wide enough to survive normal pullbacks, tight enough to keep most of the move. On the GBPUSD run, the trail survived three pullbacks of 80–120 pips and finally exited 60 pips below the top. I left $600 on the table and kept $6,200. That's the deal trend following offers: you never sell the top, and you don't need to.
The psychological trick that finally worked for me: I stopped thinking of open profit as "my money at risk" and started thinking of the trailing stop as the actual position. The stop is the trade; the floating profit is just weather. Once that clicked, letting winners run stopped feeling brave and started feeling mechanical.
Trend presets I ran through the signal calculator
Before trusting the darkwolf mt5 trend following algorithm with real money, I modeled four trend configurations on the DarkWolf MT5 Signal Calculator against two years of GBPUSD and EURUSD data:
| Preset | Entry / Exit logic | 2-year result |
|---|---|---|
| Conservative | Pullback entries only, 2.0x ATR trail | +34%, max DD 6.1% |
| Standard | Pullback + breakout, 1.5x ATR trail | +51%, max DD 8.4% |
| Aggressive | Breakout entries, 1.0x ATR trail | +47%, max DD 12.2% |
| With pyramiding | Standard + add on 2nd pullback | +68%, max DD 9.7% |
I run the "with pyramiding" preset live. The aggressive preset's extra 4% drawdown wasn't worth the lower return — tighter trails get shaken out of trends that the standard trail survives. And pyramiding added 17 points of return for only 1.3 points of extra drawdown, because adds only trigger when the trend has already proven itself. The compounding growth strategy guide shows how I size those pyramid adds without blowing up the risk budget.
Trend classification: how the EA scores momentum
The darkwolf mt5 trend following algorithm doesn't just detect trends — it classifies them into early, mature, and late phases, and treats each phase differently. This is the core mechanic most EAs skip.
Early trend: fresh structure break with expanding momentum. The EA takes full-size entries here with the widest trail — this is where the 800-pip moves are born. Mature trend: established direction, orderly pullbacks. Full size continues, but no new pyramid adds. Late trend: parabolic moves, shrinking pullbacks, momentum divergence. The EA reduces size by 50% and tightens the trail to 1.0x ATR. It doesn't predict the top; it just charges less rent for staying in a tired trend.
The classification uses three inputs: moving-average slope and separation, the sequence of swing highs/lows, and momentum divergence between price and the EA's internal oscillator. When I backtested phase-aware vs phase-blind versions of the same logic, the phase-aware version kept 23% more of each trend's total move — mostly by not giving back the late-phase gains. If you want a visual companion for trend phases, the daily bias dashboard shows the same classification in real time.
Riding GBPUSD's 800-pip move on a $100,000 account
The full anatomy of the July 2026 run, on a $100,000 demo I was using to validate the pyramiding preset:
| Stage | Price | EA action | Floating P&L |
|---|---|---|---|
| Entry | 1.2680 | Long 2.0 lots, early-trend signal | $0 |
| Pullback 1 | 1.2840 | Add 1.0 lot (pyramid trigger) | +$3,200 |
| Mature phase | 1.3120 | Trail tightened, no new adds | +$11,400 |
| Late phase | 1.3390 | Size effectively halved via tighter trail | +$18,900 |
| Exit | 1.3420 | Structure break, both positions closed | +$19,600 realized |
A 19.6% account gain from one trend sequence. The honest footnote: this was the best trend of the year, not the average one. My average trend trade makes 2–4%. But the darkwolf mt5 trend following algorithm only needs two or three of these a year — the math of trend following is that a handful of outsized winners pay for dozens of small stopped-out attempts. My gold strategy breakdown shows the same phase logic applied to XAUUSD's sharper, shorter trends.
The trend-strength equation behind each entry
Every darkwolf mt5 trend following algorithm entry carries a strength score. Below 60, no trade. Here's the simplified version of what the EA computes:
| Variable | Meaning | Example value |
|---|---|---|
| Slope | MA angle and separation, 0–1 scale | 0.85 → 29.75 pts |
| Structure | Higher highs/lows sequence quality, 0–1 | 0.9 → 36 pts |
| Momentum | Impulse vs pullback strength, 0–1 | 0.7 → 17.5 pts |
| Trend Score | Total of weighted components | 83.25 — strong early-trend entry |
Structure carrying the most weight (40) is deliberate: slope can be faked by one sharp candle, but a clean sequence of higher highs and higher lows can't. The ECB's research on trend persistence and BabyPips' trend-trading school both converge on the same finding — structure-based entries outperform indicator-cross entries over long samples.
Chasing late trends: $620 of tuition
The mistake that taught me to respect the phase system. In August, EURUSD had already rallied 350 pips when I noticed it. The EA scored the trend at 58 — below the 60 threshold — and refused entry. I overrode it manually: "the trend is obvious." I bought at the top of the late phase. Price reversed 90 pips over the next two days. My manual trade: -$620. The EA's non-trade: $0.
The journal entry I wrote: "The EA wasn't being cautious. It was being correct." Late-phase trends have the worst risk-reward in trend following — you're buying the move everyone already owns, with a full-sized stop below and limited room above. The darkwolf mt5 trend following algorithm's phase system exists to keep you out of exactly that trade. I now have a rule: if the EA scores it below 60, I don't take it manually either. The machine's discipline has to cover my FOMO, not just its own entries.
The deeper lesson: most of my losing manual trades were late-phase chases. Trend following works because of early entries with wide trails. Everything else is expensive entertainment.
Trend following algorithm — FAQs
How does the EA know a trend is starting?
The darkwolf mt5 trend following algorithm scores structure breaks: MA slope and separation, a fresh higher-high/higher-low sequence, and momentum confirmation. A score above 60 triggers early-phase entries at full size. It doesn't predict — it reacts to confirmed structure, which means it enters later than a human guesser but with far fewer false starts.
What's the difference between early, mature, and late trend phases?
Early: fresh break, full size, widest trail — where big moves begin. Mature: established direction, full size continues but no new pyramid adds. Late: parabolic or divergent, size effectively halved via tighter 1.0x ATR trail. The phase system kept 23% more of each trend's move in my backtests versus phase-blind exits.
Should I use fixed take-profits or trailing stops?
Trailing stops, without question, for trend following. A fixed TP caps your best trades — the GBPUSD run would have made a third as much with a fixed 200-pip target. The 1.5x ATR trail survived three 80–120 pip pullbacks and captured 740 of the 800-pip move. Fixed targets are for ranging strategies.
How does pyramiding work in the trend preset?
In the darkwolf mt5 trend following algorithm, after the initial entry the EA adds one position (default half size) on the second confirmed pullback, only in early-to-mature phases. Adds share the same trailing stop logic. My two-year test showed pyramiding added 17 points of return for 1.3 points of extra drawdown — the adds only fire when the trend has proven itself.
What markets suit the trend algorithm best?
GBPUSD and EURUSD on H4/D1 — clean, persistent trends with manageable pullbacks. Gold trends harder but reverses violently; I run gold at reduced size. Avoid ranging pairs like EURCHF. The algorithm tells you itself: if scores rarely clear 60 on a pair, that pair doesn't trend enough to bother.
How many trend trades per month are normal?
With the darkwolf mt5 trend following algorithm, fewer than you think: 4–10 per pair. Trend following is a low-frequency game — most of the profit comes from 2–3 big winners per quarter. If your instance takes 30 trend trades a month, the phase filter is too loose and you're trading noise with trend-sized stops.
Can trend following work on small accounts?
Yes. The darkwolf mt5 trend following algorithm is size-agnostic — a $500 account runs the same scores and phases. Use 0.5% risk and expect the dollar amounts to be small; the percentages are what matter. My first profitable quarter on trend mode was on a $1,000 account making $180. The process scaled cleanly to six figures later.
What kills trend-following performance most often?
For the darkwolf mt5 trend following algorithm, the killer is manual interference — closing winners early. My journal showed four separate urges to close the GBPUSD run early; each would have cost thousands. Second is chasing late-phase trends manually when the EA correctly refuses. Set the preset, configure the trail, and let the structure decide. Your job is patience, not prediction.