Pitt Petruschke / Perfil

Free tool: Expert Advisor Statistics by Magic Number (MT5)

You run several EAs on one account, plus a few trades by hand. The balance shows the total, not which EA earns the money and which one gives it back. Magic Stats reads the deal history and shows one row per magic number: trades, win %, net profit, profit factor, maximum drawdown, recovery factor and open P/L. Tabs for today, week, month, year, all or a custom date, sorting by column and a CSV export.

Trades are counted per position, and the magic number comes from the opening deal. Stop-loss and stop-out closes often carry magic 0, so an EA's losing trades do not end up under "Manual". Swap, commission and fees are included. It is an indicator: read-only, no internet functions, no DLL, and it never trades.

The drawdown is closed-trade drawdown. The drawdown while a trade was open is not in it and can be much larger. Magic Stats sees what the terminal's history contains, so load the full history once on a fresh terminal. Free and fully working, no time limit.

Product: https://www.mql5.com/en/market/product/200695
Complete guide: https://www.mql5.com/en/blogs/post/777293

Free tool: Daily Loss Limit and Equity Lock for MT5

A daily loss limit only helps if it is enforced, also on the day you would rather ignore it. You set one line for the day, in money or in % of the day reference (default 5 %), with a reset at a server time you choose. The tool runs on one chart and watches the whole account: manual trades, the mobile app and every EA.

When equity reaches the line, pending orders are deleted first, then positions are closed, largest loser first. The account stays locked until the next daily reset; new trades opened while locked are closed again. An optional daily profit target works the same way, and a warning comes at 70 % of the limit used. The day reference and the lock survive restarts, and deposits or withdrawals do not count as profit or loss.

Gaps and slippage can take equity below the line, and it only works while the terminal runs with Algo Trading on, so a VPS is recommended. It is a daily guard, not a prop-firm rule monitor. Free and fully working, no time limit. It never opens a trade.

Product: https://www.mql5.com/en/market/product/200696
Complete guide: https://www.mql5.com/en/blogs/post/777292

Free tool: Economic News Filter and Trading Pause for MT5

High-impact news such as Non-Farm Payrolls, CPI or a rate decision can move a market in seconds: spreads widen, prices gap, pending orders fill far away. News Filter reads the economic calendar built into MetaTrader 5 and shows the next 8 events on a panel. Around every event that counts it opens a news window, by default 15 minutes before and after. One instance on one chart watches all symbols.

Only the alert is on by default. If you switch them on, it deletes pending orders at window start, closes positions (all or only losing ones), closes trades opened inside the window and publishes a pause flag your own EAs can read. At most 100 close or delete requests per day (adjustable). It never shows CLEAR while the calendar is stale or the terminal is offline.

News Filter cannot stop another EA from sending orders; MetaTrader 5 does not allow that. It can only alert, delete and close afterwards. The MT5 calendar does not exist in the Strategy Tester, and its data depends on your broker. Free and fully working, no time limit. It never opens a trade.

Product: https://www.mql5.com/en/market/product/200697
Complete guide: https://www.mql5.com/en/blogs/post/777291

Recently released: Drawdown Minimizer – old losers paid off with profit already earned (MT5)

Drawdown Minimizer is a utility that runs next to your other EAs and never opens a trade. It keeps a savings book of the net realized profit of the whole account and uses it to close the oldest losing positions of the EAs you choose, slice by slice. It was built because one of our grid EAs sat in drawdown for months while the account kept booking profits elsewhere.

Only what has been earned is spent. Deposits, withdrawals, credit and bonus never count. You select the positions by magic number, magic range or comment text; winners and all other positions are never touched. Spending starts at a position age, at an account drawdown level or when you press START, and PAUSE always wins. Dry run is on by default, so the journal first shows what it would close.

Closing a loser with saved profit does not change your equity: a floating loss becomes a realized one. What changes is the risk: less open volume, less margin, no more swap on old positions. If the market would have come back, that way back is gone for the closed slices. Check in dry run how your EA reacts.

Product: https://www.mql5.com/en/market/product/201728
Complete guide: https://www.mql5.com/en/blogs/post/778023

How RecoveryPilot works: paying off a losing trade part by part (MT5)

RecoveryPilot works a losing position back instead of hoping for a reversal. In its main mode, Partial Recovery, it locks the losing trade with an opposite position, splits the volume into up to 8 parts and pays the frozen loss off part by part with small recovery orders in the direction of the trend. It can take over your manual trades or use its own entries. Grid averaging and zone recovery are included as alternatives.

Before a cycle starts, the EA calculates what it can lose. If that number does not fit your account, the cycle does not start. The auto profile sets step, lot and order count per symbol, and a symbol check rates each symbol from GOOD to NOT SUITABLE. Every recovery order has its own broker stop, and no martingale is used by default. If the recovery orders lose more than 5 % of the balance (default), they are closed and the lock stays in place.

A recovery system does not turn a losing trade into a sure win. It needs time and free margin, the lock pays swap, and a strong trend against the recovery orders costs money. Please start on a demo account and read the level plan in the Journal.

Product: https://www.mql5.com/en/market/product/199239
Complete guide: https://www.mql5.com/en/blogs/post/776887

Snapback Pro EA: trading the snap back after an overstretched day (MT5)

Snapback waits for a day on which price stretches far beyond the previous day's range, by default more than four times half of that range. It lets the first real pullback show, then enters against the stretch. If price keeps going, it works the trade as a linear grid: lots grow by one start lot every three orders, no exponential martingale. The basket is closed with a trailing stop. One chart trades up to 20 pairs, with at most 5 pairs at risk at the same time.

MT5 real-tick backtest (RoboForex ECN, 1 Jan 2025 – 9 Oct 2026, 0.01 start lot, no equity stop): the 3-pair set made +2,637 USD with a maximum drawdown of 818 USD and a profit factor of 1.68 (recommended minimum equity 2,500 USD). The 7-pair set made +3,606 USD with 1,147 USD drawdown and a profit factor of 1.59.

Snapback is a grid EA. Single positions have no stop loss, and the set files use no equity stop. All 20 pairs together on 3,000 USD reached a stop out in March 2026, so use the capital recommended for your set. An optional equity stop is built in. Please start on a demo account.

Product: https://www.mql5.com/en/market/product/202250
Complete guide: https://www.mql5.com/en/blogs/post/778006

How Dawnbreak trades: one morning range, one breakout, one trade a day (MT5)

Dawnbreak measures the high and low of a fixed time window, for example 03:05–06:05 on gold or 07:00–11:00 on the DAX (server time). It places a BuyStop above and a SellStop below that range; the first to fill cancels the other. One trade per day and chart, a stop loss on every trade, a fixed close time. No grid, no martingale.

The lot size follows the stop distance and your risk per trade. Version 3.7 fixed an oversized volume on some index CFDs. MT5 real-tick backtest (Darwinex, 1 Jan 2025 – 14 Oct 2026, 10,000 USD, 1 % risk, no take profit): gold 413 trades, +41.65 %, maximum drawdown 20.77 %, profit factor 1.21. DAX 431 trades, +40.28 %, maximum drawdown 12.44 %, profit factor 1.20.

Only gold and the DAX passed the test. EURUSD, GBPUSD, USDJPY, EURJPY and US30 missed the pass mark. The DAX curve has been falling since May 2026. Gold at 1 % risk needs about 7,500 USD or more. Please test with your broker's data and on a demo account first.

Product: https://www.mql5.com/en/market/product/178401
Complete guide: https://www.mql5.com/en/blogs/post/776683
Live account (my own, same strategy, since Feb 2026)*: https://www.mql5.com/en/signals/2372831

* Different EA build. The account mixes this strategy with two others, so its result is not a track record of Dawnbreak's code.

Product spotlight: Sextant Trade Panel – every trade sized from your risk (MT5)

Sextant is a trade panel, not an EA. You set how much you want to risk, place the stop where the chart says, and Sextant calculates the lot size, sends the order and manages the trade: break-even, partial close, five trailing methods and bulk actions. Everything sits in one tabbed panel: TRADE, MANAGE, GUARD and INFO.

The lot size is rounded down, so the loss at the stop never exceeds your risk. Before sending, Sextant checks the broker's minimum stop distance, volume limits, free margin and your maximum spread. The guard adds a daily loss limit, a daily target, trade caps and a room check that refuses a trade if its stop could break today's limit. On a hit it closes everything and blocks new trades until the next server day. All 17 built-in self-test checks passed in the MT5 Strategy Tester with real ticks.

A panel helps you keep your rules; it does not make trades profitable. Hidden stops and the guard only work while the terminal is running, so a VPS is recommended. You can try every button in the visual Strategy Tester first.

Product: https://www.mql5.com/en/market/product/199309
Complete guide: https://www.mql5.com/en/blogs/post/778504

Recently released: Relay – Dawnbreak, LongRunner and Second Wind as one EA (MT5)

Relay runs three strategies from one chart: Dawnbreak (morning range breakout), LongRunner (daily index long without swap) and Second Wind (Turnaround Tuesday). It has 13 slots, one strategy on one symbol each, with its own magic number and the published example set of the single EA. One overview panel shows the whole account, a profit matrix per symbol and strategy, and a pause switch per slot. The overview never sends, changes or deletes an order.

Built-in sizes: Dawnbreak risks 1 % per trade. LongRunner and Second Wind trade without a stop loss, and on a Monday the two together can hold about 2.8 × the balance in indices. Every size can be changed per slot. In the MT5 real-tick check (IC Markets, 1 Jan – 9 Oct 2026) 12 of 13 slots were identical deal for deal with the single EAs. That shows Relay trades like them; it is not a result of the example sets.

Please read the Dawnbreak slots: in the latest test only gold and the DAX passed. GBPUSD is off by default; EURJPY, USDJPY, US30 and BTCUSD stay on although they did not pass or were not tested. You can switch each off with one input. Relay needs a hedging account. Test on a demo account first.

Product: https://www.mql5.com/en/market/product/201730
Complete guide: https://www.mql5.com/en/blogs/post/778503

How Second Wind works: Turnaround Tuesday on stock indices (MT5)

Weekend news often pushes indices lower on Monday. When that happens while the index is already in a pullback, Tuesday has tended to recover. Second Wind buys on Monday while price is below its moving average (default: 20-day SMA) and sells at a fixed time on Tuesday. At most one trade per week and symbol, long only, no grid, no martingale, no averaging.

The 21-year check (Python simulation on daily data, Jan 2005 – Oct 2026, position value 0.5 × balance, spread and one night of swap deducted) gave a profit factor of 1.39 to 1.50 on four indices, about 2.5 to 3.2 % a year, and a maximum drawdown of 9.0 to 13.2 %. The moving average filter beat "buy every Monday" on all four indices, and with doubled costs the profit factor stayed at 1.34 to 1.43.

The returns are modest: the position is half the balance and the EA is in the market about one day a week. The MT5 real-tick tests from 2025–2026 look much better, but that was a short and strong period for index dips. Please test on your broker's data and on a demo account first.

Product: https://www.mql5.com/en/market/product/200107
Complete guide: https://www.mql5.com/en/blogs/post/778434
Live account (my own, same strategy, since Feb 2026)*: https://www.mql5.com/en/signals/2372831

* Different EA build. The account mixes this strategy with two others, so its result is not a track record of Second Wind's code.

Product spotlight: LongRunner – one index long a day, flat before the rollover (MT5)

LongRunner buys a stock index at a fixed server time and sells it again the same server day, before the broker's daily rollover. Indices have risen over the long run, but holding them as a CFD costs swap at every rollover. LongRunner takes the daily move and leaves that cost out. One trade per day and symbol. No grid, no martingale, no averaging.

The optional trend filter only buys while the last daily close is above its 200-day SMA. In the 26-year Nasdaq-100 simulation it cut the maximum drawdown from 56 % to 16 %. MT5 real-tick backtest (RoboForex, Dec 2017 – Jul 2026, 10,000 USD, trend filter on, 0.5 × balance): 1,645 trades, profit factor 1.10, +38.8 %, maximum drawdown 15.7 %.

Spread decides whether the edge survives: the same Dow trade was profitable at one broker and lost at another. There is no S&P 500 set, it showed no edge after costs. Without a stop loss a crash day hurts: the worst test day was −10.9 % on the Nasdaq-100. Please test with your broker's real ticks and on a demo account first.

Product: https://www.mql5.com/en/market/product/198979
Complete guide: https://www.mql5.com/en/blogs/post/778433
Live account (my own, same strategy, since Feb 2026)*: https://www.mql5.com/en/signals/2372831

* Different EA build. The account mixes this strategy with two others, so its result is not a track record of LongRunner's code.

New on the Market: Fairway – AUDCAD mean reversion grid with a basket stop (MT5)

Fairway is a grid EA for AUDCAD on M5. It opens a basket when the price is stretched beyond an H1 Envelopes band, adds orders at a growing distance while the price keeps running, and closes them in profitable pairs on the way back. BUY and SELL baskets run independently, so it needs a hedging account.

Grid EAs have one weak spot: a long trend against the basket. Fairway closes a basket completely once it reaches a set loss (default 150 USD per 3,500 USD of balance), then waits for the next signal. In the MT5 real-tick backtest (RoboForex, 3,300 USD, 1 Jan 2024 to 17 Oct 2026, default set) it made +226.54 USD with a profit factor of 1.16, 1,371 trades, 9.0 % maximum equity drawdown and 6 basket stops. The same logic without the stop lost 539.02 USD with a 57.8 % drawdown.

The stop has a price: 2024 ended slightly negative (-96 USD) because three baskets were stopped. Fairway is still a grid EA and can hold a losing basket for days. Please test it on your broker's real ticks and on a demo account first.

Product: https://www.mql5.com/en/market/product/203336
Complete guide: https://www.mql5.com/en/blogs/post/778374
Live account with the default set: https://www.mql5.com/en/signals/2396880

Conta real do Fairway , o meu EA de grelha para AUDCAD M5 com um stop de cesta fixo. Iniciada em 23 de outubro de 2026 com 1.000 USD na RoboForex (ECN) – acompanhe cada operação ao vivo desde a primeira semana. Nesta conta opera apenas o Fairway, com as configurações padrão: AUDCAD M5, retorno à média da banda Envelopes no H1 (EMA 14, 0,2 %) Lote inicial 0,01, novas ordens a uma distância crescente (20, 24, 28,8 … pips), fechadas em pares com lucro Stop de cesta: uma cesta que atinge −150 USD é