SomaOil EA Review (2026): Twenty Breakout Strategies on WTI Crude, Two Live Signals and a Price That Rises With Every Ten Sales

53.3 / 100

weighted verdict score on eight published criteria

PF 1.88 and 1.93

profit factors of the two live MQL5 signals, over 134 and 76 trades

$500 or $60/mo

Market price to buy or rent for a month on 28 Oct 2026, with a free demo; the vendor's launch price was $100

The verdict at a glance: the best-explained oil EA we have read, with the youngest record

SomaOil earns points for candour about how it was built. The listing names the optimisation ranges, the number of strategies that survived each filter, the meaning of every magic-number digit and the four risk presets. It loses points for time. Sixteen weeks and 134 trades is a good start and not a record, the two signals run different presets, and the way the strategies were chosen leaves the backtest less informative than it looks. The scorecard and radar show how the total was built.

What SomaOil actually is

SomaOil is an Expert Advisor for MetaTrader 5, sold on the MQL5 Market by Andrii Soma of Ukraine, whose seller profile shows a 4.4 rating from 22 votes, 6 products and 16 signals. It was published on 17 July 2026, showed version 3.5 and was last updated on 7 October 2026, with 10 activations per licence. It is built 'exclusively for WTI crude oil (XTIUSD)'. One chart, one EA, twenty independent strategies running together as a single portfolio.

The construction is the interesting part. The author ran optimisations over two complementary history ranges, 2018 to 2024 and 2020 to 2026, so each range's out-of-sample window falls inside the other's in-sample window. He crossed five swing timeframes (D1, H12, H8, H4 and H1) with three breakout breadths (broad, medium and narrow), optimised each combination for return against balance drawdown and then against equity drawdown, and filtered the results. Strategies that correlated too strongly with another survivor were removed, and so were those that failed to match real ticks or failed their out-of-sample window. Twenty survived: seven from the first range and thirteen from the second, spread over three D1, five H12, two H8, six H4 and four H1 strategies. A proprietary Python optimiser then set each strategy's lot size from return against equity drawdown, giving less weight to strategies that draw down together.

The runtime features are broad. There are four built-in lot presets: V2.1 Conservative (the default), V2.2 Medium, V2.3 High Risk and the older V1.0, which balances by balance drawdown. A built-in news calendar blocks entries, cancels pending orders and optionally closes positions around NFP, CPI and rate decisions, with windows tuned per strategy. A spread filter stops new pending orders when oil spreads widen, an optional compliance layer adds a daily drawdown lockout and a pending-order cap, a margin-call guard reserves margin for same-side pending orders, and a random pip offset can vary entries for prop-firm accounts. A dashboard shows spread, drawdown, slippage, swap and commission over several windows.

How we judged it, and who has something to sell

Evidence on EAs is never neutral. Here is every party near this product and what each stands to gain, ours included.

• The developer sells the EA, rents it, sells signal subscriptions (30 USD a month on both signals, with one subscriber on the first) and runs the buyer groups. His record is public, which counts for something, and it is also his sales page.
• MQL5 and MetaQuotes host the Market, the reviews and the signals. The signal statistics and every notice on their pages are the platform's own.
• Buyers who wrote reviews are three, all positive, one of whom mentions a 'sketchy start' and a recovery. Testimony, not data, and each is unverifiable from outside.
• ForexCopilotEA sells a listing for this EA. We gain if you buy it, which is why this verdict opens with that fact and applies the same rules as every other.

What the two live signals show, number by number

Platform-computed signal statistics are the best public evidence a vendor can show. They still have limits: the vendor runs the account, the broker is one broker, the records are short, and the two signals use different lot presets. With those limits in mind, here is what the two SomaOil signals show on 28 October 2026.

Two profiles side by side show what the presets do. The older V1.0 preset earned more and drew down more: a 30.2% balance drawdown, a worst trade of minus 137.91 USD and a six-trade losing run worth minus 400.13 USD, on a 1,000 USD account. The V2 Conservative preset, the default for new buyers, has a 6.6% balance drawdown and a worst trade of minus 65.73 USD, roughly half. Its equity drawdown is nonetheless 15.4%, more than twice its balance drawdown, because the strategies hold positions for a day on average and floating losses sit under water before they close. Anyone sizing from the balance drawdown alone underestimates the swings they will see on screen.

The platform's notes add caution. On the Conservative signal MQL5 states that 80% of the growth came within three days, which is 4.11% of its 73 days. A breakout portfolio on oil earns most of its money in a handful of large moves and idles or bleeds between them. The profit factors of about 1.9 on both signals are respectable, the Sharpe ratios of 0.23 and 0.27 are low, and the average loss exceeds the average win on both. With a 67% to 70% win rate the arithmetic works: 0.7014 times 18.00 is 12.63 against 0.2985 times 22.45 at 6.70, which leaves the 5.92 USD per trade the platform reports. It would stop working if the win rate fell towards 55%.

How a twenty-strategy oil portfolio makes and loses money

A breakout strategy waits for price to take out a recent swing high or low and enters with a pending order, betting that the move continues. Crude oil is a good market for it because oil trends hard on inventory data, supply news and geopolitical shocks, and it is a hard one because oil also gaps, spikes and reverses and its spreads widen around rollovers and news. SomaOil's answer is breadth. Twenty strategies on five timeframes mean that when a slow D1 swing and a fast H1 swing both break, the portfolio participates at two scales, and when one side whipsaws the others may not trigger.

Diversification across timeframes does not remove regime risk. All twenty strategies are breakouts on one instrument. In a long, quiet range they all wait or all lose a little, and in a sharp reversal after a spike they can lose together. The vendor's lot optimiser tries to reduce that by giving smaller allocations to strategies that draw down together, which works on history and is not guaranteed to hold when the market changes. The signals, at 16 and 11 weeks, are the only evidence of how it behaves live, and the older signal's 30% drawdown shows that the portfolio can give back a third of an account in a bad stretch.

The backtest claims: well described, and less clean than they sound

Most listings say 'tested over many years' and stop there. This one gives the in-sample and out-of-sample ranges, the grid, the optimisation targets and the filters, and that is worth recognising. It is also worth reading closely. The two ranges are complementary halves of the same history, so every part of 2018 to 2026 was in-sample for one range and out-of-sample for the other. The vendor then kept only the strategies that passed, and he says so: strategies that failed their out-of-sample window were removed. Once a window is used to decide which strategies survive, it is part of the selection, and a survivor's performance in it is no longer an honest estimate of the future.

This is not a criticism of the method; it is the normal state of any portfolio chosen from a large grid, and the vendor's cross-validation is better than a single in-sample optimisation. The consequence for a buyer is simple. Treat the 2018 to 2026 equity curve as a fitted curve, expect live results to be worse than it, and rely on the live signals and your own demo. For how fitted curves differ from tests that predict, see our guide to optimising without overfitting.

Risk: stops and filters, a portfolio that can still draw down 30%

SomaOil is a better-behaved design than a grid. It has exits and stops, it is not described as a martingale or a grid, and its worst trades are 138 and 66 USD on 1,000 USD accounts. The risks that remain are the ones an oil breakout portfolio carries.

• Preset risk. The V1.0 preset had a 30.2% balance drawdown in 16 weeks and the High Risk tier is higher still. Start with V2.1 Conservative and judge it before moving up.
• Equity versus balance. Equity drawdown on the Conservative signal is 15.4% against 6.6% by balance. Plan for the figure you see on screen.
• Stacked positions. Twenty strategies can hold pending orders and positions at once; the listing's margin-call guard exists because same-side orders can over-commit free margin.
• Oil-specific execution. Spreads, slippage and gaps on crude are larger than on major forex pairs. The spread filter protects pending orders only, and once a trade is open the normal exits run.
• Regime and version risk. The records cover one stretch of oil, and the product is already at version 3.5 after launching on 17 July. A signal records what one preset and one version did.

What buyers say

The listing shows a rating of 5 from three reviews, which is too few to carry weight. One, from 14 October 2026, calls it a 'good EA' with a 'great recovery after sketchy start' and 'big potential', and adds 'and not XAU', a pointed comparison with the gold EAs that dominate the Market. One, from 22 July, praises the vendor's transparency about methodology, the multi-broker testing and the diversification of twenty strategies on higher timeframes, while warning that there may be months without gains. The third says it 'seems well put together' and that 'live trading will tell'.

The 'sketchy start' fits the older signal's 30% drawdown, and the 'live trading will tell' review is the most accurate of the three. With a few reviews and four months of history, the honest reading is that the early signs are encouraging and the sample is small.

Price: $500 to buy, $60 to rent, and a ladder that rises with sales

The Market price on 28 October 2026 was $500, with a one-month rental at $60 and a free demo of 727 downloads. The vendor says the price started at $100 and rises by $100 for every ten copies sold, at most once a day. That is a clearly stated ramp, and it means the price tracks demand, not the record. It rewards the earliest buyers, who took the most risk on the least evidence, and charges later buyers more as the record grows, which is a defensible logic. It also means the price today does not tell you what the product is worth.

The rental option is the most useful pricing feature we have seen on a young EA. For $60 a buyer can run the live product for a month on a demo or a small account, which is cheap insurance against a bad fit with a broker's oil symbol. A month will not show a regime, so use it to check execution and spreads, not to judge the strategy. Remember the other costs: a VPS, and an oil account with spreads the filter will accept. If you buy from a source other than the Market, such as our own listing, the price, version and support are different. Our listing is a separate purchase with its own terms, shows version 3.3 against 3.5 on the Market and ships without a user manual.

The score, criterion by criterion: 53.3 out of 100

The total is 53.3 out of 100. The grade A on evidence is earned by two platform-built signals; the score stays at 5 because both are vendor-run, at one broker, under four months old and run different presets. Risk, at 5, gets credit for exits, filters and no stated grid or martingale, and loses it for a 30.2% balance drawdown on one preset and a 15.4% equity drawdown on the default. Trust, at 6, is the best of the criteria because the vendor explains his method in unusual detail, and it is held back by the out-of-sample selection issue.

Who SomaOil fits, and who it does not

• It fits a trader who wants a documented, diversified oil breakout portfolio, runs a VPS on an account with controlled oil spreads, starts on the Conservative preset at a small risk multiplier and judges it over many months.
• It does not fit anyone who needs steady monthly returns, who will switch to the High Risk preset after one good week, or who cannot tolerate a 15% equity swing on screen.
• It is risky for prop-firm challenges with tight daily limits unless the compliance layer and a low multiplier are used: stacked pending orders around news can fill badly.
• It is a poor choice for a buyer who is paying mainly because the price is about to rise. Rent it for a month, or run the demo, first.

How it compares with other EAs we have scored

Most EAs we have scored trade gold, such as Gold House and Gold Snap. SomaOil is one of the few built for crude oil, and its multi-timeframe portfolio is a different design from a single breakout system. Its 53.3 reflects excellent documentation, a young record and a drawdown on one preset that is larger than the gold systems show. For how breakout, grid and scalper designs differ, see our strategy comparison and our EA categories.

Questions to put to the developer before you pay

1. Which version and preset does each signal run, and is the Conservative preset the one a new buyer receives?
2. Can you publish tester reports on real ticks for each preset, with the out-of-sample windows marked?
3. What happens to the portfolio in a prolonged oil range, and what is the longest flat period in the live record?
4. How were the other brokers tested, and can you show them?
5. What risk multiplier do you advise for a 1,000 USD account?
6. Why did the V1.0 preset reach a 30% balance drawdown, and what changed in V2?
7. How does the ramping price change if the signals underperform?