Smart Golden Portfolio EA Review (2026): A $299 MT5 Gold EA With a Two-Month Live Record and Claims Still to Prove

40.3 / 100

weighted verdict score on eight published criteria

$299

MQL5 Market price on 29 Oct 2026, product 193479

1 Sep 2026

date the listing was published; the live signal was started the same day

The verdict at a glance: a credible seller, a sensible pitch and a record that has barely started

Most people searching for a Smart Golden Portfolio review want five answers: is it legit, does it work, how risky is it, what does it cost and what broker and settings does it need. Here are the short answers, with the detail below. It is legit in the sense that matters first: a real, purchasable MT5 product from a seller with a public history on the MQL5 Market and other live signals, not an anonymous file sold through a messaging app. Whether it works is not yet knowable from outside. The product has been on sale since 1 September 2026, and its only live record started that day. The risk is not documented in a form we could check, so we label it High until it is. The price on the MQL5 Market is $299. The broker named on the live signal is IC Markets; we could not confirm the recommended settings or minimum deposit, which on this seller's other gold EAs are shared after purchase.

That combination produces a low score with a specific meaning. The weight of our method sits on evidence and risk, which together carry 40 of the 100 points, and both need a record long enough to contain a bad stretch. Eight weeks of trading cannot contain one reliably. The scorecard and radar show where the points came from and where they are missing.

What Smart Golden Portfolio actually is

Smart Golden Portfolio is an Expert Advisor for MetaTrader 5, sold on the MQL5 Market as product 193479. The seller is Barbaros Bulent Kortarla, whose seller page lists it alongside four other products. The listing went live on 1 September 2026 and showed a price of $299 when we checked it on 29 October 2026. It trades XAUUSD, the symbol most brokers use for spot gold against the US dollar.

The opening of the listing sets out the pitch in three parts. First, the purpose: Smart Golden Portfolio is 'designed for traders who want a stable strategy that can remain part of a trading portfolio for the long term'. Second, the proof offered: 'You can follow the live IC Markets signal results from today here', with a link to MQL5 signal 2389307. Third, the origin: 'The core strategy was developed before 2025 as part of a much larger research process involving thousands of generated and tested strategies. From that pool, only a small number were selected for extended' testing; the text we read was cut off at that point, so we do not quote the rest of the sentence or the details of what the extended testing involved.

Two things follow from the pitch. The word 'portfolio' in the name is about how the developer expects the EA to be used: as one strategy among several in a trader's account, not as a single system that carries the whole account. And the origin story describes a strategy-generation workflow, in which many candidate systems are produced and tested automatically and a few survivors are kept. That workflow can produce robust strategies, and it can also produce strategies that fit the past by chance. We come back to that below because it is the central question about this product.

Who is behind it: the seller's track record on MQL5

When a product is new, the seller's history is the best context available. Barbaros Bulent Kortarla's seller page on 29 October 2026 listed five products. The oldest, Smart Funded HFT for MetaTrader 4, was published in January 2024 and is described as an EA for passing high-frequency-trading challenges at prop firms that allow it. Easy Funded MT5 followed in May 2024 with the same purpose. Smart Gold Hunter, the seller's main gold EA for live and classic prop accounts, was published in March 2026, Smart Gold Impulse in mid-2026, and Smart Golden Portfolio on 1 September 2026.

The averages above are our arithmetic from the vote counts and vote totals on the seller page, rounded to one decimal. Three points stand out. The seller has been on the Market for close to three years and has kept products on sale, which is more than many gold EA vendors manage. The seller publishes live MQL5 signals for the gold products; the seller page linked five signal pages when we checked. And the ratings are good but not uniform: the two most-reviewed gold products sit around 4.0 to 4.1, which is a realistic spread for EAs whose results vary by broker. The 5.0 on Smart Golden Portfolio comes from eight votes in the first eight weeks, when buyers have seen the least.

Our earlier verdict on Smart Gold Hunter is relevant here because it is the same seller's longest public gold record. That review found a real live signal with a positive profit factor, a moderate balance drawdown, and platform warnings that most of the growth came within a few days. The lesson carried over is not about Smart Golden Portfolio's results, which are separate, but about how to read this seller's signals: look at concentration of profit, look at monthly growth after the launch period, and check that the signal belongs to the product you are buying.

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 and runs the live signal. Publishing a live account from launch day is a good habit; it is also the sales page, and the developer chooses the broker, the settings and the risk on it.
• MQL5 and MetaQuotes host the Market, the ratings and the signals. Signal statistics on MQL5 are computed by the platform from the broker's trade history, which is why a signal with enough months behind it can earn our top evidence grade.
• Buyers who left ratings have had the product for at most a few weeks. A rating in the first two months mostly measures installation, communication and the first trades, not survival through a bad regime.
• ForexCopilotEA sells a listing under this name. We gain if you buy it, which is why this verdict opens with that fact and applies the same rules as every other.

Does Smart Golden Portfolio work? What the live record can and cannot show

The honest answer to 'does it work' is that the public record is too young to say. The listing links one live signal, at IC Markets, and introduces it with the words 'from today', which places its start at the launch on 1 September 2026. On the day we checked, that is about eight weeks of trading. We did not read the signal's statistics for this verdict, so the table below records what is known and what a buyer should look up, rather than numbers we did not verify.

Why does age matter so much? Because a short sample can look excellent by chance, especially on gold. Gold has spent long stretches trending strongly, and a strategy that buys strength or fades extremes can look very good for a month or two and then give much of it back in a different regime. For a system that trades a few times a week, eight weeks is perhaps a few dozen trades, and a few dozen trades cannot tell a real edge from luck. Our method's anchors expect a live record of at least several months with enough trades before evidence can score in the middle of the range, and a year or more across different conditions before it scores well.

When you open the signal, read it in this order. Start with the age and the number of trades. Then look at the maximum drawdown by balance and by equity: if equity drawdown is much larger than balance drawdown, the EA is holding losing positions open, which is a warning sign. Then the profit factor and the average win against the average loss, which tell you the payoff shape. Then the longest losing streak, which tells you what to size for. Finally, check whether MQL5 shows notices about growth concentrated in a few days or too-frequent trades; both appeared on this seller's Smart Gold Hunter signal and both change how the headline growth should be read. Our comparison of grid, martingale, scalper and trend EAs explains how each design shows up in these numbers.

The 'thousands of generated strategies' claim, explained

The most distinctive thing the listing says is that the core strategy came out of 'a much larger research process involving thousands of generated and tested strategies', from which only a small number were selected for extended testing. This is a common professional workflow. Software generates many rule combinations, tests each one on historical data, discards the weak ones and sends the survivors to further checks: testing on unseen data, on other brokers' data, with higher costs, and forward on demo or live accounts. Done carefully, it can find simple, robust rules a human would not think of.

Done carelessly, it does the opposite. If you test thousands of strategies on the same history, some will look excellent purely by chance, in the same way that if thousands of people toss a coin ten times, a few will throw ten heads. Selecting the best-looking candidates from a large pool is exactly how curve-fitting happens, and the larger the pool, the stronger the effect. The defences are well known: keep a long period of data the generator never saw, test the survivors on it once, require them to work across several brokers and cost assumptions, and then forward-test for months before selling. The listing's phrase 'only a small number were selected for extended' testing points in that direction. We could not see the details, so we cannot say how strict the selection was.

The claim that the strategy was 'developed before 2025' is useful if it is true and documented, because it means 2025 and 2026 act as out-of-sample data the strategy was not built on. Gold moved through very different conditions in that period, so a strategy that held up through it would be meaningful. The way to show that is a dated, unedited record from before 2025 to now. A backtest run today over 2025 and 2026 is not the same thing, because nothing stops a strategy from being adjusted after the fact. This is the first question we would put to the developer, and it is in the list below.

• What would raise our evidence score: a forward-test record that started before the product launched, ideally on a platform-verified signal or a Myfxbook account with verified track record and trading privileges.
• What would help on risk: a strategy tester report with real tick data and realistic costs across several years, plus the maximum drawdown and longest losing streak of the live signal after six months or more.
• What would not change the score: more five-star ratings, a longer description, or a higher headline growth figure on a short record.

How risky is Smart Golden Portfolio? What we know and what we do not

We label Smart Golden Portfolio High risk. This is a statement about uncertainty, not a finding that the EA is reckless. The part of the listing we read does not state the trade structure: whether every trade carries a stop loss, whether positions are ever added to a loser, how many trades can be open at once, or what risk per trade the default settings use. The seller's other gold EAs are explicit about these points. The Smart Gold Hunter listing, for example, lists 'No grid, No martingale, No recovery, No hedging, No dangerous averaging, Single entry / One Shot logic, Real Stop Loss and Take Profit'. That is a good sign about the seller's design habits. It is not evidence about this product, and we do not assume it carries over.

Gold itself is the second reason. XAUUSD routinely moves several hundred points in a session and can gap at the weekly open and around major US data releases. Any gold EA, however careful, is exposed to slippage on stops in fast markets, to spread widening at the daily rollover and to periods in which its pattern simply stops working. A portfolio-style EA is meant to be sized small inside a larger account, and that framing is the right one: it assumes losing periods will happen and are absorbed by the rest of the portfolio.

• Unknown drawdown. No verified maximum drawdown was available to us. Assume the first serious drawdown has not happened yet.
• Unknown streaks. Without a losing-streak figure, size as though ten or more losses in a row are possible, which is normal for many gold systems.
• Selection risk. A strategy chosen from thousands of candidates can underperform its history more than a strategy designed from a single idea. Expect live results to be worse than any backtest you are shown.
• Broker dependence. The live signal runs at IC Markets. Your spread, commission, slippage and server time may differ, and short-hold gold strategies are especially sensitive to that.
• Version risk. New products are updated often in their first months. A signal records one version; check which version you are buying.

Settings, broker and deposit: what we could and could not confirm

Searchers often want the best settings, the right timeframe and the minimum deposit. We could not confirm any of them for Smart Golden Portfolio from the material we read, and we will not guess. What we can say is where the information comes from on this seller's products. The Smart Gold Hunter listing asks buyers to 'send a short message via the comment section after your purchase to receive your guides, set files, and access to the live chat group', and the Smart Gold Impulse listing says set files of the running signal account and a live chat link are shared only with buyers. If Smart Golden Portfolio follows the same pattern, the settings that match the live signal are a post-purchase item. Ask before you pay whether that is the case.

On broker choice, the one firm fact is that the developer runs the live signal at IC Markets. That does not make IC Markets a requirement, but it is the only broker where the EA's live behaviour can be compared with yours. If you use a different broker, compare the gold spread, the commission per lot, the server time zone (which shifts daily candles and session filters) and the symbol name. The Smart Gold Hunter listing gives sensible general advice that applies to any gold EA from this seller: low spread, low slippage, fast execution, stable conditions and a VPS for live accounts.

• Platform: MetaTrader 5. The listing is in the MT5 section of the Market; we did not see an MT4 version.
• Symbol: XAUUSD or your broker's gold symbol. Confirm that the EA accepts suffixes such as XAUUSDm or GOLD before buying.
• Broker: the live signal is at IC Markets. Match its account type as closely as you can if you want comparable results.
• Deposit and lot size: not confirmed. Ask the developer what the live signal's starting balance and risk per trade are.
• Timeframe and set files: not confirmed. Ask whether the settings on the signal are the defaults or a post-purchase set file.
• VPS: strongly advisable for any live EA so that trades are not missed when your computer is off.

What buyers say so far

On 29 October 2026 the seller's product list showed eight rating votes for Smart Golden Portfolio with an average of 5.0. We did not read the individual review texts, so we do not quote any. A perfect average from eight votes is common for a product in its first weeks: early buyers tend to be followers of the seller's other products, the first trades are usually the ones people write about, and unhappy buyers usually take longer to post. The seller's more established gold EAs, Smart Gold Hunter with 79 votes and Smart Gold Impulse with 26, average about 4.0 and 4.1, which is a more realistic guide to where a gold EA's rating settles once results vary across brokers and months.

Read reviews for specifics, not stars: which broker the reviewer uses, how long they have run the EA, what risk setting, and whether they mention a drawdown. A review that says 'great support, good start' is about the developer, not the strategy. A review written after six months that reports a drawdown and still recommends the product is worth more than ten early five-star ratings.

Price: is $299 fair for Smart Golden Portfolio?

The MQL5 Market price on 29 October 2026 was $299. That sits between the seller's Smart Gold Impulse at $199 and Smart Gold Hunter at $349, and below many gold EAs that sell for $500 or more. In absolute terms it is a moderate price for a Market EA. In terms of what the buyer is paying for, it is a price for a story and a two-month record. That is why cost scores 5 rather than higher: the amount is reasonable, but the evidence it buys is thin.

We did not confirm whether a rental option or a free demo is offered for Smart Golden Portfolio. If a demo is available on the Market page, run it in the Strategy Tester over as many years as your data allows, with realistic spread and commission, before paying. A demo cannot prove a strategy works, but it can show you a strategy that does not, and it costs nothing. Add the running costs to the licence price: a VPS, and possibly a raw-spread account with commission. If you buy from a seller other than the MQL5 Market, including our own listing, the price, version, updates and support may differ; read the terms before you pay.

The score, criterion by criterion: 40.3 out of 100

The total is 40.3 out of 100. Evidence, at 3, is capped by our rule that a product without a grade A source cannot score above 4, and it sits below that cap because the only live record is about eight weeks old and was not read for scoring. Risk, at 4, reflects that no drawdown, streak or trade-structure figure could be checked; it is not lower because nothing we read points to a grid or martingale, and the seller's other gold EAs are built around single entries with stops. Profitability, at 3, scores the absence of a measurable edge, not a measured loss. The middle criteria, trust, fit, cost and support at 5, credit an established seller, a sensible portfolio framing and a moderate price.

Two things would move the score most. A live record of six to twelve months on the linked signal, with a drawdown and losing streak we can read, would lift evidence, risk and profitability together, because those three criteria carry 55 of the 100 points. And a documented forward test from before the 2026 launch, as the 'developed before 2025' claim implies, could lift evidence and sustainability sooner. Neither would guarantee a higher score: a longer record could also show a weak edge.

Who Smart Golden Portfolio fits, and who it does not

• It may fit an experienced trader who already runs several EAs, wants to add a small gold component, is willing to forward-test on demo or a small live account for several months, and will judge it on the signal's record rather than the description.
• It does not fit anyone looking for a single EA to carry their whole account, anyone who needs steady monthly results, or anyone who would raise the lot size after a good first month.
• It is risky for prop-firm challenges until the drawdown and losing-streak behaviour is known, because daily and maximum loss limits leave no room for discovering them during the evaluation.
• It is a poor choice for a first EA purchase. A new trader learns more, and risks less, from a product with a year of verified history.

How it compares with other gold EAs we have scored

Against the same seller's Smart Gold Hunter, Smart Golden Portfolio is cheaper and has a cleaner pitch, but a far shorter record; Smart Gold Hunter's signal had hundreds of trades when we scored it. Against Gold House, a breakout EA with two live signals of 27 and 37 weeks, it has much less evidence and a much lower price. Against grid-based gold EAs such as Wave Rider, it is not known to carry the same structural tail risk, but that is not yet shown either. Its 40.3 reflects the evidence gap, which time can close. For how grid, martingale, scalper and trend designs differ, see our strategy comparison and our gold EA category.

Questions to put to the developer before you pay

1. Is there a dated forward-test or live record of the core strategy from before 2025, and can it be verified on MQL5 or Myfxbook?
2. How many strategies were generated, what data was held out from the generator, and how were the survivors tested before launch?
3. Does every trade carry a stop loss, and does the EA ever use a grid, martingale, averaging or recovery logic?
4. What risk per trade and starting balance does signal 2389307 use, and are its settings the defaults?
5. What is the expected maximum drawdown and longest losing streak in your own testing?
6. What minimum deposit, timeframe and broker conditions do you recommend?
7. What do buyers receive after purchase: set files, a manual, a chat group, updates?