Copy trading and an automated system get filed in the same mental folder, and they should not be. They share exactly one property: you are not the one deciding when to click. Underneath that, they are opposites. One outsources the decision to a person you cannot see. The other outsources it to a set of rules you can read, test and rerun.
Choosing between them is not really a choice about convenience. It is a choice about what you are willing to not know.
You link your account to somebody else's. When they buy, your account buys, scaled to your size. When they sell, your account sells. The platform handles the plumbing and takes a cut: a subscription, a share of profits, a spread markup, or some combination.
The pitch is intuitive and it is not stupid: somebody good at this does the work, you get the results, minus a fee. It is the instinct behind buying a fund. The difference is that a fund is a regulated structure with disclosure requirements, and a leaderboard on a copy trading platform is a leaderboard.
When you copy a person, you are buying their future decisions, and you cannot inspect what you are buying. When you run a rule-based system, you are buying a fixed set of rules, and you can test them against years of history before you pay. Both can lose money. Only one of them can be examined first.
None of these is an accusation. They are structural features of the arrangement, and they apply to honest traders as much as to dishonest ones.
You can audit a rule. You cannot audit a mood.
A rule-based system is a written procedure: these conditions, this entry, this stop, this target, this window. Because it is written down, three things become possible that are simply not possible with a person.
You can test it over history: load it into NinjaTrader 8's Strategy Analyzer, set a period, and get the same report we publish. You can see the bad parts, because the report prints the maximum drawdown and the losing months in the same font as the profits. And you can know it will not change, because software does what it was written to do until somebody rewrites it, and if we rewrite it you get a version number.
The main published the Algomyth engine report covers January 2022 to September 2026 on a mid-size account: $87,359 net with costs already deducted, 1,164 trades, a profit factor of 1.46 and a worst drawdown of $5,273. Those are hypothetical results from a simulation over historical data, with no real money at risk. Publishing the drawdown next to the profit lets you decide in advance whether you could sit through it. No leaderboard hands you that decision in advance.
It does not make the ownership problem go away. It relocates it.
The system will not decide whether to stay switched on. That is yours, and it matters precisely when it is hardest: four losing days in a row, the account down, and a switch sitting right there. A system built around a target three times its risk produces losing streaks as arithmetic, not malfunction. Turning it off in the valley is the most common way people lose money with a perfectly functional system, and no automation can take that decision away from you.
It also does not stop you tampering. Widening the target because the market looks lively, tightening the stop after a bad fill: every adjustment turns what you tested into something with no track record. Copy trading has a mirror version: unfollowing after a drawdown and refollowing after a run, which is buying high and selling low with extra steps.
| Copy trading | Rule-based automated system | |
|---|---|---|
| What you are buying | A person's future decisions | A fixed, written procedure |
| Can you inspect it beforehand? | No: only the outcomes they show you | Yes: the logic, the period, the drawdown |
| Can you audit the track record? | No | Yes, every signal is archived live |
| Can it change without warning? | Yes, at any moment, silently | Only if you change it |
| Emotional risk | Theirs and yours | Yours only |
| Ongoing effort | Choosing and re-choosing who to follow | Keeping a machine on and hands off |
| Typical cost structure | Subscription, profit share, or wider spreads | One-time or licensed software, plus commissions |
| Main failure mode | They change, or they were never good | You switch it off at the wrong time |
| Works on a funded futures account? | Often restricted or prohibited | Yes, this is the normal use |
One more difference worth naming: in US futures, taking compensation to trade for other people or advise them generally requires registration with the CFTC and the NFA as a commodity trading advisor. Selling software you run in your own account does not. That regulatory line is roughly the line between "someone is deciding for you" and "you are running a tool." That is worth knowing before you hand an account to a stranger with a good-looking chart.
Here is where the vocabulary trips people up. Inside the funded-account world, "copying" usually means something completely unrelated to following a stranger. It means trade replication across several accounts that all belong to you.
The setup: you run one instance of the system on one platform, and the entries and exits are mirrored into two, three or five accounts you hold yourself. NinjaTrader 8 supports this natively through account groups. There is no other trader involved. You are not inheriting anybody's judgment. You are running the same rules in parallel because a single account has a size ceiling.
Why anyone does it: funded account sizing is capped by the account. A 50k evaluation is where the published backtest sits, at one contract. A 100k allows up to two. A 250k allows up to five. Rather than concentrate everything into one large evaluation, many people run several smaller ones side by side, which also spreads the risk that any single account gets closed out on a drawdown rule.
Three constraints before you set it up. Firms cap how many accounts one person can hold. Several restrict or prohibit copying between different firms, and some require a declared master account. And replication multiplies costs as well as results: every account pays its own commissions and its own evaluation fee. Mechanics in copying trades across funded accounts, ground rules in how funded accounts work.
The full backtest report, the losing stretches, the maximum drawdown, and the live signal log that archives every new signal with its outcome in NinjaTrader 8.
It would be dishonest to pretend there is none. Copy trading makes sense if you want exposure to a discretionary style no rule set captures (a macro trader whose whole value is judgment about events) and you accept that you are betting on a person. It makes sense if running software is genuinely not something you will do and the alternative is not trading at all. And there are regulated structures: managed accounts run by registered advisors with real disclosure documents, which are a far more serious thing than a social trading leaderboard.
What does not make sense is picking a name off a leaderboard because the last three months looked good. That is not copy trading. That is momentum-chasing another person's luck.
Ask it out loud before you commit money to either: if this loses for four months, will I be able to tell whether it is broken?
With a rule-based system you can answer it. You compare the current stretch against the system's own history: the backtest tells you how deep its worst drawdown ran and how long its flat periods lasted, so you can separate "this is within normal" from "this has stopped behaving like itself." With a copied trader you have nothing to compare against except their equity curve, which is the same information as your loss.
That is the difference between a decision and a hope. The wider comparison of everything sold as automation is in best trading bots, and the checklist for evaluating any of it is in how to choose a trading bot.
Hypothetical performance. The Algomyth figures cited above come from a backtest over historical data, not from a live account. Simulated results are prepared with hindsight, carry no financial risk, and cannot fully reflect real execution, slippage or liquidity. Past performance, real or simulated, does not guarantee future results.
Copying trades in your own account is legal, but the person or platform on the other end may need to be registered. In US futures, taking compensation to trade or advise on other people's accounts generally requires registration with the CFTC and membership of the NFA as a commodity trading advisor. Before you link an account, check that whoever is generating the trades is operating within that framework, because an unregistered signal provider has no obligations to you at all.
Usually not in the way people expect. Most futures prop firms require that you are the one trading your account, and prohibit following an outside signal provider or sharing accounts. What they generally do allow, within limits, is replicating your own trades across your own accounts at the same firm. Read the specific rulebook, because getting this wrong is grounds for closing the account and voiding payouts.
No, and anyone claiming otherwise has not run one through a drawdown. It removes emotion from execution, which is where most damage happens: no hesitating on entries, no moving stops, no revenge trades. What it cannot remove is the decision to keep it running when the account is down, which is a genuinely hard thing to do and the main reason people fail with systems that work.
A signal service alerts you and you place the order yourself; copy trading places it automatically. The difference matters more than it sounds, because manual execution of somebody else's signals combines the worst of both worlds: you still cannot see their rules, and now your own hesitation degrades the result. If you are going to follow someone, following them automatically is at least internally consistent.
Technically yes, in separate accounts, but think about what you would learn. Two uncorrelated sources of profit and loss in one portfolio make it much harder to judge either, and the temptation becomes shifting money toward whichever is hot. If you try both, keep them fully separate, give each a fixed amount of capital and a fixed evaluation period, and do not rebalance on feel. Our view on sizing is in risk.
Everything we write comes back to one idea: verify, then decide. The live terminal exists for exactly that.
Unlock Algomyth Live