Leverage decides how much of your money the broker locks up while a position is open. It does not decide how much you make or lose — the size of the position does that, and you choose the size.
Why it matters
Leverage is routinely described as the thing that makes trading risky, which puts the blame in the wrong place. It sets the margin requirement and nothing else; the risk on a trade comes from lot size and stop distance, both of which the EA controls independently.
✓It changes what you can open, not what you lose. A one-lot position loses the same amount per pip at 1:30 and at 1:500.
✓It sets the floor under the account. Higher leverage means less margin locked, which means more free margin standing between an open drawdown and a stop-out.
✓It is a broker and jurisdiction property, so the same EA meets different caps in different places, and a strategy that needs margin headroom may simply not fit.
✓It is the single most misread number in retail trading, and the misreading is expensive in both directions — oversizing because leverage is available, and undersizing an account until it stops out on noise.
How brokers define it
Leverage is the ratio of notional position size to required margin. At 1:100, one standard lot of a USD-quoted pair, notionally $100,000, requires $1,000 of margin.
It is granted per account and frequently stepped down by exposure: the first few lots at the headline ratio, larger positions at a lower one.
It may also be reduced per symbol. Metals, indices and exotic pairs typically carry lower leverage than major currency pairs on the same account.
Brokers commonly reduce leverage around weekends and major releases, which changes margin requirements on positions that are already open.
Regulated caps are set by the regulator, not the broker: ESMA and the FCA cap retail major-pair leverage at 1:30, while offshore entities of the same brands quote far higher.
What differs between brokers
The headline ratio, which ranges from 1:30 under EU and UK rules to four figures at offshore entities.
The exposure tiers at which the ratio steps down, and whether the tiers apply per symbol or per account.
Per-symbol reductions, which are where an EA trading metals or indices meets a much lower ratio than the marketing figure.
Whether leverage is reduced automatically ahead of weekends or high-impact news, and by how much.
Whether the account can request a different ratio, and whether doing so is possible while positions are open.
Impact on EA performance
Leverage sets the margin an EA must have free to open its next position. It does not appear anywhere in the profit and loss calculation.
Too little leverage produces order failures rather than losses: the EA computes a valid lot, the terminal refuses it for insufficient margin, and the strategy silently trades less than it should.
Strategies that hold several positions at once — grids, baskets, multi-leg portfolios — consume margin in proportion to the number of open legs, so they are the ones that meet the cap first.
The Strategy Tester runs at the leverage the trade server grants, which is not always the number typed into the dialog — and a backtest run at higher leverage can open positions that a 1:30 live account will refuse.
Higher leverage raises the stop-out cushion by lowering used margin, which is the one genuine safety argument for it — and it only holds if position size is left unchanged.
What to confirm before funding
The ratio that applies to the specific symbol the EA trades, not the headline figure on the home page.
The exposure tiers, and where the EA's largest expected open exposure falls within them.
Whether leverage is cut before weekends or news, and whether that can trigger a margin event on positions already open.
The stop-out level that goes with it, since leverage and stop-out together decide how much adverse movement the account survives.
That the tester leverage used to produce a published backtest matches the account the EA will run on.
Typical risks
Treating available leverage as a sizing instruction. The broker offering 1:2000 is not suggesting a position two thousand times the deposit.
Opening an account at the minimum ratio a regulator allows and then running an EA sized for a margin regime it does not have, producing constant refused orders.
Meeting a tiered reduction mid-drawdown, when margin requirements rise exactly as free margin is falling.
Reading a backtest produced at high tester leverage as achievable on a capped account, when some of its trades could never have been opened.
Example
One standard lot of EUR/USD, opened on a $5,000 account at two different ratios. The trade is identical; only the margin locked up differs.
Position
1.00 lot EUR/USD
Notional $100,000 at a rate near 1.00.
Margin at 1:30
$3,333
Leaves $1,667 of free margin on a $5,000 account.
Margin at 1:500
$200
Leaves $4,800 of free margin on the same account.
Loss on a 50-pip adverse move
$500 in both cases
Leverage does not appear in this number at all.
The higher ratio did not make this trade riskier. It made the account harder to stop out while the trade was open.
Set position size from risk per trade and stop distance, then choose leverage that leaves the margin requirement comfortably below the account balance. Never the other way round.
Range
What it means
Margin usage under 10% of equity
Ample cushion. Ordinary drawdown cannot approach a stop-out.
10–30%
Workable for multi-position strategies if the maximum expected exposure has been counted.
30–50%
A normal drawdown starts to interact with the margin level. Reduce size or raise the ratio.
Over 50%
The account is being managed by the stop-out level rather than by the strategy.
Size the position from risk and stop distance first. Leverage then only has to be high enough that the margin fits.
Count the maximum number of positions the EA can hold at once, not the typical number — margin is consumed by the worst case.
Check leverage on the traded symbol. A 1:1000 account can be 1:100 on gold and 1:20 on an index.
Match tester leverage to the live account before trusting a backtest's trade list.
Read leverage together with the stop-out level. Either number alone says nothing about how much adverse movement the account survives.
Broker pages in this catalogue record maximum leverage as a field, including the account types each figure applies to, so the headline number is not the only one visible.
Common mistakes
✕Believing high leverage increases risk per trade
A one-lot position loses ten dollars per pip regardless of the ratio. Raising leverage while keeping lot size fixed strictly reduces the chance of a stop-out, because less margin is locked. The danger is behavioural: high leverage makes an oversized position possible, and traders then open one.
✕Sizing from the deposit and the ratio
Multiplying the balance by the leverage produces the largest position the broker permits, which has nothing to do with the largest position the strategy should take. Risk per trade times account equity, divided by stop distance and pip value, produces the number that matters.
✕Reading only the headline ratio
Tiered reductions by exposure and lower ratios on non-FX symbols are standard, and both bite exactly when the account is largest or most concentrated. The applicable ratio is the one for the symbol at the size actually traded.
✕Backtesting at a leverage the account will not have
The tester refuses orders on insufficient margin just as a live account does. A report produced at 1:500 and deployed at 1:30 can contain trades the live account would have skipped, which changes the equity curve rather than merely scaling it.
In depth
The leverage we asked for, and the leverage the tester used
Across the 14 EAs published here
Figure
Backtests configured at 1:500 in the tester dialog
14
Backtests the server actually ran at 1:100
14
Starting deposit on every run
$10,000
Lot size on every leg of every record
0.10, fixed
Listings whose stated requirement reads “1:100 or higher”
14
Every run manifest behind this catalogue carries two fields, leverage_requested and
leverage_effective, and they exist because the two numbers refused to stay equal. We
set 1:500 in the tester dialog for all fourteen runs; the server granted 1:100 every
time, and every published record was produced at the granted ratio, not the typed one.
The gap matters more than it looks. Five times less leverage means five times the
margin locked per position, and for a strategy that sizes near its margin ceiling that
is the difference between a trade list that executes and one with holes in it. These
fourteen records absorbed the substitution because every leg is a fixed 0.10 lot
against a $10,000 deposit, which keeps margin usage in the lowest band of the table
above — a portfolio record like Cairn runs two dozen legs, and even its worst-case
stack stays clear of the ceiling. A grid sized to exploit 1:500 would not have been so
lucky, and nothing in the tester’s results tab would have said why trades went missing.
The practical reading: when a vendor quotes the leverage behind a backtest, the number
that shaped the trade list is the one the server granted, and almost nobody records it.
It is also why each listing here states its requirement as “1:100 or higher” — the
figure matches what the tests actually ran on, not the figure we asked for.
Frequently asked questions
Does higher leverage mean higher risk?
Not by itself. Leverage sets the margin requirement; lot size and stop distance set the loss. Holding position size constant, raising leverage frees margin and makes a stop-out less likely. It becomes risky only because it permits positions large enough to be dangerous, and the decision to open one is the trader's.
What leverage does an EA need?
Enough that the margin for its largest expected simultaneous exposure sits well below account equity. Single-position strategies on major pairs are comfortable at 1:30. Grids, baskets and multi-leg portfolios hold several positions at once and typically need a few hundred to one to avoid refused orders.
Why is my EA failing to open trades?
Insufficient free margin is the most common reason, and it looks like an EA that has stopped working rather than an error. Check the margin the position requires against free margin, the leverage that applies to that specific symbol, and whether an exposure tier has stepped the ratio down.
Is 1:30 too low to run an EA?
It is ample for one position at a time on a major pair with sensible sizing. It becomes limiting for strategies that hold many positions at once, where the same account balance supports far fewer open legs than the strategy expects.