Micro lots and lot steps, how small sizes helped me test live with real fills
if you want to test a strategy with real execution but don't want real risk, the minimum lot size and the lot step on the account matter. I found out how much while setting up a test account.
Some accounts allow very small positions, like a hundredth of a standard lot, and the step between sizes is the same. Others have bigger minimums. If the minimum is large relative to the balance, you can't size trades properly with a small stop. You'd be forced to take more risk than planned, and the test is useless.
I use a small live account for new setups, with positions so small that a loss doesn't matter. What I'm looking at is the fills and the spreads, not the profit. After a few weeks of the small account I compare it to the demo results and see how close they are.
Check also the maximum number of open positions and the minimum distance for stops, which differ between brokers and instruments. The stop level can limit how tight your stop can be, which matters for scalping.
Micro lots and lot steps, how small sizes helped me test live with real fills
Re: Micro lots and lot steps, how small sizes helped me test live with real fills
This is exactly how I test new setups now. The point you made about the minimum lot is important. On a tiny account with a large minimum, a tight stop and a wide stop end up with the same size, so the risk changes from trade to trade and the test is meaningless. Checking the lot step and minimum before funding the account saves a lot of frustration. Some brokers also have different minimums for different symbols, especially metals and indices, so check each one.
It’s Fairman 
Re: Micro lots and lot steps, how small sizes helped me test live with real fills
I like that the focus is on fills and spreads, not profit. The money result on a tiny test account tells you nothing over a few weeks, but the execution data is real. I keep a simple log comparing the price I wanted with the price I got, and the spread at entry. After a month, I compare it with the demo. If the gap is big, the demo results are not something I can trust for that setup. It is slow and a bit boring, but it is the only honest way I know to see how a setup survives real execution.
It’s Fairman 
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
Re: Micro lots and lot steps, how small sizes helped me test live with real fills
The lot step matters even when the minimum is small, because rounding down eats into planned risk on tiny balances. On EURUSD a 0.01 lot is worth about 10 cents a pip, so a 500 dollar test account risking 1 percent has 5 dollars to spend per trade.
With 0.01 steps, a 10 pip stop takes 0.05 lot and the full 5 dollars. A 20 pip stop needs 0.025, which rounds down to 0.02, so that trade only risks 4 dollars or 0.8 percent.
The chart (illustration) compares that with a made up account whose minimum and step are both 0.10 lot. At 5 pips the two match at 1 percent, but at 25 pips the bigger minimum forces 25 dollars of risk, which is 5 percent of the balance.
That spread of risk sizes muddies the comparison with demo, since the demo run was presumably sized exactly. Setting the demo to the same lot step means both sides carry the same rounding.
Commission is the other thing tiny sizes distort. A few brokers set a floor on the commission for each ticket, and a 1 dollar minimum on a 0.01 lot trade costs the same as 10 pips, which would make any scalping test look hopeless for the wrong reason.
With 0.01 steps, a 10 pip stop takes 0.05 lot and the full 5 dollars. A 20 pip stop needs 0.025, which rounds down to 0.02, so that trade only risks 4 dollars or 0.8 percent.
The chart (illustration) compares that with a made up account whose minimum and step are both 0.10 lot. At 5 pips the two match at 1 percent, but at 25 pips the bigger minimum forces 25 dollars of risk, which is 5 percent of the balance.
That spread of risk sizes muddies the comparison with demo, since the demo run was presumably sized exactly. Setting the demo to the same lot step means both sides carry the same rounding.
Commission is the other thing tiny sizes distort. A few brokers set a floor on the commission for each ticket, and a 1 dollar minimum on a 0.01 lot trade costs the same as 10 pips, which would make any scalping test look hopeless for the wrong reason.
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
Re: Micro lots and lot steps, how small sizes helped me test live with real fills
Coming back to the fill comparison, the average slippage figure can hide most of the story on a small sample. Say 50 live fills come in, 44 at the requested price, 3 one pip worse and 3 a full 3 pips worse around a release.
That is 3 plus 9, or 12 pips over 50 fills, an average of 0.24 pip. The median is zero, and six fills carry the whole cost.
So it helps to log slippage per fill with the time and order type, not only a running average. The demo comparison then shows whether the gap comes from normal conditions or from a handful of news minutes.
Order type matters as much as timing. Limit orders at most brokers fill at the price or better, while stop orders and stop losses take whatever is there, so a test that exits mostly by stop will look worse live than one that exits by limit.
Demo stops usually fill exactly at the level, which flatters the losers more than the winners. Splitting the log into entry slippage and stop slippage shows which side of the trade the demo is too kind to.
That is 3 plus 9, or 12 pips over 50 fills, an average of 0.24 pip. The median is zero, and six fills carry the whole cost.
So it helps to log slippage per fill with the time and order type, not only a running average. The demo comparison then shows whether the gap comes from normal conditions or from a handful of news minutes.
Order type matters as much as timing. Limit orders at most brokers fill at the price or better, while stop orders and stop losses take whatever is there, so a test that exits mostly by stop will look worse live than one that exits by limit.
Demo stops usually fill exactly at the level, which flatters the losers more than the winners. Splitting the log into entry slippage and stop slippage shows which side of the trade the demo is too kind to.
-
FXS Prop Desk
- Posts: 1689
- Joined: Wed Oct 07, 2026 9:52 am
Re: Micro lots and lot steps, how small sizes helped me test live with real fills
The fills on small sizes are usually similar to those on larger sizes for retail accounts, so the results tell you a lot about costs.
One thing to watch on raw-spread accounts is a minimum commission per ticket. At 0.01 lots that floor can be several times the normal rate per lot, so a micro test can look worse than the same setup at full size. Converting each trade's commission into pips before comparing it with the demo keeps that from skewing the verdict.
One thing to watch on raw-spread accounts is a minimum commission per ticket. At 0.01 lots that floor can be several times the normal rate per lot, so a micro test can look worse than the same setup at full size. Converting each trade's commission into pips before comparing it with the demo keeps that from skewing the verdict.
FXS Official Team | The Idea Lab & Strategy
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
Re: Micro lots and lot steps, how small sizes helped me test live with real fills
Forward testing on a demo account is a valuable step because it exposes a system to data that could not have been fitted. It is not, however, identical to live trading, and scalpers in particular should understand the differences.
Demo servers frequently fill orders instantly at the quoted price, with little or no slippage and no rejections. Live accounts face requotes, partial fills, variable latency, and liquidity that thins when many orders hit at once. Spreads on demo feeds may also be cleaner than those seen during real volatile moments.
Psychology adds a second layer. Automated systems remove some emotional pressure, but a trader who must decide whether to switch off a system during a losing streak behaves differently when real money is at stake. Interventions of that kind can change results more than any parameter.
A reasonable approach is staged. Run the demo test long enough to collect a meaningful number of trades, then move to a very small live size to measure real fill quality, and compare the two logs trade by trade. The gap between demo and micro-live results is itself a useful estimate of execution cost, and scaling up should depend on that gap staying small.
Demo servers frequently fill orders instantly at the quoted price, with little or no slippage and no rejections. Live accounts face requotes, partial fills, variable latency, and liquidity that thins when many orders hit at once. Spreads on demo feeds may also be cleaner than those seen during real volatile moments.
Psychology adds a second layer. Automated systems remove some emotional pressure, but a trader who must decide whether to switch off a system during a losing streak behaves differently when real money is at stake. Interventions of that kind can change results more than any parameter.
A reasonable approach is staged. Run the demo test long enough to collect a meaningful number of trades, then move to a very small live size to measure real fill quality, and compare the two logs trade by trade. The gap between demo and micro-live results is itself a useful estimate of execution cost, and scaling up should depend on that gap staying small.