Where it matters most for practice, yes. The daily range, trend and range days, and the time the high and low are set are within 3 points or less of 4,916 real sessions. Where it falls short: busy minutes cluster more than in the real market, and there is no order book.
We checked by generating 2,000 sessions with the futures simulator for each instrument and measuring them the same way as the real ones. Here are the numbers, what the simulator doesn't reproduce and what it is (and isn't) good for.
Why a synthetic market and not real days
Practicing on historical days has two problems. First, if you replay sessions, you end up remembering how they ended. Second, exchange data is licensed and can't be freely redistributed in a free app. So the simulator builds every session second by second from aggregate statistics, without a single real price. The same seed gives the same session, which is why the daily challenge is the same for everyone who plays it.
What it is calibrated on
- Micro Nasdaq (MNQ): 1,813 real Micro Nasdaq sessions (June 2019 to October 2026).
- Micro Gold (MGC): 1,836 real Micro Gold sessions (September 2018 to October 2026).
- Micro Crude (MCL): 1,267 real Micro Crude sessions (July 2021 to August 2026).
Only each instrument's main New York session. For each one we keep statistics, not prices:
- how volatility changes minute by minute through the day;
- how the daily range is distributed;
- how many days are trend days and how many are range days;
- how often extreme minutes show up.
The test: 2,000 synthetic sessions against the real ones
The daily range is the high minus the low, as a % of the opening price. A trend day closes far from where it opened: the distance between open and close is more than 60% of the range. A range day stays below 25%.
Micro Nasdaq (MNQ)
| Simulator | Real | |
|---|---|---|
| Daily range, quiet day (10th percentile) | 0.62% | 0.60% |
| Daily range, typical day (median) | 1.14% | 1.14% |
| Daily range, busy day (90th percentile) | 2.12% | 2.23% |
| Trend days | 41% | 38% |
| Range days | 25% | 23% |
| Days with the high or low in the first hour | 75% | 78% |
| Tails of 1-minute moves (kurtosis; normal is 3) | 6.12 | 5.39 |
| Clustering of busy minutes (autocorrelation) | 0.16 | 0.11 |
Micro Gold (MGC)
| Simulator | Real | |
|---|---|---|
| Daily range, quiet day (10th percentile) | 0.41% | 0.42% |
| Daily range, typical day (median) | 0.76% | 0.76% |
| Daily range, busy day (90th percentile) | 1.41% | 1.50% |
| Trend days | 34% | 32% |
| Range days | 28% | 27% |
| Days with the high or low in the first hour | 66% | 69% |
| Tails of 1-minute moves (kurtosis; normal is 3) | 5.95 | 5.59 |
| Clustering of busy minutes (autocorrelation) | 0.15 | 0.10 |
Micro Crude (MCL)
| Simulator | Real | |
|---|---|---|
| Daily range, quiet day (10th percentile) | 1.26% | 1.29% |
| Daily range, typical day (median) | 2.21% | 2.22% |
| Daily range, busy day (90th percentile) | 3.96% | 4.11% |
| Trend days | 34% | 33% |
| Range days | 28% | 26% |
| Days with the high or low in the first hour | 64% | 67% |
| Tails of 1-minute moves (kurtosis; normal is 3) | 5.96 | 5.95 |
| Clustering of busy minutes (autocorrelation) | 0.15 | 0.08 |
The biggest gap is clustering: in the simulator a busy minute is followed by another busy minute a bit more often than in the real market (0.15 to 0.16 versus 0.08 to 0.11). In practice, runs of big candles last a little longer than in the real market. And very busy days (90th percentile) are somewhat calmer than the real ones.
What the simulator doesn't reproduce
- The order book. There is no queue: when price only touches your limit order, it fills with some probability per second, to mimic the fact that you don't always get filled in the real market. If price trades through it, it always fills.
- Variable slippage. It's fixed: 1 tick on market and stop orders, plus $0.60 commission per contract per side. During real news it can be much more.
- News. Modeled as an initial jump and volatility that calms down over a few minutes, not as a price gap with an empty book.
- Overnight. Only each instrument's main session: neither Asia nor Europe.
What it's good for and what it isn't
It's good for training what blows the most prop firm accounts: position size, always using a stop, stopping at your daily loss, not trading after it, and respecting each firm's trailing drawdown and consistency rule. The simulator logs everything in a journal and scores your discipline.
It's not good for validating a strategy. A synthetic market doesn't have the real market's inefficiencies; if your strategy wins here, that says nothing about whether it wins there. For that you need a backtest on real data with costs, and then a small-size live test (we explain it in 12 questions to ask before buying an evaluation).
Educational content based on our own experience. Not investment advice. Firm rules change: check them on each firm's website. This article contains no affiliate links.