Usage Quotas (Bars, SR/Hour)
What it is
Each Strateda plan defines two primary usage quotas that govern how much data you can process and how frequently you can run computations. These quotas apply to all job types — single backtests, parameter optimizations, and walk-forward optimizations.
Quota limits by plan
Bars per simulation
The bar limit determines the maximum number of candles (bars) loaded per backtest or optimization run. More bars mean more historical data coverage.
| Plan | Bar Limit |
|---|---|
| Free | 1,000 |
| Plus | 4,500 |
| Pro | 15,000 |
| Premium | 37,500 |
The bar limit applies per individual simulation request. For optimization jobs with multiple parameter combinations, each combination uses one simulation request worth of bars.
Strategy runs per hour (SR)
The SR quota caps how many simulation requests you can execute per hour. This resets on a rolling hourly basis.
| Plan | Strategy Runs / Hour |
|---|---|
| Free | 4 |
| Plus | 20 |
| Pro | 45 |
| Premium | 90 |
If your quota is exhausted, queued jobs are automatically rescheduled to the next hour when quota resets. See Job System for details on how jobs queue and retry.
Live strategy limits (EA connections)
Each live strategy requires one MT5 Expert Advisor terminal connection.
| Plan | Live Strategies |
|---|---|
| Free | 0 |
| Plus | 0 |
| Pro | 2 |
| Premium | 4 |
How to interpret it
- Bar limits and interval choice interact. A 1-minute interval consumes bars far faster than a daily interval. On the Free plan (1,000 bars), M1 covers about 2.5 trading days; D1 covers about 4 years. See Candle Intervals for a full coverage table.
- Optimization is quota-efficient. Each parameter combination in an optimization job consumes significantly less quota than an individual backtest — historical data is cached and calculations are batched on the backend, making large grid searches quota-efficient. The job system handles this automatically — queued combinations execute as quota becomes available.
- Reloading results does not consume quota. Completed job results can be reloaded from the job table within the 7-day retention window without using any bars or strategy runs.
Getting the most from your bar quota
Three platform behaviours are worth understanding clearly — they directly affect how much research you can do within your plan's bar limit.
Trading hours restriction extends calendar coverage
When you configure trading hours in the strategy builder (for example, 08:00–20:00 instead of 24 hours), bars outside those hours are not loaded at all. This directly reduces bar consumption per backtest, extending how far back your calendar coverage reaches within the same bar budget.
For example, a forex strategy restricted to London and New York session hours (08:00–20:00 UTC) loads approximately half the bars of a 24-hour strategy over the same date range. On a Pro plan at M30, this can extend coverage from ~10 months to ~20 months within a single backtest request.
This makes trading hours restriction a practical research tool — not just a signal filter, but a way to access more calendar history within your quota. See Candle Intervals for coverage estimates by interval and plan.
WFO spans much more history than a single backtest
The bar limit applies per window in a Walk-Forward Optimization job — not to the total WFO date range. A WFO job with 7 rolling windows, each consuming 15,000 bars at M30, accesses 7 sequential periods of history in a single job submission. The total calendar span of the WFO can cover many years of broker data, even though each individual window respects the per-request bar limit.
This means WFO is not just a validation methodology — it is also the most efficient way to research strategy behaviour across long historical periods. A Premium user running WFO on M30 forex data with 7 windows can span 6+ years of history in a single job, while each window individually fits within the 37,500 bar limit. See Walk-Forward Optimization for full details.
Manual out-of-sample testing on all paid plans
Walk-Forward Optimization automates rolling IS/OOS validation and is available on Premium. But all paid plans can perform manual out-of-sample testing using the start and end date inputs in the strategy builder.
The workflow is straightforward: run your optimization or strategy development on one date range (in-sample), then test the resulting parameters on a separate, later date range (out-of-sample) without modifying them. This is genuine OOS validation — the out-of-sample period was not used during development, so results on that period reflect real generalisation, not overfitting.
For example:
- In-sample: January 2018 – December 2021 (optimize and refine parameters)
- Out-of-sample: January 2022 – present (validate with fixed parameters, no further adjustment)
This approach is available to Plus and Pro users who do not have access to automated WFO. It requires discipline — the out-of-sample period must be selected before development begins and never used for parameter tuning — but it provides a meaningful validation step at any plan tier.