Plan Limits
Your plan sets a few caps on backtesting. The two most important ones decide whether a scenario will run at all, so size your window against them before you launch — otherwise the run is rejected (and a long run can waste real time getting there).
Two backtest limits, and they are independent
There are two separate time limits. They are easy to confuse, so keep them distinct:
- Historical data (recency) — how far back a backtest can reach. This is the earliest start date your plan can use.
- Max backtest length — how long a single run can span (end date minus start date), regardless of where it starts.
These two limits are independent. On Backtester you can start a run up to 5 years ago (recency), but any single run can cover at most 1 year (max length). To test a 5-year history you'd run several backtests, each ≤ 1 year — not one 5-year run.
| Plan | Historical data (how far back) | Max backtest length (per run) |
|---|---|---|
| Sandbox | last 6 months | 180 days |
| Backtester | last 5 years | 1 year |
| Analyst | unlimited | 5 years |
The error you'll see if a run is too long
If a scenario's window is longer than your plan's max backtest length, the run is rejected with:
Time range exceeds plan limit (180 days). Upgrade for longer backtests.
The number reflects your plan (180 days on Sandbox). Shorten the window or upgrade — see Pricing for the upgrade path.
Other per-plan caps
| Limit | Sandbox | Backtester | Analyst |
|---|---|---|---|
| Backtests per month | 15 | 500 | 2,000 |
| Kept strategy versions | 5 | 50 | Unlimited |
| Saved configurations | 3 | 10 | Unlimited |
| Multi-instrument backtesting | — | — | Yes |
| Multi-scenario Tektii Boards | — | — | Yes |
"Kept strategy versions" counts versions you've pinned with Keep; creating versions is free and unlimited, and each version's image is stored free for 30 days from upload.
How many you can submit at once
Admission is gated by your remaining monthly quota, not a fixed batch size. A batch submit is bounded only by how many backtests you have left this month — there is no separate per-batch cap. The quota check is all-or-nothing: if a batch would take you past your remaining monthly quota, the whole batch is rejected before anything launches, so you never get a partial submit.
The rejection error tells you exactly how many runs remain this period, alongside your usage so far:
Batch needs 40 backtest run(s) but only 25 remain this period (475/500 used). Upgrade for more.
Single submits pass the same gate — with no quota left, a single run is rejected with Backtest limit reached (500/500). Upgrade for more. The quota counts against your current billing period and resets when the next period starts.
Shrink the batch to fit what remains (or wait for the reset, or upgrade), then retry.
Sizing a scenario up front
- Pick your instruments and confirm the data actually exists for the dates you want — see Available Instruments for each symbol's
dataStartDate→dataEndDaterange. - Check that your start date is within your plan's historical data reach and your window is within the max backtest length above.
- Author the scenario — see
tektii scenario runin the commands reference.