Strategy slots, sports, and resource points

Three caps, one checkpoint

The activation fails. You still have a free slot, and the strategy is on a sport you already use. So what stopped you? Almost certainly the third cap, the one nobody sees coming.

Every plan carries three limits, all checked at the same moment: when you activate a strategy. Strategy slots cap how many saved, active strategies you can run at once. Sports cap how many distinct sports you author on (three football strategies count as one sport, not three). Resource points cap the combined computational weight of everything you have active. That last one behaves differently from the other two: it depends on what is inside your strategies, not just how many you have.

None of these caps touches your drafts. Sketch twenty ideas across five sports and you have spent nothing. The check happens when a draft becomes a saved, active strategy, the moment it starts consuming detection resources on live matchups.

What a resource point measures

Filters are not equally expensive to run. When a matchup goes live, the detection engine has to evaluate your conditions against it. Some conditions are almost free; others crunch real history.

Checking whether the moneyline favorite is at home (Favorite side), or whether the match falls on a Tuesday (Day of week), is a single lookup: 1 point each. Form filters have to walk back through a team's recent matches, and their cost roughly tracks how much history they aggregate. Result streak costs 2, Last N result rate costs 3, Goals average costs 4. At the top of the scale, the head-to-head filters (H2H result rate, H2H result streak) cost 5, because they do more than read a team's history: they reconstruct its record against this specific opponent, a different slice of data for every matchup.

Context filters are cheap, form filters are mid-priced, opponent-specific filters are premium. That is the whole pricing table in one line.

Your strategy's weight

A strategy's resource weight is the sum of the costs of every condition in its filter tree. Take a concrete tree: Favorite side (1) plus Day of week (1) plus Result streak (2) plus H2H result rate (5). That strategy weighs 9 points. Swap the head-to-head condition for a Last N result rate and it drops to 7. Small edits move the number, and the expensive filters dominate it.

The weight is computed when you save, then frozen. A saved strategy's configuration is locked anyway, so its weight never drifts: whatever you activated is what counts against your cap, until you archive it.

Notice what this implies for your portfolio. Your cap is spent by conditions, not by strategies. Five lean strategies can cost less than two head-to-head-heavy ones. If your ideas lean on opponent history, the resource-point ceiling will bind long before the slot count does.

The cheap tree is usually the better tree

Here is the part worth internalizing: resource points push you in the same direction good practice does. Every filter you add is one more parameter fitted to the past, and as Backtesting and overfitting explains, each added condition is another chance to fit noise instead of signal. A tree with four sharp conditions almost always generalizes better than one with ten, and it also costs fewer points. Treat the cap as a prompt, not a wall. If a strategy is bumping your ceiling, ask which conditions actually earn their keep in the backtest. Removing the ones that do not usually improves both numbers at once.

Freeing space without deleting anything

When you need room, archive. Archiving a saved strategy releases its slot and its resource points immediately, while keeping the strategy and its full record (bets, stats, history) intact. It stops firing picks and disappears from detection and listings, but nothing is deleted. Two details worth knowing. Only saved strategies can be archived: a draft occupies nothing, so there is nothing to free. And unarchiving re-runs the same three cap checks as activation, so you cannot bring a heavy strategy back if it no longer fits under your current plan.

Downgrades, drafts, and the actual numbers

If you downgrade your plan, nothing you already run is touched. Existing active strategies are grandfathered and keep firing; the lower caps only bite on your next activation. That gives you time to decide what to archive rather than forcing a fire sale.

Because drafts cost nothing, the sensible workflow follows naturally: iterate freely in draft, backtest, prune, and only activate the version that earns it. The caps constrain what you run, never what you explore.

The actual figures, how many slots, sports and points each plan includes, are not fixed constants of the platform. They live on the plans page, which is always the source of truth.

Where to go next

• Which plan fits you?: how slots, points and the other capabilities map to real usage profiles.
• Anatomy of a strategy: what actually lives inside the filter tree you are pricing.
• Backtesting and overfitting: why fewer, sharper conditions beat many weak ones.