Brute Force
An exhaustive grid sweep. Guarantees the best within your ranges.
- Quality
- Highest
- Speed
- Very slow
- Budget
- O(N^k)
How it works
Brute Force does exactly what it says: it builds every possible combination of every enabled parameter's values and runs them all.
Because it leaves nothing untested, it is the only method that can guarantee it found the best settings inside the ranges you gave it. There is no luck and no sampling involved.
The catch is arithmetic. The number of cycles is the product of how many values each parameter can take, so it grows explosively as you add parameters. It is the right tool for a small, important search and the wrong tool for a big one.
Under the hood
Cycle count is the cartesian product of the per-parameter value counts: a numeric parameter contributes floor((max - min) / step) + 1 values, a boolean contributes two. Three parameters with ten values each is already a thousand cycles.
Because the total is fixed by the grid, Brute Force overrides your cycle-budget setting with the true grid size so the sweep is never cut off half way. The traversal order is deterministic, so the run is fully reproducible.
Use it as the definitive cross-check on a small space, or to validate that a smarter method actually found the global optimum.
- Cycles
- Product of value counts
- Budget
- Overridden
- uses the true grid size
- Order
- Deterministic
When to use it
Reach for it when
- Three or fewer parameters with small ranges
- When you need a definitive, provably-best answer
- To validate that a smarter method found the true optimum
Reach for something else when
- More than a handful of parameters (the grid explodes)
- Wide ranges or fine steps that blow up the cycle count
Auto-selected: Auto-selected when 3 or fewer parameters are enabled.
Want to see optimization in action? Browse our real strategy backtests vs buy and hold on NVDA, TSLA, and AAPL.
New to this? Read how to run your first optimization or learn how to choose the right optimization method for your strategy.
Related methods
-
Bisection (Smart Search)
Brackets the best region of each parameter and zooms in, deterministically and resumably.
Quality 4 of 5Speed 5 of 5O(N) How it works -
Recommended
TPE (Bayesian)
Builds a probabilistic model of what works and spends your cycles where the payoff is most likely.
Quality 5 of 5Speed 5 of 5O(N) How it works