Pineify Strategy Optimizer: Plan a Parameter Grid
Pineify Strategy Optimizer automates changes to exposed strategy inputs in a TradingView browser tab, then collects the resulting Strategy Tester metrics. Its useful output is a comparison of specified parameter combinations on a specified dataset. It does not establish the best settings for future trading.
Count the search before running it
A grid tests the Cartesian product of the values selected for each parameter. For inclusive numeric ranges that land exactly on the end value, the count is (end - start) / step + 1.
| Illustrative parameter | Values | Count |
|---|---|---|
| EMA length | 10, 15, 20, 25, 30, 35, 40, 45, 50 | 9 |
| RSI threshold | 30, 40, 50, 60, 70 | 5 |
| Trailing-stop switch | Off, on | 2 |
The first two ranges produce 45 combinations, not 25. Adding the switch produces 90. Adding three chart intervals would produce 270. These counts describe workload, not an expected improvement in returns.
Select ranges for a reason. A fast average longer than the slow average may contradict the strategy's intended definition. If some combinations are invalid, change the grid or the strategy's validation rules rather than treating those rows as meaningful candidates.
Prepare one reproducible baseline
Load a strategy that produces trades in TradingView's Strategy Tester. Record its script version, symbol and exchange, chart type and interval, data period, capital, sizing, commission, slippage, and execution settings. Save one manual baseline result before starting automation.
The extension supports exposed numeric, checkbox, dropdown, and interval selections. It cannot infer the economic purpose of a field or tune a hard-coded constant that is absent from the strategy's inputs. Test a small selection first to check that the intended controls change and the results match a manual run.
The optimizer operates TradingView's own tester. Its fills and results retain that tester's assumptions and data limits. TradingView explains the broker emulator, order timing, costs, and testing limitations in its strategy documentation.
Run the grid and account for incomplete results
Install the extension through the link on the Strategy Optimizer page, sign in as required, and check current plan eligibility. Pineify's documented optimizer access is Advanced or higher; current checkout terms determine price and entitlement.
Keep the target strategy and chart context stable during the run. The extension updates inputs, waits for calculations, and collects results sequentially. A browser session or calculation can be interrupted. Review completion status and the number of collected rows before assuming the whole planned grid finished. There is no universal runtime: script complexity, data, page responsiveness, and retries all matter.
The extension performs browser automation, but this is not an offline workflow. TradingView remains an online service, and Pineify's extension has account requests and usage telemetry. Browser automation does not justify a promise that no data ever leaves the computer. Optional AI tools also have their own data flows.
Read the table as a set of trade-offs
The following figures are invented examples, not Pineify results or evidence of strategy performance:
| Candidate | Net profit | Maximum drawdown | Closed trades |
|---|---|---|---|
| A | $4,250 | 6.2% | 42 |
| B | $5,100 | 14.7% | 18 |
| C | $3,800 | 4.1% | 88 |
B has the highest sample profit but also the largest drawdown and fewest trades. C has more trades, which still does not prove independence or stability. No row can be selected responsibly from these three fields alone: the test period, exposure, costs, loss distribution, and rules matter.
Choose acceptance criteria before sorting the table. Review neighboring parameter values and whether gains concentrate in one short market episode. A broad region of similar results is useful evidence about sensitivity, not proof of an enduring edge.
Export a research record
CSV export includes the available result rows and parameter configuration. The file can contain a Parameters section before the Results table, so a spreadsheet or Python import may need to locate the Results header rather than treating the first line as the data header.
Keep the CSV together with the baseline settings, dates, and script version. Optimization summary rows are not automatically the underlying trade-by-trade history. Export the relevant TradingView report separately if detailed trade analysis is needed.
Keep selection separate from evaluation
Testing more combinations increases opportunities to select a historical accident. Grid search does not prevent overfitting. Reserve a later period before searching, freeze the chosen rules, and evaluate that period without retuning. If you revise the rules after seeing the result, that period has become part of development.
The optimizer does not automatically conduct this holdout process. A walk-forward study requires explicitly defined training windows, evaluation windows, and a consistent reselection rule. Paper execution can then expose timing and operational problems absent from historical reports.
How many tests are in an EMA 10 to 50 by 5 and RSI 30 to 70 by 10 grid?
There are nine EMA values and five RSI values, giving 45 combinations. Additional independent selections multiply that count.
Does testing every combination prevent overfitting?
No. It makes the search systematic but increases the number of candidates examined. Use separate evaluation data and preserve a record of all selection decisions.
Is this a separate backtesting engine?
No. The extension automates strategy inputs and collects results from TradingView's Strategy Tester.
Is the optimizer entirely offline?
No. Automation runs in the browser while TradingView and Pineify account services remain online. Do not infer a universal privacy guarantee from local browser execution.