Amibroker vs Pineify Backtest Reports: Strategy Validation Compared
AmiBroker and Pineify Backtest Deep Report work at different stages of strategy evaluation. AmiBroker runs strategies and portfolio simulations in its own environment. Pineify analyzes an imported TradingView trade list. Both offer useful reporting, and AmiBroker has native Monte Carlo simulation. Neither a report nor a simulation proves that a strategy has a durable edge.
This comparison uses the vendors' documentation and Pineify's implemented report workflow. It does not report a controlled cross-platform performance test.
Reading AmiBroker's Backtest Report
AmiBroker's report reference describes separate all-trade, long and short statistics, return and drawdown measures, exposure, profit factor, payoff ratio and other metrics. Its built-in exposure measure is capital exposure relative to portfolio equity, averaged over bars; it is not simply the percentage of time any position was open.
| Measure | Question it helps answer | Comparison constraint |
|---|---|---|
| Net profit and annual return | What happened to capital in the tested period? | Same fees, sizing, cash flows and date range |
| Drawdown | How far did equity fall from a prior peak? | Closed-trade equity differs from intrabar or marked-to-market equity |
| Profit factor | How do total gains compare with total losses? | Undefined or unstable with few/no losses |
| Payoff ratio | How large is the average winner relative to the average loser? | Different from profit factor and incomplete without win frequency |
| Sharpe or Sortino | How do returns compare with a volatility measure? | Return interval, annualization, risk-free rate and downside definition |
| Ulcer measures | How persistent and deep were observed drawdowns? | Sampling frequency and equity construction |
Color-coded thresholds are heuristics. The AmiBroker manual itself describes its colors as arbitrary guidance. A Sharpe ratio above a selected cutoff or a green cell is not a pass/fail certificate for live trading.
What Pineify Adds to a TradingView Trade List
Pineify Backtest Deep Report parses a supported TradingView CSV and computes metrics and visualizations from its trade records. It is a report-analysis workflow, not a replacement for the original strategy engine or its market data.
A useful import procedure is:
- Export the relevant TradingView strategy trade list and retain the strategy settings and date range.
- Import the file and verify parsed trade count, direction, timestamps, currency and net P/L against the original.
- Set initial capital, risk-free assumptions and treatment of open trades deliberately.
- Inspect summary, distribution, rolling and trade-excursion views.
- Save an exported workbook with the original CSV and assumptions.
The current implementation includes a 20-trade rolling window and an Excel workbook with separate overview, trades, calendar returns, distribution, rolling and Monte Carlo sheets. These features help inspect the imported sample; they do not independently rerun the strategy or check whether its signals used future information.
Compare Methods, Not Checkmarks
| Capability | AmiBroker | Pineify Backtest Deep Report |
|---|---|---|
| Strategy execution model | Backtesting engine with AFL and portfolio features | Analyzes a supplied trade list |
| Monte Carlo | Native documented simulator | Built-in bootstrap analysis of imported trade P/L |
| Custom analysis | AFL/custom backtester workflows | Implemented report calculations and views |
| Charts | Includes documented Monte Carlo charts and report outputs | Browser equity, distribution, rolling and excursion views |
| Portfolio context | Can model simultaneous positions within the source test | Imported records do not reconstruct a complete portfolio execution engine |
| Data requirements | Strategy, compatible data and settings | Supported CSV plus report settings |
It is incorrect to describe AmiBroker as a text-only report with no native simulation or charts. Likewise, a longer checklist of named metrics does not establish that Pineify is more accurate. Use each product's documented definitions and the installed version when comparing less common statistics.
Monte Carlo: What the Simulations Mean
AmiBroker's Monte Carlo guide documents trade-list bootstrap sampling and a portfolio-equity-change mode. Its settings include position sizing and displayed equity curves. Overlapping positions require care because sequentially resampling trades can lose their original overlap.
Pineify's current implementation runs 1,000 draws of a trade sequence, sampling the imported net P/L values with replacement. An original trade can appear several times or not at all. This differs from merely shuffling the same trades once each. The chart retains 50 example paths; it does not plot every simulated outcome or all possible future paths.
The sampled amounts are added to initial capital. The routine does not rerun entries, prices, liquidity or dynamic position sizing under a new market history. It also does not preserve serial dependence between trades. A strategy with clustered losses, overlapping positions or changing size can therefore need a different stress model.
The displayed drawdown percentiles describe the model and input sample. They are not guaranteed future loss bounds. In the current routine, the percentage-drawdown value updates at the largest absolute drawdown event; that event need not be the largest percentage drawdown. Independently recompute percentage drawdowns before using those values for a capital decision.
The current “risk of ruin” output counts paths whose final equity is zero or below. It does not count every path that crosses zero and later recovers, and it does not simulate broker margin liquidation. Interpret it as an endpoint statistic, not a complete probability of bankruptcy. After changing input data, direction filters or capital, rerun the analysis and verify which inputs produced the displayed result.
Distribution, Rolling Metrics and MFE/MAE
A histogram can reveal asymmetry in the observed sample. A normal-curve overlay does not show that returns are normally distributed or that unseen tail losses are impossible. Pineify's historical VaR/CVaR calculations use imported per-trade percentage returns; do not equate them with a daily portfolio loss forecast.
A rolling window describes recent observations. Overlapping 20-trade windows are correlated, and a rising value does not establish a structural improvement. Confirm the return units and annualization before comparing a rolling ratio with a full-period ratio from another platform.
MFE and MAE describe excursions recorded during an existing trade. They cannot, by themselves, determine what would happen after an earlier stop, a later exit or a different entry. Testing a changed exit requires the underlying price path and a new simulation. The scatter plot is a way to formulate a hypothesis, not proof that a particular stop is optimal.
Privacy and Optional AI Analysis
The CSV parsing and numerical report calculations run in the browser in the reviewed implementation. If you explicitly invoke the AI-analysis workflow, it sends metrics and summarized monthly, rolling and Monte Carlo information to a server endpoint. That payload excludes the large raw trade and equity arrays, but it still contains information derived from the report.
Therefore “nothing ever leaves your device” and “completely safe” are inaccurate descriptions of the entire feature. Check the current privacy policy and decide whether the optional analysis suits the sensitivity of your data.
Choosing a Workflow
Choose based on where strategy logic and market data live. An AFL portfolio strategy belongs in an engine that can model its positions and capital constraints. A TradingView user with an existing trade export may find Pineify useful for additional inspection and a portable workbook. These workflows can complement one another, but metrics must be reconciled before comparing outcomes.
TradingView plan requirements for export, history and Deep Backtesting should be checked in the current plan table. More history does not automatically make a model more dependable if the strategy was repeatedly tuned on it.
Frequently Asked Questions
Does AmiBroker have native Monte Carlo simulation?
Yes. Its official guide documents the simulator, sampling methods, position-sizing settings, percentile tables and charts.
Does Pineify reproduce the full original backtest?
No. The report analyzes imported trades. It cannot reconstruct missing market data, unrecorded signals or a complete portfolio execution model from a trade list alone.
Is Pineify report processing entirely local?
Parsing and numerical analysis run in the browser. The optional AI-analysis action sends report metrics and summaries to a server, so the entire workflow is not exclusively local.
Does a 95th-percentile drawdown cap future losses?
No. It is a statistic of a model and sample, with implementation and sampling limitations. Future losses can exceed it.
Can SQN or Sharpe prove a strategy is good?
No single threshold establishes reliability. Sample size, dependence, selection bias, costs and the method used to calculate the metric all matter.
