Trixter LOF

Local Outlier Factor (LOF) Algorithm is a machine learning method for finding anomalies in data that cannot be detected using global thresholds. Unlike simple methods, LOF looks for local outliers by comparing the density of points in their immediate vicinity.

How it works

The main idea of LOF is simple: an anomaly is a point whose surrounding object density is significantly lower than the density around its neighbors.

1. Selecting the neighborhood. For each point, the algorithm finds  k  nearest neighbors.

2. Calculating local density. It computes how "tightly" the point is surrounded by its neighbors.

3. Comparison with neighbors. The algorithm compares the density around the point with the average density around its neighbors.

4. Final score. The final LOF value indicates how many times the density around the point is lower than around its neighbors.

   • LOF ≈ 1: the point's density is comparable to its surroundings (normal).

   • LOF >> 1: density is significantly lower (potential anomaly). The typical threshold for an anomaly is > 1.5.

What it is used for

LOF is widely applied in problems where anomalies can exist within individual groups of data, rather than only against the overall background.

• General anomaly detection: Identifying suspicious transactions, equipment failures, or abnormal user behavior.

• Industry and energy: Analysis of anomalous energy consumption patterns to prevent losses and transition to predictive equipment maintenance.

• Time series: Finding anomalous points or entire segments (subsequences) in time-series data (e.g., using an adapted version.

The Trixter LOF bot is built using this approach.

The Trixter LOF algorithm implements pattern search using the Local Outlier Factor (LOF for short). We assume that the gold market is sufficiently efficient and that most price fluctuations constitute market noise that cannot be predicted. However, we can identify fragments of the gold chart (so-called anomalies) in which it is most predictable. In this way, I have isolated 30% of chart sections that are well predictable.

Key principles

• Analysis of gold volatility across multiple timeframes and indicator periods.
• Grouping volatility to find effective patterns (from 100 to 3000).
• Statistical verification: high mathematical expectation of profit and statistical significance.
• Patterns that fail the verification are marked as noise — in such situations, trades are not opened.
• Mandatory 2-year forward test before being allowed on a live account.

Trading logic features

• Opening trades with limit orders at a selected distance from the price — minimal slippage.
• If the limit order does not trigger (price does not touch the level) and the signal disappears — the order is deleted.
• Open positions are accompanied by stop-loss and take-profit, and are closed at market price upon a reverse signal.
• The robot does not use martingale, arbitrage, or other high-risk strategies.
• Recommended timeframes: from M5 to H1. Optimal — H1 XAUUSD.
• Flexible adjustment of trading activity is provided (parameters below).

Main settings

• Allow BUY signals, Allow SELL signals — allow or prohibit buy/sell orders when confident in the market direction.
• Signals sensitivity, Filter sensitivity — threshold for signal sensitivity and noise filtering. Default value is 0.5. Can be increased to 0.7–0.9 for cleaner signals.
• Filter by trading hours — trading time restriction. By default — 24/7.

Money management settings

• Distance in points for limit order — distance from price to limit order. Default: 250. Lower → more trades, higher → more conservative.
• Max positions + orders number — maximum number of simultaneously open positions (without averaging). Default: 3.
• Additionally configurable: delay between openings, maximum spread control, stop-loss and take-profit levels.