Understanding Model Tiers
How the Models Work
When you select a chart, the model reviews the recent price pattern and generates several possible price paths for the next 50 candles. These paths are based on what the model learned during training, namely how prices tended to move after similar patterns in the past. Each path represents one plausible outcome rather than a single fixed prediction.
Training Periods by Model
Each model is trained on a different length of market history:
- Light: trained on the last 1 year of market data
- Standard: trained on the last 3 years of market data
- Plus: trained on the last 7 years of market data
- Pro: trained on the full market history available since 2010, approximately 16 years
All four models cover the same 49 markets across currencies, cryptocurrencies, stocks, indices, and commodities, and all four produce the same type of output.
Choosing a Model
The most suitable model depends on current market conditions:
- When market behavior has changed recently, Light or Standard may be more suitable
- For general, everyday use, Standard or Plus provides a balanced view
- When the market is unusually volatile, Plus or Pro may be more suitable
- When you are unsure, comparing the results of two models can provide additional insight
If two models generate similar price paths, the outlook is more consistent. If their paths differ, the outlook is less certain.
Why Multiple Models Are Offered
No single model performs best in every market condition. Offering four training periods allows you to choose between a view focused on recent behavior and a view informed by long-term history, or to compare both. Pro is not a more advanced model than Light; it has simply been trained on a longer period of data.
Summary
Light, Standard, Plus, and Pro share the same design and differ only in their training period: 1 year, 3 years, 7 years, and the full history since 2010. Shorter training periods reflect recent market behavior, while longer periods cover a wider range of conditions. The generated price paths reflect how prices moved after similar patterns in the past and should be used as one input in your decisions, not as a guarantee of future results.