Three steps, no hidden layer.
Every seasonal statistic on TimingAX follows the same pipeline. There's no proprietary "secret sauce" obscuring where a number comes from — it's straightforward, and it's described here in full.
(this month − last month) ÷ last month. Each return is tagged with its calendar month (Jan–Dec) and the year it occurred.That's it. There is no smoothing, no curve-fitting, and no selective date ranges chosen to make a pattern look stronger. The window is always "as many of the last 15 years as data exists for," applied identically to every asset.
What VERIFIED and
MODELLED mean.
Every analysis on TimingAX displays a small badge telling you exactly which kind of data you're looking at. We show this everywhere — the Analyzer, Backtest, Screener, Watchlist, and Correlations — because the difference matters.
The statistics were computed directly from real historical monthly closing prices for that specific asset, using the pipeline above. This is the default for major indices, large-cap stocks, ETFs, and Bitcoin/Ethereum.
Real price history wasn't available for this specific asset (typically smaller international tickers without reliable free data feeds), so the statistics come from a sector-level seasonal model instead. We show this badge rather than silently substituting — you always know which one you're looking at.
What's verified today.
Real-data coverage depends on what free public price feeds carry reliably. Here's the current state, honestly:
How each of the
9 tools uses data.
What this isn't.
Seasonality is a real, well-documented statistical phenomenon — but it is a tendency, not a law. Here's what we want every user to understand before relying on it:
TimingAX is one input among many. We built it to make a real, measurable effect visible — not to replace fundamental analysis, risk management, or your own judgement. Questions about the methodology? We read every message.