Reading seasonal curves

Seasonality describes how a market has tended to move through the calendar across previous completed years. TDLab stores only non-reversible aggregates generated offline; the production module does not expose or distribute raw OHLC history.

Why the curve starts at 100

Starting each seasonal path at 100 removes the original price level. Daily historical percentage returns are compounded from the same base, making stronger and weaker portions of the calendar easier to compare. A curve at 106 does not predict a market price of 106.

Choose the history you want to inspect

• 5Y: the most recent completed five-year window.
• 10Y: a broader cycle with more observations.
• 20Y: the longest available structural view.

Average and median curves are shown separately. The average is more sensitive to exceptional years; the median represents the middle observation and is more resistant to extremes. Monthly summaries show compounded return and the percentage of positive years.

Seasonality at trade entry

For a trade in year Y, TDLab uses only years completed by Y−1. The classification looks at the historical five-session path around the entry date and requires three directional anchors. It never uses the future realized price after that trade.

Aligned, opposed and neutral trades remain split by symbol and Long or Short direction. Missing vintages or an incomplete historical sample are shown as Insufficient data.