Input_Ticker_List_by_Theme.csv — Every ticker is assigned to one or more of a
structured set of themes and sub-themes, which the rotation model then weights by
sub-theme market capitalization.
| Metric | Count |
|---|---|
| Unique tickers | 1,394 |
| Themes | 34 |
| Sub-themes | 175 |
| Total assignments (rows) | 1,935 |
| — Primary assignments | 1,397 (one per ticker) |
| — Secondary assignments | 538 |
Each company has exactly one Primary sub-theme (its core business) and may carry one or more Secondary sub-themes where it has meaningful second-order exposure. Because a ticker can appear in several sub-themes, per-sub-theme ticker counts sum to more than the 1,397 unique names; a theme-level total counts each company once even when it holds multiple sub-themes within that theme.
The universe now spans the full large- and mid-cap core of the U.S. market:
GOOG, FOX, NWS), which are folded into their primary listings
(GOOGL, FOXA, NWSA).Membership reflects the S&P 500 and S&P MidCap 400 rosters as constituted through the 2026 index rebalances.
The CSV has five columns:
| Column | Description |
|---|---|
Ticker |
Exchange symbol |
Company Name |
Issuer name |
Theme |
One of 34 top-level themes |
Sub Theme |
One of 175 sub-themes nested under a theme |
Assignment |
Primary or Secondary |
SubThemes_by_Theme_Counts.pdf — the full theme → sub-theme tree with the
unique-ticker count for each sub-theme.Weekly_Theme_Updater_v01.py — Python Program that take the Input Ticker List
an assigns Market Cap and Price Change History to each line.Tickers_Themes_SubThemes.csv — The output of the Python Program that is
feed into Claude as a Context file.Claude_rotatation_instructions — The Claude instruction file to produce the
weekly report.I started with an AI scan of the FinViz Theme heatmap, but a lot of companies (e.g. Amazon) participate in a lot of different markets. So then I had AI narrow it to the two largest. I actually went back and forth between Claude and Grok on this to triangulate the results. Then went back and added in the S&P Large Caps and Small Caps that were missing. Decided to use a 60/40 split for market cap when there was both a primary and secondary sub theme. Then manually combined and sub-themes with less than 5 tickers. All in all a pretty arbitrary process. May not be perfect, but it is directional.