Track record
Every call, verified 30 days later
When a bill, contract or filing gets a bullish or bearish call, the call is logged that day with the closing prices of SPY and the named tickers. Thirty days later it is checked against SPY. This page is that log, misses included. None of it is a backtest.
Read those four together. About half of the calls beat the index over 30 days — no better than a coin flip — and the median call is roughly flat against SPY. A small number of large moves carry much of the positive excess. That is the record; it is published so it can be checked, not so it can be marketed.
Sector ETF level
The funnel
| Calls logged | 1090 |
| Still inside the 30-day window | 236 |
| Insufficient data at logging time | 143 |
| Verification errors | 15 |
| Verified | 696 |
| Verified but excluded: zero-move SPY read (artifact) | 179 |
| Verified but no ticker prices recorded | 128 |
| Usable at ticker level | 389 |
| Usable at sector-ETF level | 315 |
By call direction
| Direction | Calls | Beat SPY | Mean excess (adj.) | Median excess (adj.) |
|---|---|---|---|---|
| bullish | 307 | 46.6% | +0.08% | -0.55% |
| bearish | 82 | 54.9% | -1.10% | +1.42% |
By sector
| Sector | Calls | Beat SPY | Mean excess (adj.) | Median excess (adj.) |
|---|---|---|---|---|
| Technology | 73 | 61.6% | +3.60% | +4.33% |
| Healthcare | 70 | 50.0% | +1.03% | 0.00% |
| Energy | 49 | 40.8% | +0.12% | -1.70% |
| Defense | 39 | 48.7% | -1.21% | -0.60% |
| Unknown | 24 | 37.5% | -2.44% | -5.79% |
| Finance | 24 | 41.7% | -4.23% | -0.69% |
| Infrastructure | 24 | 50.0% | -1.70% | -0.36% |
| Utilities | 17 | n < 20 | — | — |
| Agriculture | 15 | n < 20 | — | — |
| Transportation | 11 | n < 20 | — | — |
| Consumer | 11 | n < 20 | — | — |
| Manufacturing | 10 | n < 20 | — | — |
| Real Estate | 7 | n < 20 | — | — |
| Telecommunications | 7 | n < 20 | — | — |
| Materials | 4 | n < 20 | — | — |
| Space | 4 | n < 20 | — | — |
By month the call was made
| Month | Calls | Beat SPY | Mean excess (adj.) | Median excess (adj.) |
|---|---|---|---|---|
| 2025-12 | 2 | n < 20 | — | — |
| 2026-02 | 1 | n < 20 | — | — |
| 2026-03 | 2 | n < 20 | — | — |
| 2026-04 | 8 | n < 20 | — | — |
| 2026-05 | 92 | 53.3% | +0.61% | +0.87% |
| 2026-06 | 39 | 46.2% | -2.72% | -0.46% |
| 2026-07 | 141 | 53.2% | +1.25% | +1.61% |
| 2026-08 | 74 | 37.8% | -1.44% | -1.38% |
| 2026-09 | 30 | 33.3% | -3.33% | -4.78% |
Inclusion rules
- Only verified rows count. Pending, insufficient-data and error rows appear in the funnel and nowhere else.
- Ticker-level numbers require a recorded average ticker return and a recorded SPY return; rows without ticker prices are reported separately, not as wins or losses.
- A row whose recorded 30-day SPY return is exactly 0.00 is a verification artifact and is excluded from every rate.
- Beat is direction-aware: a bullish call beats when the tickers outperform SPY over 30 days, a bearish call beats when they underperform.
- Excess return is shown raw (ticker minus SPY) and direction-adjusted (sign flipped for bearish calls). The mean is reported next to the median because a few large moves carry it.
- Per-sector and per-month rates are printed only at n ≥ 20. Smaller groups show the count alone.
- Nothing here is a backtest. Every row was logged when the signal landed and verified 30 days later from prices recorded at both ends.
What this is not. It is not a portfolio return: calls are unweighted, overlapping, and not traded. It is not investment advice. Historical backtests on this site are kept separate from this page and are never mixed into these numbers. The same figures are available as JSON at /api/public/track-record, with the rules attached. Computed 10/17/2026, 9:14:33 PM ET.