Met 42 criteria | Higher 21 57% | Lower 16 43% | Median +1.3% change | Average +0.9% change |
| Date Met | Stock | Sector | Close Then | Latest Close | Change |
|---|---|---|---|---|---|
| 22 Oct | Members | Energy | $122.32 | $122.32 22 Oct | — |
| 22 Oct | Members | Healthcare | $20.48 | $20.48 22 Oct | — |
| 21 Oct | Members | Technology | — | $25.32 22 Oct | — |
| 21 Oct | Members | Healthcare | — | $4.78 22 Oct | — |
| 21 Oct | Members | Energy | — | $108.42 22 Oct | — |
| 19 Oct | Members | Healthcare | $67.08 | $69.16 22 Oct | +3.1% |
| 16 Oct | Members | Healthcare | $152.64 | $157.36 22 Oct | +3.1% |
| 16 Oct | Members | Healthcare | $31.89 | $33.32 22 Oct | +4.5% |
| 15 Oct | Members | Industrials | $14.40 | $14.55 22 Oct | +1.0% |
| 15 Oct | Members | Basic Materials | $56.88 | $55.36 22 Oct | -2.7% |
| 15 Oct | Members | Real Estate | $19.36 | $20.67 22 Oct | +6.8% |
| 12 Oct | Members | Healthcare | $397.77 | $401.37 22 Oct | +0.9% |
| < -15% | 1 | |
| -15 to -10 | 3 | |
| -10 to -5 | 7 | |
| -5 to 0 | 5 | |
| 0 to 5 | 10 | |
| 5 to 10 | 4 | |
| 10 to 15 | 3 | |
| > 15% | 4 |
| After | Signals | Average Gain | Vs Same-Day Scanned | Vs S&P 500 | Vs Nasdaq-100 |
|---|---|---|---|---|---|
| 1 day | 37 | -0.3% | -0.1% t -0.3 | -0.4% t -0.7 | -0.3% t -0.6 |
| 3 days | 35 | +1.0% | +1.4% t 1.8 | +0.7% t 0.9 | +0.9% t 1.2 |
| 5 days | 34 | +0.2% | +1.2% t 1.3 | -0.3% t -0.3 | +0.4% t 0.4 |
| 10 days ● | 23 | +0.1% | +2.3% t 1.1 | -0.8% t -0.4 | +1.2% t 0.6 |
| 15 days | 16 | +0.4% | +3.7% t 1.4 | +0.1% t 0.0 | +3.6% t 1.5 |
- Record so far: 19 trading days. 23 signals have a full 10-day result. Next scheduled review: day 40.
- Over 10 trading days the stocks that met the criteria gained +0.1% on average, 2.3 points more than stocks scanned the same day that did not meet them (t = 1.1).
- There are too few matured signals to draw a conclusion. Any gap, good or bad, is still within what chance produces.
- Against the market over the same dates (average excess): S&P 500 -0.8% (t = -0.4); Nasdaq-100 +1.2% (t = 0.6).
What This Page Is
A running log of every stock that My Radar’s automated scans rated Criteria Meet, with what happened to its price afterwards, compared with the market and with stocks the scan did not select.
The aim is to find out, in the open, whether the scan adds information. It is a research record, not a list of recommendations.
How A Signal Is Counted
A stock counts once, on the first scan day it is rated Criteria Meet since the record began (28 Sep 2026).
The hypothetical entry price is that day’s official closing price. Results are shown for 1, 3, 5, 10 and 15 trading days later, and for the latest close.
What It Is Compared With
Same-day scanned stocks: stocks the scan looked at on the same day but did not rate Criteria Meet (up to 400 per day, picked by a fixed rule based on the ticker so nobody can choose favourable ones).
The S&P 500 (SPY) and the Nasdaq-100 (QQQ) over exactly the same dates.
Excess = the stock’s return minus the comparison’s return over the same dates. A positive number means the stock did better than the comparison.
How We Judge It
The verdict uses one holding period, 10 trading days, chosen in advance (23 Oct 2026) so that we cannot pick the best-looking period afterwards. The others are shown for context only.
t-statistic = average excess ÷ its standard error. It measures whether the gap is bigger than luck alone would usually produce. We call a result significant only when |t| is at least 2.5 (2.0 after 130 trading days of records). The bar is higher at first because we look at the results repeatedly.
Below that bar the honest reading is ‘cannot be told apart from chance’, not ‘no effect’.
How Much Data Is Needed
With the spread seen so far (about 8% between stocks), detecting a real average edge with about 80% confidence needs roughly: +3% edge, 60 signals (about 2 months); +2%, 110 signals (about 3 months); +1.5%, 240 signals (about 6 months); +1%, 530 signals (more than a year). Signals on the same day share the same market, so real needs are higher, about 1.5 times these.
Scheduled reviews: 40, 65 and 130 trading days of records.
What This Cannot Tell You
It covers one short period and one market condition. A small sample can look good or bad by luck.
Results are hypothetical: they assume entry at the close, ignore commissions, spreads and slippage, and no real money was invested.
Stocks that stop trading or drop out of the data leave no price and are missing from both sides of the comparison.
The scan rules can change. Every change is listed below, and results from different rule versions should not be mixed without saying so.
Even a significant result describes the past. It does not predict future results and is not a recommendation.
Data And Prices
Prices are official daily closing prices, adjusted for splits and dividends, from a market data provider. Each price shows the date of the close it comes from; a warning mark means no newer price could be retrieved.
The log is updated each trading day after the US close and is not real-time.