Trend new_52w_high_low

52-Week New High / New Low

Detects when price makes a new 52-week (252-day) high or low. Standard institutional definition: today's high ≥ prior 252-day max (bullish), or today's low ≤ prior 252-day min (bearish). Used for market breadth and regime analysis (Hindenburg Omen, etc.).

Signal family

Trend — Signals that fire when price is continuing or reversing an established directional move. Momentum-following by nature.

Parameters

Name Description Default Range
period Lookback period (days) 252 126–504

Historical context

2,273,612 triggers on 21,080 tickers, 1989-02-08 → 2026-05-01. Universe: global — all covered exchanges (mcap ≥ $100,000,000, price ≥ $1). Long-only convention: BUY at open T+1, hold the horizon, compare to S&P 500 Equal Weight over the same window.

Methodology footnotes

Benchmarks shown in the detail tables: spxew (S&P 500 Equal Weight — primary, median-stock view, avoids the 2020+ megacap-concentration distortion), spx (S&P 500 cap-weighted, distorted post-2020), msci (MSCI World USD). Per-stock regime tags: trending = ADX(14) ≥ 25, high vol = 20d realized annualized vol ≥ 20%. 1d return = intraday T+1 open→close; 20d = open T+1 to close T+20.

At a glance — alpha vs S&P 500 Equal Weight, global universe

Holding-period sensitivity. Bullish columns: positive = signal worked (long the trigger beat the index). Bearish columns: negative = signal worked (the flagged stock underperformed).

Horizon Bullish α Bearish α
5-day +0.04% +0.09%
20-day +0.52% +0.39%
60-day +1.60% +0.79%
1-year +7.25% +3.06%
Random-date null check (20-day): Bullish: beats random (p=0.005).
Bearish: worse than random (p=1.000).

Where does NEW_52W_HIGH_LOW actually fire?

The bucket distribution often reveals what the signal really is, regardless of its textbook label. Heavy concentration in "non-trending + high vol" = it's mostly a chop-market event. Heavy in "trending + low vol" = it picks up the smooth grinds. Read the chart before the alpha numbers — context shapes everything that follows.

52-Week New High / New Low (new_52w_high_low) — trigger count distribution by per-stock regime quadrant (trending/non-trending × high/low realized volatility) for bullish and bearish triggers, global universe — bullish half — measured on new_52w_high
Bullish half — measured on new_52w_high
52-Week New High / New Low (new_52w_high_low) — trigger count distribution by per-stock regime quadrant (trending/non-trending × high/low realized volatility) for bullish and bearish triggers, global universe — bearish half — measured on new_52w_low
Bearish half — measured on new_52w_low

Does it work in every regime?

Trigger alpha split by the host stock's own regime on the trigger date — trending or ranging, high-vol or low-vol. The 20d alpha you'd actually capture if you took the trade. Bars matching your direction's "right" sign (positive for bullish, negative for bearish) = the signal worked in that regime; opposite sign = avoid it there. A signal with one strong-positive bar and three flat ones isn't a "20d alpha" signal — it's a "20d alpha when the stock is X" signal.

52-Week New High / New Low (new_52w_high_low) — mean 20-day alpha versus S&P 500 Equal Weight by per-stock regime quadrant, bullish and bearish triggers side by side — bullish half — measured on new_52w_high
Bullish half — measured on new_52w_high
52-Week New High / New Low (new_52w_high_low) — mean 20-day alpha versus S&P 500 Equal Weight by per-stock regime quadrant, bullish and bearish triggers side by side — bearish half — measured on new_52w_low
Bearish half — measured on new_52w_low
Trending + Low vol
Stock in a clean directional move with low realized volatility. Textbook "trend-following paradise" — smooth grind with little whipsaw risk.
Trending + High vol
Violent directional moves — parabolic rallies, crisis selloffs. Trend exists but the path is noisy. Signal timing may be imprecise.
Non-trending + Low vol
Quiet chop, summer doldrums, consolidations. No directional bias but also no big swings — small edges become reliable if they exist at all.
Non-trending + High vol
Choppy and violent — the classical "whipsaw zone" for momentum signals. Crossovers and breakouts fire repeatedly without follow-through.

Does it work in every era?

A multi-year average can hide major instability. The sample splits into three windows: 2015–2019 (pre-COVID), 2020–2022 (pandemic + 2022 bear), and 2023+ (post-ZIRP + AI megacap rally). All three matching your direction's "right" sign = the signal is durable. One era doing all the work = a regime-specific edge that may not repeat. The bigger the variance across eras, the smaller the position you should run. Long-history signal: requires 260 trading days of prior data per ticker. The earliest era may show fewer triggers as a result.

52-Week New High / New Low (new_52w_high_low) — 20-day alpha split by historical sub-period (2015-2019, 2020-2022, 2023+) to check consistency across market regimes — bullish half — measured on new_52w_high
Bullish half — measured on new_52w_high
52-Week New High / New Low (new_52w_high_low) — 20-day alpha split by historical sub-period (2015-2019, 2020-2022, 2023+) to check consistency across market regimes — bearish half — measured on new_52w_low
Bearish half — measured on new_52w_low

Longer-horizon views

This signal carries a long lookback window (260 trading days of prior history required per ticker), suggesting it's designed to catch moves that play out over months, not days. The charts below repeat the quadrant and sub-period analyses at the 60-day and 1-year (252-day) horizons so you can see how the signal's relationship with the benchmark evolves with holding period.

60-day alpha by stock regime

52-Week New High / New Low (new_52w_high_low) — mean 60-day alpha versus S&P 500 Equal Weight by per-stock regime quadrant — bullish half — measured on new_52w_high
Bullish half — measured on new_52w_high
52-Week New High / New Low (new_52w_high_low) — mean 60-day alpha versus S&P 500 Equal Weight by per-stock regime quadrant — bearish half — measured on new_52w_low
Bearish half — measured on new_52w_low

60-day alpha by era

52-Week New High / New Low (new_52w_high_low) — 60-day alpha split by historical sub-period — bullish half — measured on new_52w_high
Bullish half — measured on new_52w_high
52-Week New High / New Low (new_52w_high_low) — 60-day alpha split by historical sub-period — bearish half — measured on new_52w_low
Bearish half — measured on new_52w_low

1-year alpha by stock regime

52-Week New High / New Low (new_52w_high_low) — mean 1-year (252 trading day) alpha versus S&P 500 Equal Weight by per-stock regime quadrant — bullish half — measured on new_52w_high
Bullish half — measured on new_52w_high
52-Week New High / New Low (new_52w_high_low) — mean 1-year (252 trading day) alpha versus S&P 500 Equal Weight by per-stock regime quadrant — bearish half — measured on new_52w_low
Bearish half — measured on new_52w_low

1-year alpha by era

52-Week New High / New Low (new_52w_high_low) — 1-year alpha split by historical sub-period — bullish half — measured on new_52w_high
Bullish half — measured on new_52w_high
52-Week New High / New Low (new_52w_high_low) — 1-year alpha split by historical sub-period — bearish half — measured on new_52w_low
Bearish half — measured on new_52w_low

1-year observed lift vs random-date null — bullish side

Observed 1-year lift (vertical line) against the null distribution of random-date firing. If the line is deep inside the null cloud, the signal adds no information. If it sits in the right tail, the signal is doing real work in that direction; in the left tail it ran inverted — random dates served the bullish case better than its own triggers did.
52-Week New High / New Low (new_52w_high_low) — bullish 1-year observed lift versus the random-date permutation null distribution

1-year observed lift vs random-date null — bearish side

Observed 1-year lift (vertical line) against the null distribution of random-date firing. If the line is deep inside the null cloud, the signal adds no information. If it sits in the left tail, the signal is doing real work in that direction; in the right tail it ran inverted — random dates served the bearish case better than its own triggers did.
52-Week New High / New Low (new_52w_high_low) — bearish 1-year observed lift versus the random-date permutation null distribution

↑ Bullish triggers

Bench Metric 1d 5d 20d 60d 252d
spxew Stock % -0.03% +0.20% +1.06% +3.56% +16.03%
Bench % +0.03% +0.16% +0.62% +1.96% +8.68%
Alpha % -0.06% +0.04% +0.52% +1.60% +7.25%
Median alpha -0.09% -0.23% -0.59% -1.15% -3.89%
Hit rate (α>0) 47.2% 47.3% 46.8% 46.6% 44.7%
p (naive) <0.001 <0.001 <0.001 <0.001 <0.001
p (HAC) <0.001 <0.001 <0.001 <0.001 <0.001
N 1,290,043 1,242,906 1,237,259 1,192,246 1,019,669
spx Stock % -0.03% +0.20% +1.06% +3.56% +16.03%
Bench % +0.01% +0.17% +0.85% +2.81% +12.27%
Alpha % -0.04% +0.03% +0.27% +0.74% +3.54%
Median alpha -0.08% -0.28% -0.92% -2.14% -7.73%
Hit rate (α>0) 47.4% 46.7% 45.1% 43.8% 40.0%
p (naive) <0.001 <0.001 <0.001 <0.001 <0.001
p (HAC) <0.001 0.0013 <0.001 <0.001 <0.001
N 1,301,253 1,258,669 1,248,137 1,210,590 1,032,379
msci Stock % -0.03% +0.20% +1.06% +3.56% +16.03%
Bench % +0.03% +0.16% +0.69% +2.29% +9.35%
Alpha % -0.07% +0.04% +0.44% +1.27% +6.42%
Median alpha -0.11% -0.27% -0.73% -1.57% -4.73%
Hit rate (α>0) 46.7% 46.8% 46.0% 45.3% 43.8%
p (naive) <0.001 <0.001 <0.001 <0.001 <0.001
p (HAC) <0.001 <0.001 <0.001 <0.001 <0.001
N 1,292,012 1,249,741 1,244,155 1,195,538 1,026,576
Observed 20-day lift (vertical line) against the null distribution of random-date firing. If the line is deep inside the null cloud, the signal adds no information. If it sits in the right tail, the signal is doing real work in that direction; in the left tail it ran inverted — random dates served the bullish case better than its own triggers did.
52-Week New High / New Low (new_52w_high_low) — bullish 20-day observed lift versus random-date permutation null (200 iterations)
Permutation null detail — all horizons × each benchmark
200-iteration null: for each ticker, sample N random dates from its history (matching observed trigger count) and compute the same alpha. Both observed and null are baseline-centered per ticker (each ticker's own baseline alpha is subtracted), which removes the universe-selection lift that all surviving names share, so the comparison is about the trigger's timing. It does not put the null at zero: that baseline is a median, while a random date's expectation is the ticker's mean, so the null settles at the gap between the two — the right-skew of the ticker's own alpha distribution, positive for equities. Read observed minus null, not the absolute position of either column — which is also why these figures do not match the α columns in the tables above (those are raw, uncentered means). pperm = one-sided fraction of null iters with mean in the "signal was right" tail (right for bullish, left for bearish), so it runs from a floor of 0.005 (trigger dates beat every random draw in the claimed direction) to a 1.000 ceiling (every random draw served the claimed direction better — a reliably inverted signal, not an absent one). The floor is 1/(iterations + 1) and moves with the iteration count; the ceiling does not. Either end says the result was reliable, not that it was large: the gap between the line and the cloud can be a fraction of a percent and still reach an endpoint, and the transaction-cost floor in the caveats below would swallow a gap that small. Read the size off the α and hit-rate columns, the reliability off pperm.
Horizon Bench Observed lift Null mean Null 95% CI pperm
1d spxew +0.06% +0.06% [+0.06%, +0.07%] 1.000
1d spx +0.06% +0.07% [+0.07%, +0.08%] 1.000
1d msci +0.07% +0.08% [+0.07%, +0.08%] 1.000
5d spxew +0.32% +0.30% [+0.29%, +0.31%] 0.005
5d spx +0.35% +0.32% [+0.31%, +0.33%] 0.005
5d msci +0.35% +0.33% [+0.32%, +0.34%] 0.005
20d spxew +1.12% +1.01% [+0.98%, +1.03%] 0.005
20d spx +1.12% +1.05% [+1.03%, +1.07%] 0.005
20d msci +1.16% +1.06% [+1.04%, +1.08%] 0.005
60d spxew +2.11% +2.23% [+2.19%, +2.27%] 1.000
60d spx +2.02% +2.32% [+2.29%, +2.36%] 1.000
60d msci +2.09% +2.32% [+2.28%, +2.35%] 1.000
252d spxew +1.41% +3.43% [+3.31%, +3.53%] 1.000
252d spx +1.88% +3.91% [+3.79%, +4.01%] 1.000
252d msci +2.28% +3.79% [+3.68%, +3.90%] 1.000

Example triggers on US large-caps (2023+, mcap ≥ $30B)

Six recent bullish NEW_52W_HIGH_LOW triggers on US mega-caps. Top three: the signal's best outcomes. Bottom three: the worst. Extreme outliers (|α| > 25%) excluded. The three best and three worst are still tail outcomes by construction — read them as the range, not the typical result.

Strongest outcomes (what NEW_52W_HIGH_LOW looks like when it works)
52-Week New High / New Low (new_52w_high_low) — strongest bullish trigger #1 on a US mega-cap, showing the stock's OHLC candles, moving averages, signal-specific overlays, and the 20-day forward window 52-Week New High / New Low (new_52w_high_low) — strongest bullish trigger #2 on a US mega-cap, showing the stock's OHLC candles, moving averages, signal-specific overlays, and the 20-day forward window 52-Week New High / New Low (new_52w_high_low) — strongest bullish trigger #3 on a US mega-cap, showing the stock's OHLC candles, moving averages, signal-specific overlays, and the 20-day forward window
Weakest outcomes (what NEW_52W_HIGH_LOW looks like when it fails)
52-Week New High / New Low (new_52w_high_low) — weakest bullish trigger #1 on a US mega-cap, showing the 20-day forward window where the signal failed 52-Week New High / New Low (new_52w_high_low) — weakest bullish trigger #2 on a US mega-cap, showing the 20-day forward window where the signal failed 52-Week New High / New Low (new_52w_high_low) — weakest bullish trigger #3 on a US mega-cap, showing the 20-day forward window where the signal failed
Stock-regime quadrants (2×2 per-stock, 20d alpha detail table)
Each quadrant groups triggers by the stock's own ADX(14) and RV(20) at the trigger date — the textbook conditioning variable (not market-level). Stock %, bench %, alpha %, and HAC p-value shown for each benchmark.
Quadrant N Stock % (spxew) Bench % (spxew) Alpha % (spxew) p (HAC) Stock % (spx) Bench % (spx) Alpha % (spx) p (HAC) Stock % (msci) Bench % (msci) Alpha % (msci) p (HAC)
Trending + Low vol Clean directional grind, low whipsaw 236,307 +0.67% +0.29% +0.45% <0.001 +0.67% +0.64% +0.09% 0.0249 +0.67% +0.46% +0.28% <0.001
Trending + High vol Crisis selloff or parabolic rally 727,791 +1.32% +0.67% +0.73% <0.001 +1.32% +0.89% +0.49% <0.001 +1.32% +0.74% +0.64% <0.001
Non-trending + Low vol Quiet chop, summer doldrums 111,905 +0.54% +0.43% +0.18% <0.001 +0.54% +0.70% -0.13% 0.0006 +0.54% +0.52% +0.07% 0.0608
Non-trending + High vol Classical "whipsaw zone" for momentum 238,443 +1.17% +0.82% +0.40% <0.001 +1.17% +0.95% +0.26% <0.001 +1.17% +0.80% +0.40% <0.001
Sub-period breakdown table (20d alpha)
Historical clustering check. If alpha concentrates in one era, the signal's robustness is questionable.
Period N Alpha % (spxew) p (HAC) Alpha % (spx) p (HAC) Alpha % (msci) p (HAC)
2015-2019 2015-01-01 → 2020-01-01 335,648 -0.09% 0.0226 -0.40% <0.001 -0.19% <0.001
2020-2022 2020-01-01 → 2023-01-01 404,735 +0.36% <0.001 +0.49% <0.001 +0.70% <0.001
2023-2026 2023-01-01 → 2099-01-01 601,192 +1.00% <0.001 +0.53% <0.001 +0.63% <0.001

↓ Bearish triggers negative alpha = signal was right (stock underperformed market)

Bench Metric 1d 5d 20d 60d 252d
spxew Stock % -0.10% +0.11% +1.87% +4.50% +15.77%
Bench % -0.02% -0.04% +1.37% +3.80% +12.87%
Alpha % -0.09% +0.09% +0.39% +0.79% +3.06%
Median alpha -0.10% -0.20% -0.62% -1.62% -7.24%
Hit rate (α>0) 47.8% 48.2% 47.2% 45.9% 41.3%
p (naive) <0.001 <0.001 <0.001 <0.001 <0.001
p (HAC) <0.001 <0.001 <0.001 <0.001 <0.001
N 897,334 865,852 855,360 840,537 803,794
spx Stock % -0.10% +0.11% +1.87% +4.50% +15.77%
Bench % -0.02% +0.08% +1.69% +4.10% +16.49%
Alpha % -0.09% +0.01% +0.18% +0.51% -0.46%
Median alpha -0.08% -0.26% -0.85% -1.96% -11.15%
Hit rate (α>0) 48.1% 47.6% 46.2% 45.2% 37.7%
p (naive) <0.001 0.4604 <0.001 <0.001 <0.001
p (HAC) <0.001 0.6043 <0.001 <0.001 0.1446
N 905,478 875,963 870,891 854,579 816,508
msci Stock % -0.10% +0.11% +1.87% +4.50% +15.77%
Bench % -0.00% +0.07% +1.50% +3.79% +14.31%
Alpha % -0.09% +0.01% +0.32% +0.89% +1.39%
Median alpha -0.10% -0.26% -0.70% -1.58% -9.02%
Hit rate (α>0) 47.7% 47.5% 46.8% 46.0% 39.6%
p (naive) <0.001 0.2456 <0.001 <0.001 <0.001
p (HAC) <0.001 0.4148 <0.001 <0.001 <0.001
N 901,092 870,367 862,786 848,505 810,057
Observed 20-day lift (vertical line) against the null distribution of random-date firing. If the line is deep inside the null cloud, the signal adds no information. If it sits in the left tail, the signal is doing real work in that direction; in the right tail it ran inverted — random dates served the bearish case better than its own triggers did.
52-Week New High / New Low (new_52w_high_low) — bearish 20-day observed lift versus random-date permutation null (200 iterations)
Permutation null detail — all horizons × each benchmark
200-iteration null: for each ticker, sample N random dates from its history (matching observed trigger count) and compute the same alpha. Both observed and null are baseline-centered per ticker (each ticker's own baseline alpha is subtracted), which removes the universe-selection lift that all surviving names share, so the comparison is about the trigger's timing. It does not put the null at zero: that baseline is a median, while a random date's expectation is the ticker's mean, so the null settles at the gap between the two — the right-skew of the ticker's own alpha distribution, positive for equities. Read observed minus null, not the absolute position of either column — which is also why these figures do not match the α columns in the tables above (those are raw, uncentered means). pperm = one-sided fraction of null iters with mean in the "signal was right" tail (right for bullish, left for bearish), so it runs from a floor of 0.005 (trigger dates beat every random draw in the claimed direction) to a 1.000 ceiling (every random draw served the claimed direction better — a reliably inverted signal, not an absent one). The floor is 1/(iterations + 1) and moves with the iteration count; the ceiling does not. Either end says the result was reliable, not that it was large: the gap between the line and the cloud can be a fraction of a percent and still reach an endpoint, and the transaction-cost floor in the caveats below would swallow a gap that small. Read the size off the α and hit-rate columns, the reliability off pperm.
Horizon Bench Observed lift Null mean Null 95% CI pperm
1d spxew +0.06% +0.08% [+0.08%, +0.09%] 0.005
1d spx +0.04% +0.09% [+0.08%, +0.10%] 0.005
1d msci +0.08% +0.09% [+0.09%, +0.10%] 0.005
5d spxew +0.58% +0.38% [+0.36%, +0.39%] 1.000
5d spx +0.53% +0.39% [+0.38%, +0.41%] 1.000
5d msci +0.53% +0.40% [+0.38%, +0.41%] 1.000
20d spxew +1.85% +1.22% [+1.19%, +1.25%] 1.000
20d spx +1.87% +1.25% [+1.22%, +1.28%] 1.000
20d msci +1.88% +1.26% [+1.23%, +1.29%] 1.000
60d spxew +4.03% +2.65% [+2.60%, +2.69%] 1.000
60d spx +4.47% +2.72% [+2.67%, +2.76%] 1.000
60d msci +4.41% +2.73% [+2.69%, +2.78%] 1.000
252d spxew +9.62% +5.65% [+5.56%, +5.73%] 1.000
252d spx +9.98% +5.99% [+5.90%, +6.07%] 1.000
252d msci +9.61% +5.92% [+5.84%, +6.01%] 1.000

Example triggers on US large-caps (2023+, mcap ≥ $30B)

Six recent bearish NEW_52W_HIGH_LOW triggers on US mega-caps. Top three: the signal's best outcomes. Bottom three: the worst. Extreme outliers (|α| > 25%) excluded. The three best and three worst are still tail outcomes by construction — read them as the range, not the typical result.

Strongest outcomes (what NEW_52W_HIGH_LOW looks like when it works)
52-Week New High / New Low (new_52w_high_low) — strongest bearish trigger #1 on a US mega-cap, showing the stock's OHLC candles, moving averages, signal-specific overlays, and the 20-day forward window 52-Week New High / New Low (new_52w_high_low) — strongest bearish trigger #2 on a US mega-cap, showing the stock's OHLC candles, moving averages, signal-specific overlays, and the 20-day forward window 52-Week New High / New Low (new_52w_high_low) — strongest bearish trigger #3 on a US mega-cap, showing the stock's OHLC candles, moving averages, signal-specific overlays, and the 20-day forward window
Weakest outcomes (what NEW_52W_HIGH_LOW looks like when it fails)
52-Week New High / New Low (new_52w_high_low) — weakest bearish trigger #1 on a US mega-cap, showing the 20-day forward window where the signal failed 52-Week New High / New Low (new_52w_high_low) — weakest bearish trigger #2 on a US mega-cap, showing the 20-day forward window where the signal failed 52-Week New High / New Low (new_52w_high_low) — weakest bearish trigger #3 on a US mega-cap, showing the 20-day forward window where the signal failed
Stock-regime quadrants (2×2 per-stock, 20d alpha detail table)
Each quadrant groups triggers by the stock's own ADX(14) and RV(20) at the trigger date — the textbook conditioning variable (not market-level). Stock %, bench %, alpha %, and HAC p-value shown for each benchmark.
Quadrant N Stock % (spxew) Bench % (spxew) Alpha % (spxew) p (HAC) Stock % (spx) Bench % (spx) Alpha % (spx) p (HAC) Stock % (msci) Bench % (msci) Alpha % (msci) p (HAC)
Trending + Low vol Clean directional grind, low whipsaw 124,333 -0.48% +0.91% -1.35% <0.001 -0.48% +1.16% -1.58% <0.001 -0.48% +1.03% -1.45% <0.001
Trending + High vol Crisis selloff or parabolic rally 498,175 +3.14% +1.75% +1.27% <0.001 +3.14% +2.07% +1.04% <0.001 +3.14% +1.87% +1.21% <0.001
Non-trending + Low vol Quiet chop, summer doldrums 56,541 -0.90% +0.81% -1.67% <0.001 -0.90% +1.05% -1.91% <0.001 -0.90% +0.93% -1.78% <0.001
Non-trending + High vol Classical "whipsaw zone" for momentum 248,993 +1.14% +1.05% +0.00% 0.9760 +1.14% +1.33% -0.18% 0.0006 +1.14% +1.19% -0.08% 0.1135
Sub-period breakdown table (20d alpha)
Historical clustering check. If alpha concentrates in one era, the signal's robustness is questionable.
Period N Alpha % (spxew) p (HAC) Alpha % (spx) p (HAC) Alpha % (msci) p (HAC)
2015-2019 2015-01-01 → 2020-01-01 285,844 +0.09% 0.0721 +0.10% 0.0412 +0.33% <0.001
2020-2022 2020-01-01 → 2023-01-01 312,377 +0.27% <0.001 +0.12% 0.0390 +0.37% <0.001
2023-2026 2023-01-01 → 2099-01-01 331,155 +0.78% <0.001 +0.33% <0.001 +0.27% <0.001

Methodology and caveats

How to read. Entry at open of T+1 (one trading day after the signal fires on close of T). 20d = open T+1 to close T+20. Alpha = stock return − benchmark return over the same window (Convention A, single-sided, textbook). For bullish triggers, POSITIVE alpha = signal was right. For bearish triggers, NEGATIVE alpha = signal was right (stock underperformed market). No sign-flipping; the direction of the bet determines what "good" looks like. Per-stock regime is each stock's own ADX(14) and RV(20) at the trigger date — not market-wide state.

Three p-values, three robustness tests. (a) p_naive: scipy one-sample t-test on winsorized alphas. Optimistic because overlapping 20d windows on the same ticker inflate effective N. (b) p_hac: Newey-West HAC with lag = horizon — corrects for the overlap and is the academic-finance standard. (c) p_perm: one-sided fraction of 200 random-date null iterations falling in the “signal was right” tail (mean ≥ observed for bullish; mean ≤ observed for bearish). Tests whether the signal beats random date selection at all. A signal that clears all three (pnaive, phac, pperm all < 0.05) has real information; a signal that fails pperm has not beaten random timing whatever the t-test says — and because the test is one-sided, a pperm up at its 1.000 ceiling is not "no edge" but inverted edge: every random draw served the claimed direction better than the trigger dates did.

Caveats. (i) Universe reflects today's active tickers; delisted losers pruned → survivorship bias. (ii) Mcap ≥ $100M filter uses today's snapshot, not point-in-time — mild lookahead on which stocks enter the sample, not on returns. (iii) Means and p-values use winsorized alphas (1/99 percentile) to prevent data errors from dominating. Medians and hit rates use raw data. (iv) Zero transaction costs assumed. Realistic bid-ask + commissions remove 20–40bps from 20d alpha on US large-caps, more on small-cap. Sub-20bps alpha is noise in practice. (v) Past performance does not predict future results.

How to use this

1 · When to reach for this signal

Use 52-Week New High / New Low bullish as a long-side screening tile. Bullish 20d alpha is +0.52% and beats random (permutation test, 200 iterations). Bearish 20d alpha is +0.39% — worse than random : firing on random dates would have done better.

These verdicts are 20-day holds vs S&P 500 Equal Weight. Longer horizons can differ in either direction — check the permutation detail tables below before extrapolating.

2 · When it works — the setups that drive it

  • Best bullish setup: Trending + High vol — alpha +0.73% / 20d on 727,791 historical triggers.
  • Best bearish setup: Non-trending + Low vol — alpha -1.67% / 20d on 56,541 historical triggers.
  • Best era for bullish: 2023-2026 — alpha +1.00% / 20d on 601,192 triggers.
  • Least-bad era for bearish: 2015-2019 — alpha +0.09% / 20d on 285,844 triggers — still wrong-signed; no era produced negative alpha.

3 · When it fails — common false positives

  • Weakest bullish cell: Non-trending + Low vol — alpha +0.18% / 20d on 111,905 triggers.
  • Weakest bearish cell: Trending + High vol — alpha +1.27% / 20d on 498,175 triggers.
  • Worst era for bullish: 2015-2019 — alpha -0.09% / 20d on 335,648 triggers.
  • Worst era for bearish: 2023-2026 — alpha +0.78% / 20d on 331,155 triggers.

Signal-specific failure patterns

Late confirmation is structural — the move is mature by the time the trigger prints
The signal fires when today's intraday high touches or exceeds the maximum high of the prior 252 trading days (lookback adjustable 126-504 in the report builder). By construction, a stock can only print a 52-week high after a substantial advance is already behind it, so the trigger marks a mature trend, not an early entry. Whether buying that strength still beats the equal-weight benchmark from the trigger date forward is an empirical question that moves with the sample — see the at-a-glance table and the permutation-null line above for the current answer rather than assuming the 'momentum buys winners' intuition holds.
Consecutive-day re-triggering concentrates the sample in runaway names
A stock in a strong uptrend re-fires this trigger day after day — each session's high exceeds the prior 252-day maximum again. Raw trigger counts are therefore dominated by long runs in a handful of trending names rather than independent events, which inflates the trigger count and clusters the risk. The related fresh 52-week high signal adds a 20-trading-day cooldown so only the first trigger of each run counts; compare both signals' tables before concluding the difference is anything more than trigger bookkeeping.
Selection bias flatters the index comparison — the permutation null is the honest test
Names printing 52-week highs are, by construction, past winners: higher-momentum and often higher-beta stocks. Simple alpha versus an index mixes what kind of stock the signal picks with when it fires. The random-date permutation test — re-firing the same number of triggers per ticker at random dates — isolates the timing content, so read the permutation line above, not just the raw alpha, before treating a fire as tradable evidence. Note also that there is no candle confirmation: an intraday spike to a marginal new high that closes weak still counts as a trigger.
The decline is mature by the time the trigger prints
The signal fires when today's intraday low touches or breaks the minimum low of the prior 252 trading days (lookback adjustable 126-504 in the report builder), with no close confirmation. By construction, a stock can only print a 52-week low after a substantial decline is already behind it. From that point, forward behavior splits between continued distress and base-building recovery, and which dominates at each horizon changes with the sample — check the at-a-glance table and the permutation-null line above for the current answer at each horizon rather than assuming new lows keep falling.
Short-squeeze and base-building mechanics can invert the read at longer horizons
Stocks making fresh 52-week lows attract short interest, and heavily shorted names are squeeze candidates: forced-cover rallies in the two-to-three-month window after a low are a classic market-mechanics phenomenon, and washed-out sellers plus bargain hunters can start a base regardless of fundamentals. This is a timeless structural reason the bearish read can weaken or invert as the holding period extends. Compare the short-horizon columns against the 60-day-and-beyond columns in the current tables before extending any short-side hold.
Regime dependence: continuation indicator in some markets, bounce indicator in others
Whether a wave of new lows marks further weakness or a contrarian washout has historically flipped with the market regime — breadth, volatility, and how concentrated the selling is. This is why the tables above show sub-period rows and a regime split (trending vs. non-trending by ADX, high vs. low realized volatility). Do not extrapolate a single sub-period's result to the current market; weight the most recent rows and treat older regimes as context.

4 · Pairing inside a screen

The statements below describe how this signal relates to others by construction — which indicator family it belongs to, and where same-family redundancy might reduce the independence of evidence inside a Daily Report. These are taxonomic classifications drawn from standard technical-analysis texts; they are not pairing backtests. This signal was not part of the tested set in the pair backtest run, so its pairing behaviour remains unmeasured.

Breakout-family redundancy

New 52-week high, new 20-day high, and fresh 52-week high are breakout signals at different lookbacks — each fires when price reaches the maximum of the prior N bars, though the 20-day variant additionally requires a non-down candle (close at or above open) while the 52-week variants use the bare institutional definition with no candle filter (Edwards & Magee, Technical Analysis of Stock Trends, 11th ed. 2018; Kirkpatrick & Dahlquist, Technical Analysis, 3rd ed. 2015; Bulkowski, Encyclopedia of Chart Patterns, 3rd ed. 2021). Stacking two or more in the same direction within a single Daily Report produces correlated rather than independent evidence.

Breakdown-family redundancy

New 52-week low, new 20-day low, and fresh 52-week low are breakdown signals at different lookbacks — each fires when price reaches the minimum of the prior N bars, though the 20-day variant additionally requires a non-up candle (close at or below open) while the 52-week variants use the bare institutional definition with no candle filter (Edwards & Magee, Technical Analysis of Stock Trends, 11th ed. 2018; Kirkpatrick & Dahlquist, Technical Analysis, 3rd ed. 2015; Bulkowski, Encyclopedia of Chart Patterns, 3rd ed. 2021). Stacking two or more in the same direction within a single Daily Report produces correlated rather than independent evidence.

What would likely rescue this signal

This block calls out the data or conditions that could turn a technically weak signal into a usable one in a composite screen. Based on signal mechanics and the observed failure patterns above; individual combinations are not yet backtested.

  • Require consolidation structure — A breakout to a new high from a tight base (for example, a narrow trailing 20-day trading range) is a structurally different setup from a straight-line advance that happens to print another daily high. Classical pattern literature treats flat-base breakouts more favorably than extended continuation. Measurable from the same OHLC data; testable without new sources.
  • Add a fundamentals condition — A 52-week high backed by earnings or revenue growth is a structurally different trade than one without. The live fundamentals filter in the report builder can require this directly — for example, earnings growth or revenue growth above a chosen year-over-year threshold alongside the new high — so the combination is screenable today rather than a hypothetical.
  • Sector-relative filter — A new high that is ALSO a sector-relative high (the stock outperforming its sector over the trailing month) is a cleaner leadership signal than an absolute new high alone. Derivable from existing data.
  • Time-stop discipline — If the current tables show the bearish edge decaying or inverting beyond the short horizon, exit on a fixed time stop near the horizon where the edge fades rather than holding for more. The structural reason is squeeze and base-building risk, which grows with time since the low.
  • Regime-gate on breadth — New lows behave differently in narrow-breadth and broad-breadth markets. Conditioning the signal on a market-breadth measure — for example, only acting on bearish fires when broad participation is already weak — would make it regime-adaptive. The breadth dashboard provides the input; the sub-period rows above are the check on whether the gate earns its keep.
  • Skip low-mcap names — Microcap 52-week lows are often liquidity-driven rather than fundamental — thin order books print marginal new lows on small sell orders. Require a market-cap floor (e.g. $500M) or use the LIQUID variant of the universe to filter out that noise.

See also Why technical-only signals don't survive on their own for the broader argument.

5 · Before you act — a 5-point checklist

  1. Normal trading day? Rule out earnings (within ±3 days), ex-dividend, or known corporate-action dates — the signal is almost certainly reading noise, not momentum, in those windows.
  2. Where is price vs its own 50 / 200 DMA? A trend signal is only as credible as the underlying trend it claims to confirm. Check the 200DMA orientation before acting.
  3. What's the sector breadth doing? An isolated signal in a broadly down-trending sector is a lower-confidence setup than one firing with the rest of its peer group.
  4. Is ADV20 enough for your size? If the trigger is on a $500M name and you want to move $1M notional, you're the tape. Consider adv20d ≥ 5% of your intended position.
  5. What invalidates you? Define a price level (for longs: a close below the trigger-day low; for shorts: close above the trigger-day high) and honor it. The backtest alpha is an average; any one trade can be at either tail.

Execution notes

Best use is as a universe filter for a broader screen rather than a standalone entry trigger: isolate the leadership subset, then apply a secondary condition (fundamentals, sector-relative strength, consolidation structure, volume). The backtest convention is entry at the next session's open (T+1) after a trigger at the close of day T. Direction guidance should follow the live data: if the bullish side clears the random-date permutation null in the current tables, treat the trigger as a leadership screen tile; if it does not, treat 52-week-high fires as descriptive context about where momentum sits, not as an entry list. Either way, this is a historical tendency, not a recommendation. New lows side: The backtest convention is entry at the next session's open (T+1) after a trigger at the close of day T; the trigger itself fires on an intraday touch of the prior 252-day minimum. Direction guidance should follow the live data: if the bearish side beats the random-date permutation null at short horizons in the current tables, the signal can serve as a short-side screen tile with a strict time stop; if the longer-horizon columns weaken or invert, that is the squeeze and base-building window — do not extend holds mechanically. If the bearish side fails the null outright, treat fires as distress context for regime and breadth reading rather than as trade triggers. Historical tendency, not a recommendation.

See a live screen of this signal

Curated Daily Reports that screen for this signal, refreshed at the open, midday and close — free to view, no account needed: