Trend macd

MACD Crossover

Bullish: MACD line crosses above signal line. Bearish: MACD line crosses below signal line.

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
fast Fast EMA period 12 5–50
slow Slow EMA period 26 10–100
signal_period Signal line period 9 3–30

Historical context

3,243,150 triggers on 24,078 tickers, 1988-04-19 → 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 -0.10% +0.01%
20-day +0.04% +0.01%
60-day +0.38% +0.34%
1-year +2.20% +2.21%
Random-date null check (20-day): Bullish: inside null (p=0.945).
Bearish: beats random (p=0.005).

Where does MACD 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.

MACD Crossover (macd) — trigger count distribution by per-stock regime quadrant (trending/non-trending × high/low realized volatility) for bullish and bearish triggers, global universe

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. This is the 20-day alpha a trade taken on the trigger would have captured. Bars carrying the sign that favours the trade (positive for bullish triggers, negative for bearish) mark the regimes where the signal worked; the opposite sign marks regimes to avoid. One strong bar beside three flat ones is not a "20-day alpha" signal — it is a "20-day alpha when the stock is X" signal. Bar labels carry the sample size; a cell built on a handful of triggers is noise, not a regime finding.

MACD Crossover (macd) — mean 20-day alpha versus S&P 500 Equal Weight by per-stock regime quadrant, bullish and bearish triggers side by side
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 carrying the sign that favours the trade means the signal is durable; one era doing all the work means a regime-specific edge that may not repeat. The greater the variance across eras, the smaller the position it justifies.

MACD Crossover (macd) — 20-day alpha split by historical sub-period (2015-2019, 2020-2022, 2023+) to check consistency across market regimes

↑ Bullish triggers

Bench Metric 1d 5d 20d 60d 252d
spxew Stock % +-0.00% +0.19% +0.98% +2.91% +12.34%
Bench % +0.05% +0.28% +0.99% +2.52% +10.15%
Alpha % -0.06% -0.10% +0.04% +0.38% +2.20%
Median alpha -0.13% -0.38% -0.86% -1.86% -6.29%
Hit rate (α>0) 46.7% 46.0% 45.7% 44.7% 42.1%
p (naive) <0.001 <0.001 <0.001 <0.001 <0.001
p (HAC) <0.001 <0.001 0.0002 <0.001 <0.001
N 1,561,702 1,506,578 1,491,077 1,458,261 1,296,775
spx Stock % +-0.00% +0.19% +0.98% +2.91% +12.34%
Bench % +0.03% +0.31% +1.18% +3.17% +13.86%
Alpha % -0.03% -0.12% -0.16% -0.27% -1.59%
Median alpha -0.11% -0.42% -1.10% -2.57% -10.21%
Hit rate (α>0) 46.9% 45.5% 44.5% 42.9% 37.9%
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 1,571,311 1,521,874 1,512,718 1,476,353 1,317,668
msci Stock % +-0.00% +0.19% +0.98% +2.91% +12.34%
Bench % +0.07% +0.31% +1.03% +2.75% +11.40%
Alpha % -0.07% -0.12% -0.01% +0.16% +0.70%
Median alpha -0.15% -0.43% -0.96% -2.16% -7.87%
Hit rate (α>0) 46.0% 45.4% 45.2% 43.9% 40.4%
p (naive) <0.001 <0.001 0.4358 <0.001 <0.001
p (HAC) <0.001 <0.001 0.4989 <0.001 <0.001
N 1,564,570 1,514,869 1,500,340 1,465,824 1,302,252
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.
MACD Crossover (macd) — 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 +0.07% +0.08% [+0.07%, +0.08%] 0.861
1d spx +0.09% +0.09% [+0.08%, +0.09%] 0.269
1d msci +0.08% +0.09% [+0.09%, +0.09%] 1.000
5d spxew +0.28% +0.34% [+0.33%, +0.35%] 1.000
5d spx +0.30% +0.36% [+0.35%, +0.37%] 1.000
5d msci +0.30% +0.36% [+0.35%, +0.37%] 1.000
20d spxew +1.09% +1.11% [+1.09%, +1.12%] 0.945
20d spx +1.13% +1.14% [+1.12%, +1.15%] 0.836
20d msci +1.16% +1.15% [+1.13%, +1.17%] 0.080
60d spxew +2.44% +2.38% [+2.35%, +2.41%] 0.005
60d spx +2.54% +2.45% [+2.42%, +2.48%] 0.005
60d msci +2.55% +2.47% [+2.44%, +2.50%] 0.005
252d spxew +4.73% +4.60% [+4.54%, +4.66%] 0.005
252d spx +5.01% +4.95% [+4.89%, +5.01%] 0.030
252d msci +5.00% +4.90% [+4.85%, +4.96%] 0.005

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

Six recent bullish MACD triggers on US mega-caps, ranked within this sample of six: the top three and the bottom three, with extreme outliers excluded. Ranking is relative, not absolute — where the signal did well across the sampled names, even the bottom three can have beaten the benchmark, so the alpha printed on each panel is what settles it. Both groups are tail outcomes by construction; read them as the range, not the typical result.

Best three of the six sampled
MACD Crossover (macd) — 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 MACD Crossover (macd) — 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 MACD Crossover (macd) — 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 three of the six sampled — not necessarily losses
MACD Crossover (macd) — weakest bullish trigger #1 on a US mega-cap, showing the 20-day forward window where the signal failed MACD Crossover (macd) — weakest bullish trigger #2 on a US mega-cap, showing the 20-day forward window where the signal failed MACD Crossover (macd) — 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 101,975 +0.43% +0.64% -0.15% <0.001 +0.43% +0.92% -0.45% <0.001 +0.43% +0.77% -0.29% <0.001
Trending + High vol Crisis selloff or parabolic rally 510,937 +1.27% +1.21% +0.12% <0.001 +1.27% +1.39% -0.08% <0.001 +1.27% +1.19% +0.10% <0.001
Non-trending + Low vol Quiet chop, summer doldrums 207,810 +0.52% +0.61% -0.07% <0.001 +0.52% +0.87% -0.34% <0.001 +0.52% +0.73% -0.19% <0.001
Non-trending + High vol Classical "whipsaw zone" for momentum 802,556 +1.02% +0.99% +0.06% 0.0002 +1.02% +1.16% -0.10% <0.001 +1.02% +1.02% +0.03% 0.0802
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 → 2020-01-01 489,732 -0.27% <0.001 -0.45% <0.001 -0.27% <0.001
2020-2022 2020-01-01 → 2023-01-01 477,836 -0.08% 0.0003 +0.26% <0.001 +0.43% <0.001
2023-2026 2023-01-01 → 2099-01-01 655,143 +0.37% <0.001 -0.25% <0.001 -0.13% <0.001

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

Bench Metric 1d 5d 20d 60d 252d
spxew Stock % -0.02% +0.17% +0.74% +2.74% +12.02%
Bench % +0.02% +0.16% +0.79% +2.46% +9.86%
Alpha % -0.05% +0.01% +0.01% +0.34% +2.21%
Median alpha -0.07% -0.21% -0.81% -1.79% -6.23%
Hit rate (α>0) 48.3% 47.7% 45.9% 44.9% 42.1%
p (naive) <0.001 0.0103 0.4075 <0.001 <0.001
p (HAC) <0.001 0.0121 0.4811 <0.001 <0.001
N 1,555,328 1,499,445 1,491,245 1,455,890 1,300,554
spx Stock % -0.02% +0.17% +0.74% +2.74% +12.02%
Bench % +0.00% +0.18% +0.98% +3.14% +13.67%
Alpha % -0.02% -0.02% -0.20% -0.38% -1.75%
Median alpha -0.05% -0.25% -1.07% -2.57% -10.22%
Hit rate (α>0) 48.6% 47.2% 44.6% 42.8% 37.8%
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 1,570,637 1,515,379 1,508,282 1,474,405 1,315,044
msci Stock % -0.02% +0.17% +0.74% +2.74% +12.02%
Bench % +0.03% +0.18% +0.87% +2.75% +11.31%
Alpha % -0.04% -0.01% -0.09% +0.06% +0.55%
Median alpha -0.07% -0.25% -0.95% -2.15% -7.92%
Hit rate (α>0) 48.0% 47.3% 45.1% 43.9% 40.3%
p (naive) <0.001 0.2364 <0.001 0.0008 <0.001
p (HAC) <0.001 0.2463 <0.001 0.0437 0.0003
N 1,561,935 1,508,905 1,501,606 1,465,933 1,309,100
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.
MACD Crossover (macd) — 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 +0.08% +0.08% [+0.07%, +0.08%] 1.000
1d spx +0.09% +0.09% [+0.08%, +0.09%] 0.985
1d msci +0.10% +0.09% [+0.08%, +0.09%] 1.000
5d spxew +0.37% +0.34% [+0.33%, +0.35%] 1.000
5d spx +0.38% +0.36% [+0.35%, +0.37%] 1.000
5d msci +0.39% +0.36% [+0.35%, +0.37%] 1.000
20d spxew +1.04% +1.11% [+1.09%, +1.13%] 0.005
20d spx +1.07% +1.14% [+1.12%, +1.16%] 0.005
20d msci +1.06% +1.15% [+1.13%, +1.17%] 0.005
60d spxew +2.41% +2.38% [+2.35%, +2.41%] 0.975
60d spx +2.43% +2.45% [+2.41%, +2.48%] 0.124
60d msci +2.45% +2.47% [+2.43%, +2.50%] 0.129
252d spxew +4.76% +4.60% [+4.54%, +4.67%] 1.000
252d spx +4.92% +4.95% [+4.89%, +5.02%] 0.164
252d msci +4.90% +4.90% [+4.84%, +4.97%] 0.428

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

Six recent bearish MACD triggers on US mega-caps, ranked within this sample of six: the top three and the bottom three, with extreme outliers excluded. Ranking is relative, not absolute — where the signal did well across the sampled names, even the bottom three can have beaten the benchmark, so the alpha printed on each panel is what settles it. Both groups are tail outcomes by construction; read them as the range, not the typical result.

Best three of the six sampled
MACD Crossover (macd) — 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 MACD Crossover (macd) — 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 MACD Crossover (macd) — 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 three of the six sampled — not necessarily losses
MACD Crossover (macd) — weakest bearish trigger #1 on a US mega-cap, showing the 20-day forward window where the signal failed MACD Crossover (macd) — weakest bearish trigger #2 on a US mega-cap, showing the 20-day forward window where the signal failed MACD Crossover (macd) — 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 112,897 +0.28% +0.51% -0.18% <0.001 +0.28% +0.81% -0.52% <0.001 +0.28% +0.67% -0.36% <0.001
Trending + High vol Crisis selloff or parabolic rally 562,962 +1.10% +0.77% +0.40% <0.001 +1.10% +1.00% +0.14% <0.001 +1.10% +0.88% +0.27% <0.001
Non-trending + Low vol Quiet chop, summer doldrums 205,330 +0.19% +0.67% -0.43% <0.001 +0.19% +0.90% -0.68% <0.001 +0.19% +0.77% -0.54% <0.001
Non-trending + High vol Classical "whipsaw zone" for momentum 738,674 +0.72% +0.88% -0.11% <0.001 +0.72% +1.02% -0.25% <0.001 +0.72% +0.93% -0.17% <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 → 2020-01-01 485,684 -0.36% <0.001 -0.55% <0.001 -0.38% <0.001
2020-2022 2020-01-01 → 2023-01-01 482,552 -0.14% <0.001 +0.10% <0.001 +0.21% <0.001
2023-2026 2023-01-01 → 2099-01-01 651,055 +0.41% <0.001 -0.16% <0.001 -0.09% <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

Not a standalone entry trigger at 20 days. Bullish 20d alpha is +0.04% — inside the null : indistinguishable from random timing. Bearish beats random on timing, but the raw 20d alpha is +0.01% — the flagged stocks still outperformed the benchmark, so it is not a short trigger on its own. Fires are screening context inside a composite (section 4), not entries.

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 +0.12% / 20d on 510,937 historical triggers.
  • Best bearish setup: Non-trending + Low vol — alpha -0.43% / 20d on 205,330 historical triggers.
  • Best era for bullish: 2023-2026 — alpha +0.37% / 20d on 655,143 triggers.
  • Best era for bearish: 2015-2019 — alpha -0.36% / 20d on 485,684 triggers.

3 · When it fails — common false positives

  • Weakest bullish cell: Trending + Low vol — alpha -0.15% / 20d on 101,975 triggers.
  • Weakest bearish cell: Trending + High vol — alpha +0.40% / 20d on 562,962 triggers.
  • Worst era for bullish: 2015-2019 — alpha -0.27% / 20d on 489,732 triggers.
  • Worst era for bearish: 2023-2026 — alpha +0.41% / 20d on 651,055 triggers.

Signal-specific failure patterns

The laggard trap (bullish)
MACD is built from two exponential moving averages (12-day minus 26-day at default settings), with the trigger firing when that difference crosses its own 9-day EMA. Every layer is a lag on price, so a bullish cross on a stock already trending cleanly typically prints after a pullback has resolved — the market has moved, and the stock may merely catch up without exceeding the benchmark's move. This is the structural reason a textbook-favourable setup can still fail to produce alpha. Whether it currently does, and in which trend/volatility regime, is what the regime split in the tables above is for — read it rather than assuming.
Capitulation bounce (bearish)
Bearish crosses cluster in violent selloffs: the fast EMA collapses through the signal line right around short-term lows, so the trigger often prints in high-volatility, non-trending tape near the point where sharp mean-reversion bounces are most likely. A bearish print at that moment looks like confirmation of weakness but is mechanically closest to an oversold extreme. Check the volatility-regime rows and the permutation-null line above before treating bearish crosses as short entries.
Whipsaw when the EMAs run flat
When price moves sideways, the 12- and 26-day EMAs converge and run nearly parallel, and the MACD line hovers around its signal line. Tiny price wobbles then produce rapid alternating bullish and bearish crosses with no directional content. There is no minimum-separation or zero-line condition in the trigger, so ranging tape generates a stream of low-information fires in both directions. Any single cross on a quiet chart is an anecdote; the signal's properties only exist in aggregate.
Earnings-week contamination
MACD crossovers that fire on news-driven gap days are reacting to information, not technical structure. The backtest behind the tables above does not exclude earnings weeks, so some of the measured behaviour — in either direction — is contaminated by event-driven moves. A practical mitigation when using the signal in a screen is to disregard triggers within a few trading days of scheduled earnings, or to require confirmation from a non-price read such as the live fundamentals filter.

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. Measured pair results — same-day co-fires put through the pair backtest — follow under “Measured pairings” below.

Trend-following momentum construction

MACD is the difference between two exponential moving averages of closing price (Appel, Technical Analysis: Power Tools for Active Investors, 2005) and is classified as a trend-following momentum indicator (Murphy, Technical Analysis of the Financial Markets, 1999; Kirkpatrick & Dahlquist, Technical Analysis, 3rd ed. 2015). Its construction overlaps with moving-average crossover signals, which also derive from differences of moving averages of price; stacking MACD with MA crossover in the same direction produces correlated rather than independent evidence.

Measured pairings — Bonferroni survivors

Beyond the literature pairings above, these are the same-day co-fire combinations involving MACD Crossover that cleared the pair backtest's Bonferroni cut on the full 2016–2026 sample (549 pairs × 5 horizons = 2,745 hypotheses), on universes filtered to ADV ≥ $5M, price ≥ $5 and market cap ≥ $100M. That cut is two-sided: it asks only whether the co-fire's α is reliably different from zero, in either direction, so a pair can survive by reliably underperforming — 2 of the 22 rows below do exactly that (negative full-sample α). The same run holds out 2023+: the Test columns are that held-out window, printed for every row with enough held-out co-fires to measure, so a survivor that did not repeat out of sample is visible rather than hidden. All α figures here are for holding the stock long after the co-fire — no shorting assumed, and no sign flip for bearish legs. So positive α means the co-fire was followed by outperformance and negative α by underperformance, whichever way either leg points — a bearish leg does not flip the reading. Survivors are rare by design — absence of a pair here means it did not clear the cut, not that it was untested. Ranked by held-out (2023+) α. Historical tendencies, not recommendations.

US (NYSE / NASDAQ / AMEX)

Pair (same-day co-fire, long) Full α Test α (2023+) Test N p_perm test
macd bearish + new_20d_low bearish +0.55% +0.85% 9,573 0.002

1 of this universe's 18 surviving pairs involves this signal · α vs ^SPXEW.

Europe — 2 surviving pairs
Pair (same-day co-fire, long) Full α Test α (2023+) Test N p_perm test
macd bearish + rsi bearish +0.77% +0.80% 800 0.006
macd bullish + new_20d_high bullish +0.37% +0.23% 4,112 0.110

2 of this universe's 20 surviving pairs involve this signal · α vs ^STOXX.

Hong Kong — 3 surviving pairs
Pair (same-day co-fire, long) Full α Test α (2023+) Test N p_perm test
macd bearish + volume_breakout bullish +5.41% +6.12% 24 0.052
macd bullish + volume_breakout bullish +1.52% +2.29% 897 0.002
macd bullish + new_20d_high bullish +0.94% +1.56% 1,255 0.002

3 of this universe's 23 surviving pairs involve this signal · α vs ^HSI.

China A-shares — 16 surviving pairs
Pair (same-day co-fire, long) Full α Test α (2023+) Test N p_perm test
hh_hl_structure bearish + macd bullish +1.61% +3.03% 799 0.002
ma_crossover bullish + macd bearish +2.20% +2.49% 436 0.002
failed_double_bottom bullish + macd bullish +1.42% +1.82% 756 0.002
failed_double_top bearish + macd bearish +1.57% +1.38% 887 0.002
macd bullish + stochastics bullish +0.66% +1.36% 5,230 0.002
macd bullish + weekly_change bullish +0.51% +1.25% 15,314 0.002
macd bullish + williams_r bullish +0.50% +1.10% 10,608 0.002
cci bullish + macd bullish +0.74% +1.01% 9,273 0.002
macd bullish + rsi bullish +1.09% +0.87% 1,588 0.002
hh_hl_streak bullish + macd bullish +1.01% +0.83% 11,383 0.002
macd bearish + weekly_change bearish +0.76% +0.72% 9,155 0.002
macd bullish + vwap_cross bullish +0.29% +0.61% 32,576 0.002
macd bearish + vwap_cross bearish +0.38% +0.50% 30,953 0.002
macd bearish + new_20d_low bearish +0.41% +0.50% 16,288 0.002
macd bullish + new_20d_high bullish -0.61% -0.55% 18,940 0.002
double_top_breakout bullish + macd bullish -1.85% -1.75% 1,011 0.002

16 of this universe's 138 surviving pairs involve this signal · α vs 83188.HK.

China A-share survivor α runs large but skews toward small-caps, where trading costs and thin liquidity claim a large share of any measured edge — screening context, not a capturable spread.

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.

  • Use crosses as a stack member, not a standalone trigger — Even where a side clears the permutation null in the current tables, a single EMA crossover is a weak, widely watched event. The coherent use is conjunction: require the MACD cross AND an independent read in the same direction — close relative to the 50DMA, a fresh 52-week extreme, sector breadth — so the surviving triggers are a concentrated population rather than every crossover on the tape.
  • Regime-gate the bullish side — Crossover momentum behaves differently when index returns are driven by a handful of concentrated leaders versus broad participation. If the regime and sub-period rows in the tables above show the bullish side working only in some environments, a breadth-based gate — for example, only taking bullish crosses when market breadth is broad — is the natural refinement. The platform's breadth history makes this testable directly.
  • Pair with volume confirmation — MACD fires on a mathematical EMA relationship that ignores volume. A bullish crossover on heavy volume (say 2x the 20-day average) is plausibly a different population than one on thin volume, and volume is already available on every chart in the platform. Implementing the conjunction as a composite screen is straightforward and discards the thin-volume majority by design.

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

5 · Before acting — 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 the intended size? A $1M notional order in a $500M name moves the tape by itself. A useful floor is adv20d ≥ 5% of the intended position.
  5. What invalidates the trade? Define a price level in advance (for longs: a close below the trigger-day low; for shorts: a close above the trigger-day high) and honour it. The backtest alpha is an average; any single trade can land at either tail.

Execution notes

The trigger is a one-day cross event: the MACD line (12-day EMA minus 26-day EMA of the close, defaults) crossing its 9-day EMA signal line, in either direction, at any level — there is no zero-line or histogram condition. The backtest behind the tables measures entry at the open on T+1 after a trigger on the close of T; earlier intraday entry was not tested and our prior is that it is noisier due to late-day fade dynamics on momentum names. Treat direction tradability as an empirical question: take a side as a screen tile only if it beats the random-date null in the current tables at your horizon, and choose the horizon from the live horizon columns rather than a remembered ranking. 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: