MACD Crossover: does it work?
Last updated
In-sample, not a promise. Every figure we measured here comes from a backtest on past data.
Did MACD Crossover beat the benchmark in Tickfloor's test?
Not reliably. We tested MACD Crossover on 1 ETF, two US stock universes (171 stocks and 40 stocks), 18 crypto pairs, 2005 to 2025. After costs, it returned anywhere from 8.3 percentage points a year less than the benchmark to 17.7 more. None held up once we allowed for the many other rules we tested. Historical test, not advice.
The MACD(12,26,9) signal-line crossover is one of the most widely taught trend-following entries, often pitched on its own without any RSI confirmation.
The rule we tested. Four long-only versions of the MACD(12,26,9) rule in two codings. The classic coding goes long on the cross above the signal line and to cash on the cross back below (GLD, and 171 US stocks). The other is long while the MACD line sits above its signal line and in cash otherwise (40 US stocks and 18 crypto pairs).
- Classic crossover, GLD −3.0 pts/yr vs the benchmark, 203 trades, p=1.000, corrected q=1.000
- MACD above signal, 40 US stocks −8.3 pts/yr vs the benchmark, 3,290 trades, p=1.000, corrected q=1.000
- Classic crossover, 171 US stocks −6.3 pts/yr vs the benchmark, 14,068 trades, p=1.000, corrected q=1.000
- MACD above signal, 18 crypto pairs +17.7 pts/yr vs the benchmark, 1,382 trades, p=0.275, corrected q=1.000
Receipt: parameters, data, window and result file
Classic crossover, GLD
- Run id
- trend-macd-12-26-9-classic, run 6 October 2026
- Frozen parameters
- timeframe 1d; fast=12, slow=26, signalN=9
- Data source
- vendor not recorded in the result file; data fingerprint sha256 d100bb23b9571f69...
- Instruments and coverage
- 1 asset: GLD
- Date window
- 4 January 2005 to 14 March 2025 (discovery cutoff 15 March 2025)
- Sample
- 203 completed trades over 4 January 2005 to 14 March 2025 (20.2 years)
- Cost rule
- 0.05% per side; charged on every change in position/weight; benchmark pays none; no funding/borrow on shorts
- Fill rule
- not recorded
- Benchmark
- equal-weight long of the same assets, rebalanced daily across assets trading that day (no costs)
- Rule source (public)
- research/hunt2/strategies/trend-macd-12-26-9-classic.mjs
- Run record (public)
- research/hunt2/ledger.jsonl, line with id trend-macd-12-26-9-classic
- Full result file
- not published yet
MACD above signal, 40 US stocks
- Run id
- t100-macd-s, run 6 October 2026
- Frozen parameters
- timeframe 1d; fast=12, slow=26, sig=9
- Data source
- vendor not recorded in the result file; data fingerprint sha256 7b5251fa65d8de0e...
- Instruments and coverage
- 40 assets: AAPL, NVDA, AMZN, GOOGL, META, TSLA, JPM, V, UNH, XOM, JNJ, COST and 28 more (full list in the result file). 9 corrupt vendor bar-day(s) across 9 assets were dropped, never repaired
- Date window
- 15 September 2016 to 14 March 2025 (discovery cutoff 15 March 2025)
- Sample
- 3,290 completed trades over 15 September 2016 to 14 March 2025 (8.5 years)
- Cost rule
- 0.05% per side; charged on every change in position/weight; benchmark pays none; no funding/borrow on shorts
- Fill rule
- not recorded
- Benchmark
- equal-weight long of the same assets, rebalanced daily across assets trading that day (no costs)
- Rule source (public)
- research/hunt2/strategies/t100-macd-s.mjs
- Run record (public)
- research/hunt2/ledger.jsonl, line with id t100-macd-s
- Full result file
- not published yet
Classic crossover, 171 US stocks
- Run id
- wide-trend-macd-12-26-9-classic, run 18 September 2026
- Frozen parameters
- universe US_STOCKS_WIDE; timeframe 1d; fast=12, slow=26, signalN=9
- Data source
- vendor not recorded in the result file; data fingerprint sha256 9c3348762d6d47fb...
- Instruments and coverage
- 171 assets: AAPL, NVDA, AVGO, ORCL, CRM, AMD, INTC, ADBE, QCOM, TXN, AMAT, PANW and 159 more (full list in the result file)
- Date window
- 15 September 2016 to 14 March 2025 (discovery cutoff 15 March 2025)
- Sample
- 14,068 completed trades over 15 September 2016 to 14 March 2025 (8.5 years)
- Cost rule
- 0.05% per side; charged on every change in position/weight; benchmark pays none; no funding/borrow on shorts
- Fill rule
- not recorded
- Benchmark
- equal-weight long of the same assets, rebalanced daily across assets trading that day (no costs)
- Rule source (public)
- research/hunt2/strategies/wide-trend-macd-12-26-9-classic.mjs
- Run record (public)
- research/hunt2/ledger.jsonl, line with id wide-trend-macd-12-26-9-classic
- Full result file
- not published yet
MACD above signal, 18 crypto pairs
- Run id
- t100-macd-c, run 6 October 2026
- Frozen parameters
- timeframe 1d; fast=12, slow=26, sig=9
- Data source
- vendor not recorded in the result file; data fingerprint sha256 e7098c5c15a8401c...
- Instruments and coverage
- 18 assets: BTCUSDT, BNBUSDT, XRPUSDT, ADAUSDT, DOGEUSDT, LTCUSDT, LINKUSDT, TRXUSDT, BCHUSDT, ATOMUSDT, ETCUSDT, DASHUSDT and 6 more (full list in the result file)
- Date window
- 18 August 2017 to 14 March 2025 (discovery cutoff 15 March 2025)
- Sample
- 1,382 completed trades over 18 August 2017 to 14 March 2025 (7.6 years)
- Cost rule
- 0.10% to 0.17% per side; charged on every change in position/weight; benchmark pays none; no funding/borrow on shorts
- Fill rule
- not recorded
- Benchmark
- equal-weight long of the same assets, rebalanced daily across assets trading that day (no costs)
- Rule source (public)
- research/hunt2/strategies/t100-macd-c.mjs
- Run record (public)
- research/hunt2/ledger.jsonl, line with id t100-macd-c
- Full result file
- not published yet
No. None of the versions we tested passed. None held up once we allowed for how many other rules we tested. That covers the versions, costs and benchmark described here, not every version of the rule.
Tested a rule of your own? Check whether your backtest is just luck. It is free and needs no account.
This is one line in a bigger check. The research desk has run 1,346 strategies, most against an equal-weight benchmark of the same assets rebalanced daily that pays no costs (the strategies pay theirs). After correcting for how many we tried (the luck check, Benjamini‑Hochberg, 1,778 tests), 0 passed. That does not prove no strategy works, and it says nothing about versions we did not test.
Short answers: the verdict and the rule, as questions
Does the classic MACD crossover work?
No. None of the versions we tested passed. After costs: Classic crossover, GLD: −3.0 pts/yr vs the benchmark; MACD above signal, 40 US stocks: −8.3 pts/yr vs the benchmark; Classic crossover, 171 US stocks: −6.3 pts/yr vs the benchmark; MACD above signal, 18 crypto pairs: +17.7 pts/yr vs the benchmark. None held up once we allowed for how many other rules we tested. That covers the versions, costs and benchmark described here, not every version of the rule.
What exact rule did Tickfloor test?
Four long-only versions of the MACD(12,26,9) rule in two codings. The classic coding goes long on the cross above the signal line and to cash on the cross back below (GLD, and 171 US stocks). The other is long while the MACD line sits above its signal line and in cash otherwise (40 US stocks and 18 crypto pairs).
Is this financial advice?
No. This measures a publicly claimed strategy rule, not a recommendation. General information only, not personal advice.
See the full numbers on the Trend following, Top 100 retail strategies family pages, or the full method and every result.
Other rules we've tested
Terms explained
Next steps
- See a full worked example (free, no account)
- Start lesson 1 of the course (free)
- Run your own version of this rule in the Backtest Lab (needs a pass)
General information only. It doesn't consider your objectives, finances or needs. Tickfloor holds no financial services licence and never places trades.