Tickfloor

Fear & Greed Timing: does it work?

Last updated

In-sample, not a promise. Every figure we measured here comes from a backtest on past data.

Did Fear & Greed Timing beat the benchmark in Tickfloor's test?

Not reliably. We tested Fear & Greed Timing on two crypto universes (18 pairs and 1 pair), 2017 to 2025. After costs, it returned anywhere from 59.6 percentage points a year less than the benchmark to 40.6 more. None held up once we allowed for the many other rules we tested. This was a historical test, not advice.

Beyond the plain buy-fear, sell-greed rule, traders use the crypto Fear & Greed Index as a timing layer: only run momentum when sentiment is greedy, or add an override after extreme fear.

The rule we tested. Three ways to use the index as a timing input, each tested net of costs on the same coins: momentum switched on or off by the reading, a four-week momentum rule with an extreme-fear override, and staying invested while the reading is neutral or above. The contrarian version is included for reference.

Receipt: parameters, data, window and result file

Two-week momentum only in greed, BTC in fear

Run id
s29-cx-fng-cond-mom, run 29 September 2026
Frozen parameters
rebalance weekly; timeframe 1d; greed=60, fear=40, lookback=14, top=4
Data source
vendor not recorded in the result file; data fingerprint sha256 59abf6bb185e5bad...
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
186 holdings changes across 393 rebalances (a holdings change is a rebalance that altered what the portfolio held, not a completed trade) 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)
Run record (public)
research/hunt2/ledger.jsonl, line with id s29-cx-fng-cond-mom
Full result file
not published yet

Crypto 4-week momentum plus a 90-day extreme-fear override

Run id
r27-cx-tsmom28-fngfear, run 6 October 2026
Frozen parameters
timeframe 1d; lookback=28, fearLevel=10, fearDays=90
Data source
vendor not recorded in the result file; data fingerprint sha256 59abf6bb185e5bad...
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
1,197 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)
Run record (public)
research/hunt2/ledger.jsonl, line with id r27-cx-tsmom28-fngfear
Full result file
not published yet

Crypto held while Fear & Greed is 50 or above

Run id
r27-cx-fng-trend, run 6 October 2026
Frozen parameters
universe CRYPTO_DAILY; timeframe 1d; threshold=50
Data source
vendor not recorded in the result file; data fingerprint sha256 59abf6bb185e5bad...
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
1,335 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)
Run record (public)
research/hunt2/ledger.jsonl, line with id r27-cx-fng-trend
Full result file
not published yet

Crypto Fear & Greed Index contrarian entries

Run id
mrsa-fear-greed-contrarian-crypto, run 6 October 2026
Frozen parameters
timeframe 1d; fearThreshold=20, greedThreshold=80, consecutiveDays=3, exitIndexLevel=50, maxHoldDays=30
Data source
vendor not recorded in the result file; data fingerprint sha256 446daf36b0035f83...
Instruments and coverage
1 asset: BTCUSDT
Date window
18 August 2017 to 14 March 2025 (discovery cutoff 15 March 2025)
Sample
72 completed trades over 18 August 2017 to 14 March 2025 (7.6 years)
Cost rule
0.10% 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)
Run record (public)
research/hunt2/ledger.jsonl, line with id mrsa-fear-greed-contrarian-crypto
Full result file
not published yet

Not reliably. Some versions finished ahead of the benchmark, but not by enough to pass our full test. 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.

Stress tests on the discovery data

Same data, harder conditions. Excess is CAGR minus the equal-weight benchmark's CAGR.

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 Fear & Greed timing work?

Not reliably. Some versions finished ahead of the benchmark, but not by enough to pass our full test. After costs: crypto pairs +39.5 pts/yr vs the benchmark; crypto pairs +40.6 pts/yr vs the benchmark; crypto pairs +0.6 pts/yr vs the benchmark; crypto pair −59.6 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?

Three ways to use the index as a timing input, each tested net of costs on the same coins: momentum switched on or off by the reading, a four-week momentum rule with an extreme-fear override, and staying invested while the reading is neutral or above. The contrarian version is included for reference.

Is this the same as the Fear & Greed contrarian page?

No. The contrarian rule (buy extreme fear, sell extreme greed) has its own page, linked below. This one covers using the index as a filter or timing layer on other rules.

Did it hold up in stress tests and on later data?

Two-week momentum only in greed, BTC in fear showed +32.1 pts/yr with costs doubled; Crypto 4-week momentum plus a 90-day extreme-fear override showed +35.6 pts/yr with costs doubled. None of this is an edge: Tickfloor's correction for multiple testing left 0 survivors out of 1,778 tests.

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 Momentum, Sentiment extremes, Pairs and stat arb family pages, or the full method and every result.

Linked from this page

Other rules we've tested

Terms explained

Next steps

General information only. It doesn't consider your objectives, finances or needs. Tickfloor holds no financial services licence and never places trades.