Strategies that survived the vetting bar
Tested against the hypothesis, on the app's own engine — full evidence attached, including the ugly parts.
Strip-best-trade
Profitable even without its best trade.
Two windows
Works across independent multi-year regimes.
Cross-asset
Holds on more than one instrument.
Net of fees
Realistic costs, plus a stressed rerun.
Sane drawdown
Survivable drawdowns, roughly ≤ 20%.
Beats random entry
Out-earns 400 random-timing runs of the same exit.
Tokyo-Fix Gotobi Long — 5m
Japanese importers settle on gotobi days and buy their dollars at the 09:55 Tokyo fix, so banks buy ahead of it — long from 03:00 to 09:55 Tokyo on those mornings. A calendar of cash flows, not a chart pattern: the first sub-daily entry to clear all six checks in both windows.
Funding Harvest (Cash-and-Carry)
No price prediction at all: long spot + short perp, equal size, and collect the funding payments shorts receive every 8 hours while positioning is crowded long. Exit when the heat fades.
BTC Funding-Squeeze — 1h
Entry #1's short-squeeze fade on 1h at real futures fees — the first sub-daily strategy to survive the research sweeps. BTC only; needs a ≤0.05%/side taker tier.
Return
+13.1%
Profit factor
1.36
Win rate
44%
Double-7s Channel Dip-Buy
Buy an uptrending index ETF below its prior 7-day low, sell above its prior 7-day high — the channel cousin of the RSI(2) dip-buy (same family: don't stack both).
Return
+103.8%
Profit factor
2.22
Win rate
75%
SMA-200 Trend Timing
Own the index while it trends, stand aside while it doesn't — in above the 200-day line, out below it. The crash years are spent in cash.
Return
+136.5%
Profit factor
3.06
Win rate
28%
Index-ETF RSI(2) Dip-Buy
Panic dips in a diversified index overshoot and snap back — buy the dip only while the long-term trend is up, sell the bounce days later.
Return
+36.4%
Profit factor
2.17
Win rate
72%
BTC Funding-Squeeze Long
When funding turns negative, shorts are crowded and paying longs — buy the oversold bounce and ride the short squeeze with a trailing stop.
Return
+28.9%
Profit factor
2.32
Win rate
55%
Few entries by design — one is added when a strategy clears the bar, not on a schedule. Sixteen sweeps, seven survivors — see what the bar rejected →
Strategies that worked for a while, then stopped
These failed the bar above. Each one beat random entry timing in at least one of its two test windows, most of them in only one, so each is shown with its whole record: the window it worked in, the other one, and every year. One window is a weaker test than two — run enough tests and a few will clear one window by luck — so read the record before the result.
Required in the window it claims
Profitable in one window
Net of fees and funding, in one of the two test windows.
Strip-best-trade
Still profitable there without its best trade.
Beats random entry
Out-earns 95% of 400 random-timing runs, in that same window.
Enough trades
At least 20 in that window, so two or three winners can't decide it.
Shown, not required: the other window, the drawdown and the other instruments.
Adoptable — paper-trade them first
BTC RSI Back Above 50 in a Daily Uptrend — 4h
While the daily trend points up, buy each time the 4h RSI turns back above 50 and trail a stop two ATRs under the high. It caught the steepest legs of the 2020–21 and early-2023 rallies, and has given money back since.
2020–23
+111.5%
Beat 99% of random entries
2023–26
−15.4%
The other window
Last 12 months
−11.8%
25 trades
By year of entry · 2020 from April, 2026 through September
Three quarters made all of its money: October 2020 to March 2021 (+104.1%) and January to March 2023 (+30.7%). The other 20 quarters with trades lost 38.7 points between them.
BTC RSI Pullback to 50 in a 4h Uptrend — 1h
While the 4h trend points up, buy each time the 1h RSI dips below 50 and trail a stop two ATRs under the high — the classic multi-timeframe pullback. In 2023–26 it made a little where random entries lost a lot.
2023–26
+9.2%
Beat 98% of random entries
2020–23
+5.4%
The other window
Last 12 months
−2.4%
132 trades
By year of entry · 2020 from April, 2026 through September
A knife-edge: at RSI 45 or 55 instead of 50, the same 2023–26 window lost 43–47%.
BTC Squeeze Breakout — 1h
When BTC's last 48 hours have traded inside a 3% range, buy the 1h close that breaks above it and trail a stop two ATRs under the high. The breakouts ran from 2018 to 2020; since 2023 they have mostly failed.
2018–22
+51.8%
Beat 99% of random entries
2022–26
−37.3%
The other window
Last 12 months
−18.4%
52 trades
By year of entry · 2018 from August, 2026 through July
It has lost money every year since 2023, 2026 so far included: 156 trades, −40.4% between them. Adopting it is a bet that BTC's tight ranges break out again the way they did from 2018 to 2020.
BNB Donchian Breakout Above the 200-Day Average — 4h
While BNB's 4h close is above its 200-day average, buy a close above the prior 55 bars' high and trail a stop three ATRs under the high. Most of its money came in early 2021 and early 2024; the filter kept it out of most of 2018 and 2022.
2018–22
+199.9%
Beat 97% of random entries
2022–26
+56.4%
The other window
Last 12 months
+13.9%
10 trades
By year of entry · No trades from August 2018 to January 2019; 2026 through July
Two quarters made most of its money: January to March 2021 (+171.9%) and January to March 2024 (+45.5%). The other 24 quarters with trades netted +38.9 points.
BNB Funding-Squeeze — 15m
Entries #1 and #5's short-squeeze setup on BNB's 15-minute chart: while funding is negative, buy RSI(14) crossing back above 30 and trail a stop two ATRs under the high. Its 2018–22 result came from one quarter.
2018–22
+46.0%
Beat 99% of random entries
2022–26
+4.1%
The other window
Last 12 months
+2.7%
16 trades
By year of entry · 2020 from February, when BNB's funding history starts; 2026 through July
One quarter made all of its money: April to June 2021, +50.5% on 16 trades. The other 23 quarters with trades netted −0.4 points.
Tokyo-Fix Gotobi Give-Back Short — 5m
Short USD/JPY at the 09:55 Tokyo fix on gotobi days and buy it back at noon — the give-back after the morning's dollar buying. Entry #7 trades the morning into the fix; this trades the two hours after it.
2018–26
+3.3%
Beat 98% of random entries
2010–18
+11.8%
The other window
Last 12 months
−0.7%
45 trades
By year of entry · 2026 through September, with no data from late April to late July
It is small — +3.3% over 507 trades in 2018–26, under a pip a trade — and it assumes fills on the minute at the 09:55 fix and at noon, at 0.006% a side.
Crypto RSI(2) Dip-Buy With a Resting Limit Order — 1h
Buy sharp 1h dips above the 200-hour average (RSI(2) under 10) with a limit order 0.5% below the close; sell the first close above the 5-hour average. Bought at market at 0.1% a side, the same signal lost money in five windows of six.
2018–22
+179.5%
Beat 100% of random entries
2022–26
+32.4%
Also beat 95% of random entries
Last 12 months
−2.3%
60 trades
Also cleared the bar in the same window on BTC (+59.0%, 95.5th percentile) and ETH (+112.1%, 99.8th).
By year of entry · 2018 from August, 2026 through July
Most of its money came from 2019 to 2021: +184.9% of +211.9% on BNB. Since 2022 it has made +35.1%, and since January 2025 it has lost 2.5 points.
Records only — the app can't run these yet
VIX Term-Structure Capitulation — 1d
Buy an index ETF when the VIX falls back below VIX3M — spot fear dropping under three-month fear after a panic — and trail a stop two ATRs under the high.
2010–18
+68.3%
Beat 100% of random entries
2018–26
+26.9%
The other window
Last 12 months
+5.2%
6 trades
Also cleared the bar in the same window on QQQ (+91.6%, 100th percentile) and IWM (+83.7%, 100th).
By year of entry · 2026 through July
In 2018–26 it kept making money on all three — +26.9% on SPY, +44.0% on QQQ, +23.7% on IWM — but no more than random entries with the same exit did (72nd–80th percentile).
BTC Open-Interest Deleveraging Exhaustion — 1h
After a day of unusually heavy position closing on Binance — open interest down by a bottom-decile amount for the past four weeks — buy BTC's 1h RSI turn up from 30 and trail a stop two ATRs under the high: a bet that the forced sellers are done.
2024–26
+14.3%
Beat 98% of random entries
2022–24
−21.6%
The other window
Last 12 months
−2.1%
16 trades
By year of entry · 2026 through September
The windows disagree: the same rule lost 21.6% in 2022–24, at the 3.8th percentile of random entries.
BNB Taker-Selling Exhaustion — 4h
After a day of unusually one-sided aggressive selling on BNB's perpetual, buy the 4h RSI turn up from 30 and trail a stop two ATRs under the high.
2023–26
+50.7%
Beat 100% of random entries
2020–23
+26.2%
The other window
Last 12 months
+19.7%
8 trades
By year of entry · 2020 from April, 2026 through September
Much of it is BNB, not the gate: on the same bars the 4h RSI turn alone made +47.6% in 2020–23 and +45.1% in 2023–26.
BNB Perp Discount — 1h
When BNB's perpetual trades at an unusual discount to spot, buy the 1h RSI turn up from 30 and trail a stop two ATRs under the high — the raw premium that the funding rate averages.
2021–24
+49.1%
Beat 98% of random entries
2024–26
−16.9%
The other window
Last 12 months
−16.1%
38 trades
By year of entry · 2021 from May, 2026 through September
One trade of +46.1%, bought on 23 May 2021, is nearly all of its 2021–24 result: without it the window made +3.0%.
BTC Whale Shorts — 4h
When Bitfinex's BTC margin shorts are in the top tenth of the last four weeks, buy BTC's 4h RSI turn up from 30 and trail a stop two ATRs under the high — a bet on the shorts being squeezed.
2024–26
+31.6%
Beat 99% of random entries
2021–24
+9.2%
The other window
Last 12 months
−4.5%
9 trades
By year of entry · 2021 from May, 2026 through September
It failed in the window that contains 2022: +9.2% in 2021–24 only because of its best trade (−13.4% without it), with a 31.7-point drawdown.