FUNDING REGIME BACKTEST: Funding Regime: Every Flip & What BTC Did Next

Median BTC move after a flip, by destination regime
Extreme Long2 flips
7d
+3.3%
50% down
14d
+4.5%
50% down
30d
-0.5%
50% down

Longs deeply crowded — funding is paying shorts at an elevated clip. A stretched-positioning extreme; the cells show how price actually resolved.

Long-Heavy17 flips
7d
+1.1%
47% down
14d
+3.5%
24% down
30d
+9.0%
41% down

Net-long positioning with a positive carry cost — the typical state through an uptrend. The cells show the realized follow-through, wins and losses.

Balanced15 flips
7d
+1.9%
33% down
14d
+1.2%
40% down
30d
-0.2%
53% down

Funding near zero — no crowded side. Usually consolidation or the early part of a move.

“% down” = share of occurrences where BTC was lower at that horizon. Medians shown to resist single-outlier skew. Every resolved flip is counted — this is not a filtered highlight reel. Read the flip counts: a two- or three-occurrence regime is an anecdote, not a tendency.

How this is computed (methodology)

The live Funding Regime Score reads today's crowd positioning across 8 major perps. This is its full history: every time the regime flipped bands, and the realized BTC spot move over the next 7, 14, and 30 days — every occurrence, wins and losses. Observed record, not a prediction.

The funding-regime composite is reconstructed from Binance perpetual funding history using the identical weights, tanh scoring, and band thresholds as the live tool (8 of 8 perps had data over the window), resampled to a daily score.

A flip is a day whose regime band differs from the running regime and holds for at least 2 days (to filter boundary noise), dated at the first day of the new regime. For each flip we take BTC's daily close and its close 7 / 14 / 30 days later; a horizon still in the future is marked pending and excluded from the aggregates.

Window: 2023-01-14 → 2026-08-24 · 34 flips (34 with at least one resolved horizon) · lookback 1,500 days · updated 2026-10-09. All figures computed from public market data — none are estimated or illustrative.

Complete flip log — 34 occurrences, newest first
DateFlipBTC+7d+14d+30d
2026-08-24Long-Heavy→Balanced$78,993-0.5%+0.2%+6.8%
2026-08-22Balanced→Long-Heavy$77,075+1.5%+3.6%+12.4%
2024-12-10Long-Heavy→Balanced$96,593+9.9%+2.1%-4.2%
2024-11-07Balanced→Long-Heavy$75,858+15.1%+29.6%+31.6%
2024-06-08Long-Heavy→Balanced$69,310-4.5%-7.3%-18.2%
2024-05-27Balanced→Long-Heavy$69,436-0.9%+0.2%-12.3%
2024-05-24Long-Heavy→Balanced$68,550-1.5%+1.2%-7.8%
2024-05-21Balanced→Long-Heavy$70,148-2.5%+0.6%-7.5%
2024-04-13Long-Heavy→Balanced$63,925+1.6%-0.7%-1.5%
2024-04-08Balanced→Long-Heavy$71,620-11.5%-6.7%-14.6%
2024-04-05Long-Heavy→Balanced$67,821-1.0%-5.9%-5.6%
2024-04-02Extreme Long→Long-Heavy$65,464+5.6%-2.6%-9.8%
2024-03-31Long-Heavy→Extreme Long$71,280-2.7%-7.9%-14.9%
2024-03-15Extreme Long→Long-Heavy$69,500-8.2%+0.5%-5.5%
2024-02-29Long-Heavy→Extreme Long$61,131+9.3%+16.8%+13.8%
2024-02-15Balanced→Long-Heavy$51,880-1.1%+17.8%+25.9%
2024-01-04Long-Heavy→Balanced$44,151+5.0%-6.4%-2.6%
2023-12-20Balanced→Long-Heavy$43,669-0.6%-1.9%-4.6%
2023-12-13Long-Heavy→Balanced$42,869+1.9%+1.3%-0.2%
2023-11-29Balanced→Long-Heavy$37,855+15.6%+13.3%+11.1%
2023-11-22Long-Heavy→Balanced$37,408+1.2%+17.0%+17.5%
2023-11-20Balanced→Long-Heavy$37,449-0.6%+12.1%+16.6%
2023-11-17Long-Heavy→Balanced$36,614+3.0%+5.7%+13.0%
2023-11-02Balanced→Long-Heavy$34,942+5.0%+3.5%+12.9%
2023-10-28Long-Heavy→Balanced$34,081+2.9%+9.0%+9.3%
2023-10-24Balanced→Long-Heavy$33,923+2.1%+4.4%+9.9%
2023-04-18Long-Heavy→Balanced$30,380-6.8%-5.6%-11.7%
2023-04-16Balanced→Long-Heavy$30,305-9.0%-3.5%-10.8%
2023-03-16Long-Heavy→Balanced$24,999+13.2%+12.1%+21.2%
2023-03-14Balanced→Long-Heavy$24,670+13.9%+10.5%+23.1%
2023-01-25Long-Heavy→Balanced$23,061+2.9%-0.4%+0.5%
2023-01-21Balanced→Long-Heavy$22,784+1.1%+2.4%+9.0%
2023-01-17Long-Heavy→Balanced$21,135+7.1%+9.4%+11.3%
2023-01-14Balanced→Long-Heavy$20,955+8.7%+9.9%+3.9%

This is a record of what has already happened, presented in full — not a signal to act on and not a promise about the future. Funding positioning is context, one input among many. The live regime updates on the Funding Regime Score, and dated calls we publish resolve on the Track Record. Information, not financial advice.

XTelegramRedditLinkedIn

Frequently asked questions

What is the Funding Regime Backtest?
The Funding Regime Backtest is a historical track record of every time Liquid State's 8-perpetual crypto funding-regime composite flipped from one positioning band to another — Extreme Short, Short-Heavy, Balanced, Long-Heavy, or Extreme Long — paired with the realized BTC spot price move over the following 7, 14, and 30 days. It records every occurrence, wins and losses alike, rather than a curated highlight reel, so it is a report of what has already happened and not a signal to act on.
How is a regime "flip" defined in the backtest?
A flip is dated to the first day the composite funding-regime score crosses into a new band, and it only counts once the new band holds for at least two consecutive days — a persistence filter meant to exclude single-day boundary noise from the record. The underlying composite is reconstructed using the same eight perpetual futures markets, weighting, and tanh-based scoring as the live Funding Regime Score, resampled to a daily reading across available Binance funding history.
What do the median BTC move figures after a flip mean?
For each destination regime, the backtest reports the median BTC spot price move at 7, 14, and 30 days after every historical flip into that band, alongside the share of occurrences where BTC was lower at that horizon. Medians are used to resist distortion from a single outlier occurrence, and every resolved flip in the window is counted rather than a filtered selection — a regime with only two or three flips is an anecdote, not a statistically meaningful tendency, and the flip count is shown alongside every figure for that reason.
Where does the funding-regime backtest data come from?
Liquid State reconstructs the backtest from Binance's public perpetual futures funding-rate API across the same eight major perpetual markets used by the live Funding Regime Score, with no account or API key required to view it. The backtest is recomputed periodically from that public history and republished free for educational and market-monitoring purposes, not as investment advice.
THE BRIEF · FREE

Get the funding regime read + when it flips

Free market reads — plus the Three-Lens Framework the desk runs on every chart. No spam. Unsubscribe anytime.

🔒 No spam. One-click unsubscribe. Free framework on signup.