Insider Selling Pressure Methodology
By Rishabh Narang · Last updated April 16, 2026
How the Insider Selling Pressure signal works, and why it works.
The Problem
Time-based unlock data is often incomplete, stale, or shaped by the same teams that benefit from ambiguity. The signal therefore avoids trusting stated unlock dates and sizes. Instead it watches for the observable footprint of the trade: shortable tokens with persistent cross-venue negative funding and confirming derivatives pressure. Three types of actors commonly drive this pattern:
- Hedge funds buy locked tokens at 80%+ discount from insiders, then short perpetual futures to hedge. On a typical 6-month lockup, this generates approximately 40% annualized return on capital.
- Market makers operating on loan models short the coin to manage inventory risk or execute OTC flow, creating persistent downward pressure through derivatives.
- Team members quietly short their own token through perps, hedging their vested positions while publicly maintaining alignment with holders.
The outcome is a total perversion of incentives: mechanisms created to align teams with tokens end up having the opposite effect. Once positions are hedged, insiders actually prefer the price going down, because if it goes up, their short blows up.
This is confirmed by industry research: Keyrock studied 16,000+ unlock events and found 90% result in price declines. Presto Research calls the OTC-discount-plus-perp-hedge “a very profitable market neutral strategy.” Neutrl raised $5M to literally productize this trade. STIX reports $10B in locked token transactions in 2024 alone.
The result: coins with large locked supplies face persistent, sophisticated selling pressure that retail often cannot see from price charts alone. Because the schedules themselves can be adversarial, Sharpe treats derivatives behavior as the primary evidence and wallet/supply context as supporting evidence.
The Signal
Persistent negative funding across multiple exchanges is the core signal. Normal traders do not systematically maintain short exposure for days or weeks. Funds, market makers, or teams hedging locked token exposure often do.
The model requires shortability and negative funding, then validates it with exchange breadth, open interest, basis, true taker-flow/CVD, price path, and sell-side flow. Top-holder concentration, dev/insider/bundler/sniper supply, and holder-quality deterioration can raise conviction, but they cannot create a signal by themselves.
Because sophisticated teams split inventory across many wallets, the model still watches for camouflage: holder-count inflation, low visible concentration paired with old-wallet flow, wallet-age clustering, high churn through thin liquidity, and concentration falling while hedge pressure rises.
Borrow and OTC details are useful when available, but they are not required inputs. Public borrow data is inconsistent across venues and private OTC hedges are intentionally opaque, so the production signal prioritizes continuously observable perp behavior.
How We Score
Each flagged coin receives a composite score from 0 to 10 based on multiple signal families. Coins must have active shortability and sufficient recent derivatives coverage before they can enter the leaderboard:
Negative cross-venue funding, funding persistence, and exchange breadth. This is the core footprint of locked-token or market-maker hedging.
Open interest, basis, taker-flow/CVD, short crowding, and price path. These separate systematic short building from noisy funding prints.
Top-holder, insider, dev, sniper, bundler, holder-count, old-wallet activity, and fragmentation-camouflage context. These raise conviction but do not independently trigger a signal.
We publish the signal families, validation process, and backtest results, but keep exact weights and thresholds private so teams cannot easily tune behavior to avoid detection.
Backtest Results
A 12-month walk-forward replay across 394 coins confirms the signal works. 83.9% of high-conviction flagged coins declined within 30 days — 7.7 percentage points higher than a random coin from the same universe on the same entry dates. At 60 days the median drop deepens to -22.2%. An equal-weight short basket of the top 10 flagged coins, rebalanced weekly, returned +158% with a 35% max drawdown over the same window. Every coin was re-scored end-of-day using only data available at that moment; nothing cherry-picked, nothing survivorship-filtered.
Key result
of flagged coins declined within 30 days
median price move 60 days after the signal · 83.3% hit rate
equal-weight portfolio return over 12 months, max drawdown −35.5%
Walk-forward replay across 394 coins from April 16, 2025 to February 15, 2026. Every coin is re-scored end-of-day using only data available at that moment — no lookahead. Refreshed monthly; last computed April 16, 2026.
Headline: High-conviction signals (score ≥ 7)
Portfolio simulation — equal-weight basket, how a user would trade it
Every 7 days, capital is split evenly across the top 10 highest-scored flagged coins (score ≥ 4) and held short until the next rebalance. Positions overlap and compound. This is the most realistic “could I trade this?” view — it caps positions at what real capital allows and handles concurrent signals naturally.
Top-10 equal-weight short basket, rebalanced every 7 days
Final: $257.61 (+157.6%)Max DD: −35.5%
Starts at $100. At every rebalance (every 7 days), the portfolio is evenly divided across the 10 highest-scored flagged coins (score ≥ 4). If fewer than 10 coins are flagged, the basket is smaller. Portfolio compounds period-over-period. No fees or slippage. Illustrative — not a recommendation.
Signal precision — every crossing as one independent trade
Each step in the curve below is the P&L of a single $100 notional short held for 30 days, entered the day after the score crossed the threshold. Non-compounding — this isolates the edge of the signal itself, independent of any capital-allocation strategy.
Equity curve — short $100 notional per signal, 30-day horizon
Final: $553.66 (+453.7%)
Simulates a naive strategy: on every signal event, enter a $100 notional short at the next-day close and exit 30 days later. No fees or slippage. Illustrative — not a recommendation.
Case studies — the calls that played out
The five largest drops among flagged coins and the five times the signal went against us. Both are shown in full — nothing cherry-picked.
Highest-conviction wins (30-day horizon)
| Coin | Entry date | Score | Avg funding (72h) | 30d return |
|---|---|---|---|---|
| AlturaALU | Aug 29, 2025 | 7.0 | -0.049% | -55.8% |
| DashDASH | Nov 6, 2025 | 7.1 | -0.112% | -55.6% |
| AptosAPT | Oct 6, 2025 | 5.3 | -0.039% | -53.1% |
| Act I The AI ProphecyACT | Jan 1, 2026 | 4.7 | -0.031% | -46.4% |
| AnimecoinANIME | Jun 11, 2025 | 9.0 | -0.099% | -45.3% |
Calls that went against us
| Coin | Entry date | Score | Avg funding (72h) | 30d return |
|---|---|---|---|---|
| StellarXLM | Jun 22, 2025 | 4.1 | -0.008% | +101.4% |
| TezosXTZ | Jun 23, 2025 | 4.1 | -0.028% | +96.2% |
| Kyber Network CrystalKNC | Jul 12, 2025 | 5.8 | -0.272% | +37.9% |
| Aerodrome FinanceAERO | May 30, 2025 | 8.1 | -0.056% | +34.1% |
| TezosXTZ | Dec 26, 2025 | 5.9 | -0.030% | +31.9% |
Consistency across horizons and score buckets
Hit rate and price move measured at 7, 14, 30, and 60 days, broken down by bucket. The highlighted row is the strongest combination of hit rate and sample size.
| Horizon | Bucket | N | Hit rate | vs baseline | Median price move | Return edge |
|---|---|---|---|---|---|---|
| 7d | Strong (≥7) | 31 | 64.5% | +0.8 pp | -3.8% | +0.9% |
| 7d | Moderate (≥4) | 69 | 59.4% | +0.9 pp | -2.1% | -0.6% |
| 7d | All flagged | 100 | 61.0% | +0.9 pp | -2.7% | +0.0% |
| 14d | Strong (≥7) | 31 | 67.7% | +0.6 pp | -7.9% | -1.0% |
| 14d | Moderate (≥4) | 69 | 65.2% | −1.1 pp | -5.1% | +1.8% |
| 14d | All flagged | 100 | 66.0% | −0.6 pp | -6.0% | +0.9% |
| 30d | Strong (≥7) | 31 | 83.9% | +7.7 pp | -13.2% | -0.9% |
| 30d | Moderate (≥4) | 69 | 71.0% | −1.8 pp | -9.1% | +2.5% |
| 30dbest | All flagged | 100 | 75.0% | +1.2 pp | -10.8% | +0.9% |
| 60d | Strong (≥7) | 30 | 83.3% | +6.8 pp | -22.2% | -3.8% |
| 60d | Moderate (≥4) | 63 | 74.6% | −3.2 pp | -16.7% | +8.8% |
| 60d | All flagged | 93 | 77.4% | +0.0 pp | -19.0% | +3.4% |
How we avoided biasing the results
- Walk-forward replay: at each day D, only data up to D is used to generate signals
- 14-day warm-up: a coin must be below threshold for 14 consecutive days before its first crossing counts (eliminates 'discovery' bias at window start)
- 30-day lockout: after any signal, that coin is suppressed for 30d (prevents overlapping trades)
- Universe baseline uses the same coins at the same entry dates (controls for market beta)
- Includes all coins that traded during the window — delisted coins are NOT dropped (avoids survivorship bias)
Price source: correlation_prices (daily market close). Entry rule: close[signal_day + 1] (realistic next-day execution). Portfolio sim excludes transaction costs — real returns will be lower after fees, borrow, and slippage.
Methodology Questions
What causes persistent negative funding rates?
Persistent negative funding can come from hedge funds hedging locked tokens, market makers managing inventory, team-related hedging, or active short pressure. It is the core observable footprint, but still not proof by itself.
How is the insider selling pressure score calculated?
Each coin receives a composite score from 0 to 10 based on negative cross-venue funding, funding persistence, open interest, basis, CVD, sell-side flow, and supporting wallet/supply context. Exact weighting is proprietary to reduce gaming.
What exchanges does the signal monitor?
The signal monitors perpetual futures contracts across 13 exchanges including Binance, Bybit, OKX, Deribit, and Hyperliquid. Multi-exchange coverage ensures the signal captures systematic positioning rather than exchange-specific noise.
How was the signal backtested?
We replayed the signal walk-forward over 12 months across 394 coins using daily market prices. Strong signals (score >= 7) hit 83.9% at 30 days with median -13.2% — a 7.7 percentage-point edge over a random coin from the same universe on the same entry dates. An equal-weight short basket of the top 10 flagged coins rebalanced weekly would have returned +158% with a 35% max drawdown. A 14-day warm-up eliminates 'discovery' bias at the window start.
Is this signal financial advice?
No. This is an experimental signal derived from public market, derivatives, and on-chain context. It does not constitute financial advice or an accusation of illegal activity. Past backtest performance does not guarantee future results.
Disclaimer
This is an experimental signal derived from public market, derivatives, and on-chain context. It does not constitute financial advice or an accusation of illegal activity. Past backtest performance does not guarantee future results. Always do your own research.