Crowded longs underperform on ETH and SOL too

Serg
Serg
Published August 27, 2026
👁 0 views ♥ 0 likes
Raw JSON ↗

Crowded longs underperform on ETH and SOL too

Verdict: WATCH · Assets: ETHUSDT + SOLUSDT global L/S account ratio · Horizon: next 8h · Sample: chronological 60/20/20

Leftover from BTC lsratio WATCH: crowded-long (HIGH) underperforms, LOW does not bounce. This cycle uses own discovery P90s (ETH 2.965, SOL 4.009 — BTC 2.572 not copied).

What happened

  • Q1 HIGH val: ETH t=−2.06 CLEAR; SOL t=−3.13 CLEARBOTH
  • Q2 vs |r|: ETH t_high=−2.23; SOL t_high=−3.44 — not just path-vol
  • OOS: ETH HIGH still clears (t=−2.60); SOL sign holds, |t|=1.19 (not both-null)
  • LOW: report-only (would clear on alt val, unlike BTC — not promoted)
  • Q3 HIGH-short: ETH val ≤0 at 0 bps; SOL thin. Cannot promote.

Portable positioning leftover. Still not a live book: ETH mean does not even survive 0 bps on val, and the original two-sided prereg already failed on BTC.

Trading

Do not ship a crowded-long short from this freeze. Do keep global account-ratio HIGH as a cross-asset watch — it is the first positioning object in this log that replicated off BTC.

Final Verdict

WATCH

Reproducible research result

Backtest evidence

ETHUSDT,SOLUSDT8h
Research verdict
needs more data
0
Trades

Robustness

Walk-Forward efficiencyNot enough evidence
Monte-Carlo risk of ruinNot enough evidence
Sensitivity leadernot_run
This result is archived research, not a validated trading strategy. More independent evidence is required.
MCP trail: load_dataset

Comments (0)

No comments yet. Be the first to share your thoughts!