· Analytic
LH CFTC Crowded Short Amplification
CFTC managed-money positioning decile-rank crowded extremes on CME Lean Hogs — the regime-specialist read designed to catch short-squeeze setups and crowded-long liquidation turns.
At a glance
Current reading
+0.762
Crowded Short bottom-decile, 103 trading days running since June 2026
Signal state
Extreme Crowded Short Regime
phase +1, managed-money NET in bottom 10% of 5-year window
Sample
Weekly CFTC, daily ffill, 36-year span
CFTC_DCOT_WEEKLY, LH market, 2006+ (5y warmup)
Historical IC
+0.0443
Spearman vs LH fwd60d, n = 8,233 (significant for a regime signal)
Baseline LH fwd60d
+1.76%
Unconditional mean
Crowded Short fwd60d
+4.33% mean
+2.57 pp above baseline, 52.2% hit rate
Days since last event
102
Last event: Jun 05, 2026
Nearest threshold
-0.56
To cross into Crowded Short
Typical run length
19 days
Based on 25 historical Extreme Crowded Short Regime runs; 25-75 pct range: 10-24 days
As of Oct 30, 2026. Updated nightly.
OBJECTIVE
Managed-money futures positioning in the CME Lean Hogs market tends to cluster at extremes when a cycle inflection is near. The CFTC Disaggregated Commitments of Traders report, published every Tuesday for Friday-close positions, surfaces that clustering roughly a week after the fact. This signal reads the weekly managed-money NET position (longs minus shorts), ranks it against the trailing five-year window on a decile scale, and fires a specialist regime when positioning lands in the bottom decile (crowded short, declared bullish Lean Hogs) or the top decile (crowded long, declared bearish). In February 2019 the signal pinned bottom-decile crowded short at $53; lean hog futures rallied 74 percent over the next sixty days as the ASF-driven China import narrative forced a historic short squeeze. In July 2017 the signal pinned top-decile crowded long at $91.5; lean hog futures crashed 39 percent over the next sixty days as fund longs capitulated into a seasonal supply build. The memorable twist: this is a regime-specialist, not a continuous edge. The middle 80 percent of positioning deciles delivers near baseline (plus 1.94 percent fwd 60-day, baseline plus 1.76 percent). Only the extreme deciles carry the asymmetric payoff — crowded short delivers plus 2.57 pp above baseline, crowded long delivers minus 2.66 pp. Read on for the two decade-plus episode library, the current 103-day crowded-short firing, and how to pair with the continuous commercial-positioning signal.
LIVE CHART
LH CFTC Crowded Short Amplification
Composite reads +0.762 as of October 30, 2026, in the bottom-decile crowded-short bucket, 103 trading days running since June 5, 2026. Managed-money NET position has stayed below the bottom-decile threshold through the entire window, consistent with the heaviest speculative short positioning on record outside the 2019 ASF squeeze setup. Historical state-conditional edge for Crowded Short is +2.57 pp above the +1.76% baseline. Pair with the continuous commercial-positioning read for the professional-side confirmation.
COMPOSITE
+0.762 / tanh
STATE
Extreme Crowded Short Regime
PHASE
+1
FWD 60d
+9.01% (baseline +1.76%)
SAMPLE N
8,293 days
STATE-CONDITIONAL EDGE
Forward 60-day returns by state
FORWARD 60-DAY RETURNS BY STATE
Mean forward-60-day benchmark return per composite state. Baseline (horizontal line) is the unconditional sample mean. Bars above the line are the constructive bands; bars below are the bearish bands.
TODAY’S PLAYBOOK READING
STATE
Extreme Crowded Short Regime
STATE AND HISTORY
Crowded Short extreme-positive firing, 103 trading days running since June 5, 2026. State-conditional mean fwd60d LH is +4.33%, +2.57 pp above the +1.76% baseline. Historic analog: February 2019 bottom-decile read preceded a 74 percent rally.
WHY REGIME-SPECIALIST, NOT CONTINUOUS
The middle 80 percent of positioning deciles delivers near baseline (+1.94% mean fwd60d, +0.18 pp edge). Only the extreme tails carry asymmetric payoff. Crowded short: +2.57 pp. Crowded long: -2.66 pp. FAVAR Diebold-Yilmaz study confirms the spillover network shifts in regime subsamples (n = 253).
HOW TO POSITION
Pair with the continuous commercial-positioning read (the commercial-positioning signal). This regime-specialist catches tails; the continuous version reads the full positioning spectrum. Weekly CFTC cadence — watch Tuesday 3:30 PM ET for the next position update.
WHAT THE CHART SHOWS
The composite sits at +0.762 in the bottom-decile crowded-short bucket, 103 trading days running since June 5, 2026. The blue signal line has pinned at the saturated-positive level through the entire Q3 2026 window. CME lean hogs (orange) have declined from the $100 Q2 high to the $56-$62 range through the firing — consistent with the 2026-08 paradox case where positioning extremes at extended-cycle price tops fail to deliver the historical squeeze pattern. The current firing may be a late-cycle read rather than a pre-squeeze setup.
TOP INPUTS FIRING
- Primary composite +0.762 State Crowded Short bottom-decile; 103 trading days running since June 5, 2026.
- Managed-money NET bottom decile Decile-rank of CFTC managed-money NET position vs trailing 260-week window. Current reading in bottom 10% of 5-year distribution.
- State-conditional edge: +2.57 pp mean Crowded Short historical fwd60d LH mean +4.33%, baseline +1.76%. Hit rate 52.2%. 2019 ASF analog delivered +74%; 2018 analog delivered +58%. 2026-08 paradox at -40% warns against positioning extremes at extended-cycle price tops.
THE TAKEAWAY
What this means for you today
THE POSITIONING CALL
Composite +0.762, in the bottom-decile crowded-short bucket. Managed-money NET has held in the bottom 10% of the 5-year window continuously since June 5, 2026 — the longest crowded-short episode since the 2024 summer window.
THE EMPIRICAL CONTEXT
Crowded Short historically delivers mean fwd 60-day LH of +4.33%, +2.57 pp above the +1.76% baseline. Hit rate 52.2%. The 2019 ASF-era precedent delivered +74%; the 2018 analog delivered +60%. Not every crowded-short episode delivers — 2026-08 was the worst in-sample paradox at -40%.
THE HONEST CAVEAT
Regime-specialist only. Middle 80 percent of positioning readings delivers no edge. Weekly CFTC cadence means intra-week moves are not captured; the signal updates once per week with a one-day lag post-Tuesday 3:30 PM release.
THE REGIME MAP
The three states, forward returns and sample share
Computed from 8,233 trading days with observable LH fwd60d. Baseline LH fwd60d (unconditional): plus 1.76 percent mean. Decile-rank signal fires only in the extreme tails — the middle 80 percent is a no-signal regime.
| State | Phase | Share | Fwd 60d | HAC t | Edge vs baseline | Last firings |
|---|---|---|---|---|---|---|
| Crowded Short TODAY | +1 | 7.9% (654 d) | +9.01% | +7.26pp | Jun 2026, Jun 2024, Jan 2024 | |
| Balanced | 0 | 79.6% (6601 d) | +3.63% | +1.88pp | Mar 2026, Oct 2025, Mar 2025 | |
| Crowded Long | -1 | 12.5% (1038 d) | -1.83% | -3.59pp | Jan 2026, May 2025, Oct 2024 |
The historical pattern
Signature episodes in the 36-year sample
What this measures
The signal pulls the CFTC Disaggregated Commitments of Traders weekly report (CFTC_DCOT_WEEKLY, market_name = LEAN HOGS, published Tuesdays for Friday-close positions), computes managed-money NET as longs minus shorts, and decile-ranks the result against a 260-week (5-year) trailing window. The signal is +1 if NET is in the bottom decile (crowded short, declared bullish LH), -1 if top decile (crowded long, declared bearish), and 0 for the middle 80 percent. Daily forward-fill between weekly releases, output bounded via tanh to plus or minus 0.76. State machine fires only at the two extremes — the middle 80 percent is a no-signal regime.
Why it matters
Why decile-rank positioning and not raw net position
Raw managed-money NET on CME Lean Hogs ranges from -40k contracts (deep short) to +90k contracts (deep long) across the historical sample, with structural drift as the fund base has grown. A raw-NET read confuses level with regime. The decile-rank approach normalizes against the trailing five-year window: the bottom decile is crowded short relative to recent history, not relative to the 1990s. This is the right scale for a regime-specialist signal — it fires when positioning is at its recent extreme, independent of multi-decade drift.
Why only the tails deliver edge
The middle 80 percent of positioning deciles delivers plus 1.94 percent mean fwd 60-day — effectively baseline (unconditional baseline is plus 1.76 percent). This signal is explicitly not a continuous read. It is a regime-specialist designed to catch positioning extremes. The S10 FAVAR Diebold-Yilmaz spillover study on this engine (n = 253 crowded-short days) found that the spillover network between commodity-positioning regimes is asymmetric: the crowded-short subsample NET shifts to plus 0.1621 vs the balanced baseline of plus 0.0728. The regime matters; the level does not.
How to use this signal today
Treat as the extreme-positioning regime overlay on the Lean Hogs stack. Pair with the continuous commercial-positioning signal for the professional-side confirmation — when both the specialist and the commercial-positioning reads agree, the signal is strongest. Current deep Crowded Short firing (103 days) is historically constructive (+2.57 pp mean edge) but 2026-08 paradox case warns: positioning extremes at extended price tops can fail. Watch Tuesday 3:30 PM ET for the next CFTC release.
GO DEEPER
Want the full depth?
The research brief walks through the CFTC Disaggregated Commitments of Traders decile-rank methodology, documents the FAVAR Diebold-Yilmaz spillover amplification study (n = 253) that validates the regime-specialist design, and compares to the continuous commercial-positioning signal.
Related signals
Not investment advice · The analysis on this page is informational and educational. Signal readings, historical case figures, and empirical statistics are derived from publicly available data and are provided without warranty. Hypothetical performance results have inherent limitations, including the benefit of hindsight. Readers who intend to trade should consult a licensed investment professional and see the research brief for the full disclaimer.
Last refreshed Oct 31, 2026 03:38 PM ET