COT Jun 23 – Treasury Curve Splits at Extremes

Nine markets reached positioning extremes this week — the 95th percentile threshold or beyond — but the real story is in the Treasury complex, where large speculators have built the most polarised curve exposure in two years.

For someone new to Commitment of Traders data: these reports show how hedge funds and other large speculators position in futures markets each week. When positioning reaches the 95th percentile or higher, it means the current net long or short exposure is larger than 95 out of the past 100 weeks — a signal of crowding, not necessarily of imminent price direction.

This analysis uses CFTC Commitment of Traders data (Legacy Report format) with dual-window percentile ranking: 2-year lookback (104 weeks) as the primary reference, and 10-year (520 weeks) for structural context. Data sourced from the CFTC Commitment of Traders Report. Full Methodology → | Download this week's data (JSON) | View Interactive Dashboard →

Top 10 Most Extreme Positions — Jun 23, 2026

Rank	Market	Net Position	2Y Pct	10Y Pct	Z-Score
1	5Y Treasury	-1,301,269	99th	46th	1.79
2	Bitcoin	+3,524	99th	99th	2.51
3	British Pound	-105,719	1st	1st	-1.98
4	Lean Hogs	-57,209	1st	1st	-2.53
5	30Y Treasury	-176,043	2nd	8th	-2.29
6	New Zealand Dollar	-54,844	2nd	1st	-1.39
7	Nasdaq 100	-16,272	3rd	1st	-2.14
8	Japanese Yen	-146,104	4th	4th	-1.91
9	Cotton	+83,558	95th	81st	2.52
10	Copper	+71,620	94th	97th	1.83

Treasury Curve Split — Five-Year at Record Shorts, Long Bond Near Exhaustion

The Treasury complex is showing the most divergent positioning I've seen in this dataset. 5Y Treasury net shorts hit -1,301,269 contracts — the 99th percentile over two years, though only the 46th over 10 years, meaning this level of short exposure has precedent in earlier rate cycles. The build was sustained: -250k over the past four weeks, -331k over 13 weeks. Non-commercial shorts represent 20.9% of open interest, which is concentrated but not wildly extreme.

Meanwhile, 30Y Treasury sits at -176,043 contracts (2nd percentile over two years, 8th over 10 years) — still in persistent extreme territory after eight consecutive weeks below the 10th percentile. That's the eighth week running at the same extreme, so it's not news this week, but the contrast with 5Y is.

The 2Y and 10Y notes sit in the middle: 32nd and 44th percentile respectively. So the curve split is specifically at the 5Y belly (heavy short crowding) and 30Y tail (speculative shorts largely washed out). This positioning pattern is consistent with a view that the Fed's most sensitive point on the curve (the 5Y) gets sold while the long end finds some technical support from short-covering exhaustion.

I don't have a strong take on what this means directionally — COT data doesn't predict price. But when one part of the curve shows 99th percentile short crowding and another shows 2nd percentile (near-exhaustion), that divergence has historically sometimes lined up with periods where curve steepeners or flatteners see heightened volatility as one side unwinds faster than the other.

FX Extremes — Sterling and Kiwi at Multi-Year Lows

British Pound net shorts reached -105,719 contracts (1st percentile over both two years and 10 years), with speculative shorts representing 35.6% of open interest. That's heavily concentrated. The build was sustained: -211k over four weeks, -195k over 13 weeks. Commercials hold the other side at +123,431 contracts (41.5% of OI), which is normal — this is classic hedger-versus-speculator opposition, not an alignment configuration.

New Zealand Dollar net shorts sit at -54,844 (2nd percentile over two years, 1st over 10 years), with crowding at 53.1% of open interest. Again, sustained build: similar magnitude over four and 13 weeks, suggesting structural conviction rather than a single-week event-driven spike.

Both of these currency shorts are at or near the lowest speculative exposure in the full 10-year dataset. Within this sample, readings at the 1st–2nd percentile have sometimes preceded mean reversion, but low positioning can persist for months if the macro narrative remains one-sided. The key variable is whether price stalls while positioning stays extreme — that's when tactical unwinding risk tends to show up.

Bitcoin — Still at 99th Percentile, Quietly Holding

Bitcoin net longs sit at +3,524 contracts (99th percentile over both two years and 10 years), representing 17.1% of open interest. The position has been steady for weeks: the four-week and 13-week averages are nearly identical to the current level, meaning this isn't a fresh spike — it's a sustained hold.

Commercial shorts sit at -3,253 contracts (15.8% of OI on the other side). This is standard two-sided positioning, not an alignment signal. The lack of movement week-to-week suggests speculative longs are comfortable here, which can mean either strong conviction or complacency depending on how price behaves in the next few sessions.

Lean Hogs and Agriculture — A Pair of 1st Percentile Shorts

Lean Hogs net shorts reached -57,209 contracts (1st percentile over both two years and 10 years, Z-score of -2.53). Non-commercial shorts represent 19.7% of open interest. The build was sustained: roughly similar drawdowns over four and 13 weeks. This is the most extreme net short positioning in the full 10-year dataset for this contract.

Cocoa sits at -19,789 (9th percentile over two years, 4th over 10 years), which is low but not quite at the 5th percentile threshold. Coffee and Cotton remain at persistent extremes (8+ weeks) but aren't new stories this week.

What's notable in the ag complex is the absence of fresh extremes elsewhere. Grains are mostly mid-range. Soybeans, corn, wheat — all sitting between the 50th and 75th percentile. The positioning drama this week is concentrated in the meats and softs, not the row crops.

Equity Index — S&P 500 Alignment Signal Activated

S&P 500 shows a Both Short alignment this week: large speculators hold -32,156 contracts (-1.6% of OI) while commercials hold -88,575 contracts (-4.4% of OI). That means small speculators — retail public — are holding the entire net long position as the sole counterparty to both groups.

The COT percentile for S&P 500 is at the 79th, so this isn't an extreme positioning level on its own, but the alignment structure is worth noting. Historically, when both large specs and commercials have positioned on the same side, the third group (small specs) has sometimes experienced elevated volatility in their positioning, particularly if price fails to confirm their view. This is a structural observation, not a directional forecast.

Australian Dollar also shows a Both Short alignment (large specs -13,012, commercials -863), though the COT percentile is only at the 64th, so it's less amplified by extremity.

What Different Scenarios Might Mean — 5Y Treasury, British Pound, Lean Hogs

5Y Treasury (99th percentile net short, sustained build):

• If 5Y yields rise further while net shorts remain above the 95th percentile, then this is consistent with a positioning trend that has historically sometimes persisted for multiple weeks when macro narratives remain one-sided — though crowding at 20.9% of OI means the trade is concentrated enough to unwind quickly if yields stall.

• If 5Y yields flatten or fall while net shorts stay at this extreme, then this may indicate a crowded consensus short vulnerable to tactical covering — particularly given the sustained nature of the build (not a single-week spike), which suggests conviction that could reverse sharply if the narrative shifts.

• If net shorts begin to decline week-over-week without a sharp move in yields, then this is consistent with slow repositioning rather than a forced unwind — watch the speed: a 100k+ weekly reduction would signal urgent exit, while 20-30k would suggest orderly profit-taking.

British Pound (1st percentile net short, 35.6% of OI crowding):

• If GBP continues lower while net shorts stay at the 1st percentile, then this is consistent with a trend exhaustion setup where speculative shorts are already maximally positioned and further downside depends entirely on fresh sellers — though "exhaustion" can last longer than intuition suggests.

• If GBP stabilises or rallies while net shorts remain at this extreme, then the concentrated crowding (35.6% of OI) suggests a sharp short-covering rally is possible — similar configurations in past cycles have sometimes preceded multi-week unwinds, though timing cannot be determined from positioning data alone.

• If net shorts begin covering (moving up from the 1st percentile) but price remains weak, then this may indicate repositioning into options or cash, rather than a change in directional view — COT only captures futures, so it's an incomplete picture.

Lean Hogs (1st percentile net short, Z-score -2.53):

• If hog prices decline further while net shorts stay at the 1st percentile, then speculative positioning is as short as it has been in 10 years, meaning downside is increasingly dependent on commercial hedgers or physical market conditions rather than speculative flow.

• If prices stabilise or bounce while net shorts remain at this extreme, then the 19.7% of OI crowding is meaningful but not wildly concentrated — a technical rally could see some covering, but it would likely need a catalyst (weather, disease, export demand) rather than positioning alone.

• If net shorts start to cover quickly (50%+ reduction within two weeks), then this would be consistent with a fast unwind typical of agricultural contracts when a weather event or supply shock changes the fundamental narrative — lean hogs have a history of sharp reversals from positioning extremes when the physical market shifts.

What I'm Watching Next Week

The Treasury curve split is the positioning story this week. If 5Y shorts start to unwind while 30Y stays pinned near lows, that would suggest tactical profit-taking at the belly while the long end remains structurally under-owned. If both move in parallel, it's a simpler Fed repricing story.

Sterling and Kiwi are at multi-year lows in speculative positioning. The build has been sustained, not spiked, which means these are conviction shorts, not event-driven trades. They can persist, but they're also large enough to move fast if the macro narrative shifts.

S&P 500 alignment (Both Short) isn't at an extreme percentile, but it's a configuration worth tracking. If small specs continue holding the full net long into next week while price stalls, that's a setup that has historically sometimes coincided with retail positioning volatility.

This analysis is for educational purposes only and does not constitute financial advice.

Next week's report covers the period ending 2026-06-30.

Explore More: Interactive Dashboard | Methodology | Download Data | Previous Analysis