1. Bottom Line & Directional Bias
Call: SHORT Natural Gas (NG=F). Invalidation: a daily settle above 3.05 (R2).
Three reasons underpin the call. First, price action is weak: the 2026-10-01 settle was 2.97, down 1.95% on the day and 11.96% over five sessions, and the market sits at the 13.2% position of its 20-day 2.9–3.4 range — a breakdown profile, not a base. Second, the curve offers no support: the M1–M2 spread is -0.345 (-10.43%), a contango that pays shorts to roll and confirms there is no prompt scarcity. Third, positioning is not yet a contrarian buy: managed-money net was -65,547 contracts as of 2026-09-22, but the three-year crowding percentile is only 27%, meaning the short base can still grow.
The main risk to the call is seasonal. The next 20 sessions have averaged +5.76% over the last 15 years (median +1.71%, up 9 of 15), so this is a tactical short, not a structural one. A settle above 3.05 would break the 20-day range top and invalidate the view; a settle below 2.9 (S2) would open the 52-week low at 2.48.
2. Price Action & Technical Analysis
The prior session settle (2026-10-01) was 2.97, down 1.95% on the day. Over five sessions the contract has lost 11.96%, and over 20 sessions it is down 3.98%. The 20-day range is 2.9–3.4, and at 2.97 the market sits at the 13.2% position — near the bottom of the range but not yet broken. The 52-week range is 2.48–7.83, so the market is in the lower third of its annual envelope.
In early Asian trade on 2026-10-02 (07:00), the last print was 2.95, down 0.64% versus the settle, with the session high and low both at 2.95 — a narrow, illiquid Asian range that offers little information beyond confirming the soft tone. All levels below are from the settle-based snapshot.
ATR14 is 0.122, or 4.12% of price on a full daily range basis. That is a wide expected daily range and means stops must be placed with room; a stop inside roughly one ATR is noise. RV20 is 42.7%, high in absolute terms but below the 51.69 reading on ^OVX (WTI implied vol, 51st percentile), so there is no obvious vol dislocation to trade in gas itself.
Pivots from the settle: P 2.98, R1 3.01, S1 2.94, R2 3.05, S2 2.9. Price is trading just below the pivot, which keeps the intraday bias negative. The first real support is S1 2.94, then S2 2.9, which coincides with the 20-day low. A break of 2.9 would confirm the range breakdown and target the 52-week low at 2.48. On the upside, R1 3.01 caps the immediate bounce, and R2 3.05 is the invalidation level.
The last completed weekly bar (2026-09-21 to 2026-09-25) opened at 3.04, high 3.4, low 2.98, and closed at 3.23, up 5.98% w/w. That was a strong week, but the current week (from 2026-09-28, four sessions) is not closed and shows a last price of 2.97, down 8% — no weekly-close conclusions can be drawn from an unfinished bar. The weekly structure is therefore a failed push into 3.4 followed by a return to the range floor, which is consistent with the bearish call.
View: bearish while below 3.01 (R1); a settle below 2.9 targets 2.48.
3. Supply-Demand Balance & Fundamental Drivers
The curve is the clearest fundamental signal. The M1–M2 spread is -0.345, or -10.43%, with a slope of -0.0105 and a roll yield of -125.19%. This is a steep contango: the front month trades at a deep discount to the second month, which tells us the market is not pricing any near-term shortage. For a short position, contango is a tailwind — the position earns roll yield as the front contract converges upward toward the second month, or equivalently the short benefits from the curve carry. For a long, it is a cost. The structure is consistent with a market that has adequate storage and no immediate supply shock.
The COT data show open interest of 1,837,146 contracts as of 2026-09-22, with managed-money longs at 266,563 and shorts at 332,110, for a net of -65,547. The net short narrowed by 34,658 contracts from the prior week, meaning some shorts covered into the 2026-09-22 print. That covering coincided with the strong weekly bar (2026-09-21 to 2026-09-25, +5.98%), so it looks like a short-covering rally rather than new long conviction. Since then, price has fallen back to 2.97, which suggests the covering was exhausted and the downtrend has resumed.
Macro transmission is indirect but relevant. The US 10-year yield (^TNX) is 5.237, down 1.06% on 2026-10-01, while the dollar index (DXY) is 102.04, up 0.58%. A stronger dollar is a mild headwind for dollar-denominated commodities, including gas, though gas is primarily a domestic US market and the dollar effect is second-order. The more important macro channel is demand: a 5.24% 10-year yield implies restrictive financial conditions, which weighs on industrial demand and power burn at the margin. The week-ahead calendar includes the September ISM Services PMI (BJT 10-05 22:00 | ET 10-05 10:00, forecast 54 vs prior 55.4), and a print below 52.6 would reinforce the demand-side bear case.
The contango and the price action are sufficient to establish that the near-term balance is loose. The key fundamental driver to watch is whether the front spread narrows toward flat; a move to -0.2 or better would signal prompt tightness and would argue for covering shorts.
View: contango and a still-growing short base keep the fundamental bias bearish; a narrowing front spread is the first warning sign.
4. Positioning & Fund Flows
Managed-money net positioning was -65,547 contracts as of 2026-09-22, up from -100,205 on 2026-09-15. The week-over-week change of +34,658 contracts was a significant short-covering event, and it aligns with the 5.98% weekly gain in the last completed week (2026-09-21 to 2026-09-25). In other words, the rally was driven by shorts reducing exposure, not by longs adding. That is a weak foundation for a sustained advance, and the subsequent decline to 2.97 confirms it.
The crowding percentile is 27% on a three-year window, based on netPct of -3.57% (managed-money net divided by open interest). A 27th percentile is not extreme — it is below the middle of the three-year distribution. This matters because it means the short trade is not crowded. There is ample room for additional managed-money selling before positioning becomes a contrarian buy signal. The CTA trend-following proxy is 74, which is high and suggests trend followers are already positioned for weakness; that is a risk if a sharp reversal forces them to cover, but it also confirms the trend is down.
Hedge pressure is 27.41%, a moderate reading that does not indicate a producer-hedging wall that would cap rallies aggressively. Open interest rose to 1,837,146 from 1,820,003 the prior week, so the covering happened into rising open interest — a sign that new shorts were entering even as old shorts exited. That is a healthy bearish rotation.
RV20 for gas is 42.7%, which is elevated. Without a gas-specific IV, we cannot say whether options are cheap or expensive, so we do not trade vol here. The positioning data alone support the bear case: shorts are not crowded, the covering rally has faded, and open interest is rising on the decline.
View: positioning is bearish but not stretched; further short accumulation is possible, and the 27th percentile crowding argues against a contrarian long.
5. Cross-Asset Relative Value
The most relevant cross-asset signal for gas is the dollar. DXY is 102.04, up 0.58% on 2026-10-01, and the 10-year yield is 5.237, down 1.06%. A firm dollar and high real rates are a mild drag on commodity demand expectations. The VIX is 16.39 (35th percentile), indicating no broad risk-off event that would typically lift gas via safe-haven or supply-disruption channels.
Within the energy complex, ^OVX at 51.69 (51st percentile) shows WTI implied vol is mid-range, not signaling a supply panic. Gas contango of -10.43% is a steeper carry cost than typical for a market in balance, which reinforces the relative-value case for short gas versus long crude if one were constructing a pair, though we are not recommending that here. The gold vol (^GVZ 23.32, 16th percentile) and silver vol (^VXSLV 37.23) are not directly relevant to gas but confirm that the broader commodity vol complex is not in crisis.
The key relative-value takeaway is that gas is underperforming its own curve: the front month is deeply discounted to the second month, and the roll yield of -125.19% is a strong incentive for holders to stay short or avoid long exposure. There is no cross-asset signal that offsets the bearish gas-specific structure.
View: cross-asset backdrop is neutral-to-bearish for gas; the contango is the dominant relative-value signal.
6. Historical & Seasonal Patterns
Seasonality is the strongest argument against the short. Over the last 15 years, the next 20 sessions from the same calendar start have averaged +5.76%, with a median of +1.71%, and the market was up in 9 of 15 years. The best year was 2020 (+37.57%) and the worst was 2021 (-10.06%). This is a small sample and should be treated as context, not a forecast, but it does mean the base rate favors a bounce over the next month.
The tension is clear: the seasonal tailwind is positive, but the current price action, curve, and positioning are negative. In similar setups, the seasonal tendency often manifests as a choppy bottom rather than an immediate rally, especially when the front spread is in deep contango. The 2021 worst case (-10.06%) is a reminder that seasonality can fail outright when the fundamental balance is loose.
Given the 9-of-15 hit rate and the +5.76% mean, a short position should be managed tactically with a tight stop and a defined target, rather than held through the seasonal window. The 2.9 target is consistent with a move to the 20-day low, and the 3.05 invalidation respects the seasonal risk.
View: seasonality is a headwind for shorts, which is why this is a tactical short with a tight invalidation at 3.05, not a structural position.
7. Scenario Analysis (Base / Bull / Bear)
Base case (50%): grind lower to 2.9. The market remains below the 2.98 pivot and the 3.01 R1, with the front spread staying in contango near -0.345. Managed-money shorts add to positions, and price tests S1 2.94, then S2 2.9. A settle below 2.9 confirms the range breakdown and opens the 52-week low at 2.48. Action: stay short, trail the stop to 3.02 once price settles below 2.94.
Bull case (25%): seasonal bounce to 3.05–3.1. The 20-session seasonal tailwind (+5.76% mean, 9 of 15 up) kicks in, and a short-covering rally lifts price through R1 3.01 and toward R2 3.05. If the front spread narrows toward -0.2, the prompt-tightness signal would strengthen the bounce. A settle above 3.05 invalidates the short. Action: cover on a settle above 3.01 and stand aside; do not add to shorts into a seasonal rally.
Bear case (25%): breakdown to 2.48. A macro demand shock — for example, the ISM Services PMI printing below 52.6 on BJT 10-05 22:00 | ET 10-05 10:00 — or a further widening of the contango beyond -0.4 triggers accelerated selling. Price slices through 2.9 and targets the 52-week low at 2.48. Action: add to shorts on a settle below 2.9 with a stop at 2.98, targeting 2.48.
The probabilities sum to 100%. The base case agrees with the section 1 call: bearish, with 2.9 as the first target and 3.05 as the invalidation.
8. Trading Strategies & Risk Management
Strategy 1: Tactical short (conviction 7/10). Entry 2.98 (pivot P), stop 3.06 (above R2 3.05, roughly 0.08 or about two-thirds of ATR14 0.122 — placed beyond the invalidation level), target 2.9 (S2 / 20-day low), horizon 1–5 days. Size: 1.5% risk of portfolio equity, given ATR14 of 4.12% of price. Rationale: price is below the pivot, the curve is in contango, and positioning is not crowded. The stop is beyond the 3.05 invalidation, so a settle above 3.05 would stop the trade and invalidate the call.
Strategy 2: Breakdown add (conviction 6/10). If price settles below 2.9 (S2), add to the short at 2.89, stop 2.98 (back above the pivot), target 2.48 (52-week low), horizon 5–10 days. Size: 1% risk. This is a momentum add that only triggers on confirmation of the range breakdown. Do not pre-position ahead of the break.
Risk management: the main risk is the seasonal tailwind (+5.76% mean over the next 20 sessions). Keep the stop tight and do not average up into a rally. If the front spread narrows to -0.2 or better, reduce the short regardless of price. The house brief is also short (entry 3.01, stop 3.11, target 2.9); our entry is lower and our stop is tighter, which reflects the 2.98 pivot and the 3.05 invalidation. We agree with the direction and the 2.9 target.
9. This Week's Data Calendar
- BJT 10-02 20:30 | ET 10-02 08:30 — US Average Hourly Earnings m/m (forecast 0.3%, prior 0.3%; surprise outside ±0.1%), Non-Farm Employment Change (forecast 89K, prior 162K; surprise outside ±73K), Unemployment Rate (forecast 4.1%, prior 4.1%; surprise outside ±0.1%). Watch for a weak NFP print below 16K, which would weigh on the dollar and could support gas via demand expectations.
- BJT 10-05 22:00 | ET 10-05 10:00 — ISM Services PMI (forecast 54, prior 55.4; surprise outside ±1.4). A print below 52.6 would reinforce the demand-side bear case.
- BJT 10-08 02:00 | ET 10-07 14:00 — FOMC Minutes. A hawkish tone would support the dollar and weigh on gas.
This report is generated automatically from public quantitative and macro data for research and market tracking only. It does not constitute investment advice or a recommendation to trade.