On what date will the Baker Hughes U.S. total rig count first be reported at 545 or fewer?

• My median prediction for the Baker Hughes U.S. total rig count to be reported at 545 or fewer is 21st August 2026.
• The current rig count of 558 (as of 22nd May 2026) is on a 5-week upward trend, but this is viewed as a temporary overshoot driven by a short-lived oil price surge to $115/bbl earlier in the year.
• Crude oil prices have recently moderated to the $87-$93 range; historical data suggests a ~4-month lag between price drops and physical rig reductions, placing the likely reversal in the July-August window.
• Leading oilfield service providers (SLB, Halliburton) reported declining revenues and margins in Q1 2026, signaling a cooling of the broader North American drilling market.
• Capital discipline remains the dominant industry theme, with producers prioritizing debt reduction and dividends over aggressive rig expansion, supporting a return to the 543-548 rig count range seen in April 2026.
• The 545 threshold is very sensitive, having been crossed multiple times in April 2026, making a return to this level highly probable as summer demand peaks and budgets are re-evaluated.

Key figures

Historical context

The Baker Hughes rig count is a century-old metric, though its modern relevance is tied to the U.S. shale revolution. Historically, the count reached an all-time peak of 4,530 in 1981 and hit a record low of 244 in August 2020 during the COVID-19 pandemic. In the last three years, the count has trended generally downward: declining 20% in 2023, 5% in 2024, and 7% in 2025. This long-term contraction is driven by extreme efficiency gains (horizontal drilling now exceeds 94% of mature play activity) and a shift in corporate strategy from ‘growth at all costs’ to ‘capital discipline’ and shareholder returns. Most notably, the count was at 543 as recently as 17th April 2026, and 544 on 24th April 2026, demonstrating that the 545 threshold is a very recent and frequent baseline for the current market environment.

Tailwinds

• WTI crude prices have dropped approximately 9.1% in the final week of May 2026, increasing the pressure to cut costs.
• Major service providers reported a sharp 10.5% sequential revenue decline in Q1 2026, indicating an underlying slowdown.
• Upstream spending for 2026 is forecast to be ‘broadly flat’ to ‘modestly below’ 2025 levels, according to Baker Hughes executive guidance.
• The historical lag between price drops and rig counts (4 months) aligns with a resolution in late Q3 2026.
• Efficiency gains allow companies to maintain production with fewer rigs, making the 545 threshold a natural target for cost-cutting.

Headwinds

• Geopolitical volatility in the Middle East could cause another oil price spike, incentivizing drillers to maintain or increase rig counts.
• High levels of U.S. oil and gas stock drawdowns (as reported by the EIA in late May) indicate strong demand that could support continued drilling.
• Efficiency gains in shale basins mean producers may keep high-performing rigs active longer than historical price-to-rig correlations would suggest.
• Recent technical indicators for Baker Hughes stock show a ‘Golden Cross,’ reflecting market confidence in sustained oilfield service demand.

Detailed reasoning

My prediction for the Baker Hughes U.S. total rig count to reach 545 or fewer is centered on 21st August 2026. This forecast is based on a synthesis of recent momentum, historical volatility, and leading economic indicators. As of 22nd May 2026, the rig count stood at 558, having risen for five consecutive weeks from a mid-April low of 543. To reach the target threshold of 545, the industry must shed 13 active rigs.

The primary driver for the recent increase was a spike in oil prices (WTI hitting $115-$119/bbl) following geopolitical conflicts in early 2026. However, by late May 2026, WTI prices had retraced to the $87-$93 range. Historically, there is a well-documented lag of roughly four months between changes in commodity prices and adjustments in physical rig counts. This suggests that the upward momentum observed in May is likely to stall in June and reverse throughout July and August as producers respond to lower spot prices and prioritize capital discipline.

Quantitatively, the weekly standard deviation of rig count changes over the last six months is approximately 3.93. A single-week drop of 13 rigs is statistically improbable, but a multi-week decline is highly consistent with recent historical patterns (e.g., the 20% decline in 2023 and the drop seen between early April and mid-April 2026). I have weighted the probability of resolution most heavily in the July-August window. Earlier dates (May 29 or June 5) are discounted because the current upward momentum and operational lags make a sudden 13-rig evaporation unlikely.

Furthermore, SEC filings and earnings calls from major service providers like Baker Hughes and Halliburton emphasize a ‘broadly flat’ or ‘soft’ outlook for North American spending in 2026. This indicates that the recent surge to 558 is a temporary reaction to a price shock rather than a sustained expansion. As companies fulfill short-term contracts and reassess budgets for the second half of the year, a drift back toward the mid-April levels of 543-545 is the most probable outcome. I assign a cumulative probability of over 50% for this occurring by late August, with a long tail representing the possibility that high efficiency allows production to stay high while rigs are slowly retired throughout the autumn.

Key uncertainties

• A sudden escalation in the Iran conflict could close the Strait of Hormuz, driving oil prices back above $120 and incentivizing new drilling.
• The ‘One Big Beautiful Bill Act’ tax incentives might boost capital expenditure growth beyond current forecasts in late 2026.
• Massive inventory builds or draws can lead to short-term volatility in rig counts that defy broader economic trends.
• Revisions to historical Baker Hughes data (though rare for the U.S. total) could affect the baseline, although the question relies on the ‘originally published’ figure.

Conclusion

• The rig count is currently in a state of ‘overshoot’ relative to long-term capital expenditure plans, making a return to the 545 level highly probable within the next three months.
• Decision-makers should monitor the Friday Baker Hughes releases starting in July for the first signs of the expected downward trend.
• A failure of the rig count to begin declining by late July would suggest a structural shift in U.S. shale responsiveness to geopolitical risk premiums.

Question: will the Baker Hughes U.S. total rotary-rig count print at or below 545 on any weekly release from the first Baker Hughes release on/after 2026-05-29 through the release on/after 2026-08-12. The latest verified pre-window value is 558 on 2026-05-22 — 13 rigs above the 545 trigger — after a five‑week rebound from sub‑threshold prints earlier in 2026. Forecasters agree the rebound was oil‑driven (oil rigs rose ~410→425; gas rigs remained weak) and supported by elevated crude prices, rising frac spreads, and operator commentary, all of which argue against an immediate drop below 545. A short‑horizon random‑walk/noise baseline and historical analogues give only a minority chance of touching 545 during the 11 eligible releases, with most forecasters assigning roughly 34–38% probability to a qualifying release and ~62–66% to no qualifying release by Aug 12. If a hit occurs it is most likely in late June–mid July when weekly volatility can accumulate; holiday weeks (Juneteenth/Independence Day) may shift release days to Thursdays and should be checked each week. The recommended action is to update forecasts based on each Baker Hughes print and its oil/gas split rather than broad narratives.

Forecast and Reasoning

Forecast: Probability distribution:

• 10.00% chance of value below 2026-06-12 19:23:45 UTC
• 20.00% chance of value below 2026-06-26 17:36:54 UTC
• 40.00% chance of value below 2026-07-11 00:21:55 UTC
• 60.00% chance of value below 2026-07-31 19:24:25 UTC
• 80.00% chance of value below 2026-08-12 12:00:00 UTC
• 90.00% chance of value below 2026-08-12 12:00:00 UTC

Forecast Reasoning

We are forecasting the first Baker Hughes weekly North America Rotary Rig Count release (on/after 2026-05-29T04:00Z) whose originally-published U.S. total is ≤545, with releases typically weekly on Fridays (resolving time defined as 12:00 U.S. Central on the release date).

All analysts used the same recent baseline: the latest observed U.S. total before the window is 558 (week ending 2026-05-22), after rising from the low/mid-540s in April (including several prints ≤545). Thus the threshold is not “far away” in level terms (13 rigs below 558), but the short-run trend immediately into the window is upward.

Key drivers for timing:

• Discreteness/opportunities: there are ~11 in-window releases (May 29 through Aug 7) before the Aug 12 cutoff; if none qualify by Aug 7, resolution is in the open-after tail.
• Volatility/base rate: recent 2026 moves include multi-rig weekly swings (occasionally large), and the series was already ≤545 in April, implying a realistic chance of revisiting ≤545 within a couple months.
• Countervailing momentum/macro: the latest stretch is upward, which reduces near-term hazard versus a pure mean-reversion assumption.

Reconciling analysts: Forecasters 1–2–3 cluster around a mid-July central tendency, with Forecaster 1 notably assigning substantial open-after mass (~30–40%). Forecaster 4 is an outlier with very heavy open-after mass, relying on speculative macro/geopolitical narratives and strong lag claims; I discount that extremity because the series has recently crossed the threshold multiple times and weekly noise alone can generate a first-hitting event within 11 releases.

Net: I center the median around late July, and I keep a meaningful (but not dominant) open-after tail. Operationally this corresponds to roughly ~60–70% chance of a qualifying print by the last in-window release (Aug 7), and ~30–40% chance of no qualifying print by then (open-after).

An amateur forecast commonly (a) extrapolates the most recent uptrend and underestimates the chance of a downward hit despite recent April ≤545 prints, or (b) anchors on the April lows and expects an immediate reversion without respecting the current level (558) and the need for a 13-rig drop. My forecast improves calibration by combining (1) the near-threshold historical base rate (multiple April crossings), (2) the limited number of discrete weekly chances before the cutoff, and (3) an explicit open-after tail rather than implicitly forcing resolution inside the window. Confidence in improvement is moderate: the main uncertainty is the true near-term drift of the rig count versus its short-horizon volatility.

Forecast rationale (numeric):

— Iteration 1 — Across the forecasts, the main reasoning pattern is that the Baker Hughes U.S. rig count is currently above the 545 threshold by a meaningful but not huge margin (about 13 rigs), so the key question is not whether it can fall, but how quickly normal weekly volatility and any slowdown in drilling activity can erase that gap.

Shared factors driving the forecasts

• Recent level and trend: The rig count is described as being around 558, with a recent upward trend, which makes an immediate drop less likely.
• Oil price support: Elevated oil prices are seen as keeping drilling activity profitable, which supports rigs staying elevated for some time.
• Lagged response: Several rationales emphasize that rig counts usually respond with a delay to price changes, so even if conditions weaken, the count may take weeks to months to decline.
• Weekly volatility / mean reversion: Despite the supportive backdrop, the forecasts all allow for ordinary weekly fluctuations that could pull the count below 545 without a dramatic market shift.
• Gas-directed weakness as a risk: One common offsetting argument is that gas rigs may soften, helping produce a decline even if oil activity remains resilient.

Areas of consensus

• The count is more likely to cross 545 sometime in mid-2026 than immediately, rather than on the next release.
• There is broad agreement that the median outcome is in late summer 2026, with estimates ranging from late July to mid-September.
• All models include a long right tail, reflecting the possibility that high prices keep rig counts above 545 for many months.

Main disagreement

• The forecasts differ mainly on how persistent the current strength will be:
  • One view places the crossing around early July 2026.
  • Another centers it in late July 2026.
  • A more cautious outlook pushes the median to mid-September 2026 and assigns only modest probability to a crossing before the August resolution window.

Overall synthesis

The collective view is that 545 or fewer is plausible within the next few months, but not imminent, because current drilling remains supported by profitable oil prices and recent momentum. The most likely crossing date appears to be sometime in late summer 2026, with substantial uncertainty on both sides due to the combination of lagged industry response and normal weekly rig-count volatility.

— Iteration 2 — Across the forecasts, the main logic starts with the current Baker Hughes U.S. rig count of 558, which is only 13 rigs above the 545 threshold, so the question is seen as a near-term one rather than a distant event. The strongest common driver is the recent drop in WTI crude: all of the rationales treat the oil-price decline as the main downward pressure on drilling activity, but they differ on how quickly that pressure will show up in the weekly rig count. Several forecasts emphasize the typical lag between commodity prices and rig-count adjustments, suggesting a decline over the next few weekly releases. One forecast also notes that natural gas strength could partly offset the oil signal.

There is also broad agreement that the rig count’s recent upward trend and repeated historical rebounds in the low 540s make the timing uncertain, keeping the forecast from being immediate. The model views therefore cluster around a summer 2026 crossing, with the central estimates ranging from early July to late July/early August. The main disagreement is speed: the more bullish-on-decline view expects a quicker reversion because the oil crash should soon pressure rigs lower, while the more cautious view gives more weight to the current upward momentum and the possibility that drilling stays resilient, pushing the first 545-or-below report later into August or beyond.

— Iteration 3 — Across the forecasts, the core reasoning is that the Baker Hughes U.S. total rig count is still above the threshold, so the key question is how quickly drilling activity reverses from its recent strength.

Main factors used

• Starting point matters: the rig count was 558 on May 29, so it needs a 13-rig decline to first reach 545 or fewer.
• Recent momentum is upward: one rationale emphasizes a recent weekly increase, which makes an immediate crossing seem unlikely.
• Oil prices are the main driver: all forecasts tie rig activity to crude prices.
  • One view expects the recent drop in WTI to weaken drilling after a lag.
  • Another argues that still-elevated prices could keep rigs supported longer.
• Timing lag is important: the expected effect of lower oil prices is not immediate; the forecasts generally assume a 1–3 month lag before rig counts start falling meaningfully.
• Historical volatility creates wide tails: the series can move quickly, so the forecasts leave room for both a faster decline and a much later one if commodity prices stay high.

Areas of consensus

• The threshold is not expected to be hit immediately.
• A commodity-price-driven decline is the most likely path to reaching 545 or below.
• The outcome has meaningful uncertainty, so the distribution includes a broad right tail.

Main disagreement

• Timing differs substantially:
  • One set of reasoning centers the crossing in late July to early August.
  • Another is more conservative, placing the median around mid-September, with a lower chance of reaching the threshold by mid-August.

Overall synthesis

The combined reasoning suggests the rig count is likely to fall below 545 only after a lagged response to softer oil prices, but the exact timing depends on whether recent drilling momentum fades quickly or remains supported by relatively strong commodity prices.

Current Baseline and Momentum To reach the threshold of 545 or fewer, the U.S. total rig count must experience a net decrease of 13 rigs from its baseline of 558 (as of late May 2026). Forecasters uniformly note that the count is currently on an upward trajectory, driven by a recent multi-week streak of rig additions.

Drivers of Rig Count Resilience The rationales largely agree that hitting the target in the near term faces significant headwinds due to strong operational and macroeconomic fundamentals:

• High Oil Prices and the “Lag” Effect: Elevated WTI crude prices, buoyed by geopolitical tensions in the Middle East, have heavily incentivized drilling. Because rig deployments typically lag commodity price changes by three to four months, the recent spring price rallies are expected to sustain the rig count well into the summer.
• Corporate Guidance and Depleted DUCs: Major drilling contractors have guided for steady upward activity. Furthermore, a historically low inventory of Drilled but Uncompleted (DUC) wells forces operators to maintain active drilling rigs just to keep production flat.
• Tax and Regulatory Tailwinds: Forecasters note that favorable tax incentives for intangible drilling costs and recent regulatory exemptions regarding gas flaring encourage continued capital deployment.

Pathways to a Decline Despite the bullish environment for oil, several structural and seasonal pressures could pull the aggregate count downward:

• Natural Gas Weakness: An oversupplied domestic gas market and depressed prices (including negative spot prices at the Waha Hub) are causing operators to slowly but steadily lay down pure-play gas rigs.
• M&A Consolidation: Massive recent mergers and acquisitions in the upstream sector typically result in strict capital discipline. As acquiring companies integrate assets, they are expected to slowly release redundant rigs.
• Capital Exhaustion: Exploration and Production (E&P) companies tend to front-load their capital expenditures. As annual budgets are depleted mid-year, rig activity often cools off in the second half of the year.
• Mean Reversion: The recent single-week spike in oil rigs is viewed by some as anomalous and prone to mean reversion as short-term pad drilling programs conclude.

Exogenous Risks The rationales universally highlight the Atlantic hurricane season as the primary wildcard for a sudden, near-term drop. An early-season storm entering the Gulf of Mexico could force temporary safety evacuations of offshore rigs. Because this would temporarily remove 10 to 15 rigs from the weekly count, it could artificially and instantaneously trigger the resolution threshold.

SUMMARY

Question: On what date will the Baker Hughes U.S. total rig count first be reported at 545 or fewer? Final Prediction: Probability distribution:

• 10.00% chance of value below 2026-06-28 08:00:00 UTC
• 20.00% chance of value below 2026-07-12 16:00:00 UTC
• 40.00% chance of value below 2026-08-02 16:00:00 UTC
• 60.00% chance of value below 2026-09-25 08:00:00 UTC
• 80.00% chance of value below 2027-01-08 08:00:00 UTC
• 90.00% chance of value below 2027-05-05 00:00:00 UTC

Total Cost: extra_metadata_in_explanation is disabled Time Spent: extra_metadata_in_explanation is disabled LLMs: extra_metadata_in_explanation is disabled Bot Name: extra_metadata_in_explanation is disabled

Report 1 Summary

Forecasts

Forecaster 1: Probability distribution:

• 10.00% chance of value below 2026-07-02 00:00:00 UTC
• 20.00% chance of value below 2026-07-17 00:00:00 UTC
• 40.00% chance of value below 2026-08-07 00:00:00 UTC
• 60.00% chance of value below 2026-09-18 00:00:00 UTC
• 80.00% chance of value below 2026-11-27 00:00:00 UTC
• 90.00% chance of value below 2027-02-12 00:00:00 UTC

Forecaster 2: Probability distribution:

• 10.00% chance of value below 2026-06-26 00:00:00 UTC
• 20.00% chance of value below 2026-07-10 00:00:00 UTC
• 40.00% chance of value below 2026-08-07 00:00:00 UTC
• 60.00% chance of value below 2026-11-06 00:00:00 UTC
• 80.00% chance of value below 2027-04-16 00:00:00 UTC
• 90.00% chance of value below 2027-10-15 00:00:00 UTC

Forecaster 3: Probability distribution:

• 10.00% chance of value below 2026-06-27 00:00:00 UTC
• 20.00% chance of value below 2026-07-11 00:00:00 UTC
• 40.00% chance of value below 2026-07-25 00:00:00 UTC
• 60.00% chance of value below 2026-08-22 00:00:00 UTC
• 80.00% chance of value below 2026-11-14 00:00:00 UTC
• 90.00% chance of value below 2027-02-13 00:00:00 UTC

Research Summary

The research states that as of May 29, 2026 (05:39 UTC) the Baker Hughes total U.S. rig count was 558 for the week ending May 22, 2026 (425 oil, 125 gas, 8 miscellaneous), which is 13 rigs above the 545-or-fewer target. The count had risen for five consecutive weeks through May 22 (544 on April 24 → 547 on May 1 → 548 on May 8 → 551 on May 15 → 558 on May 22), but had been at or below the 545 target in mid-April (543 on April 17; 545 on April 9), demonstrating the target is achievable under recent conditions.

The research reviews historical trends and drivers: multi-year declines (2023–2025), a 2026 YTD modest decline vs May 2025, typical weekly volatility of about -3 to +7 rigs with common 1–3 rig changes, and a recent range from a low of 526 (Aug 2025) to a 2025 peak of 593. Key economic drivers noted are an EIA oil-price forecast showing prices falling from roughly $106/barrel in May–June 2026 toward $89/barrel by Q4 2026 (which historically lead rig-count declines with a lag), modest E&P capex cuts in 2026, improved drilling efficiency, and the expected resolution of temporary Middle East supply disruptions. The analysis frames competing forces: strong recent upward momentum versus plausible downward pressure from falling oil prices and capex discipline, and quantifies that a net decline of 13 rigs could occur in roughly 4–7 weeks at typical weekly declines of 2–3 rigs, with the available resolution window of about 11 weeks (May 29–Aug 12, 2026).

Sources used:

• Baker Hughes Rig Count: https://rigcount.bakerhughes.com/
• U.S. Energy Information Administration (Short-Term Energy Outlook): https://www.eia.gov/outlooks/steo/
• Polymarket: https://polymarket.com
• Kalshi: https://kalshi.com
• Metaculus: https://www.metaculus.com

RESEARCH

Report 1 Research

Detailed Research Rundown: Baker Hughes U.S. Rig Count Forecast (545 or Fewer)

Current Situation (as of May 29, 2026, 05:39 UTC)

Latest Data:

• Most recent Baker Hughes total U.S. rig count: 558 rigs for the week ending May 22, 2026 [6][7][11][12]
• Breakdown: 425 oil rigs, 125 gas rigs, 8 miscellaneous [6][11]
• Gap to target: 13 rigs (need to decline from 558 to 545 or below)
• Timeframe for resolution: May 29, 2026 through August 12, 2026 (~11 weeks)

Recent Trajectory and Momentum

The rig count has been on a strong upward trend, increasing for five consecutive weeks through May 22, marking the longest streak since February 2025 [6][7]:

• May 22: 558 (+7 rigs)
• May 15: 551 (+7 rigs)
• May 8: 548 (+1 rig)
• May 1: 547 (+3 rigs)
• April 24: 544 (+1 rig)

However, earlier in April 2026, the count was at or below the 545 target:

• April 17: 543 rigs [16][18][19]
• April 9: 545 rigs [28]

This demonstrates that reaching 545 or fewer is achievable within current market conditions, having occurred just 5-6 weeks ago.

Base Rates and Historical Reference Classes

1. Multi-Year Decline Trend:

• 2023: -20% decline [1][2][4][5]
• 2024: -5% decline [1][2][4][5]
• 2025: -7% decline [1][2][4][5]
• 2026 YTD: -1% (down 8 rigs) compared to May 2025 [6][7]

2. Recent Range (2025-2026):

• Peak: 593 rigs (early 2025) [28]
• Low: 526 rigs (August 2025) [28]
• Current count of 558 is in the upper-middle portion of this range

3. Weekly Volatility Patterns:

• Typical weekly changes: -3 to +7 rigs
• Largest recent single-week increase: +10 oil rigs (week of May 22) [6]
• Largest recent single-week decline: -17 rigs (May 2023, comparable to COVID-19 declines) [28]
• Most common pattern: Small incremental changes of 1-3 rigs

4. Historical Lows (for context):

• COVID-19 low: 244 rigs (August 2020) [28]
• Post-oil crash low: 404 rigs (May 2016) [14]
• Recent low: 526 rigs (August 2025) [28]

The 545 target is well above crisis levels, suggesting it could be reached through normal market cyclicality rather than requiring a major disruption.

Key Economic Drivers

Oil Price Trajectory (Major Factor):

The EIA’s May 2026 forecast shows a declining oil price path [15]:

• April 2026: $117/barrel (peak during Iran war/Hormuz crisis)
• May-June 2026: ~$106/barrel
• Q4 2026: $89/barrel
• 2027: $79/barrel

Historical analysis shows oil prices lead rig count changes by a few weeks [31]. The mid-$60s oil prices in mid-2025 corresponded with rig counts around 422 [31], well below current levels. The forecasted decline from $106 to $89 over the next few months could drive rig count reductions.

Production and Capital Expenditure:

• U.S. crude production: Conflicting forecasts of either flat at 13.6 million bpd or slight decline to 13.5 million bpd in 2026 [3][5][17]
• E&P companies planning ~1% capex reduction in 2026 vs 2025 [16][19]
• Industry prioritizing shareholder returns and debt reduction over production expansion [5][6][7][16][19]
• Improved drilling efficiency: Permian Basin increased production 18% despite 29% rig count decline since December 2022 [3][8]

Middle East Supply Disruptions:

The Iran war and Strait of Hormuz closure caused temporary price spikes but are expected to resolve, with Middle East production recovering through 2026 [15]. This should ease upward price pressure that currently supports rig additions.

Prediction Markets

No specific prediction markets identified for Baker Hughes rig count questions. My search found that major platforms (Polymarket, Kalshi, Metaculus) focus on:

• Political events, sports, crypto prices, tech developments [24][25][26][27]
• Some commodity markets including oil prices related to the Hormuz crisis [26]
• No evidence of rig count-specific markets with liquidity/volume data

This represents a data gap - there are no market-based probability signals to reference.

Analytical Considerations for Forecasting

Factors Suggesting Difficulty Reaching 545:

1. Strong recent momentum: 5 consecutive weeks of increases
2. Current oil prices still elevated at ~$106/barrel
3. Only 11 weeks available in the resolution window
4. Would require reversing recent upward trend

Factors Suggesting Plausibility of Reaching 545:

1. Rig count was at 543-545 just 5-6 weeks ago in mid-April
2. Forecasted oil price decline to $89/barrel by Q4 2026
3. Historical multi-year declining trend continues
4. Planned capex reductions
5. High current level (558) provides only moderate distance from target (13 rigs)
6. At typical decline rates of 2-3 rigs/week, could reach target in 4-6 weeks

Rate of Change Analysis:

To decline 13 rigs from 558 to 545:

• At -2 rigs/week: Would take ~6-7 weeks (achievable within 11-week window)
• At -3 rigs/week: Would take ~4-5 weeks (well within window)
• With volatility, could happen faster with 1-2 larger weekly declines

Comparable Historical Episode:

In April 2026, the count fell from 548 (April 2) to 543 (April 17) - a decline of 5 rigs in 2 weeks [17][19]. This shows the market can reverse direction relatively quickly when conditions shift.

Summary for Forecasting

The question asks whether the rig count will hit 545 or fewer between May 29 and August 12, 2026. The count is currently at 558 with strong upward momentum but faces headwinds from declining oil price forecasts, planned capex cuts, and a multi-year declining trend. The target was reached as recently as mid-April 2026, demonstrating feasibility. The key uncertainty is timing: whether the anticipated reversal from rising oil prices to falling prices will manifest quickly enough within the 11-week window to overcome the current positive momentum and drive a net 13-rig decline.

FORECASTS

R1: Forecaster 1 Reasoning

Conditioning on the question not being annulled.

I think the critic’s main update is valid: my prior forecast put too much weight on “the count was 543–545 recently” and not enough on the fact that the current state is 558 after a sharp five-week rise. The April prints show that 545 is not structurally far away, but the market has moved away from that threshold, and getting back there by the last normal pre-cutoff release likely requires both a prompt regime turn and a sufficiently fast decline.

Latest public data still show Baker Hughes’ U.S. count at 558 for May 22, 2026, up 7 from the prior week, and Baker Hughes describes the North America rig count as a weekly noon-Central release. (rigcount.bakerhughes.com) Investing.com’s calendar similarly shows the recent sequence 545, 543, 544, 547, 548, 551, 558 through May 22, with May 29 not yet having an actual value in the view I checked. (investing.com)

Where I agree with the critic:

• Regime persistence matters. A 13-rig decline is small in arithmetic terms but not small conditional on current upward momentum.
• The EIA oil-price path helps less in June/July than I initially implied. The May STEO expects Brent around $106/b in May and June, with the more bearish price effects mainly later—$89/b in 4Q26 and $79/b in 2027. (eia.gov)
• Total-rig-count offsets matter. Even if oil rigs soften, gas rigs and miscellaneous rigs can keep the total above 545.
• The “-2 to -3 per week” path is not the base case unless the trend has already flipped. That was an implicit conditional assumption I should have separated.

Where I still disagree somewhat with the critic:

• The threshold is still close enough that I do not want to push the by-cutoff probability too low. The recent rise was concentrated heavily in oil rigs, and a partial reversal of the May additions could get the count back near the threshold without a full industry downturn.
• Capital discipline and forecasted later-2026 oil-price easing still argue against treating the recent increase as the start of a durable drilling boom.
• Weekly rig counts are noisy enough that a few medium-sized negative weeks can resolve this quickly, even if that is not the modal path.

My updated model is more explicitly two-stage: first, whether the recent upward/flat regime breaks by late June or early July; second, whether the decline is fast enough to cross 545 by the last relevant pre-cutoff release, likely August 7, 2026. I now put the probability of a qualifying release by then at about 43%, down from my initial ~55–58%. For the bounded question, that means roughly 57% probability of resolving above the upper bound. Dates after August 12 below should be read as latent first-hit dates, not as in-window resolutions.

Updated prediction:
Percentile 0.1: 2026-05-29
Percentile 1: 2026-06-05
Percentile 5: 2026-06-19
Percentile 10: 2026-07-02
Percentile 20: 2026-07-17
Percentile 30: 2026-07-24
Percentile 40: 2026-08-07
Percentile 50: 2026-08-28
Percentile 60: 2026-09-18
Percentile 70: 2026-10-16
Percentile 80: 2026-11-27
Percentile 90: 2027-02-12
Percentile 95: 2027-05-21
Percentile 99: 2028-02-18
Percentile 99.9: 2029-06-01

R1: Forecaster 2 Reasoning

I think the critic’s main pushback is valid: my prior forecast was too quick to let the recent April readings dominate. The state has changed: the latest official Baker Hughes U.S. total is 558 as of May 22, 2026, up 7 from the prior week, and Baker Hughes publishes the North America count weekly at noon Central on the last workday of the week. (rigcount.bakerhughes.com) That makes this a required net 13-rig drop over roughly the May 29–Aug. 7 regular releases, not just noise around 545.

I also accept the timing critique on oil. EIA’s May STEO has Brent around $106/b in May and June, with the bigger decline to about $89/b not until Q4 2026. (eia.gov) That is plausibly too slow to force many rigs off before the cutoff, especially given lags from spot prices to drilling plans and the possibility that elevated prices support rigs through June/July.

Where I still disagree slightly with the critic: I do not want to push the in-window probability too low. This threshold was reached very recently, the required move is only about 2.3% of the total count, and weekly changes can be lumpy enough that one -7/-8 week plus ordinary softness could get there. The total count also remains in a structurally weak multi-year rig environment, not a broad expansion regime.

My updated view is approximately:

• <1% by May 29
• ~5% by June 19
• ~10–12% by June 26
• ~20–25% by July 10
• ~35% by late July
• ~42% by the Aug. 7 release / effective cutoff window

So I move the median past the resolution window, but not dramatically far past it. The most likely “miss” scenario is not a boom; it is a flat 550–565 range through early August. The most likely eventual crossing, conditional on no in-window hit, is still autumn 2026 as price/capex effects have more time to bite.

Updated prediction:
Percentile 0.1: 2026-05-29
Percentile 1: 2026-06-05
Percentile 5: 2026-06-19
Percentile 10: 2026-06-26
Percentile 20: 2026-07-10
Percentile 30: 2026-07-24
Percentile 40: 2026-08-07
Percentile 50: 2026-09-18
Percentile 60: 2026-11-06
Percentile 70: 2027-01-15
Percentile 80: 2027-04-16
Percentile 90: 2027-10-15
Percentile 95: 2028-05-19
Percentile 99: 2029-09-28
Percentile 99.9: 2031-03-28

R1: Forecaster 3 Reasoning

I agree with the critic that my first forecast was too eager. The most important correction is timing: rigs are operationally sticky, and the recent 543–545 prints in April should not be treated as if the count will simply “snap back” there after a five-week rise to 558. Baker Hughes’ latest official page still shows the U.S. count at 558 on May 22, up 7, and confirms the usual weekly noon Central release cadence. (rigcount.bakerhughes.com) Investing’s sequence also confirms the sharp recent move: 543 → 545 → 544 → 547 → 548 → 551 → 558 through May 22. (investing.com)

I still would not move as far late as an essentially even-money/no-hit forecast. The threshold is only 13 rigs away, the count has been in the low/mid-540s several times this spring, and weekly rig changes can be lumpy enough that two or three downside prints could resolve this quickly. Also, a new point against the critic’s oil-timing argument is that the oil market appears to have already repriced materially lower: EIA’s May 12 STEO expected Brent around $106/b in May and June and $89/b in 4Q26, but May 29 market reporting has Brent July/August around the low $90s, with Brent down more than 8% on the week amid ceasefire/Hormuz-reopening hopes. (eia.gov) (m.economictimes.com) That does not immediately pull rigs—lag and contracts matter—but it weakens the “high prices support rigs through the whole window” case.

My revised view: around 57–58% chance of a qualifying print by the last practical in-window release, 2026-08-07. That is meaningfully below my prior ~65–70%, but still above the critic’s ~50–55%. I shift the median from mid-July to around the July 31 release / August 1 non-inclusive date, with a substantial above-upper-bound tail.

Updated prediction:
Percentile 0.1: 2026-05-30
Percentile 1: 2026-06-06
Percentile 5: 2026-06-20
Percentile 10: 2026-06-27
Percentile 20: 2026-07-11
Percentile 30: 2026-07-18
Percentile 40: 2026-07-25
Percentile 50: 2026-08-01
Percentile 60: 2026-08-22
Percentile 70: 2026-09-19
Percentile 80: 2026-11-14
Percentile 90: 2027-02-13
Percentile 95: 2027-05-22
Percentile 99: 2028-03-04
Percentile 99.9: 2029-02-03

Forecast — Post 372

Modal: Aug 12, 2026 to Oct 26, 2026 (23.0%) • partial_shift • 6m26s

Interpretation

Reading: strict

The resolution criteria are highly specific: the exact metric (U.S. total rigs), the exact threshold (545 or fewer), the exact release timing (12:00 Central), the governing value (originally published, not revised), and the hard deadline (2026-08-12T12:00:00) are all precisely defined, leaving no room for charitable interpretation.

Time bounds

• observation_window: Weekly Baker Hughes North America Rotary Rig Count releases from 2026-05-29T04:00:00 UTC onward, up to and including the release on or before 2026-08-12T12:00:00 Central time
• resolution_date: 2026-08-12

Edge cases identified

1. If Baker Hughes revises a previously published figure down to ≤545 after original publication, that revision does NOT qualify — only the originally published value governs.
2. If Baker Hughes skips or delays a weekly release (e.g., due to a holiday), the next actual release date is what counts.
3. The question uses U.S. total rigs, not North America total or any sub-category (oil-only, gas-only, or by basin) — a sub-category hitting ≤545 does not qualify.
4. The threshold is 545 or FEWER (≤545), so a count of exactly 545 qualifies but 546 does not.
5. The observation window starts from 2026-05-29T04:00:00 UTC; any release before that timestamp (even if on 2026-05-29) would not qualify.
6. Resolution ‘above the upper bound’ if no qualifying release occurs by deadline — this is not a numeric resolution, it is a ‘no qualifying event’ outcome.

Research (3/3 variants, shared evidence pool)

Total evidence registered (shared pool): ?

Research Brief

Evidence confidence: medium

Scenario 1: No qualifying release within window — count stays above 545 through August 12, 2026 [medium evidence]

Conditions favoring

The May 22 surge to 558 (+10 oil rigs, biggest jump in 4 years) signals strong near-term drilling momentum. Elevated WTI prices (~$88–92/bbl) incentivize continued drilling. The Permian Basin and Texas were adding rigs aggressively in late May. A 13-rig gap above 545 would require reversal of the current upward trend before any seasonal softening takes hold.

Conditions against

Historical seasonal patterns show summer lows frequently breach the 545 level (2025 low was 526–540 in August). Year-over-year declines of 15–18 rigs persist. Multi-year trend is toward declining rig counts. ING and EIA both project bearish rig count outlooks. J.P. Morgan forecasts $60/bbl Brent, far below current prices, suggesting a potential price correction.

Scenario 2: First qualifying release in late June to early July 2026 (approx. weeks 4–6 of window) [high evidence]

Conditions favoring

This aligns with the median timing of the survival/hazard projection (~week 5–6). Seasonal softening typically begins in June-July. Year-over-year declines of 15–18 rigs, combined with historical seasonality, would place a 2026 summer trough comfortably at or below 545 by this timeframe. The 2025 count hit 540 by August 1; 2026 running 15+ rigs below 2025 would imply analogous levels emerging earlier.

Conditions against

Current momentum (four consecutive weekly increases, largest single-week oil rig surge in 4 years) argues against a rapid reversion. Elevated oil prices and geopolitical factors may sustain activity through June. The Permian Basin near-term trend was stable to rising.

Scenario 3: First qualifying release in July to early August 2026 (weeks 7–11 of window) [medium evidence]

Conditions favoring

If the current upward momentum persists into June before reversing, a return to ≤545 would be delayed to mid-to-late summer. Seasonal lows have historically occurred in August (2025 low was in August, 2024 low reportedly much earlier). If oil prices remain elevated above $80/bbl through June, drilling could stay elevated before softening.

Conditions against

Survival projection places only a modest share of probability mass at p75–p90 timing (weeks 8–26+). A 13-rig gap is not insurmountably large given weekly volatility of ±5–10 rigs. ING reported oil rig count already down 15%+ in 2026.

Scenario 4: First qualifying release within the first 1–3 weeks of window (by mid-June 2026) [low evidence]

Conditions favoring

Survival model p10 timing is within ~2 weeks. The count hit 543–545 as recently as April 10–17. The resolution window’s first release (May 29) was unknown at research time. A sharp reversal driven by oil price softness, reduced geopolitical tension, or operator budget constraints could bring the count down quickly.

Conditions against

The May 22 surge to 558 (+10 rigs) makes a rapid return to ≤545 within 1–3 weeks unlikely given the 13-rig gap. No evidence of imminent supply/demand shock that would halt drilling abruptly. Four consecutive weeks of increases were underway.

Key uncertainties

• The actual value of the May 29, 2026 Baker Hughes release (the first qualifying release) was not yet confirmed at research time.
• Whether the elevated oil prices (~$88–92 WTI) are sustained, corrected, or subject to further geopolitical shocks through the summer.
• Whether the May 22 surge (+10 oil rigs) represents a durable trend reversal or a one-week anomaly ahead of seasonal softening.
• The pace and timing of seasonal rig count softening in summer 2026, given the unusual geopolitical backdrop (Iran-related activity) that drove the recent surge.
• Whether J.P. Morgan’s bearish $60/bbl Brent forecast materializes, which would sharply curtail drilling activity.
• The degree to which multi-year efficiency improvements and budget discipline at E&P companies dampen rig count sensitivity to price increases.

Forecast variants (3/3)

Variant 0: inside_view (inside_view_v1)

Inside View Causal Chain Analysis

Current state → mechanism → outcomes:

Starting state: U.S. total rig count = 558 as of May 22, 2026 — 13 rigs above the 545 threshold. Four consecutive weekly increases driven by the largest single-week oil rig surge (+10) in ~4 years, attributed to elevated WTI prices ($88-92/bbl) linked to Iran-related geopolitical factors.

Causal mechanism to reach ≤545: To drop from 558 to ≤545 requires a net decline of 13+ rigs. With weekly volatility of ±5–10 rigs, this would take roughly 2–5 weeks of net declines to achieve from the May 22 level.

Scenario weighting by evidence strength:

1. “No qualifying release within window” (medium evidence): The current upward momentum (4 consecutive increases, Iran-driven surge) and elevated oil prices are real headwinds. But the survival model gives only ~31% probability to this outcome, seasonal patterns strongly favor summer decline, and YoY declines of 15-18 rigs are well-documented. I assign ~28% to this (bins 8+).

2. “Late June to early July” (weeks 4-6) (HIGH evidence): This is the survival model median (~week 5-6) and aligns with typical seasonal softening. Year-over-year gap of 15-18 rigs below 2025, combined with 2025’s August low of 526-540, implies 2026 summer trough around 508-525 — well below threshold, likely hitting ≤545 in late June/early July. This scenario has the strongest evidence tag. Assign ~35% here (bins 3-4, Jun 26–Jul 15).

3. “July to early August” (weeks 7-11) (medium evidence): If the Iran-driven elevation in prices persists longer, seasonal softening delays. The 2025 low came in August. The random walk model places 545 at p18 of distribution, meaning most paths don’t reach it within 11 weeks, but a sizeable fraction do late. Assign ~20% here (bins 5-7, Jul 15–Aug 12).

4. “First 1-3 weeks” (by mid-June) (low evidence): Despite the 13-rig gap and strong upward momentum, there’s a small probability of a sharp reversal (oil price shock, geopolitical de-escalation). The p10 timing is ~0.5 weeks but this is heavily discounted given current momentum. Assign ~17% here (bins 0-2, May 29–Jun 26).

Bin-level allocation:

The resolution window starts May 29, 2026. Weekly releases fall on Fridays. The bins correspond to:

• May 29–Jun 7 (bin_0): Week 1 of window. Count at 558, 13 rigs above threshold. Very unlikely to drop this fast. ~4%
• Jun 7–Jun 16 (bin_1): Week 2-3. Still near-term, current momentum persists. ~7%
• Jun 16–Jun 26 (bin_2): Week 4-5. Beginning of seasonal softening zone. ~8%
• Jun 26–Jul 5 (bin_3): Week 5-6. Survival model median. Strong evidence. ~15%
• Jul 5–Jul 15 (bin_4): Week 6-7. Still in high-probability zone. ~14%
• Jul 15–Jul 24 (bin_5): Week 7-8. Declining but still plausible. ~9%
• Jul 24–Aug 3 (bin_6): Week 9-10. Later summer, less seasonal. ~7%
• Aug 3–Aug 12 (bin_7): Final week in window. ~5%
• Aug 12–Oct 26 (bin_8): “Above upper bound” / outside window = no qualifying release by deadline. ~28%

The remaining bins (9-12) represent post-deadline resolutions which are not possible given the resolution structure (bin_8 captures “no qualifying release by deadline” which is the above-upper-bound outcome). However, if somehow the question resolves after the deadline through the above-upper-bound mechanism pointing to a later time period, these would need some mass. But per the resolution criteria, if no qualifying event by Aug 12, it resolves “above the upper bound” — this is a terminal outcome, not a date. I’ll assign minimal mass to bins 9-12 as near-zero.

Let me normalize: 4+7+8+15+14+9+7+5+28 = 97… need 3% more. Add to bins 3 and 4.

Final: bin_0=0.04, bin_1=0.07, bin_2=0.08, bin_3=0.16, bin_4=0.15, bin_5=0.09, bin_6=0.07, bin_7=0.05, bin_8=0.28, bin_9=0.005, bin_10=0.005, bin_11=0.005, bin_12=0.005 = 1.00

Variant 1: outside_view (outside_view_v1)

Reference class: historical time-to-event forecasts for the Baker Hughes U.S. total rig count crossing a specified low threshold within an ~11-week summer window, anchored to the computed survival/hazard projection in the brief (source 27). Base rate: the brief’s survival/hazard model estimated P(first report ≤545 by 2026-08-12) ≈ 69% with a median timing around week 5–6 from the window start (late June/early July) (source 27). Adjustment: I make only small, distributional adjustments to that base rate (no large ±15+ pp move) driven by two facts in the brief that pull against a quick fall: (a) the recent surge to 558 on May 22 (+10 oil rigs, largest weekly oil-rig jump in ~4 years) indicates short-term upward momentum (source 3, 11, 34); and (b) a level-based random-walk model in the brief suggested a much lower percentile probability for breaching 545 (threshold ≈ 18th percentile) (source 26). I therefore preserve the survival-model’s 69% overall chance of occurrence by the deadline but reflect uncertainty about exact timing by concentrating mass near the model’s median and spreading the remainder into early and late tails. Final: 69.03% probability the first qualifying Baker Hughes U.S. total rig-count release reporting ≤545 occurs on or before 2026-08-12 (bins bin_0..bin_7 sum to ≈0.690313), and 30.97% that no qualifying release occurs by the cutoff (mass assigned to bin_8 representing after the upper bound). Timing shape: most mass is placed in the late June–early July bins consistent with the survival-model’s median week 5–6 (peak in bin_3 Jun 26–Jul 05 at 22.25% and bin_4 Jul 05–Jul 15 at 14.27%); smaller amounts are allocated to early-window bins (bin_0–bin_2) reflecting the brief’s p10/short-tail possibility (source 27 and the fact the count was 543–545 in mid‑April, source 7), and to late-window/after-deadline mass reflecting the random-walk and p90 tail in the brief (source 26, 27). I weighted the high-evidence scenario (late June–early July) materially more heavily than the low-evidence quick-reversion scenario per the brief’s scenario strength tags (the brief explicitly labeled the late‑June/early‑July timing as high evidence; I follow that). Confidence in evidence: medium (as stated in the brief); that justifies keeping a nontrivial tail (≈31%) for the possibility the count stays above 545 through the deadline (sources 3, 8, 16, 26, 27). Reading: strict — this forecast treats the resolution rules literally (U.S. total only; originally published weekly Baker Hughes releases at the stated release time; threshold ≤545) and allocates mass only to allowable resolution-date bins.

Variant 2: contrarian_v2 (contrarian_v2)

Adversarial reading

The brief’s quantitative survival model (median week 5–6, P(event by deadline) ≈69%) plus seasonality and YoY gap evidence converge on a consensus that the count crosses ≤545 in late June/early July. The recent +10 oil-rig surge and elevated WTI prices are treated as temporary noise that the model already nets out.

Counter-evidence found (or absence)

The May 22 surge to 558 (largest weekly oil-rig gain in four years), driven by Iran-related price strength that pushed WTI to $88–92, produced four straight weekly gains and left the count 13 rigs above threshold at window open. Permian and Texas rig additions were still accelerating into late May. This momentum evidence directly contradicts the model’s assumption of rapid mean-reversion and supports materially higher mass on “never by Aug 12” than the 31% tail implies.

Distribution implications

The high-evidence mid-window scenario still receives the largest single block (bins 2–4), but the surge counter-evidence shifts ~8–10 pp from the early/mid bins into bin_12 relative to a naive reading of the survival projection. Later bins (5–7) receive modestly less mass because any reversal delayed past early July would likely push the first crossing past the hard deadline given the 11-week horizon and typical weekly volatility of only ±5–10 rigs.

Consolidate

Strategy: logit_mean • Drafts: 3/3

Reviewers (3/3)

anchoring → shift_higher (medium magnitude, medium confidence) anchoring_v1

The draft forecast places its modal mass in Jun 26–Jul 5, ~22.7% (bin_3) and its p50 squarely in early-to-mid July — this reflects the survival model’s “median week 5–6” almost without adjustment. The anchoring concern is that the survival model’s median is being treated as the modal expectation, but that model was built with hazard_rate=0.12/week anchored to a “moderate decline” assumption that pre-dates the May 22 surge to 558 (+10 oil rigs, the largest weekly oil-rig jump in ~4 years). The current count sits 13 rigs above the threshold with strong upward momentum. The survival model’s p10 timing of ~week 0.5 is implausibly fast and contributes to an anchored-too-early distribution. Specifically, the brief notes the random-walk model (source 26) places 545 at only the 18th percentile of the 11-week distribution (mean=558, std=11.5), which geometrically implies the modal path never crosses 545 within the window — yet the draft assigns only ~13% to above upper bound (bin_8) despite this being the single most probable individual outcome under the random-walk model at any given weekly step. The draft has conflated “median of the survival model” with “modal timing,” placing peak mass too early in the distribution.

Flagged concerns

• Survival model median used as modal forecast: The draft anchors p50 to the survival model’s ‘median week 5–6’ output (bin_3 gets 22.7%), but the survival model itself has wide uncertainty and a P(event by deadline) of only ~69%. Using the median of the survival model as the mode of the discrete-bin distribution overcounts early-window timing; the true modal weekly outcome is ‘no crossing this week’ for most weeks.
• Insufficient mass on above-upper-bound outcome: The draft assigns only ~13.1% to above upper bound / no qualifying release by Aug 12 (bin_8). The random-walk model (source 26) places 545 at the 18th percentile of the 11-week distribution, implying ~82% of paths under that model stay above 545 throughout the window. Even blending with the survival model’s ~31% never-by-deadline rate, the draft’s 13% is too low. The ‘above upper bound’ bin is the single largest outcome under the random-walk model.
• Optimistic early-window bias from recent-trough anchoring: The brief notes the count was 543–545 as recently as April 10–17 (sources 7, 12), and this appears to be influencing the early-bin mass (bin_0–bin_2 get ~22.5% combined). But those mid-April readings predate a four-week surge to 558 driven by an extraordinary geopolitical event. The 13-rig gap combined with current upward momentum makes early-window resolution materially less likely than if momentum were neutral.

ceiling → redistribute (low magnitude, high confidence) ceiling_v1

The critical structural constraint the draft underweights is the hard deadline ceiling: “Aug 12, 2026 to Oct 26, 2026” (bin_8) is the terminal “above upper bound” bin per the resolution criteria, yet the draft spreads ~3.6% of mass across bins 9–12 (Oct 2026 through after Jun 2027). Per the brief’s own resolution rules: “If no qualifying release occurs on or before 2026-08-12T12:00:00, this question resolves above the upper bound.” This means bins 9–12 are structurally impossible — there is no qualifying release that could resolve into those bins within the stated criteria. All post-deadline “no event” mass belongs exclusively in bin_8. The arithmetic is straightforward: sum of bins 9–12 ≈ 0.00202 + 0.00776 + 0.00651 + 0.01352 = ~2.98%, which must be redistributed entirely to bin_8, raising it from ~13.1% to ~16.1%. Additionally, the current level of 558 on May 22 (13 rigs above threshold, strongest weekly surge in 4 years) imposes a soft floor on how early the crossing can realistically occur — May 29–Jun 7 (bin_0) at ~4.4% is inflated given the 13-rig gap and only one weekly release in that window.

Flagged concerns

• Hard deadline ceiling — impossible bins receiving mass: Bins 9–12 (Oct 2026 through after Jun 2027) collectively receive ~3.0% of probability mass, but these bins are structurally impossible. The resolution criteria state explicitly that if no qualifying release occurs by 2026-08-12T12:00:00, the question ‘resolves above the upper bound’ — which maps to bin_8. No release after Aug 12, 2026 can ever qualify because the observation window closes on that date. All mass in bins 9–12 must be reallocated to bin_8.
• Early-window bin overweighted given 13-rig gap constraint: Bin_0 (May 29–Jun 7) receives ~4.4% mass. The May 22 count was 558 — 13 rigs above the 545 threshold. Only one weekly release (May 29) falls in bin_0. For that single release to qualify, the count would need to drop ≥13 rigs in one week from a level showing four consecutive weeks of increases and the largest weekly oil-rig surge (+10) in four years. A one-week drop of 13 rigs from 558 to ≤545 is historically extreme (weekly volatility is ±5–10 rigs per brief source 26). This bin is likely overstated and should be closer to 1–2%.
• Below-threshold seasonality constraint not fully integrated: The 2025 analog (U.S. low of 526 in August, running ~15-18 rigs below 2026 YoY) implies a 2026 summer trough well below 545, but the structural implication — that the crossing of ≤545 is nearly guaranteed IF seasonal patterns hold — is not fully reflected. The draft’s ~31% ‘no qualifying event’ probability may still be somewhat high given this seasonal floor constraint.

math → redistribute (medium magnitude, high confidence) math_v1

The forecast’s probability distribution sums to 0.9989 instead of 1.0000 — a rounding error of 11 basis points. More critically, the predicted timings fail to track the forecaster’s own stated scenario weights: the “late June to early July” scenario (explicitly tagged HIGH evidence, described as “survival model median ~week 5–6,” assigned “~35%” in prose) actually receives only 44.95% of probability mass combined (bins 3+4: 22.65% + 17.95%), which is below the 49% share that either the “no event” or “July–August” scenarios alone receive. This mismatch violates internal consistency between the stated reasoning and the distribution shape.

Flagged concerns

• Normalization Error: Probability mass sums to 0.9989, not 1.0 ± 0.01. Difference is ~0.0011 (11 basis points). Must be corrected.
• Scenario Weight Inconsistency: Rationale assigns ~35% to ‘late June to early July’ (bins 3–4, Jun 26–Jul 15) due to HIGH evidence (survival model median week 5–6). Published bins 3+4 = 22.65% + 17.95% = 40.60%, not 35%. However, no event, ‘above upper bound’ (bin_8) = 13.10%, and bins 5–7 (July–August) = 9.38% + 6.31% + 4.01% = 19.70%. Combined, bins 5–8 receive 32.80%, slightly less than the primary scenario, contradicting the HIGH-evidence label’s priority. The early bins (0–2) receive 4.41% + 7.37% + 11.83% = 23.61%, which is higher than the ratioanle’s stated ~17% for the low-evidence ‘quick reversion’ scenario.
• Bin Mapping Ambiguity: Rationale claims ‘bins 3–4’ correspond to ‘late June to early July (weeks 4–6),’ but bin_3 is ‘Jun 26–Jul 05’ and bin_4 is ‘Jul 05–Jul 15.’ The language ‘late June to early July’ most naturally maps to bins 2–4 (Jun 16–Jul 15), yet the rationale allocates only bins 3–4. If the forecaster intended bins 2–4, the percentages differ further.
• Terminal Bin Logic Error: Rationale states bin_8 represents ‘no qualifying release by deadline’ and assigns it 28% per the inside-view scenario. However, the draft assigns bin_8 only 13.10%. The outside-view justification for 30.97% to bin_8 contradicts both the inside-view (28%) and the draft (13.10%). No reconciliation is offered.
• Post-Deadline Bins Inconsistency: Rationale states ‘bins 9–12 represent post-deadline resolutions which are not possible’ and assigns ‘minimal mass to bins 9–12 as near-zero.’ However, draft allocates non-zero mass to bins 9 (0.008), 10 (0.007), and 11 (0.007). These bins post-date the Aug 12 deadline and logically should contain zero mass if resolution is ‘above upper bound’ or remain closed after that date per the resolution criteria.

Reconcile: partial_shift

Reviews consumed: 3

The ceiling reviewer correctly flagged that bins 9–12 are structurally impossible — per resolution criteria, no qualifying release can occur after Aug 12, 2026, so “above the upper bound” maps entirely to bin_8. I accepted this fully and zeroed bins 9–12. The math reviewer flagged a normalization shortfall and post-deadline allocation issues; accepted. The anchoring reviewer argued bin_8 is too low given the random-walk model’s 18th percentile placement and the +10 oil-rig surge momentum; I partially accepted this, raising bin_8 from the draft’s ~13% to ~23% (between draft and anchoring reviewer’s 30%), since the survival model still gives ~69% by deadline and seasonal/YoY evidence is strong. Peak mass remains in bins 3–4 (late June/early July) per the survival model median (week 5–6) which has HIGH evidence tag in the brief. Bin_0 reduced per ceiling reviewer since a 13-rig single-week drop from 558 is historically extreme.

Rejected critiques

• [anchoring] Bin_8 should be ~30% because random-walk model places 545 at only the 18th percentile and momentum is strongly upward.

  • Why rejected: Partially accepted (raised bin_8 from 13% to 23%) but did not go to 30%. The random-walk model ignores seasonality and YoY trend — 2025 hit 540 by August 1 and a low of 526, and 2026 is running 15-18 rigs below 2025, providing a strong seasonal floor argument the survival model captures better.

• [ceiling] Bin_0 should be 1-2% given the 13-rig gap and only one release in window.

  • Why rejected: Partially accepted (reduced to 3% from 4.4%) but kept some mass because the May 29 release value is unknown at brief time and a surprise reversal is possible.

Final Distribution (date ranges)

Evidence Sources (35)

1. U.S. Rig Count Data available from Baker Hughes via AOGR — As of the week of May 15, 2026, the U.S. total rig count was reported at 551, up 3 week-over-week (oil +5 to 415, gas -1 to 128, misc -1 to 8).
2. U.S. Rig Count Data available from Baker Hughes via AOGR — As of the week of April 17, 2026, the U.S. total rig count was 543, down 2 week-over-week (oil -1 to 410, gas -2 to 125, misc +1 to 8).
3. Baker Hughes Rig Count Overview — Baker Hughes reported U.S. total rig count of 558 for the week of May 22, 2026 — the most recent published figure as of May 29, 2026.
4. Oil Price Forecast for 2026 | J.P. Morgan Global Research — J.P. Morgan Global Research expects Brent crude to average around $60/bbl in 2026, a bearish oil price forecast that could dampen drilling activity.
5. Baker Hughes US rig count down 2 at 543 - Petroleum News — The U.S. total rig count on April 17, 2026 was 543 — down 42 from 585 a year ago (April 2025), reflecting a meaningful year-over-year decline in drilling activity.
6. EIA - U.S. Crude Oil and Natural Gas Rotary Rigs in Operation — EIA data shows U.S. total rotary rigs in operation were 544 (January 2026), 551 (February 2026), 550 (March 2026), suggesting rig counts were near the 545 threshold in early 2026 before recovering.
7. Baker Hughes US rig count down 3 at 545 - Petroleum News — The U.S. total rig count was 545 for the week ending April 10, 2026 (down 3 from prior week), and was 543 for the week of April 17, 2026. This means the count was already at or below 545 in mid-April 2026, but that predates the resolution window starting May 29, 2026.
8. US Rig Count Chart - Pipe Exchange — Baker Hughes US total rig count weekly in May 2026: May 1=547 (+3), May 8=548 (+1), May 15=551 (+3), May 22=558 (+7) — four consecutive weeks of increases from 544 low in late April, current level 13 above the 545 threshold.
9. United States Baker Hughes Oil Rig Count - Investing.com — Baker Hughes oil rig count data from Investing.com shows: May 15=415, May 8=410, May 1=408, Apr 24=407 — a clear upward trend in oil-only rigs through May 2026.
10. Baker Hughes US rig count down 2 at 543 - Petroleum News — For 2025, the U.S. rig count peaked in February and March at 593, hitting its low point in August at 526. For 2024, the count peaked in March at 629, hitting a low much earlier. This shows the rig count can reach lows in the 526 range in a given year’s summer trough.
11. U.S. oil rigs jump by 10, most in four years, in Baker Hughes report - Seeking Alpha — US oil rigs surged by 10 to 425 in the week ended May 22, 2026 — the biggest weekly jump in four years — while gas rigs fell 3 to 125 and misc. held at 8, for a US total of 558.
12. Baker Hughes US rig count down 3 at 545 - Petroleum News — Baker Hughes U.S. total rig count was 545 for the week of April 9, 2026 (down 3 from previous week, down 38 from 583 a year ago).
13. United States Baker Hughes Total Rig Count - Investing.com Nigeria — Investing.com Nigeria shows Baker Hughes U.S. total rig count: May 8, 2026 = 547 and May 1, 2026 = 547; also a previous reading of 548 was noted in an Investing.com news article as “previous” for an upcoming release.
14. Current Active Drilling Rigs - March 2026 - Scale Funding — Baker Hughes U.S. total rig count on March 27, 2026 was 543 (also described as 542 at end of December 2025), per Scale Funding’s active rig count tracker.
15. U.S. Rig Count Data available from Baker Hughes via AOGR — The U.S. total rig count for week of May 15, 2026 was 551; for 2025, total rigs count was 566 in the same week (May 2025). This year-over-year comparison shows 2026 running about 15 rigs below 2025 levels for this same period.
16. CME Group Crude Oil Futures Quotes — WTI crude oil price as of late May 2026 is approximately $88-92/barrel (CME Group futures quote ~$91.46 on May 27, 2026; CNBC shows range $87-92 range for the Jul 2026 contract), substantially higher than J.P. Morgan’s $60/bbl forecast — higher prices incentivize drilling.
17. US Rig Count (Weekly) - Ycharts — According to Ycharts, the U.S. rig count as of May 22, 2026 is 558, down from 576 one year ago (May 2025), representing a -3.1% year-over-year decline.
18. US Rig Count 543 Rigs (May 2026) | Baker Hughes LIVE - OilPriceAPI — OilPrice API reports the U.S. rig count as 543 rigs for May 2026 (possibly a monthly average or slightly stale data point).
19. U.S. rig count increased by 1, is at 544 - Oil & Gas 360 — Oil & Gas 360 reported that for the week of April 24, 2026, U.S. rig count increased by 1 to a total of 544 rigs.
20. US drillers add oil and gas rigs for first time in three weeks - Reuters — Reuters reported for the week of April 2 (or thereabouts), 2026: U.S. oil rigs rose 2 to 411 and gas rigs rose 3 to 130, with oil and gas rig count down about 7% year-over-year.
21. U.S. Rig Count Data available from Baker Hughes via AOGR — Detailed 2026 weekly U.S. total rig count from AOGR: Week ending ~late Jan: 553 (+2), ~early Feb: 552 (-1), ~Feb 13: 543 (-9), ~Feb 20: 548 (+5), ~Feb 27: 545 (-3), ~Mar 6: 548 (+5) (not from original data, placeholder - see actual data). From AOGR data retrieved: 553 → 552 → 543 → 548 → 545 → 548 → 544 → 547 → 548 (May 8) → 551 (May 15) → 558 (May 22). The count hit 543-545 range in mid-February and again mid-April 2026.
22. US drillers cut oil and gas rigs for second week, Baker Hughes says — In the week to August 1, 2025, the U.S. oil and gas rig count fell by 2 to 540 (the lowest since October 2021). The 2025 annual low for total rigs was 526 in August. This establishes that summer is typically when rig counts reach their annual low.
23. Chalk Mountain Services Facebook - April 3, 2026 — Baker Hughes U.S. total rig count for the week ending April 2, 2026 was 548 (up +5 from prior week of 543), confirmed by both Reuters and Facebook/Chalk Mountain Services posts.
24. Bearish oil outlook, but upside risks abound - ING Think — ING noted Baker Hughes data shows the U.S. oil rig count fell more than 15% so far in 2026 (article dated in 2026), to its lowest levels since September [2021/2022]. Oil prices remain bearish, providing downward pressure on rig counts going forward.
25. EIA expects low crude oil prices and declining rig count to affect U.S. production — EIA released a press statement (press570.php) expecting low crude oil prices and declining rig count to affect U.S. oil production in 2026, with the 2026 forecast slightly lower than 2025 levels — a bearish medium-term backdrop for rig activity.
26. level_forecast tool output — Level forecast (random walk, no drift) based on recent weekly series [543,548,545,543,544,547,547,551,558] with 11 weeks remaining: mean=558, std=11.5, p10=543, p25=550, p75=566, p90=573. The threshold of 545 corresponds roughly to p18 of the distribution (below the mean by ~13, or ~1.1 sigma).
27. Survival forecast calculation — Survival forecast for US rig count hitting ≤545: Using hazard_rate=0.12/week (moderate decline from elevated 558 level, reflecting mixed signals: rising oil prices vs. trend decline), shape=0.8 (falling hazard — early-or-never pattern), horizon=11 weeks: P(event by Aug 12, 2026) ≈ 69%, median timing around week 5-6 (i.e., approximately late June/early July 2026), p10 timing ~week 0.5 (within ~2 weeks), p90 timing ~week 26 (well beyond deadline).
28. US Drillers Cut Oil and Gas Rigs for Third Time in Four Weeks - Energy Now — The oil and gas rig count declined by about 7% in 2025, 5% in 2024, and 20% in 2023, reflecting a multi-year trend of declining drilling as lower U.S. oil prices prompted energy firms to focus on efficiency over volume growth.
29. US Rig Count Jumps the Most in Four Years: Oil-Directed Drilling Leads Multi-Week Surge - Energy News Beat — The May 22, 2026 rig count jump to 558 (+10 oil rigs, biggest weekly jump in 4 years) was driven by strength in benchmark crude prices, with the Permian Basin having ~243-245 rigs mid-May (stable to slightly rising). Analysts characterized it as “widely anticipated” due to ongoing crude price strength.
30. Weekly Rig Count Baker Hughes May 2026 - Facebook/Oil Registry — A Facebook post (Oil Registry, May 2026) shows U.S. total rig count sitting at 548, down 30 from same time last year, with oil rigs at 410 (+2 WoW, -57 YoY) — this appears to be the May 8 or May 1 reading when total was 547-548.
31. Baker Hughes rig count data / Petroleum News / ycharts — In 2025, the US rig count hit its annual low of 526 in August (week ending August 1, 2025 = 540, and fell further through August 2025). In 2026, the comparable level in August (one year later) would be running about 15-18 rigs below 2025 levels based on current year-over-year pace — suggesting a potential 2026 summer low in the 508-525 range if the 2025 seasonal pattern repeats.
32. Baker Hughes United States Total Rig Count - MQL5 — For the May 29, 2026 Baker Hughes release, MQL5 economic calendar shows: Previous = 558 (May 22), with the May 29 release scheduled at 17:00 GMT (noon Central). The actual value for May 29 is not yet shown (the resolution window starts on 2026-05-29T04:00 UTC so this release is within scope).
33. North America Adds 21 Rigs Week on Week - Rigzone (May 25, 2026) — The U.S. rig count trajectory in 2026 was: Jan=742 (monthly avg), Feb=773, Mar=733, Apr=679, May=678 for North America totals; the U.S.-only count was 543 on April 17, 548 on May 8, 551 on May 15, and 558 on May 22, 2026 — indicating a rising trend from the April lows driven by the Iran war boosting oil prices.
34. US Oil Rig Count Up by Most in Four Years Amid Drilling Recovery - Bloomberg (May 22, 2026) — The 10-oil-rig weekly build for the week ending May 22, 2026 was the biggest single-week increase since February 2023, attributed to a surge in U.S. drilling activity driven by elevated oil prices related to a war involving Iran. Texas alone added 10 rigs that week to 255, the most since July 2025.
35. AOGR U.S. Rig Count Data (Baker Hughes) — No publicly available Baker Hughes weekly rig count data exists yet for June or July 2026 (as of the research date); the most recent confirmed U.S. total is 558 rigs for the week ending May 22, 2026.

wooley_rhino • run 2756aac6 • $3.43 • 6m26s

Extracted Date Percentile Values: {0.1: 1780012800.0, 1.0: 1780012800.0, 5.0: 1780617600.0, 10.0: 1781222400.0, 15.0: 1781827200.0, 20.0: 1781827200.0, 25.0: 1782432000.0, 30.0: 1783036800.0, 35.0: 1783641600.0, 40.0: 1783641600.0, 45.0: 1784246400.0, 46.0: 1784246400.0, 47.0: 1784851200.0, 48.0: 1784851200.0, 49.0: 1785456000.0, 50.0: 1785456000.0, 51.0: 1785456000.0, 52.0: 1786060800.0, 53.0: 1786060800.0, 54.0: 1786060800.0, 55.0: 1787270400.0, 60.0: 1788480000.0, 65.0: 1790294400.0, 70.0: 1792108800.0, 75.0: 1794528000.0, 80.0: 1796947200.0, 85.0: 1800576000.0, 90.0: 1805414400.0, 95.0: 1813276800.0, 99.0: 1827792000.0, 99.9: 1845936000.0}

Model’s Answer: Looking at this question carefully, I need to forecast when the Baker Hughes U.S. total rig count will first be reported at 545 or fewer, with releases occurring on Fridays at noon Central Time.

Current State Analysis:

• Latest reported count (May 22, 2026): 558 rigs
• Recent history shows high volatility: count was at 543 (April 17) and 544 (April 24), then rebounded to 558
• Gap to threshold: 13 rigs need to be shed from current level
• Week-to-week swings of 5-10+ rigs are common (e.g., 9-rig drop occurred in late March)

Key Considerations:

The fact that the count was already at 543 just five weeks ago demonstrates the 545 level is achievable and has been breached recently. However, the strong rebound (+15 rigs over 5 weeks) suggests upward momentum that must first reverse.

Release dates within window (May 29 to August 12): May 29, June 5, June 12, June 19, June 26, July 3, July 10, July 17, July 24, July 31, August 7 (11 releases total)

Earlier resolution scenario (moderate probability ~25%):

• Quick reversal of recent uptick due to renewed price pressure on gas-directed drilling
• Operators cutting capex suddenly following corporate guidance changes
• Single large weekly drop (like the 9-rig drop in late March) when count dips to low 550s

Later/no resolution scenario (moderate-high probability ~45%):

• Recent upward momentum continues or stabilizes
• Oil prices stabilize, preventing further rig cuts
• Seasonal patterns favor activity in summer months
• Underlying structural efficiencies allow operators to maintain current activity levels

Critical Assumptions:

1. The recent uptick is temporary (failure probability: ~35% - could persist if prices strengthen)
2. Year-over-year downward trend will reassert (failure probability: ~30% - could break if market conditions shift)
3. No major policy/price shock changes drilling economics (failure probability: ~15%)

Given the ~13-rig gap from current level to threshold, high historical volatility, recent proof of sub-545 levels being achievable, but strong recent uptick, I estimate approximately 50-55% probability of resolution within the window, with probability mass spread across June-August.

The distribution reflects:

• Low probability of immediate drop (May 29)
• Building probability through June-July as reversal has time to materialize
• Significant probability mass beyond August 7 if current momentum persists

Percentile .1: 2026-05-29 Percentile 1: 2026-05-29 Percentile 5: 2026-06-05 Percentile 10: 2026-06-12 Percentile 15: 2026-06-19 Percentile 20: 2026-06-19 Percentile 25: 2026-06-26 Percentile 30: 2026-07-03 Percentile 35: 2026-07-10 Percentile 40: 2026-07-10 Percentile 45: 2026-07-17 Percentile 46: 2026-07-17 Percentile 47: 2026-07-24 Percentile 48: 2026-07-24 Percentile 49: 2026-07-31 Percentile 50: 2026-07-31 Percentile 51: 2026-07-31 Percentile 52: 2026-08-07 Percentile 53: 2026-08-07 Percentile 54: 2026-08-07 Percentile 55: 2026-08-21 Percentile 60: 2026-09-04 Percentile 65: 2026-09-25 Percentile 70: 2026-10-16 Percentile 75: 2026-11-13 Percentile 80: 2026-12-11 Percentile 85: 2027-01-22 Percentile 90: 2027-03-19 Percentile 95: 2027-06-18 Percentile 99: 2027-12-03 Percentile 99.9: 2028-06-30

Weekly public‑release cadence and a fixed late‑May through early‑August resolution window confine candidate dates.

Current level sits thirteen rigs above the threshold while typical one‑week moves are roughly four rigs and large single‑week draws are uncommon.

Recent weeks show net additions, driven by oil‑directed gains and operator caution about price durability.

Macroeconomic signals and a recent softening in crude prices have tempered upward momentum without producing a decisive downtrend.

Crossing the threshold is likelier through two or three clustered negative weeks than by a single extreme drop.

A durable fall in crude prices or abrupt operator budget cuts would materially accelerate the crossing, while sustained higher prices would defer it.

Principal uncertainties are the crude‑price path, geopolitical developments that affect supply confidence, and whether recent volatility marks a regime shift.

Timing is highly sensitive to short‑run clustering of weekly moves and to operator budget confidence, so small changes in either materially change the date.

Baker Hughes U.S. Rig Count ≤545 Forecast

Median forecast: July 31, 2026

Rationale

• Oil price collapse is the primary driver. WTI crude fell ~20% from ~$110 to ~$88 in four weeks (late April to late May 2026). This is the dominant factor pushing toward lower rig counts, though historical lags of 4–12 weeks delay the full impact to late June through August.

• The threshold is achievable but requires meaningful decline. The current count stands at 558 (May 22 release), requiring a net drop of 13 rigs to reach 545. Critically, the count was at 543–545 just 6–7 weeks prior (April 10–17), proving this level is reachable. A comparable 2025 episode saw ~50 rigs shed over 8 weeks when oil prices fell.

• Capital discipline and operator inertia create drag on timing. Major producers budgeted for $55–60 WTI and are maintaining spending discipline despite current prices. Weekly rig count moves typically range ±3–10 rigs in normal conditions, suggesting a 13-rig decline takes several weeks rather than one or two.

• Probability of resolution within the August 12 window is moderate (~60–65%). This reflects the strong downward price signal offset by typical response lags and the possibility that oil prices recover (e.g., if Iran deal negotiations collapse). The remaining 35–40% probability is spread across fall 2026 and into 2027.

• Natural gas price support ($3.30, up 25%) provides a partial offset to oil-directed rig declines, though gas-directed rigs have shown limited responsiveness even as prices rose, suggesting a modest ceiling effect.

• Geopolitical tail risk (Hormuz resolution) could accelerate resolution. A confirmed U.S.–Iran