Crude Cut № 003 · Chord Energy · Williston basin, ND + MT
On July 20, sixteen days before Chord Energy published its second-quarter results, we wrote the release from the public record — state production filings, posted oil and gas prices, Chord's own guidance and its hedge book — and then held it next to the real one. Fifteen lines; ten landed within three percent. Then we ran the same exercise on the six quarters before it.
The pre-written release · 2Q26 · written 20 JUL · scored 05 AUG
Every number in the predicted column was public on July 20: North Dakota and Montana production filings, posted WTI and Henry Hub, Chord's own guidance ranges from May, its hedge book from the March 10-Q, and the state tax rate. Each line wears the lane it came from — that chip is the honest part of this page.
Chord Energy · second quarter 2026 · predicted vs print · $MM unless noted
| Line | Predicted · 20 JUL | Chord · 05 AUG | Δ | Lane |
|---|---|---|---|---|
| Oil volumes, MBopd | 165.5 | 165.4 | +0.1% | State filings |
| Natural gas, MMcfpd | 418 | 408 | +2.3% | State filings wellhead ≠ sold |
| NGL, MBblpd | 51.0 | 53.0 | −3.8% | Guidance midpoint |
| Total, MBoepd | 286.1 | 286.4 | −0.1% | Derived |
| Gross operated TILs | 61 – 70 | 66 | in bracket | State filings |
| Realized oil, $/bbl | 96.65 | 93.99 | +2.8% | Market + guided diff |
| Crude oil revenue | 1,456 | 1,415 | +2.9% | State × Market |
| Oil, NGL and gas revenue | 1,521 | 1,494 | +1.8% | State × Market |
| Lease operating expense | 266 | 268 | −0.8% | Guidance × volumes |
| Production taxes | 127 | 126 | +1.2% | ND tax rate × revenue |
| Hedge settlements | −90 | −70 | −$20 | 10-Q hedge book × market |
| Adjusted EBITDA | 940 | 924 | +1.8% | Sum of the above |
| E&P and other CapEx | 390 | 417 | −6.5% | Completed footage proxy |
| Adjusted free cash flow | 477 | 413 | +15% | Residual |
| Adjusted EPS | — | $6.44 vs $6.68 consensus | miss | No public signal |
RESULT: THE PUBLIC RECORD WRITES THE TOP HALF. Oil volumes, price, revenue, LOE, taxes and EBITDA all land within 3%; the Street's revenue consensus of $1.43B missed a $1.49B print that the filings had at $1.52B. Below EBITDA the errors compound (FCF is a residual of four estimates) and EPS is unreachable — DD&A, deferred taxes and derivative marks live in no filing before the print. Chord's own adjusted EPS missed consensus the same day its revenue beat it.
Seven quarters · 4Q24 – 2Q26 · the same table, every quarter with a clean post-Enerplus base
Each quarter's prediction uses only what was public two weeks before that print: two months of state filings with the third held, the quarter's average prices, the guidance Chord had issued at the start of the quarter, and the hedge book from its latest 10-Q. Across the seven, oil volumes miss by a point and a half on average, revenue by two and a half, EBITDA by four. The cell reads the prediction's miss against the print.
Predicted vs print · signed miss by line and quarter · green = within 2% · pale = within 5%
| Line | 4Q24 | 1Q25 | 2Q25 | 3Q25 | 4Q25 | 1Q26 | 2Q26 | Avg |miss| |
|---|
Hedge settlements are shown in $MM rather than percent — the base flips sign across quarters. CapEx is a footage proxy that fails the test and is shown greyed, to be honest rather than useful.
RESULT: THE TOP HALF OF THE RELEASE IS PUBLIC BEFORE THE PRINT. Seven quarters, mean absolute miss: oil volumes 1.4%, gas 2.8%, realized oil price 1.6%, revenue 2.4%, LOE 1.7%, production taxes 6.3%, hedge settlements ±$5MM, Adjusted EBITDA 4.0%. Gross wells turned in line: 9.2%; over the four quarters Chord guided them, the filings ran 8.7% off and the guided midpoints 19.6%. The worst quarter for volumes was the 2025 winter, when holding March at February's level could not see the thaw; the worst for hedges was 2Q26, when oil ran through every collar ceiling at once. Much of the table is inputs anyone can multiply: swap the filed volumes for Chord's guidance midpoint and the seven-quarter oil miss is 1.3% instead of 1.4%. The filings pull ahead on well counts and in the last three quarters, 0.5% against 1.6% on oil.
The two lines the market trades on · predicted vs print · 4Q24 – 2Q26
The same two series that headline every earnings story, as the public record had them two weeks early, drawn over what Chord printed. The prediction tracks the print through a $60 oil quarter and a $95 one.
Oil, NGL and gas revenue · $MM · by quarter
Adjusted EBITDA · $MM · by quarter
public-data prediction, two weeks before the print Chord reported
RESULT: WITHIN 5% ON EBITDA IN SIX OF SEVEN QUARTERS. Revenue predicted vs print, $MM: 1,056 / 1,064 · 1,059 / 1,103 · 984 / 950 · 991 / 967 · 886 / 877 · 1,118 / 1,151 · 1,521 / 1,494. Adjusted EBITDA: 618 / 640 · 636 / 696 · 570 / 547 · 601 / 578 · 513 / 506 · 681 / 713 · 940 / 924. The one wide miss is 1Q25, a winter quarter where the volume call ran 3.6% light.
The live one · 3Q26 · prediction posts 20 OCT · Chord reports ~04 NOV
The backtest earns the right to post a prediction before the print. On October 20, July and August will be filed in North Dakota and only September held. Three things are visible now.
June exit rate 229,802 bbl/d gross, 2.5% above the Q2 average. The guide needs the quarter to average 4% below June's exit with one frac crew.
At $82 the collars, ceilings at $77.65 and $81.92, barely bite; the $69.54 swaps still cost about $23MM. The oil book goes from a $70–90MM drag to roughly $25–30MM.
78 drilled-uncompleted wells in the ND inventory, 24 of them four-mile; the second frac crew dropped in July.
Rebuild this cut
Every state-lane number on this page traces to a query that runs verbatim in a Crude Code-connected Claude session, against the same public filings. Pull the recipe, re-run it on current data, swap in another operator, or post your own prediction before the next print.
> get_cut "003" — the recipe, in your session
crudecode.devThe prediction. Every line is built from inputs that were public two weeks before the print, in five lanes. State filings — North Dakota Industrial Commission and Montana Board of Oil and Gas monthly production and well records for every well filed under Chord Energy (state names Chord Energy, Whiting, Oasis). Oil volumes: prior quarter's reported MBopd × (state gross rate with months one and two filed and month three held at month two) ÷ (state gross rate, prior quarter), plus announced acquisition volumes (XTO: 4.0 MBopd × 61/92 days in 4Q25). Comparing rates cancels most of the gross-vs-net gap. Gas uses the same rule on wellhead gas. Wells turned in line: operated wells whose first reported production month falls in the quarter, plus wells completed in the last month with no production yet. Market — EIA daily WTI Cushing and Henry Hub spot, averaged over the quarter, plus Chord's guided differential and its guided NGL and gas realizations as a share of the benchmark. Guidance — Chord's published ranges for the quarter, at the midpoint: NGL volumes; LOE $/Boe × implied volumes; prior-quarter GPT per Boe and cash G&A carried forward. Filings — production taxes at the prior quarter's effective rate on crude revenue; hedge settlements from the derivative table in the latest 10-Q or 10-K before the print, remaining-year volumes spread evenly over the remaining quarters and settled at the quarter's spot average (swaps: strike − price; collars pay only outside the floor–ceiling band; three-ways cap the floor payout at floor − sub-floor). Proxy — prior-quarter capex × (0.35 × spud footage ratio + 0.65 × completed footage ratio); it fails the backtest at 23% mean miss and is shown greyed. Blind — no public input exists before the print: EPS, net income, DD&A, deferred taxes, derivative marks, buybacks.
The scorecard. Predicted Adjusted EBITDA = revenue − LOE − GPT − production taxes − cash G&A + hedge settlements. "Within 3%" counts 10 of 15 lines in 2Q26: oil, gas, total volumes, TILs, realized oil, crude revenue, total revenue, LOE, production taxes, EBITDA; TILs count on the 61–70 bracket, while the seven-quarter grid scores the 61 point estimate (−7.6%). Chord's 2Q26 hedge settlement is inferred as first-half cash settlements (−$88.8MM) less 1Q26 (−$18.5MM). Consensus figures (revenue $1.43B, adjusted EPS $6.68 vs $6.44 actual) are the Zacks consensus as carried at the print; other services had EPS at $6.75. The 2Q26 cutoff is 20 JUL 2026, with April and May filed and June not; for earlier quarters the cutoff is the same two-months-filed point.
The seven quarters run the identical prediction for 4Q24 through 2Q26 — every quarter whose base quarter sits clear of the Enerplus close (MAY 2024) — using each quarter's guidance from the prior release, each quarter's hedge book from the prior 10-Q or 10-K (including subsequent events disclosed there), and that quarter's prices. Grid cells are (predicted − reported) ÷ reported; hedges are predicted − reported in $MM. Averages are mean absolute miss.
What we'd push back on ourselves. Much of the table's accuracy comes from inputs anyone can multiply — Chord's own guidance ranges, posted prices, a hedge book printed in the 10-Q. Swap the state-filed volumes for the guidance midpoint and the seven-quarter oil miss is 1.3% instead of 1.4%: on volumes, the filings tie management's own forecast over two years and beat it only in the last three quarters (0.5% vs 1.6%), the three quarters Chord printed at or above the top of its range. Where the filings clearly add is the well count (9% vs 20% for the guided range) and the size of a beat when there is one. Point-in-time is asserted, not proven: the warehouse holds today's state of the filings, so filed months include late filings that were not in at the cutoff. The state series carries today's operator attribution for all history — XTO's wells read as Chord's throughout 2025; the rate ratio cancels most of that, and the 1H25 misses owe more to the winter, to volumes the state books a quarter later than Chord did, and to some 180 old wells that begin filing rows in April 2025. The spot average runs $0.30–0.80 above Chord's NYMEX reference in normal markets and $2.93 above it in 2Q26's backwardation, which is most of the realized-price and hedge miss. Gross ≠ net: new wells at lower working interest than the base make gross growth overstate net growth a little; non-operated volumes are invisible. The 3Q26 card's hedge arithmetic spreads the March 10-Q book evenly over the remaining 2026 quarters; the June 10-Q book replaces it in the October prediction. The prospective 3Q26 run, posted before the print, is the honest test.
Sources. Chord Energy 8-K earnings releases 3Q24 – 2Q26; 10-Q derivative notes SEP 2024 – MAR 2026 and 10-K notes DEC 2024, DEC 2025; 2Q26 call transcript 12 AUG 2026; ND Industrial Commission and Montana BOGC filings via the Crude Code DB; EIA spot prices. State data cut 02 SEP 2026, production through JUN 2026.