Coal & Consumable Fuels Data Provider · Head-to-head
EIA Coal Data (production, consumption, prices, reserves) vs EIA State Energy Data System (SEDS)
Which coal & consumable fuels data provider data fits your job: EIA Coal Data, or EIA State Energy Data System. API, files, or your warehouse. Daily, weekly, or hourly.
EIA Coal Data (production, consumption, prices, reserves)
EIA State Energy Data System (SEDS)
Where the fields line up
1 shared field — join on these.
| Field | EIA Coal Data | EIA State Energy Data System |
|---|---|---|
state | State facet on series disaggregated below the national line - the ladder rung that turns a US total into a Wyoming, Illinois Basin or Appalachian question. | documented |
Coverage, side by side
| EIA Coal Data | EIA State Energy Data System | |
|---|---|---|
| Geographic | United States at four depths: national, state, county and individual mine | All 50 states, District of Columbia, US territories and a US total |
| Temporal | Series-dependent: mine-level detail from the 1980s, national totals older, running through current periods | 1960-2024 for production, consumption and CO2 emissions; 1970 onward for prices and expenditures |
What each contains
They tie on 2 attributes. Pick by fit, not by loyalty.
| EIA Coal Data | EIA State Energy Data System | |
|---|---|---|
| Publisher | U.S. Energy Information Administration (EIA) | U.S. Energy Information Administration (EIA) |
| Subject lens | Coal only - reserves, production, prices, employment and productivity, distribution, stocks, imports and exports | All energy sources by state with coal among them, adding expenditures and CO2 emissions |
| Geographic coverage | United States at four depths: national, state, county and individual mine | All 50 states, District of Columbia, US territories and a US total |
| Temporal coverage | Series-dependent: mine-level detail from the 1980s, national totals older, running through current periods | 1960-2024 for production, consumption and CO2 emissions; 1970 onward for prices and expenditures |
| Frequencies | Annual, quarterly and monthly, set per series | Annual |
| Detail level | Facet-based series: state, sector, coal rank, mine type | Long-format matrix: state x energy source x sector x year, keyed by five-character MSN codes |
| Field documentation | Six documented fields, confidence graded inferred during research | Eight documented fields, all verified, plus a codebook covering every MSN code |
| Best for | Recency and sub-state resolution: mines, ranks, current-period prices, trade flows | Decade-scale state panels with emissions and expenditure context |
What each does better
EIA Coal Data
SEDS answers at annual grain, and its freshest figures arrive flagged preliminary besides.
Resolution below the state line. Counties and individual mines, surveyed since the 1980s - the only level at which questions like which basin, which operation, which seam have answers.
Trade and logistics. Imports, exports, shipments and receipts at electric power plants give the commodity a physical supply chain; SEDS consumes coal at the state boundary and stops.
EIA State Energy Data System
Depth. 1960. Wyoming mined 35,173 thousand short tons that year, peaked past 4,133,004 in 2023, settled to 3,333,635 in 2024 - one row, same code, same units, sixty-five unbroken years. Coal Data's deepest series reach the 1980s at mine level and nothing earlier. For why that length matters, see country-year panel.
All eight SEDS fields verified during research, against 1,495 of 1,744 records catalog-wide.
The system behind it is described at state energy data system.
Where they're equivalent
More than the architecture clash suggests. Same publisher, same federal statistical discipline, both records verified during the August 2026 review. Both ship structured tables with documented columns rather than charts hiding the numbers. Both score 9/10, placing them in the top band of a catalog averaging 7.81. And both treat estimation honestly: SEDS flags preliminary vintages inline, while Coal Data separates survey-based mine detail from modeled series. Where they disagree, it is because they were built for different questions - not because either is careless.
The verdict
Verdict: sample both, pick by fit - they are different instruments pointed at the same commodity.
Take EIA State Energy Data System (SEDS) if your question spans states or decades.
Investors and quants tend toward SEDS for factor panels and Coal Data for timing; market researchers and consultants usually want those priorities reversed, depending on whether the client sells into a basin or across a decade of demand.
Sample both, pick by fit. See EIA Coal Data · See EIA State Energy Data System
Or take both in one feed
Yes - they stack because they occupy different layers of one system. Let SEDS set the long-run backdrop: which states produce, burn and pay for coal, and how those curves moved after 1960.
Three alignments decide whether the merge holds. Identifiers reconcile easily - postal codes to state facets - until Coal Data drops below the state line, where SEDS has no counterpart rows. Units need conversion: dollars per short ton versus dollars per million Btu implies a heat-content assumption, so document yours. Vintage matters most: pair SEDS rows flagged final against Coal Data's current periods, and never mix a preliminary SEDS year into a concluded analysis. The shape will feel familiar to anyone who has worked with a long-format time series.
Datadory ships either record alone or both merged onto one calendar, delivered daily, weekly, or hourly - your call. Or take both in one feed.
API, files, or your warehouse. Daily, weekly, or hourly.
Fair questions
Do the two datasets cover the same ground?
Only partly. Everything below the state line - counties, mines, plants - or newer than the latest annual vintage belongs to Coal Data; everything before 1980, or beyond coal into expenditures and emissions, belongs to SEDS.
Which dataset reaches further back?
SEDS, decisively. EIA Coal Data's deepest series are mine-level surveys reaching back to the 1980s.
Can Datadory deliver both datasets together?
Yes - alone or merged onto one calendar, delivered daily, weekly, or hourly, your call. Each arrives normalized to its documented field dictionary, with sample rows for inspection before anything ships. The joining work is state-code alignment, unit reconciliation and vintage pairing, handled in the merge. Or take both in one feed.