Most incorrect energy claims in circulation are not fabrications. They are true statements that have lost the qualifier that made them true, somewhere between the original dataset and the headline. The qualifier is usually a boundary, a base year, a unit or a date.
This page sets out a six-step check that catches the majority of them. It needs no subscription, no specialist software and no technical background, and once practised it takes well under ten minutes.
1. Find the original publisher
Work backwards until you reach a body that generated the number rather than repeating it. Most figures reach you at third hand: a consultancy produced it, a think tank cited it, a newspaper reported the think tank, and a social post reported the newspaper. Each hop drops a qualifier.
If you cannot reach an original publisher, that is itself the finding. A number with no traceable origin should not be used, regardless of how often it appears.
2. Establish the boundary
This is the single largest source of apparent contradiction in energy debate. For a barrel of oil, you can count emissions from extraction only, from extraction plus processing and transport, or from the full lifecycle including combustion by the end user. These differ by roughly a factor of five.
Under international inventory rules, Canada counts emissions from producing a barrel; the country that burns it counts the combustion. Both accounts are correct and they are answering different questions. A great many arguments about whether Canadian oil is "clean" are boundary disputes that neither party has identified as such.
3. Check the base year
Percentage reductions are meaningless without the year they are measured from. A target of 75 per cent below 2014 levels is a very different commitment from 75 per cent below 2005 levels, and sectors whose emissions peaked at different times can claim wildly different progress by choosing favourably. Whenever you see a percentage change, find the base year before you accept the comparison.
4. Resolve the units
Energy reporting mixes units relentlessly. Megawatts measure capacity; megawatt-hours measure energy delivered. A gigawatt of solar and a gigawatt of gas do not produce the same annual output — capacity factors differ by a factor of three or more. Barrels per day, cubic metres, gigajoules, tonnes of CO2 and tonnes of CO2-equivalent all appear in the same argument routinely.
Capacity quoted where output is meant, or the reverse, is the most common unit error in electricity coverage and is often innocent.
5. Date the figure
Find when the number was current, not when the page was published. Evergreen web pages carry old figures indefinitely with no visible timestamp. Carbon prices, production volumes, commodity benchmarks and regulatory statuses all move fast enough that a two-year-old figure can invert an argument.
6. Identify who benefits from the framing
Last, not first. This step is a tie-breaker when the preceding five leave genuine ambiguity, and it is the step most often abused — used to dismiss a claim without doing the other five.
A number produced by an interested party is not thereby wrong. Industry-funded research and foundation-funded research are both frequently accurate. Interest tells you where to look harder, which assumptions to examine, and what has probably been left out. It does not settle anything on its own.
A worked example
Take a claim in the form: Canadian oil is among the cleanest in the world.
Origin. Trace it and you generally reach lifecycle intensity studies comparing crude grades by emissions per barrel produced.
Boundary. Production-side only. Combustion, which dominates the lifecycle, is excluded because it occurs wherever the fuel is burned.
Base year and date. Oil sands intensity has fallen materially over two decades, so which year the comparison uses changes the answer.
Units. Intensity per barrel, not total emissions. A producer can reduce intensity while increasing total emissions through higher volume. Both are true simultaneously.
Verdict. The claim is defensible on production intensity against some comparators, and unsupportable as a statement about total climate impact. Anyone using it for the second meaning has changed the question. Anyone dismissing it entirely has ignored a real improvement in production intensity.
The compressed case
Most claims now reach people as a screenshot, a chart without axes, or a single sentence with no link. The six-step method assumes a traceable source, and frequently there is not one.
Three shortcuts help in that situation. First, search a distinctive phrase from the claim — most viral energy claims have an origin document, and finding it usually takes seconds. Second, check the magnitude against something you already know: if a claim implies a province generates more electricity than seems plausible, it is probably a unit error. Third, if the claim has no traceable origin after a couple of minutes, treat it as unusable rather than as either true or false.
That last habit is the most valuable and the hardest. An untraceable claim is not evidence for anything, including for the position it appears to contradict.
Where to check
For Canadian claims, four free sources cover nearly everything. The Canadian Centre for Energy Information for production, trade and consumption statistics. The National Inventory Report for emissions by sector and province. Provincial regulators for project-level approvals and conditions. And the federal grants and lobbying registries when the question is who funded a piece of research. Each gets its own page in section three of this atlas.
Sources
- Environment and Climate Change Canada, National Inventory Report methodology.
- Canadian Centre for Energy Information, energy-information.canada.ca.
- IPCC national greenhouse gas inventory guidelines, on territorial accounting.
Organisational descriptions reflect each body’s own published material and independent reporting as of September 2026. Funding arrangements and mandates change; check the primary source before relying on anything here.