Energy transition analysis is littered with evidence that is real and stories that are not. A pilot exists, therefore the market is forming. A government grant lands, therefore the technology has been validated. An offtake agreement appears, therefore customers must be willing to pay. A large company issues a press release, therefore deployment is imminent. That chain of reasoning shows up constantly in hydrogen, carbon capture, synthetic fuels, small modular reactors, long-duration storage, aviation fuels, maritime fuels, and every other corner of climate-tech where a hopeful story can get ahead of the operating evidence. The useful habit is counterinduction: when a fact is presented as evidence for a climate-tech story, ask whether it proves the opposite. Does the pilot show readiness for scale, or how much special treatment is needed? Does the grant validate the technology, or reveal that private capital would not carry the risk? Does the offtake agreement prove market demand, or just corporate signaling at an undisclosed price? The full TFIE Strategy Briefing analysis works through the method in more detail, but the core move is simple: keep the fact, challenge the story attached to it. Take a pilot project. A hydrogen bus trial can be real. The bus can move. The depot can dispense hydrogen. The press photos can be crisp, and the mayor can say something suitably optimistic beside a ribbon. None of that proves that hydrogen buses are a good transit procurement strategy. The counterinductive question is whether the pilot is showing readiness for scale or showing how much special treatment is required to make a small fleet function. If the answer involves high fuel costs, low utilization, bespoke maintenance, fragile supply chains, and no repeat procurement without subsidy, the pilot is not evidence of an emerging market. It is evidence of a machine for converting public money into press releases. This is not cynicism. Cynicism assumes the answer before looking. Counterinduction is more disciplined than that. It treats the evidence as real, then tests whether the interpretation survives comparison with a better explanation. A grant may mean a government has identified a serious decarbonization pathway. It may also mean a ministry needed a politically attractive announcement, a company found non-dilutive capital, or a weak technology needed public money because private buyers would not carry the cost. The same fact can support very different stories. Offtake agreements are another good example. Climate-tech companies love announcing them because they sound like market demand. Sometimes they are. But the useful questions are less exciting than the headline. Is the agreement binding? Is there a price? Is there a delivery schedule? Is there a penalty for non-performance? Is the buyer taking product because it is cheaper and better, or because it helps with public commitments, compliance positioning, or option value? A non-binding memorandum for future synthetic fuel at an undisclosed price is not the same thing as a customer buying diesel, electricity, steel, or ammonia in a competitive market today. Carbon removal has produced a lot of this. A company can sell durable removal credits to a wealthy corporate buyer, and that can be a valid transaction. It does not automatically prove that the pathway is energy-realistic, climate-positive, scalable, or a good use of scarce clean electricity. The counterinductive test asks whether the purchase proves commercial readiness, or whether it proves that reputational demand from high-margin firms can support a boutique market at prices that have little to do with climate-relevant deployment. That distinction matters. My Carbon Engineering assessment kept returning to the same point: direct air capture claims only mean anything when tested against energy source, lifecycle emissions, permanence, end use, cost, and what the pathway displaces. Storage has the same problem, although the details differ. A long-duration storage project announcement can sound like evidence that a novel technology has solved a grid problem. Sometimes it may have found a real niche. But the counterinductive question is whether the project has to beat the actual comparator in the actual place. In Ontario, for example, Hydrostor’s proposed compressed-air storage project is aimed at a real grid constraint, which matters. But it still has to beat batteries on cost, efficiency, schedule, execution risk, and bankability. A real grid need does not automatically validate every technology pointed at it. Counterinduction is especially useful because energy transition debates are full of category errors. Technical feasibility is treated as commercial proof. Government interest is treated as market demand. A small project is treated as scale. A corporate strategy slide is treated as procurement. A cost target is treated as a delivered price. A future fuel is compared with today’s incumbent instead of tomorrow’s electrified alternative. These mistakes are not subtle, but they are persistent because they flatter the people who want the story to be true. This is why denominator work matters. A claim about cost has to name the unit. Cost per kilogram is not cost per useful kilometer. Cost per MWh of storage capacity is not delivered grid value. A tonne of captured CO₂ is not automatically a tonne of durable net removal. A vessel with a new fuel system is not necessarily a competitive ship. A pilot flight on synthetic fuel is not a pathway for aviation unless feedstock, electricity, capital, certification, and airline willingness to pay all survive the arithmetic. Good counterinduction is not reflexive negativity. It does not say new things never work. Solar worked. Wind worked. Batteries worked. Heat pumps work. Electric buses work when routes, depots, charging, and procurement are handled sensibly. The difference is that these technologies kept improving under the pressure of real deployment, manufacturing learning, cost decline, operational feedback, and repeat buyers. Their evidence base became boring in the best possible way. It moved from “look at this pilot” to “look at this procurement cycle.” That is the bar weak climate-tech claims keep trying to avoid. They want a grant treated like a market, a demonstration treated like infrastructure, and an announcement treated like evidence. Counterinduction is a way of refusing that shortcut. It asks whether the claim has survived contact with the comparator, the customer, the supply chain, the operating environment, and the cost denominator. Policymakers, investors, journalists, and procurement teams do not need to be hostile to new technologies. They need to be harder to impress. When a climate-tech claim arrives with a neat story and a shiny data point, the first question should not be whether the story is attractive. It should be whether the same data point makes more sense as evidence of the story’s weakness. That small reversal would prevent a lot of bad money, bad policy, and bad infrastructure from being dressed up as transition strategy.