Inversion is the mental model of approaching a problem backward: instead of asking what leads to success, define precisely what would guarantee failure and then avoid it. The lineage: the nineteenth-century mathematician Carl Gustav Jacob Jacobi solved hard problems by restating them in inverse form — man muss immer umkehren, invert, always invert — and Charlie Munger popularized the maxim as a decision discipline, arguing that many problems are easiest to solve backward and that avoiding standard stupidities outperforms seeking brilliance. Inversion has a team version with measured evidence: the premortem (Gary Klein, Harvard Business Review 2007) — assume the plan has already failed and independently write down why — whose mechanism, prospective hindsight, was shown by Mitchell, Russo and Pennington (1989) to raise correct identification of reasons for future outcomes by about 30%. The practical method: state the anti-goal (how would we guarantee this fails?), enumerate the failure routes concretely, then build the avoidances into the plan. The caution: inversion complements forward planning rather than replacing it — a plan made only of avoidances has no direction. In an argument tree, inversion is structural: the con side of every discussion is a standing invitation to state the failure case, with evidence, before the decision is made.

Some problems are hardest exactly the way you are facing them. Inversion is the deliberate about-turn: instead of asking how to succeed, define exactly how you would fail — and avoid it. A nineteenth-century mathematician's trick, an investor's favorite discipline, and the one mental model whose team version comes with a measured effect size.
Last updated: 2026-08-24
Inversion means working a problem backward: state the anti-goal (how would we guarantee failure?), enumerate those routes concretely, and build the avoidances into the plan. The maxim is the mathematician Carl Jacobi's — man muss immer umkehren, invert, always invert — popularized by Charlie Munger as a decision discipline. Its team version, the premortem, has measured evidence: prospective hindsight raises correct identification of failure reasons by about 30% (Mitchell, Russo & Pennington 1989; Klein, HBR 2007).
Carl Gustav Jacob Jacobi, one of the nineteenth century's great mathematicians, was known for cracking stubborn problems by restating them in inverse form — captured in the maxim attributed to him: man muss immer umkehren, one must always invert. Charlie Munger carried the maxim into decision-making and made it a pillar of his latticework of mental models (see Poor Charlie's Almanack): many problems, he argued, are easiest to solve backward — and a large share of long-term success comes not from brilliance but from systematically avoiding the standard ways of failing. The logic is asymmetric and that is the point: the routes to success are many, uncertain and contested; the routes to guaranteed failure are few, concrete and largely agreed on. Never talking to users, never publishing, never testing the key assumption — nobody debates that these destroy projects. Inversion converts that easy consensus into an actionable plan: avoid the known destroyers first, and you have cleared the field for the harder forward question.
Inversion runs the same way solo or in a team; the team version adds one load-bearing rule.
Phrase the failure explicitly: it is 18 months from now and this initiative has failed completely. The concreteness of the frame is what unlocks the concreteness of the answers.
Each person writes their failure causes before any discussion. Independence is the load-bearing rule: it keeps anchoring and authority pressure out of the lists, which is where the premortem's measured advantage comes from.
Convert each credible failure route into a design constraint, a tripwire, or a monitoring point. An inversion session that ends as a list of worries changed nothing; one that ends as plan changes did the job.
Inversion complements forward planning; it does not replace it. A plan made only of avoidances has no direction — you still need the forward case, which is why inversion pairs naturally with first principles (build the case) and second-order thinking (trace its consequences). And note the research nuance from the dissent literature: staged opposition underperforms — Nemeth's studies found assigned devil's advocates trigger bolstering while authentic dissent broadens thinking — which is why inversion works best as everyone's independent exercise rather than one person's role. In an argument tree, that is the default:
Every discussion carries an explicit against structure — the failure case has a permanent, legitimate home, not a social cost.
Where enabled, failure routes can be contributed without career risk — the condition the premortem tries to create, made structural.
Each projected failure route is an argument with support, so the credible risks separate from the speculative ones on the record.
Documented decisions show which failure routes materialized — the organization's own history becomes its best anti-goal checklist.
Inversion by format: the question how would this fail? is asked in every discussion, by the structure, before the decision — not after it, by the postmortem.
The evidence-graded guide to the seven models worth knowing — inversion among them.
The companion model: build the case from fundamentals, then invert it.
Forward through consequences — inversion's mirror image.
Where the premortem evidence and the authentic-dissent research live in full.
Includes Munger's invert-always-invert as a questioning discipline.
Inversion's rhetorical cousin: strengthen the opposing case before answering it.
The hub of this cluster — the process, the types, and every method for reaching a better decision.
Instead of asking how to achieve a goal, ask how you would guarantee failure — then avoid those routes. Failure paths are fewer, more concrete and easier to agree on than success paths, so the backward question produces actionable answers faster. The maxim comes from the mathematician Carl Jacobi — invert, always invert — popularized by Charlie Munger.
The maxim man muss immer umkehren — one must always invert — is attributed to the nineteenth-century mathematician Carl Gustav Jacob Jacobi, who solved difficult problems by restating them in inverse form. Charlie Munger carried it into business and investing as a core model in his latticework, documented in Poor Charlie's Almanack.
The premortem is inversion as a team ritual, described by Gary Klein in Harvard Business Review (2007): assume the plan has already failed spectacularly, then have each person independently write down why. Its mechanism, prospective hindsight, has measured evidence — Mitchell, Russo and Pennington (1989) found it raises correct identification of reasons for future outcomes by about 30%. The independent-writing step is what keeps authority and anchoring out of the lists.
No. Pessimism predicts failure; inversion prevents it. The exercise ends with plan changes — constraints, tripwires, monitoring — not with a mood. It also is not a replacement for forward planning: a strategy made only of avoidances has no direction, which is why inversion pairs with forward models like first principles and second-order thinking.
The research says no. Nemeth's studies (2001) found assigned devil's advocates tend to trigger cognitive bolstering — groups generate more reasons they were right — while authentic dissent broadens thinking. Inversion works best as everyone's independent exercise (the premortem's design) or as a structural feature of the discussion format, rather than as one person's staged role.
Give the failure case a structural home. In an argument tree, the con side of every discussion is a standing, legitimate place for failure routes — each entered as an evidenced argument, contributable anonymously where enabled, and preserved in the record so the next decision inherits the last one's lessons.
Munger, C. T. (2005). Poor Charlie's Almanack (ed. P. Kaufman)
Munger on Jacobi's maxim and inversion as a pillar of the mental-models latticework.
Klein, G. (2007). Performing a Project Premortem. Harvard Business Review, September 2007
The team inversion ritual, as described by its author.
View source →Mitchell, D. J., Russo, J. E. & Pennington, N. (1989). Back to the future: Temporal perspective in the explanation of events. Journal of Behavioral Decision Making, 2(1), 25–38
The prospective-hindsight experiment behind the ~30% figure.
View source →Nemeth, C., Brown, K. & Rogers, J. (2001). Devil's advocate versus authentic dissent. European Journal of Social Psychology, 31(6), 707–720
Why staged opposition underperforms — the caution in the practice section.
View source →Argumentree's con side is standing inversion: every discussion asks how this could fail — with evidence, before the decision.
Start Free Trial