The inchworm model — why harder problems mean real progress

Six months into a disciplined review practice, you notice something discouraging: the problems haven't stopped. You've fixed the obvious errors — the impulsive decisions, the skipped protocols, the emotional reactions. But new problems have appeared. Subtler ones. Harder to detect. Harder to fix.

It feels like you're not making progress. You are. You're just not recognizing what progress looks like from the inside.

The inchworm illusion

An inchworm moves by extending its front end forward, then pulling its back end up to meet it. At any given moment, one end is stationary while the other moves. If you only look at the stationary end, it looks like the inchworm is stuck.

Performance development works the same way. At any given moment, some part of your process is solid (the back end has caught up) and some part is stretching into new territory (the front end is reaching forward). The problems you're currently facing are at the front end — the new territory. The problems you used to face are at the back end — the solved territory.

From inside, you only feel the current problems. The solved ones are invisible because they no longer cause pain. So the experience is: "I always have problems." The reality is: the problems are different — and the fact that they're different means you've moved.

The sophistication ladder

Problems in any performance system follow a predictable sophistication curve:

Level 1: Mechanical failures. Skipping protocols. Missing checklists. Forgetting to document. No defined exit criteria. These are basic errors — failures of compliance, not of judgment. They're fixed with architecture: better checklists, better triggers, better workflow design.

Level 2: Emotional intrusions. Acting on fear. Chasing excitement. Revenge decisions after losses. Overcommitting after wins. These are state-based errors — the right process exists but gets overridden by emotional activation. They're managed with state checking, cooling periods, and physiological monitoring.

Level 3: Cognitive distortions. Confirmation bias driving analysis. Anchoring to irrelevant reference points. Resulting — evaluating decisions by outcomes. These are perception errors — the process is followed and the state is managed, but the thinking itself is biased. They're addressed with structured debiasing protocols.

Level 4: Belief-level patterns. The knowing-doing gap. Identity-driven decisions. Off-process behavior reinforced by intermittent success. These are deep-structure errors — the process, the state management, and the debiasing are all in place, but an underlying belief is generating the same behavior through new channels. They require belief-level work.

Level 5: System limitations. Edge cases the system wasn't designed for. Situations where the rules conflict. Environments where the assumptions underlying the system no longer hold. These are design errors — not failures of the person running the system, but limitations of the system itself. They require system evolution.

The inchworm moves up this ladder. Level 1 problems get solved, and Level 2 problems become visible. Level 2 problems get managed, and Level 3 problems emerge. Each level feels like a new set of failures. In reality, it's a new class of challenges at a higher altitude.

Why this feels like failure

The feeling of perpetual failure comes from three sources:

Recency bias in self-assessment. You're more aware of your current problems than your solved ones. The Level 3 issue you're struggling with today is vivid. The Level 1 issue you solved six months ago is forgotten. Your self-assessment is weighted toward current pain, not toward cumulative progress.

Rising standards. As your baseline improves, your standards rise. The behavior that seemed acceptable at Level 1 (occasionally skipping a checklist) would be alarming at Level 3. You're holding yourself to higher standards — and then judging yourself for not meeting them, without noticing that the standards themselves are evidence of progress.

Invisible solved problems. Solved problems produce no signal. When was the last time you noticed that you didn't make an impulsive decision? You didn't — because not-making-a-mistake is invisible. The absence of a problem generates no feedback, no satisfaction, no sense of achievement. It just... doesn't happen. And because it doesn't happen, it doesn't register as progress.

Measuring inchworm progress

Standard performance metrics are poor at capturing inchworm progress because they measure results, not problem sophistication. A month with no errors could mean you've solved all your problems (unlikely) or that the market was easy and your problems weren't tested (more likely).

Better metrics for inchworm progress:

Problem classification. In each review, classify the issues you encountered by level (1-5). Track the distribution over time. A healthy inchworm shows a leftward shift — fewer Level 1 and 2 issues, more Level 3 and 4 issues. The total number of problems might stay constant. The type of problems should be climbing.

New problem detection. How often are you encountering a genuinely new problem — one you haven't seen before? Early in development, new problems are frequent. As the system matures, new problems become rarer but more nuanced. The rate of genuinely new problems — not recycled versions of old ones — is a signal of frontier expansion.

Solved problem stability. Are your Level 1 and 2 solutions holding? Or are old problems recurring? If solved problems stay solved, the back end of the inchworm is stable. If they regress, the inchworm is losing ground — which usually indicates that the system upgrade wasn't structural enough or that stress is overwhelming the defenses.

Changelog sophistication. Look at your system changelog. Are the recent entries addressing more sophisticated issues than the early ones? A changelog that moves from "added basic checklist" to "refined contamination gate thresholds based on 6-month data" shows clear inchworm progress — the problems being solved are at a higher level.

The plateau illusion

There's a variant of the inchworm illusion that's even more discouraging: the plateau.

You've been at Level 3 for months. The same class of problems keeps appearing. No matter what you try, the issues persist. It feels like you've hit a ceiling — your development has stalled and you're stuck.

Two possibilities:

Genuine plateau. You've reached the limit of your current approach and need a fundamentally different intervention. Level 3 problems (cognitive biases) don't respond to Level 2 solutions (state management). If you're applying the wrong level of intervention, you'll plateau until you shift approaches.

Invisible consolidation. You're actually making progress, but the gains are happening at a scale that's too fine to notice session-by-session. The frequency of Level 3 errors has dropped from 8 per month to 5 per month — but 5 still feels like "a lot," and you don't notice the improvement because each individual error still feels significant.

The distinction matters: genuine plateaus require a change in approach. Invisible consolidation requires patience. Confusing the two leads to either premature system overhaul (when patience was needed) or prolonged stagnation (when a new approach was needed).

The diagnostic: are the same errors repeating in the same conditions, or are the conditions subtly different each time? Same conditions = genuine plateau (the solution isn't working). Different conditions = invisible consolidation (the solution is working, but the problem appears in new forms).

The reframe

The inchworm model asks you to reframe your relationship with problems entirely.

Old frame: "I still have problems. I'm not good enough." New frame: "I have different problems than I had six months ago. My problems are more sophisticated. That's the definition of progress."

This isn't positive thinking. It's accurate assessment. If your biggest issue last year was skipping the checklist and your biggest issue this year is detecting subtle confirmation bias in your analysis — you've climbed three levels. That's not stagnation. That's development.

The inchworm never reaches the top. There's always a next level, a new class of challenge, a more sophisticated problem to solve. That's not a discouraging truth — it's the nature of compounding improvement. The problems never end. But they do get better.

The person who says "I have no more problems" has stopped looking. The person who says "my problems have changed" has started growing.