It’s a fair point—until you follow it all the way through.

A defender of any accumulation-based teaching method could reasonably argue that just because someone leaves a program without graduating doesn’t mean the system failed them.

People leave learning systems for all kinds of reasons. Life gets in the way. Money runs out. Motivation fades. Priorities shift. Dropout rates alone don’t prove anything, and any honest analysis has to acknowledge that attrition isn’t automatic evidence of system failure.

But that defense has a hole in it.

When a system has high attrition and no clear endpoint and expects learners to discover their limits by hitting the wall, attrition stops being incidental. It becomes a symptom.

The “leaving doesn’t mean failure” defense only works if the system has other safeguards built in—ways to tell learners when they’re approaching overload, ways to distinguish between productive complexity and unsustainable accumulation, ways to signal “you’re done adding” before something breaks.

Without those safeguards, high attrition isn’t just people quitting. It’s people escaping.

Here’s where the logic gets uncomfortable for defenders.

If “leaving doesn’t mean the system failed,” then the same reasoning applies in reverse: graduating doesn’t mean the system succeeded.

You can’t have it both ways. You can’t say outcomes don’t reflect on the method when someone drops out, then turn around and claim full credit when someone finishes. That’s not analysis. That’s cherry-picking.

Either the method causes outcomes or it doesn’t. Pick one.

Once you accept that neither leaving nor staying settles the question, the conversation has to go somewhere else. Somewhere defenders typically don’t want to go.

The real question isn’t about the people who left. It’s about the people who stayed.

How many of them had to quietly adjust, ignore parts of the instruction, stop adding when the method said keep going, or build their own unofficial version of the system just to survive? And if that’s what success actually requires—is that adaptation taught? Supported? Even acknowledged?

Or does the method simply take credit for outcomes it didn’t engineer?

Survivors of any method can honestly say, “This works for me.” And it does—for the version they privately built. But that’s not the same as the method working. That’s the learner working. The system provided raw material. The learner figured out what to keep and what to quietly discard.

That’s not system success. That’s individual resourcefulness being mislabeled.

Systems that work well want to be measured. Measurement becomes proof. It becomes marketing. It becomes the answer to every skeptic. Confident systems publish outcomes because outcomes vindicate them.

Systems that seemingly avoid measurement are making a choice that only makes sense under one condition: they know the numbers will hurt more than help.

And if you want to go through the time, money, and hassle to try to get it, you’re going to need a FOIA request.

Schools have the data. They’ve always had the data. If the numbers helped them, you’d see them on every brochure.

That’s their strategy.

A method can say “leaving doesn’t mean failure.” It just can’t say it selectively.

If departures don’t disprove the system, then completions don’t prove it. And once that symmetry is acknowledged, there’s only one question left: what does the method itself do—explicitly, structurally, before breakdown—to help learners find their limits safely?

Until a method can answer that, “leaving doesn’t mean failure” is just another way of saying we don’t want you to know the real numbers.