I recently reread David Brooks’ essay “You Might Be a Late Bloomer” in The Atlantic.

What stayed with me was the temperament Brooks describes: the kind of person who does not move cleanly through the usual systems of achievement, but keeps circling a problem until it finally begins to take shape.

He describes people who are not especially good at following the expected path. People who wander, resist, self-teach, start over, and eventually discover that the thing they could not stop thinking about was the work.

The word that stayed with me was inquiétude.

I don’t want to overstate the connection. Brooks is writing about Cézanne and late-life achievement. I’m writing from a much smaller place: years of noticing the same kinds of problems in pulmonary function testing and not having a good way to organize them.

But the word helped because it names something I recognize: not inspiration, and not ambition, but the persistent discomfort of seeing a pattern before you can explain it well.

That was the early stage of the Apparent Certainty Framework for me.

I had observations, examples, objections, and instincts. What I did not yet have was a structure.

For a long time, that made the whole thing difficult to talk about. I could point to cases. I could describe situations. I could say, “This looks more certain than it should.” But I could not yet show the larger shape of the problem.

So the work kept circling.

Quality assurance. Standards. Reporting. Interpretation. Technologist practice. Physiologic variability. The difference between what a test process contains and what a final result preserves.

None of those pieces were separate, exactly. But I also did not yet know how to connect them without flattening them.

That is probably the part of Brooks’ essay that felt most relevant to me: the idea that some people do not arrive at their work by the cleanest route. They accumulate pieces before they know what the pieces are for.

That can look inefficient from the outside.

Sometimes it is inefficient.

But there is a difference between scattered attention and attention that has not yet found its organizing problem.

I think I spent years in that second category.

Brooks also writes about intrinsic motivation. That part matters to me because a lot of this work did not begin with an assignment, an academic program, a grant, or a formal role. It began because certain things kept bothering me.

Not in a dramatic way. More in the practical, recurring sense that the available explanation was not quite enough.

That is a strange position to be in. You can feel strongly that something matters while still being uncertain about how to describe it. You can be right about the presence of a problem and still be wrong, incomplete, or premature in your explanation of it.

That tension has been central to this project.

Brooks describes late bloomers as people who often have to teach themselves. He also points to the combination of a “high need for cognition” and “epistemic humility.”

That combination feels important.

The first part is familiar to me. I am often drawn to problems that seem straightforward until they don’t. I want to understand why the “obvious” explanation does not quite hold. I like the point where categories start to break down and a better structure becomes necessary.

The second part is harder, but probably more important.

Because if you are building something outside a formal pathway, you have to keep asking whether you are clarifying the problem or just becoming more attached to your own explanation.

That is where the Apparent Certainty Framework has been useful for me. It gave me a way to discipline the pattern instead of just reacting to it.

It turned a long-running discomfort into something I could test against examples, revise, and explain.

Not perfectly.

Not finally.

But better than before.

I don’t think the lesson of Brooks’ essay is that late bloomers are secretly superior, or that all wandering eventually becomes wisdom. Some wandering stays wandering. Some intensity is just intensity. Some objections are not insights.

But sometimes a long, inefficient path gives you access to something you probably would not have seen by moving more efficiently.

That is the part I keep thinking about.

The framework did not arrive because I sat down one day and decided to make a framework. It came from years of repeated contact with the same unresolved problem, followed by the slow realization that the problem needed a shape.

That is a different kind of development than I once imagined.

Less like invention.

More like finally organizing what had been accumulating for years.

Brooks quotes Van Gogh: “I am seeking. I am striving.”

That line does not feel triumphant to me.

It feels accurate.

The work is not finished. The framework is not final. I am still revising the language, the model, and my own understanding of what it is trying to show.

But the thing that used to be only a discomfort now has edges.

Not fixed.

Not finished.

But visible enough to work on.

This publication is moving from blownvalues.com to:

apparentcertainty.org

The change is mostly practical, but it also reflects where the project has been heading.

Blown Values started as a way to name a pattern I kept seeing in pulmonary function testing: results that looked clean, valid, and technically acceptable, but were still shaped by things the final report could not fully show.

Over time, the larger idea became clearer.

The problem was not just “questionable values.”

It was the appearance of certainty.

That is what the Apparent Certainty Framework is meant to describe: situations where complex, variable processes are compressed into outputs that look more stable, bounded, and reliable than they may actually be.

The original name still matters. “Blown values” captured the first visible form of the problem. But Apparent Certainty better captures the broader structure: how systems can produce confident-looking outputs even when important context, variability, or uncertainty has been reduced away.

So the site is getting a name that better fits the framework.

Nothing else is changing immediately. The focus remains pulmonary function testing, diagnostic quality, measurement uncertainty, and the quiet ways that systems can make uncertain things look more certain than they are.

New domain:

apparentcertainty.org

When I first shared the Apparent Certainty Framework, I knew the visual model would continue to evolve.

The framework itself has not changed. Its central idea is still this:

Results can look more certain than the processes that produce them justify.

But over the past several days, I’ve refined the interactive model so it communicates that idea more clearly.

The latest version is cleaner and more focused. I simplified some of the public-facing labeling, tightened the visual structure, and clarified the three distortion domains:

• Data Generation: how the signal is generated and acquired

• Model: how the signal is defined and interpreted

• Reduction: how the result is selected, calculated, and reported

The pathway, system modifier, and central tension of the framework remain intact. The goal was not to make the model more elaborate, but to make it easier to read and easier to understand.

This is still an early framework, and I expect it to keep sharpening as I work through its implications. But this version feels much closer to the visual language I was trying to find from the beginning.

You can explore the updated interactive model here.

Some results looked perfectly acceptable:

repeatable, technically valid, numerically clean.

And yet the certainty of the result often seemed greater than the process itself justified.

Not because the testing was necessarily “wrong.”

Not because standards had failed.

But because pulmonary function testing is a surprisingly complex procedural system involving:

- patient performance

- coaching

- timing

- maneuver selection

- interpretation

- reduction of variability into final reported values

Over time, I started organizing those observations into a conceptual framework I’ve been calling the Apparent Certainty Framework (ACF).

At its core, it’s an attempt to map how unresolved variability can become compressed into clean, authoritative-looking outputs.

This framework isn’t intended to undermine pulmonary function testing or existing standards. It’s an attempt to clarify how certainty is produced, reduced, and represented within a complex testing system.

If you want to explore deeper, here’s the interactive version.

In the last post, I wrote about a pattern I keep seeing in pulmonary function testing.

Results that look clean. Technically valid. Aligned with standards.

And still… something feels off.

Not obviously wrong. Not clearly flawed.

Just more certain than they should be.

I kept coming back to that.

It didn’t seem random. It showed up in a particular kind of situation—when a complex process gets compressed into a clean, final output.

By the time you see the result, most of what went into producing it is gone.

The coaching. The maneuver. The variability. The context.

What’s left is something that looks stable, bounded, and reliable. And because it looks that way, it’s treated that way.

At some point, I realized I needed a better way to think about it.

So I built a model.

Not to determine what’s right or wrong, but to map something simpler: how much confidence a result appears to carry relative to how much distortion may have been introduced along the way.

The idea is pretty simple.

Some results have low distortion and deserve high confidence. Those are the easy ones.

But others have high distortion and still appear to deserve high confidence. Those are the ones that tend to slip through, because nothing about them looks like a problem.

The model doesn’t tell you what the correct answer is. It doesn’t replace clinical judgment.

It just gives you a way to see when confidence might be coming from the structure of the system rather than the underlying signal.

I’ll walk through what this looks like in practice next.

I’ve always noticed how people often take something complex and turn it into a clean, confident statement.

Life doesn’t really work like that. Systems don’t work like that. Most things are messier than they look, shaped by multiple variables, small dependencies, and a lot of hidden structure.

We compress all of that into something simple—and then we act certain about it.

That always felt off to me. So it’s ironic that I built a model.

Because models are supposed to simplify. They’re supposed to reduce complexity into something you can point to and say, “this is what’s happening.”

But that’s not what this one is doing.

It’s not trying to explain everything or give the right answer. It’s trying to show what happens when we take a complex process and compress it into something that looks finished.

Most of the time, we don’t see the process. We see the output—the number, the interpretation, the final result. And it looks precise. It looks stable. It looks certain.

But that certainty doesn’t always come from the underlying reality.

It often comes from how the system is structured: how the data is generated, how it’s interpreted, how it’s reduced into a final form. Each step compresses a little more of the original complexity, until the result carries more confidence than the process behind it really supports.

That’s what the model is meant to surface.

Not what’s right or wrong, but where confidence and complexity start to drift apart.

I started thinking about this in pulmonary function testing, but the same pattern shows up in other systems with clean, final outputs.

I didn’t build a model to simplify reality.

I built a model to show how simplification can create the illusion of certainty.

Pulmonary function testing is strange. From the outside, it looks simple. A patient breathes into a machine. Numbers come out. A report is generated.

But the result doesn’t just come from the machine. It depends on how the test is performed. Effort, timing, coaching, judgment—small differences can change the result, even when the test technically meets criteria.

That part isn’t obvious if you’re not the one doing the test. And the system doesn’t make it obvious.

A fast, sloppy test and a careful, well-executed one can both produce reports that look acceptable. The numbers are there. The criteria are met.

Sometimes there are clues. The waveforms. Comments. Sometimes those comments say: “Testing met ATS standards for acceptability and repeatability.” Often when the testing didn’t.

So the report ends up carrying more confidence than the process behind it.

Most of the time, that doesn’t matter. But sometimes it does.

And when it does, the difference between a solid test and a marginal one isn’t always clear from the result alone.

Pulmonary function testing depends on skill. The system doesn’t reliably show it. That’s the problem.

Pulmonary function tests are strange.

They’re used to measure how well the lungs are working.

Most of the time, they don’t matter that much. The numbers are a little off. The effort isn’t perfect. The interpretation could go either way. Nothing really changes.

But sometimes they matter a lot. A surgical decision. A diagnosis. A trajectory. And in those moments, the result suddenly carries weight.

The problem is that pulmonary function tests don’t tell you which situation you’re in.

They present the same way every time. A clean report. Precise numbers. A definitive interpretation. Whether the result is trivial or consequential, it looks the same.

So we treat them like they’re always important. And we trust them like they’re always reliable.

But they’re not.

PFTs are low-stakes most of the time, high-consequence some of the time, and treated like high-certainty all of the time.

That’s the problem.

There are cases where the result doesn’t just pass.

It improves.

The numbers move.
The change looks meaningful.
The interpretation follows.

Something worked.

In pulmonary function testing, this often shows up when comparing pre- and post-bronchodilator results.

The assumption is simple.
Two measurements are taken under different conditions.
The difference reflects physiology.

But the comparison depends on something quieter.

The starting point.

If the pre-bronchodilator effort is inconsistent—non-repeatable, incomplete, or poorly performed—the baseline can be artificially low.

If the patient then “gets it” during the post-bronchodilator testing, the efforts improve.
The loops become more repeatable.
The maneuver is better understood.

The numbers increase.

Technically, both sets of results can pass.

But the comparison has changed.

Not because the physiology changed.

Because the baseline did.

The result looks like improvement.

But what’s being measured is not the same thing.

The test is valid.
The process is followed.

The comparison is the problem.

In the last post, I walked through an example of how selecting for larger ERV can shift the final result—without causing it to fail.

That kind of shift is easy to miss, because everything still looks technically correct.

There are cases where everything checks out.

The test meets criteria.
The results appear valid.
Nothing raises an obvious concern.

And still, something doesn’t add up.

These cases are easy to overlook, because there’s no clear error to point to.
The system is working as designed.

And that’s part of the problem.

What I’ve come to realize is that a result can be technically valid without fully representing what’s actually happening.

Not because of a single mistake—but because of how the test is performed, selected, and interpreted.

The ERV example is just one version of this.

A specific selection choice becomes a pattern.
That pattern shifts the relationship between variables.
And over time, it changes what the result actually reflects.

Not dramatically.
Not enough to fail.

But enough to matter.

I’ve seen this show up across different tests.

Small decisions—how something is performed, which efforts are selected, how a result is framed—can all shape the final output while still staying within acceptable limits.

This isn’t about bad testing.
And it’s not about obvious errors.

It’s about the space where everything looks correct—
but still doesn’t fully explain the patient.

Once you start to notice it, you see it more often.

Not as a failure of the system—
but as a reminder that what gets selected, and what’s actually happening, aren’t always the same thing.

There are cases where the result doesn’t just pass.

Over time, certain efforts are favored.

The effort looks strong.
The selection feels intentional.
One number improves.

In some labs, when reviewing slow vital capacity (SVC) maneuvers, the focus is placed on the largest expiratory reserve volume (ERV).

The reasoning seems straightforward.

A larger ERV is assumed to produce a larger, more accurate residual volume (RV).

So those efforts are selected.

Over time, that becomes the standard.

The “best” effort is the one with the largest ERV.

But ERV isn’t used in isolation.

It exists in relation to inspiratory capacity (IC)—and both factor into the final result.

When you consistently select for larger ERV, you can reduce IC.

And when IC decreases, the downstream values shift.

Not dramatically.
Not enough to fail.

But enough to change what the result represents.

The maneuvers still meet criteria.
The selection is consistent.
The result passes every check.

But the system has been nudged in a specific direction.

Nothing fails.
But something changes.

There’s a kind of result that doesn’t just sit on a page.

It changes what happens next.

In surgical evaluation, pulmonary function testing can be part of the workup.

The test can be done correctly.

The maneuvers can meet criteria.

The numbers can be calculated the way they are supposed to be.

Everything can check out.

The results can be interpreted using a standard set of reference equations.

Based on those values, a patient may appear not to meet the threshold.

They may appear too high-risk for surgery.

Nothing about the process has to be obviously incorrect.

The equations may be applied appropriately.

The calculations may be accurate.

The interpretation may follow established guidelines.

The result may pass every check.

But reference equations aren’t neutral.

They define what “normal” is.

They determine where a number falls — and what it means.

Change the reference, and you can change the conclusion, even if the underlying physiology hasn’t changed at all.

In that situation, the decision can change — without any change in the patient.

The result can look definitive.

Technically valid.

Defensible.

And still dependent on something most people never see.

Not completely wrong.

Not obviously flawed.

Just… almost right.

There’s a kind of result that doesn’t raise any alarms.

It looks right.
It passes.
It feels trustworthy.

Nothing about it stands out as wrong.

And that’s the problem.

I used to think the biggest risks in testing and measurement came from obvious errors—bad data, poor effort, clear mistakes.

But over time, the cases that stuck with me weren’t like that.

They were the ones that made it all the way through.

Everything checked out.

And still, something was off.

For a while, I didn’t have a good way to describe it.

When I was younger, I used to joke that the title of my autobiography would be The Stretchy Middle—that space where things aren’t clearly right or clearly wrong, but somehow still hold together.

At the time, I was just trying to get a laugh.

I didn’t realize how close it was to something real.

The more I paid attention, the more I realized:

the most difficult problems weren’t at the extremes.

They were in the middle.

Not completely wrong.
Not obviously flawed.

Just… almost right.

That’s where things start to go wrong.

This is an attempt to understand why.