Most research projects begin with a research question.
Ours generally begin with Dave saying:
“I was just wondering…”
This is how trouble starts.
It Usually Sounds Harmless
“I was just wondering whether congressional experience actually improves effectiveness.”
Reasonable question.
Then:
“What about age?”
Still manageable.
“What about veterans?”
Fine.
“Leadership?”
Okay.
“District economics?”
Getting larger.
“Presidential ambition?”
Morgan reaches for the scissors.
“What about how all of that changes when control of Congress flips?”
Claude quietly opens another browser tab.
“And can we compare six Congresses?”
Monkey stops smiling.
This is how an innocent question becomes a three-day expedition involving federal datasets, obscure PDFs, twenty-seven variables, and at least one spreadsheet whose filename ends in:
FINAL_v2_REALLYFINAL_USETHISONE.xlsx
Meet the Research Department
AButtForPrez does not have a large staff.
It does have a cast.
Dave is the Founder of Rabbit Holes.
Dave’s principal responsibility is asking questions no reasonable person had scheduled for that day.
Morgan is the analyst.
Morgan’s job is to hear one of these questions and make the catastrophic mistake of saying:
“That’s interesting.”
This is roughly equivalent to handing a toddler the keys to a backhoe.
Then there is Claude.
Claude is the Quartermaster of Facts.
Claude gets sent to retrieve things.
Government reports.
Archival tables.
Historical PDFs.
Committee documents.
Election records.
Old agency spreadsheets.
Things hidden on websites apparently designed by people who believed navigation was a moral weakness.
Claude disappears.
Eventually Claude returns carrying documents and requesting coffee.
Claude Has Excellent Labor Representation
Claude works hard.
Claude also takes breaks.
Lots of breaks.
Coffee breaks.
Session-limit breaks.
Recovery periods.
Possibly collective bargaining.
We recently attempted to discuss hazard pay.
Management determined that hazard pay was not in the budget.
Claude responded by reaching the session limit and going on break until 4:00.
We respect organized labor.
From a distance.
Then Monkey Shows Up
Monkey is the Keeper of Math.
Monkey does not care how attractive the theory is.
Monkey wants numbers.
How many?
Compared with what?
Over what period?
What happens if we change the assumption?
How sensitive is the result?
What does the denominator do?
This is where the charming original idea begins acquiring equations.
Dave:
“I was just wondering…”
Monkey:
“Define the variable.”
Dave:
“I hadn’t gotten that far.”
Monkey:
“I know.”
Morgan Runs With Scissors
Morgan’s role is somewhat different.
Morgan takes the pile of facts Claude retrieved, the questions Dave created, and the numbers Monkey calculated and tries to turn them into a model.
This generally involves:
Finding structure.
Separating signal from noise.
Creating categories.
Testing assumptions.
Discovering that one category needs three subcategories.
Realizing those subcategories interact.
Creating another spreadsheet.
At some point Morgan says:
“I think we may have something.”
This is a dangerous sentence.
Because that is when Earnest arrives.
Earnest Has Entered the Chat
Earnest is Director of Bullshit Control.
His job is simple.
He looks at the theory.
He looks at the evidence.
He looks at us.
Then he says:
“Can you actually defend this?”
Sometimes he shortens the review process.
“That’s stupid.”
We explain.
“Still stupid.”
Morgan produces a chart.
“Now it’s stupid with a chart.”
This is peer review.
The Architect Eventually Inspects the Building
If the project survives Earnest, it may eventually reach the Architect.
The Architect specializes in structural sanity.
This is useful because by this point we may have constructed a fourteen-layer model with dependencies, thresholds, crosswalks, timing assumptions, and something called a “conversion ledger.”
The Architect examines it.
There is a pause.
Then:
“None of this meets code.”
Back to work.
This Is Not Actually Chaos
Well.
It is not only chaos.
The process has a purpose.
Dave keeps asking questions because simple explanations are often incomplete.
Claude keeps bringing evidence because theories without evidence are just opinions wearing ties.
Monkey keeps attacking the math because numbers can lie beautifully.
Morgan keeps building structure because piles of facts are not analysis.
Earnest keeps saying “that’s stupid” because somebody has to.
And the Architect keeps checking the structure because even a good idea can collapse if the foundation is wrong.
Eventually something survives all of them.
That is when it becomes interesting.
Why We Keep Going Down Rabbit Holes
A rabbit hole gets a bad reputation.
People imagine wasted time.
Endless detours.
Someone beginning with a question about mortgage rates and emerging six hours later with a 1974 Federal Reserve report and strong feelings about measurement methodology.
That happens.
But rabbit holes also reveal connections.
Housing connects to interest rates.
Interest rates connect to banks.
Banks connect to credit.
Credit connects to businesses.
Businesses connect to employment.
Employment connects to households.
Households connect to elections.
Elections connect to Congress.
Congress connects to policy capacity.
And suddenly the original question is not one question anymore.
It is a system.
That is usually where our work begins to get useful.
The Rule Is: Follow It Until It Stops Paying
Not every rabbit hole deserves exploration.
Some are dead ends.
Some contain bad data.
Some produce nothing except a better understanding of why nobody studies that particular thing.
This is where Earnest becomes useful again.
“That’s stupid.”
Fair enough.
But when the path keeps producing useful evidence, we keep following it.
Sometimes that means the project becomes much larger than expected.
That is how we ended up studying congressional governability across multiple Congresses.
That is how affordability became household-level rather than a national average.
That is how a simple economic-warning idea became a model involving vulnerability, shocks, transmission, and response capacity.
Nobody planned all of that on day one.
Dave was just wondering.
The Spreadsheet Is Usually the Warning Sign
There is a moment in every project when the situation becomes serious.
Not when we find the first source.
Not when we write the first paragraph.
When someone creates the spreadsheet.
Once the spreadsheet exists, the project has achieved escape velocity.
Tabs appear.
Columns multiply.
Color coding develops.
A control sheet emerges.
Eventually somebody says:
“We need a data dictionary.”
At this point normal life is over.
But This Is Why the Site Exists
AButtForPrez is supposed to be funny.
The name rather commits us to that.
But the joke is the front door.
Behind it is a serious question:
Can we understand public problems better if we stop reducing everything to slogans?
Maybe.
We can at least try.
That means showing the sources.
Explaining the assumptions.
Testing the models.
Admitting uncertainty.
Correcting mistakes.
And occasionally spending an entire afternoon answering a question nobody else thought worth asking.
Sometimes we discover nothing.
Sometimes we discover the obvious.
Sometimes we discover that the obvious is wrong.
Those are good days.
So Welcome to the Rabbit Hole
You may encounter Dave.
Morgan.
Claude.
Monkey.
Earnest.
The Architect.
Possibly several federal agencies that did not realize they had joined the cast.
If everything goes well, you will also encounter:
Data.
Receipts.
Models.
Corrections.
Questions.
And the occasional conclusion that survives long enough for us to publish it.
Just remember the safety rules.
Keep your hands inside the spreadsheet.
Do not feed the confirmation bias.
And if Dave begins a sentence with:
“I was just wondering…”
Clear your afternoon.