The letters.
One letter a week on the craft: how to publish more, get cited, and build a research career you own, wherever you started from.
Each issue below is the letter as it went to subscribers, with the date it was sent. Newest first.
Start here
New to publishing, or here for a first paper? Read these three first, in order.
-
The Methodology Debt
Start with the diagnosis: why guessing your way through a paper costs more than it saves, and where your own gaps are.
-
Who Goes on the Paper
Settle the author list before the writing starts: the four ICMJE criteria and the order conventions, in plain terms.
-
The Five Statistical Mistakes That Kill Papers
The five statistical errors reviewers flag most often, so your first submission does not fall at the analysis.
All letters
-
The Retraction Economy
Retractions reached a record high. Where they come from, how paper mills and fake peer review work, and four habits that protect your name as a co-author.
-
The Two Billion Dollar Toll
Spending on article processing charges at six large publishers nearly tripled from 2019 to 2023. What open access costs, why hybrid journals charge more, and how to plan for fees and waivers.
-
The Bias That Makes Every Treatment Look Good
Immortal time bias makes treatments look protective in observational data. How it arises, and how target trial emulation and a shared time zero prevent it.
-
AI Models Are Getting Smarter. Reporting Them Is Getting Worse.
TRIPOD+AI is the reporting standard for clinical prediction models. The five items machine learning papers miss most often, and how to design for them from the start.
-
Academic Compound Interest
Talent sets the rate, time does the compounding. Why early papers are worth the most, how to reinvest every return, and why one coherent research line matters.
-
How Volume Bends the Curve
Why volume compounds instead of adding up: repetitions build skill, shrink the cost of any single failure, and build a track record that brings the next opportunity.
-
The Invisible Graveyard
Survivorship bias in research: institutions study the papers that got published and miss the projects that quietly died. Where projects fail, and how to catch them.
-
The Five Statistical Mistakes That Kill Papers
Five statistical errors reviewers flag again and again: p-values without effect sizes, the wrong test, uncorrected multiple comparisons, causal language and low power.
-
Who Goes on the Paper
Who belongs on the author list and in what order: the four ICMJE criteria, the conventions for first and last author, gift authorship, and agreeing early.
-
One Project, Five Papers
Prolific researchers design one project to yield several distinct papers. How to plan the primary, subgroup, methods, review and perspective papers before you start.
-
The Methodology Debt
Guessing your way through papers builds a methodology debt that compounds quietly. What it costs, where to find yours, and how to start paying it down.
8 more issues went to subscribers only. New letters stay with subscribers for at least 21 days.
By topic
The same letters, grouped. Oldest first within each topic, so a topic reads in the order it was written.
Your first paper 4 letters
Methods and statistics 4 letters
Authorship 2 letters
Productivity and compounding 4 letters
Publishing and journals 2 letters
Research integrity 2 letters
Career and the IMG path 2 letters
AI in research 1 letter
The next letter goes to subscribers first. Join to get it.