This blog untangles the difference between measurement precision and trial efficiency in the era of digital health.

By examining parallels between youthful political beliefs and professional lessons in pharma, the essay demonstrates that impractical, intractable, and shocking outcomes often arise from extremes.  This leads to the conclusion that flexibility is key to innovation.

Whether you lead a baseball team or a statistics bureau, your career can hinge on results you don’t fully control. This piece explores two high-profile examples and extracts three timeless rules for leaders: know what’s working, fix what’s broken, and make sure those above you know the difference.

This essay warns that unplanned interim analyses, like walking out of a movie before its twist, mislead decision-making and fuel avoidable failures in drug development.

This is the story of a statistical method dismissed by many, embraced by me, and why it’s time we judge tools by what they’re for, not what they’re not.

A humorous and reflective look at how statisticians are perceived in relationships, ethics, and popular media, with a call to raise their profile as heroes rather than grim messengers.

A tale of two subgroups, one that failed and one that changed a label, revealing why logic, not just math, must drive subgroup thinking.

A call to rethink our regulatory interpretations, especially in early-phase research, and to refocus on patient impact, not just procedural perfection.

I still have trouble believing these events.

How a Quiet Violation of Statistical Logic Undermines Your Trial Outcomes

Learn why it’s best for quantitative disciplines to collaborate, not compete.

See how a sharp-fitting suit and a sharp-fitting model have more in common than you think.

Learn about the Lilly Clinic: a place that had impact.

Moving to a more "Macro" topic. See if this logic makes sense to you.

Once you agree that the model that has the higher likelihood is the better one. There is still an issue. Are the likelihoods truly different from each other? Find out how I think you can determine this.

There are a lot of times when one has two models that they wish to compare for their "predictive performance." Without careful contemplation as to the metric of comparison one may be left in the cold.

A peak at the origins of my fascination with analytical sciences.

Conventional wisdom tends to a singular approach of drug development. This article wonders if there can be situations where there is a better alternative to what is nearly always adopted. This thought is driven by clinical pharmacology understanding.

Better language for communicating results.

A thought about determining the sample size for a study