Can an Exercise Tell You Who Should Be an Admiral?

1 The New Requirement

Defense Secretary Pete Hegseth announced a new Joint Warfighter Evaluation for eligible Army, Air Force and Marine Corps colonels and Navy captains screening for promotion to one-star. The assessment, announced in late September, is intended to measure how officers perform when presented with a military problem and required to make decisions under pressure. It supplements a promotion process that for decades has rested on career records: performance reports, command assignments, education and operational experience.

The rationale is the persistent complaint that records measure the ability to be evaluated rather than the ability to decide. The question is whether a single assessment can measure the second thing any better.

2 What Records Measure Well

Records-based boards are not irrational. Performance reports aggregate years of observed behavior from multiple supervisors; command assignments test an officer in a real billet with real consequences; education and operational experience establish a foundation of knowledge. A records board is a compilation of historical evidence with broad temporal coverage, and it benefits from having observed the candidate under conditions nobody could script.

Its weaknesses are equally well known: reports are inflated, raters differ in severity, careers are shaped by assignment luck and by who did the writing, and the resulting signal is partly a measure of the ability to accumulate favorable documentation. The board sees a curated record rather than the officer.

3 What A Practical Assessment Adds

An assessment like the JWE adds three properties a record cannot supply. Standardization, meaning every candidate faces comparable conditions rather than a unique career. Recency, in that the measured behavior is observed now rather than compiled over decades. And decision content, where the candidate is asked to do the thing the promotion implies - solve a problem under pressure - rather than to have been evaluated doing adjacent work.

Those are genuine improvements in construct validity, which is the question of whether the measure is related to the performance it is meant to predict. A promotion system that uses both a record and a standardized assessment has more independent evidence than one that uses either alone.

4 The Predictive Limits

Assessment centers have a long research history, and their reputation is better than most selection methods. The limit is that a two-day problem on a range or in a simulation is a narrow sample of an admiral's actual job, much of which involves bureaucratic navigation, coalition politics, resource competition and long-horizon institutional leadership rather than tactical decision-making under fire. If the assessment measures decision-making under pressure well but predicts the full job only partially, it will be a useful input and a poor sole criterion.

There is also the question of what the assessment cannot see: judgment developed over long service, personal integrity in ambiguous ethical circumstances, and the willingness to disagree with a superior. Those are hard to script and easy to coach around.

5 Gaming Risk

Any standardized assessment with a promotion consequence induces preparation. Candidates will study the format, preparation courses will appear, and the assessment will drift toward measuring preparation rather than capability. This is not a reason to abandon the approach; it is a reason to vary scenarios, rotate evaluators, hold the content closely, and treat the assessment as one input among several rather than as a score that determines a board's outcome. The moment the format stabilizes and becomes public, its discriminating power decays.

6 The Implementation Questions

Which scenarios, and who authors them? Who evaluates, and what training and calibration do evaluators receive? How is the result recorded and weighted against the record? Is there an appeal? What is the cost per candidate, and can the throughput support screening every eligible colonel and captain in three services? Each of these determines whether the JWE is a functioning instrument or an additional box to be cleared. The throughput question is the most practical: a standardized, human-evaluated assessment is expensive and slow at the scale of a one-star board.

7 The Bottom Line

Adding a decision-under-pressure assessment to one-star screening is a defensible remedy for a real weakness in records-based boards, and it measures something records do not. Its limits are the narrowness of what it samples, the certainty that candidates will prepare for it, and the practical difficulty of running it at scale. Used as one input with a well-calibrated record, it improves the evidence base. Used as a decisive score, it substitutes one imperfect measure for another.