By: Jerry Radich, MD, and Ted Gooley, MD
Depending on your source, the oldest university in the world is either the University of Al Quaraouiyine, in Fez, Morocco (founded in 859 AD), or the University of Bologna, Italy (1088 AD). Despite the debate, it is likely that the first academic Appointment and Promotions (A&P) Committee materialized shortly thereafter. Fortunately for the faculty of Bologna, the first bar dates back to 900 AD, so those emotionally drained by the meeting had somewhere to drown their sorrows and nurse their psychic wounds.
For those who haven’t had the pleasure, the A&P meeting is an annual event when faculty performance and productivity are reviewed, and decisions made as to whether to promote to the next rank up, retain in current rank, or demand alternate employment. Retention is based on time served “in rank” and performance, although sometimes performance—no matter how one chooses to establish it—isn’t enough for promotion if someone has been in rank long enough (this is sensible on one level, but also nonsense on other levels). Multiple factors are considered, such as number of publications and grants, teaching acumen, clinical performance, and good “citizenship” (no felonies). Discussions can sometimes be heated, or wrenching, but often are mostly filler. As Mo Udahl famously said (and this applies to many, if not most meetings in any enterprise), “Everything has been said, but not everyone has had a chance to say it.”
(Pro tip for young faculty members: like most meetings, as time drags on and on, the energy of the meeting rapidly dissipates, and once lively/heated discussion deteriorates into near zombie-like communication. The later you are on the agenda, the better. Since review is generally done alphabetically at each rank, you are better off with a surname at the end of the alphabet. Thus, to hedge your bets, change your surname to begin with a Z, or at least X or below. And if married and hyphenated, place the Z-surname before the hyphen. Thank me later.)
There are some obvious issues with the current system. For example, there is still an overreliance on first- or last-authored papers. This approach may have been appropriate in the first centuries of academia, since the work was likely done either by one person or a very small team, at best. However, many papers now require teams of people, all essential to the work. While we pitch the essential nature of such team science, the sad fact is that investigators buried in the middle of the author list are given short shrift. There is now the fashion of “co-first (or last) author” but that is largely ignored in the count of publications (despite what is said publicly, that * next to your name will not save you). Something better to evaluate performance and potential is needed.
For inspiration we reach to the institution which offers us the best foundation of history, beauty, skill, science, and morality. That is, of course, baseball.
Baseball by its very nature is a Disneyland for quantitative types, but the development of advanced statistical metrics began only in the 1970s. Bill James, a former high-school English teacher and night watchman at a pork-and-beans plant, is regarded as the father of advanced statistical analysis of baseball. Following in James’ footsteps was Nate Silver, who (before becoming a household name in the world of election prognostication) developed a performance metric commonly used in the early 2000’s known as PECOTA. Another metric, and one that is now one of the most used, is WAR (wins above replacement). Simply put, WAR is a distillation of several elements of baseball that can be objectively measured (such as hitting, fielding, base running, and pitching) into a single number. “Replacement” reflects the performance of a player coming from the minor leagues into the majors, or a player who could be easily “replaced” without adversely impacting team performance. The WAR calculation is position specific e.g., (pitcher versus shortstop, reflecting different specialties and skill sets). A player with a WAR of 3, for example, would be expected to add three wins to his team over the course of a season.
Baseball is an expensive game, and WAR has allowed front offices to use some objective measure to spend wisely (with the obvious exception of the NY Mets, but hedge-fund billionaires use dollars as coal). Given its construction, the correlation of WAR with win percentage is very high. A win is expensive, and in major league baseball (MLB), one WAR costs roughly $8 million per year. In comparison, the average MLB salary is roughly 0.5 WAR.
In order to 1) better facilitate the A&P process, and 2) allow universities to strategically build their “rosters,” we propose the adoption of “PAR” (performance over replacement). By “replacement” we mean a freshly minted faculty member just out of fellowship (note: one difference between baseball and academic replacements is that in baseball, the replacement often shuttles between the majors and the minors; in academia, the move is one way).
If we imagine the creation of PAR as an equation, the PAR value is the dependent variable, while the independent variables would consist of factors like publications, grants, clinical work, etc. Here the independent variables are easy to gather (though as discussed above, imperfect), but what is the dependent variable to signify performance? And analogous to WAR being position-specific, which faculty are the “pitchers” and which are the “shortstops”? A win is an easy concept in baseball—it’s the outcome of each game. But what is a “win” in academic medicine? A ranking in the U.S. News and World Report? Patient outcome measures? Grants or papers (h-index?). A dollar amount (for example, soft-money coverage of salary)? Maybe we feed in thousands of CVs of academic physicians, those we all agree are Hall of Famers, those that are terrific position players, and those that are solid utility players, and let the AI genies come up with a measure of and model for success (the coefficient of (E Donnall) Thomas?)?
I have faith that smart analysts could cook up a measure of PAR. This would allow A&P committees to compare home-grown talent to like-peers at other institutions. It would also allow for rational adjustment of salaries, and it would facilitate smart construction of faculty rosters—for any given amount of cash available to an institution, is the best value to get a professor with a PAR of 5 (who might have a high cost of salary, lab space, parking, etc.), or five replacement-level faculty with lower associated costs? Think “Moneyball” (some in the industry have claimed that Brad Pitt has long coveted the role of Fred Appelbaum on the big screen).
We are at the convergence of AI and predictive analytics, where we have the ability to fit complicated models that yield accurate predictions, while academic centers are faced with issues of forced-DEI realignment and decreasing grants, budgets, and overheads. With A&P decisions, it is time for fairness and frugality to meet the future. It’s time for us to get up to PAR.
Jerald Radich, MD, is professor and director of the molecular oncology lab at Fred Hutchinson Cancer Center.
Ted Gooley, PhD, is professor and director of clinical biostatistics at Fred Hutchinson Cancer Center.
Visit the SOHO 2025 meeting news page for more coverage from the meeting.
