Steve Selvin

Steve Selvin died on September 30, 2026, in a memory care facility in Oakland, California, at the age of 85. According to the San Francisco Chronicle, he had struggled with dementia for many years. No specific cause of death has been made public.

The UC Berkeley School of Public Health announced his death on October 13, 2026, in an article adapted from a remembrance by Nicholas P. Jewell for the American Statistical Association. Jewell said it was impossible to sum Selvin up in a single sentence, calling him an excellent applied statistician and a gifted teacher, particularly to students without much of a mathematical background, who bled blue and gold from his days selling football programs at age 12. Dean Michael C. Lu said his influence was everywhere in the school. The San Francisco Chronicle published an obituary on October 22, 2026.

He is survived by his wife of 60 years, the sculptor Nancy Selvin; his daughter, Liz Selvin; grandchildren Benjamin and Eli; son in law David Bradford Long; and his brother Joel Selvin. No public memorial service or funeral details had been announced at the time of writing.

Steve Selvin was born on July 4, 1941, at Mt. Zion Hospital in San Francisco, the eldest of three brothers. He grew up on Fairlawn Drive in the Berkeley Hills, in a house his father, David Selvin, had bought for $5,000 in 1938 while working as editor of San Francisco Labor, the official publication of the San Francisco Labor Council.

He attended Hillside Elementary School, Garfield Junior High (now Martin Luther King Jr. Middle School) and Berkeley High School, from which he graduated in 1959. By his own retelling, a Berkeley High algebra teacher told him he would never be any good at math and would never go to college, a prediction he enjoyed quoting once both parts had proved wrong. His daughter later said he had been a C student in high school.

His lifelong attachment to the University of California began when he was 12, selling football programs for a nickel at Cal games in Memorial Stadium, running down the hill from home to work. Family stories also recall a teenage prank involving the Cold War air raid sirens in the Berkeley Hills, in which several neighborhood teens were picked up by police while Selvin eluded arrest.

Rather than enrolling at Berkeley, Selvin went to the University of California, Riverside, then a new campus of about 2,000 students promoting a liberal arts curriculum with small classes, some held in professors' homes. He was drawn in part by the absence of a Greek system; his wife later said he was adamant about not attending a school with fraternities. Both of his younger brothers followed him there.

At 5 feet 10 inches he was too short for varsity basketball, but he made the junior varsity team, where, according to family members, he spent much of his time on the bench keeping the statistics. He graduated with a BA in mathematics in 1965.

He won a fellowship to Iowa State University, known for its statistics program, but after one year in the master's program he returned to UC Berkeley. He earned his PhD in biostatistics in 1970 under the direction of Jacob Yerushalmy, with a dissertation titled A Biostatistical Investigation of Twins and Twin-Study Methodology. Because of fears of anti-war protests, his degree was conferred at a small departmental ceremony rather than a campuswide commencement.

Selvin's first job was at the Birth Defects Institute of the New York State Department of Health in Albany, where he studied the statistics of birth defects for two years. In 1972 he returned to UC Berkeley as an acting assistant professor in the School of Public Health. He became an assistant professor in 1974, chair of the Biostatistics Division in 1977 and full professor in 1983, and he remained on the Berkeley faculty for the rest of his career, a span of about 45 years of active teaching.

His research applied statistics to a wide range of public health questions: genetics and twin studies, cancer incidence and mortality, the mapping and clustering of rare diseases, environmental exposure assessment, maternal weight gain in pregnancy, perinatal epidemiology (especially neural tube defects) and risk factors for childhood leukemia. He authored or co-authored more than 250 scientific papers.

As head of the school's Undergraduate Management Committee he was instrumental in building Berkeley's undergraduate public health program, including recruiting students to the undergraduate major added in 2004, and he helped create the MPH in Epidemiology and Biostatistics. Beyond Berkeley he was an adjunct professor of epidemiology at the University of Michigan from 1990 to 1998, a professor of biostatistics at Johns Hopkins University while on sabbatical in 2005, and a faculty member of the Johns Hopkins Graduate Summer Institute of Epidemiology and Biostatistics from 2004 to 2018.

In 1975, while watching the game show Let's Make a Deal at home in the Berkeley Hills, Selvin worked out the odds behind a three-choice scenario inspired by the show and its host, Monty Hall. He wrote it up as a short, humorous letter titled A Problem in Probability, published in The American Statistician in February 1975. In his version a contestant chooses one of three boxes, one holding the keys to a car. The host, who knows where the prize is, opens one of the other boxes to show it is empty and offers a switch. Selvin showed that switching raises the chance of winning from one in three to two in three.

Several readers challenged his reasoning, so in August 1975 he published a follow-up letter, On the Monty Hall Problem. It was the first time the phrase appeared in print, and it set out a solution based on Bayes' theorem along with explicit assumptions about how the host behaves, including that the host chooses at random between two empty boxes when the contestant has picked the prize.

Monty Hall wrote Selvin a playful letter that argued a few points and closed with the line "Next time let's play on my home grounds." Selvin worked the episode into his lectures for years. The puzzle remained a niche curiosity until Marilyn vos Savant answered a reader's version in her Parade magazine column in 1990, prompting thousands of letters, many of them insisting she was wrong. According to his daughter, Parade did not mention Selvin, so his name was not linked to the problem for many years until the internet put the connection back together, and he later reclaimed and embraced it.

The problem is now a staple of introductory probability and statistics courses and has been discussed on NPR's Car Talk, in countless videos and in its own lengthy Wikipedia article. Colleague Alan Hubbard said it showed Selvin's skill at taking one deceptively simple problem and using it to separate a real understanding of probability from technical mathematical skill.

Selvin met Nancy Brown at UC Riverside and they married in 1965. Nancy Selvin became a sculptor, and the couple were married for 60 years. Their daughter, Elizabeth (Liz) Selvin, followed him through the Berkeley public schools and became a professor of epidemiology at the Johns Hopkins Bloomberg School of Public Health and director of the Welch Center. He had two grandchildren, Benjamin and Eli, and a son in law, David Bradford Long.

He was the oldest of three brothers. His youngest brother, Joel Selvin, became a rock music writer and spent 36 years as a staff critic at the San Francisco Chronicle.

Selvin spent nearly his whole life in Berkeley and, according to his daughter, walked a steep two miles to work every day for 45 years. He was a devoted Cal sports fan, an avid San Francisco 49ers fan and, to a lesser extent, a San Francisco Giants fan, and he enjoyed untangling the statistics after each 49ers season. Family lore holds that he refused to wear bell-bottoms in 1960s Berkeley and went over the hill to Contra Costa County to find straight-legged jeans. His brother Joel summed him up as square all the way, but no egghead.

His wife described Selvin as an original democratic socialist who did not believe in social clubs or social climbing, which is why he chose a university without fraternities. His daughter said he was not drawn to campus activism in the 1960s.

In statistics he valued the practical. He wanted to apply statistics to the common good and to explain it to everyone, including, in his daughter's words, people who were C students in high school, as he had been. He liked to compress ideas into quips, such as "The odds of winning the lottery are roughly the same whether you buy a ticket or not," and he adapted the advice "Don't force your shots" from UCLA basketball coach John Wooden. His brother recalled that he had theories about baseball statistics long before analytics, arguing that runs batted in mattered more than batting averages.

Selvin's best known legacy is the Monty Hall problem, which he posed and named and which remains one of the most famous examples in probability teaching. Within his field, colleagues credit him with major contributions to epidemiology and environmental health, and with building the strength of Berkeley's biostatistics program and its undergraduate public health major.

His deepest impact may have been in the classroom. His daughter estimated that he taught tens of thousands of undergraduates, and his textbooks were assigned at universities across the United States. Former doctoral students such as Jennifer Schoendorf, who spent 30 years at the National Center for Health Statistics, have said they passed his words and wit on to their own colleagues and students. Dean Michael C. Lu said his influence lives in the strength of the biostatistics program, in the students he taught and mentored, and in the books and ideas that continue to shape the field.

Berkeley Distinguished Teaching Award (1983).

UC Berkeley School of Public Health Distinguished Teaching Award (1998).

Berkeley Citation (2011), awarded at the age of 70 and given to individuals whose contributions to UC Berkeley exceed the standards of excellence in their fields. It is one of the highest honors the university bestows.

Selvin wrote around ten books on biostatistics, epidemiology and applied statistical methods (the UC Berkeley School of Public Health counts ten; the San Francisco Chronicle counted nine textbooks). They include Statistical Analysis of Epidemiologic Data (Oxford University Press, 1991; third edition 2004), Modern Applied Biostatistical Methods Using S-Plus (1998), Epidemiologic Analysis: A Case-Oriented Approach (2001), Biostatistics: How It Works (Prentice Hall, 2004), Survival Analysis for Epidemiologic and Medical Research (Cambridge University Press, 2008), Statistical Tools for Epidemiologic Research (2011) and The Joy of Statistics: A Treasury of Elementary Statistical Tools and Their Applications (Oxford University Press, 2019).

His 1975 letters to The American Statistician, A Problem in Probability and On the Monty Hall Problem, are widely cited, including in Jason Rosenhouse's book The Monty Hall Problem (Oxford University Press, 2009). In July 2020 his daughter shared on X the letter Monty Hall had written to him after the problem's publication, which drew wide interest online. The puzzle he posed features in many popular explainer videos, including ones by Numberphile and Vox.