Sports
Ken Pomeroy, college basketball stats guru, used to do meteorology – New York Post
Thanks for contacting us. We've received your submission.
Sign up here to get Beyond the Back Page delivered to your inbox each weekday morning.
No one knew that college basketball was on the verge of changing forever. Not even the catalyst of the revolution, tinkering with statistics and mathematical formulas during his downtime from writing a thesis on atmospheric science.
Ken Pomeroy didn’t aspire to become one of the most influential people in the sport, playing a role on virtually every team in the country. He couldn’t know how many coaches would become reliant on the advanced statistics system featured on his website (kenpom.com). He couldn’t fathom becoming the godfather of a movement, changing how teams prepare, altering media coverage and influencing sportsbooks.
Pomeroy, 48, was simply a college basketball fan with no connection to the sport outside rooting interests for his alma maters (Virginia Tech, Wyoming). He’d been enamored with mathematics since he was a child. By creating proprietary algorithms to more accurately evaluate the strengths and weaknesses of every Division I team, Pomeroy found an absorbing and fulfilling hobby to keep him occupied when he wasn’t busy working as a meteorologist at the National Weather Service.
“When I first started doing this I never would’ve guessed it would’ve gotten to where it is now,” Pomeroy told Post Sports+. “There’s no doubt it’s wild.”
As a young teenager in Northern Virginia, Pomeroy developed an affinity for advanced sports statistics when he stumbled across the baseball works of Bill James at his local library. In college, Pomeroy chose pragmatism over passion, earning a civil engineering degree from Virginia Tech. He then took a job in which he was responsible for designing roads.
“I never gave it a lot of thought, but engineering made the most sense with the natural talents that I had,” Pomeroy said. “One of the big things I was doing, if there was a new road being proposed I had to go out and take measurements of an existing road and how it was impacting neighboring houses. It wasn’t bad, but I wasn’t really enjoying the job that much, so I decided to go into weather.”
Pomeroy’s long-held fascination with meteorology led him to the University of Wyoming. After earning his graduate degree in atmospheric science, Pomeroy joined the National Weather Service, where he worked for 12 years, helping make predictions regarding tornado warnings and evaluating how to improve forecasts.
“I loved doing meteorology,” Pomeroy said. “When I started doing the basketball stuff, I didn’t think it could possibly be a career. I didn’t necessarily even want it to be a career. The job I had was great and this was a cool hobby.”
Pomeroy created his website in 1999. The little-known web address provided ratings for multiple sports, including golf, boxing, college lacrosse and minor league hockey. In 2004, the site became fully dedicated to his first love, giving college basketball an unprecedented deep dive into analysis — such as points scored/allowed per possession and tempo — long obscured by clichés.
He found inspiration from a 2004 Air Force game, when a broadcaster praised the team’s defense. Pomeroy was immediately skeptical, believing the Falcons’ slow pace was the primary factor in keeping opponents point totals low. His calculations validated his opinion.
“I’d see what people were doing in baseball from an analytical standpoint and it was something I was interested in, and nobody was doing it for college basketball,” Pomeroy said. “I thought, ‘I’ll give it a shot. I could be OK at this.’
“I am not a programmer by trade. I was learning stuff on the fly and teaching myself. The one thing going for me was there wasn’t any work being done in college basketball at that point, so there were really simple things you could do to generate an interesting concept.”
A loyal, and relatively small, following exploded after Kenpom.com received the equivalent of a free Super Bowl ad. After then-Butler coach and wunderkind Brad Stevens revealed that he used Pomeroy’s system to prepare against — and nearly upset — Duke in the 2010 national title game, the meteorologist out of Salt Lake City welcomed droves of new subscribers to his site. In 2012, he determined that the side project was profitable enough to work on full-time, and left his job at the National Weather Service.
“The thing that appealed to me most was that I could work for myself,” Pomeroy said. “If I was a basketball player, they’d call me uncoachable. I didn’t get along with most of my bosses, so I had a desire to do my own thing at that point.”
Pomeroy says the site continues to experience year-over-year growth despite spending no money on advertising. Free promotion for the site occurs countless times per day, as a fixture in the writing, discussion and evaluation of college basketball teams.
“I’ve definitely been surprised that it continues to do as well as it does,” Pomeroy said. “The Brad Stevens thing, if there was one moment, that definitely would’ve been it. He makes the title game and gives me a shoutout. There were a lot of people who didn’t know my work, and that just gave me tremendous credibility. I’d get little mentions in newspapers or Sports Illustrated that were big deals at the time.”
Pomeroy, who has worked for various media outlets and been employed as a consultant for NBA and college teams, created the site as a reference point for a 358-team landscape that was too vast for any person to truly know intimately.
Pomeroy’s ratings are designed to calculate expected win percentage, emphasizing scoring margin over the quality of wins and losses.
“The system is entirely trying to make predictions. That’s its sole focus,” Pomeroy said. “You look historically, winning is slightly important to making predictions and scoring margin is extremely important in making predictions. That’s kind of the basis of the system. Sometimes people are confused by that, but if you think the system is wrong, it’s very easy to place wagers in 2022, so you can make some money if you really think the system is wrong.
“The reason I created the site was to give you an objective second opinion. It’s not always gonna be right, but you should pay serious attention to it because in most cases it’s gonna have teams slotted in the correct neighborhood. The college basketball season is very short. Some teams are gonna drastically overachieve or underachieve from their true ability, and that’s where the probabilities come into play.”
Pomeroy believes his work as a meteorologist was key to the foundation of college basketball’s most prominent advanced statistics.
“It’s most helpful in understanding the limits of your predictive ability and the need to express that,” Pomeroy said. “If you miss a weather forecast, it can affect people in really profound ways. So it’s important not to be overconfident or underconfident in your forecast. It’s important to communicate the uncertainty in certain events. That’s permeated my whole thought on predicting games or even doing studies on what matters for winning teams, trying to predict player performances. There are limits and you’re gonna be wrong on a certain amount of predictions, and the need to express how likely you are to be wrong is important.”
The former outsider is now part of the game’s fabric. A supposedly overcomplicated system is now officially incorporated into seeding the West Coast Conference Tournament and used as a tool in selecting the NCAA Tournament field.
Pomeroy was a 16-seed who went to Dayton for a play-in game and reached the Final Four. No system is needed to calculate the probability of that path.
Pomeroy’s work is now viewed like Duke or Kentucky. Regular-season success barely registers. It will be judged by what happens in the NCAA Tournament, the most unpredictable postseason in sports.
“I can’t help but have some vested interest in the tournament results being somewhat similar to what my ratings are, but part of the reason why [the tournament] is popular is because it’s a crapshoot, so for people to expect there’s some system out there to predict the results of the tournament perfectly is completely unreasonable,” Pomeroy said. “In the tournament, there’s not huge mismatches. A 3-14 game is probably an 85, 90 percent chance for the favorite. A 4-13 game is probably close to 80 percent, so you’re gonna have eight of those games combined and you’d expect at least one upset in that group every year. Good luck trying to figure out who that is ahead of time.
“The fan in me would prefer it’s definitely not the No. 1 team to win [the title] every year, that we do have upsets and surprises. That’s pretty much why we watch.”
Pomeroy’s ratings date back to 2002. Of the 19 national champions crowned since then, 13 teams finished with a higher offensive efficiency rating than defensive efficiency. All but one champion (2014 UConn) finished with a top-20 offense; all but three finished with a top-10 offense. Only one champion finished without a top-15 defense, as Baylor’s talented, 22nd-ranked defense felt the brunt of a lengthy COVID pause last season.
“Offense is slightly more important than defense when you get into the tournament,” Pomeroy said. “The way I look at it, to win a championship, you have to play like a top-20 offense. You can come in being ranked 60th or 70th and [find] it and [make] shots for six games. I wouldn’t rule out a team outside of that group winning it, but it would take a change of character when they get to the tournament.”
Gonzaga entered last year’s tournament as the favorite, looking to complete the first undefeated season since 1976. Even after getting blown out by Baylor in the national title game, Gonzaga remained the top team in Pomeroy’s system.
“If they had another game, they’d still be favored in my system,” Pomeroy said. “People are just gonna remember how that game turned out. …[I]t was a total obliteration, and history will show that Baylor was way better than Gonzaga because of that game, but the reality is they were pretty evenly matched teams. Baylor shot the lights out and were clicking defensively and Gonzaga didn’t know what hit them.
“It’s a great thought experiment that people don’t want to do.”
Though Gonzaga hasn’t been nearly as dominant this season, Pomeroy’s system sees the Bulldogs as an even bigger favorite to win their first-ever national championship.
“Gonzaga is [No.] 1 by a pretty wide margin,” Pomeroy said. “They’re not as good as last year, but there’s no Baylor from last year. There’s no obvious second-best team emerging. It’ll be interesting to see what my methods spit out in terms of the probabilities of Gonzaga winning the tournament. I imagine it’ll be in the 20 [percent range].
“Gonzaga will probably not be favored as much in real life as they are in my system — they get a bit of an inflated record beating up on their league — but they’re still gonna be the favorite to win the title, so clearly Vegas thinks they’re the No. 1 team as well.”
Rarely does the top-overall seed in the tournament finish the season on top. Pomeroy sees a handful of teams capable of delaying Gonzaga’s long-awaited parade.
“People say this every year, and it’s usually garbage, but it actually is shaping up to be a really wide open tournament this year,” Pomeroy said. “There are five or six evenly matched teams that are title contenders — Kentucky, Baylor, Arizona, Duke, Kansas, Auburn — and Gonzaga’s half a step ahead of them, but not miles ahead of them. It’s hard to pick a team from that group that’s much better than the others. They all have their flaws, but they’re all solid teams.
“Somebody could win it outside that group, but it’d be pretty unlikely. It’s silly to say, but it all comes down to who they play.”
Brooklyn or Staten Island will be in line for an NCAA Tournament berth.
No. 2 Wagner (20-5, 15-3) and No. 3 LIU (16-13, 12-6) face off Saturday night (8 p.m., ESPN3) at Spiro Sports Center in the Northeast Conference Tournament semifinals, with the winner advancing to Tuesday’s title game. LIU has won four championships since 2011 — including an upset of top-seeded Wagner in the 2018 championship — while the Seahawks are searching for their first NCAA Tournament appearance since 2003 despite winning three regular-season league titles under coach Bashir Mason.
On Friday, the Division III NCAA Tournament kicks off with No. 19 Yeshiva (25-3) playing No. 10 Johns Hopkins (22-3) in the first round at Stockton University in New Jersey. This is the first time the tournament will be held since the Maccabees reached the Sweet 16 — winning tournament games for the first time in school history — in 2020, then saw the season canceled by COVID. The game (1 p.m.) can be viewed on Macslive.com.
Advertisement
