Kevin Willis, The Greatest Lousy NBA Player

A few years ago, Jon Bois of SBNation did an episode of Chart Party where he talked about Jeff Francoeur, describing him as “my favorite worst baseball player.” The idea was that “Frenchy” had one of the most explosive first months of a career in the history of Major League Baseball, then dropped off a cliff.

Go watch it. It’s good.

The weird part is despite that, he lasted 12 years in the league, played in over 1400 major league games for his career, and all this despite frequently ending up with negative Wins Above Replacement (for those who don’t follow baseball, think basketball’s VORP.)

Francoeur finished his career with 6.8 WAR in 12 years and 1480 games played.

How bad is that? 6.8 WAR is so low that even in 1981, a year where the MLB season was truncated by 60 games due to a players’ union strike, two players—Mike Schmidt and Andre Dawson—had more WAR in 102 and 103 games, respectively, than Francoeur had in 12 years and 1480 games.

This left me wondering, who’s basketball’s Jeff Francoeur? Who’s the guy who was never actually that good but played the longest career while doing less than guys who are actually great do in a single season?

Finding an answer was easy enough. I went into Basketball Reference, set a minimum games played of 1200 (or about 15 full seasons’ worth of 82-game campaigns), then sorted by VORP, from first to worst.

The guys at the top of the list are a murderer’s row of great NBA talents—LeBron, Stockton and Malone, Dirk, KG, Kareem, Kobe…guys you can identify by just a single name or even a pair of initials.

But down at the bottom of the list of the 41 players to play in at least 1200 career games?

Kevin Willis…who in 21 NBA seasons and 1424 games played managed a remarkable minus-1.0 VORP.

A guy who lasted that long, who played until he was 44 years old, who even made an All-Star team in 1992 with the Atlanta Hawks…for his career posted advanced stats so bad that he finished his career with less overall value contributed relative to a replacement-level player than I, a sportswriter who never even played college much less pro ball, put up in zero NBA games.

In the 3-point era, 322 players have posted a worse career VORP than Willis did in his long NBA tenure. But that is out of the thousands of players to pick up a basketball for pay in the past 41 years. In 2020 alone, there were 529 guys who suited up for at least one game in the Association. Since 1979, it’s hard to get a count on how many names are on the overall list (Basketball Reference gives them out 100 at a time and I’m not doomscrolling that list just to get the exact number but it’s doubtless well over 10,000.)

The simple fact of the matter is that Willis is, at best, in the bottom 3 percent of all guys ever to play NBA basketball for a catch-all stat that has, per an earlier study on this site, a 96 percent correlation with team wins.

And what’s more, Willis was a power forward and a center throughout his career; he was listed at 7 feet in old media guides and video games.

Advanced stats tend to overrate big men; in 2019, Marc Gasol was dead last in WS/48 among centers who played meaningful minutes at .118.

Willis, for his career, finished at .102 WS/48. Sure, that’s technically starter level…but not for a center!

And in 21 years in the league (from 1984-85 to 2006-07, not counting ’88-89, which he missed with a knee injury, and 2005-06, which he spent retired at age 43), he only cracked .100 WS/48 four times.

That year he made the All-Star team? Willis posted .143 WS/48, good for 41st in the NBA. In the Eastern Conference in 1991-92, he ranked sixth among power forwards, behind (in order) Horace Grant, Larry Nance, Ed Pinckney (!), Dennis Rodman, and Detlef Schrempf.

Schrempf made the All-Star team a year later, rectifying an injustice. Nance got snubbed for the ’92 game, a gross slander on his career-best .204 WS/48 that year. Grant, overshadowed by the Bulls’ two bigger stars, got left out of the game despite 14.1 raw Win Shares and .237 WS/48, both third in the entire league that year.

How crazy-go-nuts was Grant’s year, where he didn’t even make the All-Star team? The top three guys for Win Shares were Michael Jordan, Karl Malone…and Horace Freaking Grant. For WS/48, it was David Robinson taking the silver medal between His Airness and the Bulls’ erstwhile power forward.

Even Scottie Pippen couldn’t match up with Grant in the advanced stats, although Pip did finish eighth and 12th in WS and WS/48 respectively and netted a First Team All-Defensive and Second Team All-NBA that year while taking ninth in the MVP voting.

Kevin Willis was an All-Star that year, one of the most undeserving All-Stars in NBA history.

How on Earth did Willis last until he was 44 years old despite not only finishing his career with negative VORP but doing so while not having a demonstrably good year after 1994?

He was an above-average rebounder, standing 24th in NBA history in that stat and 7th in offensive boards, and he was a decent defender, standing 57th overall in Defensive Win Shares, but he never cracked the top 20 in that latter stat after ’94 and never broke into the top ten in any year.

He was a lousy shooter for a big man, hitting just 48.7 percent of his shots for his career despite playing power forward and center and getting a ton of putbacks from all those offensive boards.

We only have data since the 1996-97 season, but it perfectly encapsulates the decline of Kevin Willis in his 30s and into his 40s. He had a positive on-court net rating exactly twice. Once, it was in his age-34 season in Houston in ’97, where he posted a +1.2 on the floor…but that was on a Rockets team that went to the Western Conference Finals and was 7.1 points better per 100 possessions with Willis on the bench.

The other time was in his last year, where he only played five games and 43 minutes in garbage time for the Dallas Mavericks, another team that was outstanding in the regular season…let’s just not talk about the 2007 playoffs.

Speaking of the Rockets and that ’97 campaign where they fell short against the Jazz for the right to get beat up by the Bulls in the Finals that year, Willis had negative Win Shares in 295 playoff minutes and 16 games for the Rockets in the postseason. He was actively hindering every effort the Rockets made to recapture the glory of the 1994 and ’95 titles, one of the factors that cost Houston a shot at a ring.

Willis played 12 of his 21 seasons between 1995 and 2007, missing one to retirement.

He had zero or negative VORP in all 12 of them, never once venturing into measurable positive territory.

He had 4.2 VORP in all or part of 11 seasons with the Hawks, who drafted him 11th overall in 1984, and negative VORP on all 7 of the other teams he played for, totaling minus-5.2.

For his career, if every team he was on just replaced him with the at-the-time equivalent of a decent player from the G-League today, all those teams combined would have won two more regular season games than they did with Willis on their rosters. The Hawks would’ve lucked into about nine wins, or one per year if you don’t count Willis’ partial-year seasons with them later in his career.

Every other team? They’d have been 11 wins in the hole.

Willis is not in the Hall of Fame, so Hall voters aren’t seduced by mere longevity.

But he does have one interesting claim-to-fame. He is, by a large margin, the worst player in NBA history who also played more career games than Michael Jordan.