Jeremy Lin’s emergence as New York’s basketball savior is thrilling for its novelty: an Asian-American Harvard grad goes undrafted and bounces around the Developmental League before sliding into the Knicks rotation, where he puts up staggering statistics instantly (despite sleeping on couches) and leads an injured team on an inspired seven-game win streak. It’s as if the script was ripped from a mawkish Disney sports movie. And I imagine there are plenty of NBA executives who are losing sleep watching it unfold, wondering how they missed the chance to sign a point guard who could turn out to be a very special player.
How did this happen? How did the guard from Palo Alto slip through the cracks? Science writer Jonah Lehrer draws one lesson from Lin’s surprising run: on the whole, professional sports franchises are terrible at identifying talent. Lehrer uses research on the NFL Combine to back up his claim; a recent analysis performed by the economists Frank Kuzmits and Arthur Adams found that there was “no consistent statistical relationship between combine tests and professional football performance.” Lin, likewise, worked out in front of eight teams during pre-draft tryouts but always in games and drills with fewer than 10 players, a set-up that probably highlights an individual’s physical attributes effectively but is not the most natural way to judge a player’s overall feel for the game. “There is talent everywhere,” writes Lehrer. “We just don’t know how to find it.”
Lehrer is a smart cookie, but I think he overstates his case slightly. Past performance, while no guarantor of future success, is still a straightforward and relatively accurate indicator of one’s athletic potential. Take this article in the journal Human Performance that was published last April. The authors, studying NFL prospects once again, determined that "collegiate performance engendered a stronger relationship with future NFL performance than a variety of physical ability tests administered during the NFL Combine." If you're good in college, in other words, there's a pretty decent shot you'll be good at the next level.
Tuesday night, after watching Lin beat the Raptors at the buzzer and reading these silly and self-serving quotes from a former Cornell player who guarded Lin a few years back, I reviewed the Harvard grad’s advanced stats from his senior year (2009-2010). As an avid college hoops fan who was personally invested in the career of Crimson coach Tommy Amamker, I was familiar with Lin at the time, but I had forgotten just how dominant he really was.
Via Ken Pomeroy’s site ($ required), Lin was on the floor for 79 percent of Harvard's available minutes and used almost 27 percent of the team’s offensive possessions. Even carrying this heavy load, Lin registered a blistering 62 percent true shooting percentage (46th nationally) and a 30.5 percent assist rate (66th nationally). He got to the free line a ton, too. If he had a fault, it was that he turned it over too much—21 percent, definitely below average for elite point guards—but he balanced out those takeaways by recording one steal for every 20 of Harvard’s defensive possessions, the 16th highest rate in the country. Yes, Lin posted these numbers against weaker competition than guards at major conference schools: the Crimson's strength of schedule was 239th out of 345 Division 1 teams in 2010. But Lin didn't fool around.
We can take it one step further by comparing Lin to a few other combo guards that excelled for big programs that season:
Jeremy Lin: 79% min, 27% usage, 62% true shooting, 30% assist rate, 21% TO rate, 68% FT rate
Evan Turner: 74% min, 34% usage, 58% true shooting, 37% assist rate, 21% TO rate, 39% FT rate
John Wall: 83% min, 27% usage, 56% true shooting, 34% assist rate, 24% TO rate, 53% FT rate
Landry Fields: 90% min, 31% usage, 56% true shooting, 19% assist rate, 13% TO, 50% FT rate
Jon Scheyer: 91% min, 23% usage, 57% true shooting, 25% assist, 11% TO rate, 41% FT rate
I didn't include rebounding numbers, as the bigger Turner laps the rest of the field. And that makes sense: he was probably the most prolific player in America that year. (Sorry, Harangody.) But Lin still stacks up, and was a decidedly more efficient player than Wall (the top pick in the NBA draft) even given a comparable usage rate. Plus, he came to play in big games, notching 25 on 10 shots against nearby Boston College and 30 points, nine rebounds, and two big dunks in a six-point loss to then-14th ranked Connecticut.
I’m not blaming NBA general managers for their early skepticism. Chris Dudley was the last Ivy Leaguer to play an NBA game, way back in 2003, and Lin’s stint in Golden State last year—in which he shot just 39 percent from the floor—proved he needed time to adjust to the tempo of the pro game. What’s clear, as unofficial Lin biographer Pablo Torre wrote this week, is that “evaluators of basketball talent, in particular, failed to see the whole picture.” It’s hard to say whether or not race played a factor. But the evidence of Lin’s playmaking ability was staring NBA execs in the face, if they cared to look.
How did this happen? How did the guard from Palo Alto slip through the cracks? Science writer Jonah Lehrer draws one lesson from Lin’s surprising run: on the whole, professional sports franchises are terrible at identifying talent. Lehrer uses research on the NFL Combine to back up his claim; a recent analysis performed by the economists Frank Kuzmits and Arthur Adams found that there was “no consistent statistical relationship between combine tests and professional football performance.” Lin, likewise, worked out in front of eight teams during pre-draft tryouts but always in games and drills with fewer than 10 players, a set-up that probably highlights an individual’s physical attributes effectively but is not the most natural way to judge a player’s overall feel for the game. “There is talent everywhere,” writes Lehrer. “We just don’t know how to find it.”
Lehrer is a smart cookie, but I think he overstates his case slightly. Past performance, while no guarantor of future success, is still a straightforward and relatively accurate indicator of one’s athletic potential. Take this article in the journal Human Performance that was published last April. The authors, studying NFL prospects once again, determined that "collegiate performance engendered a stronger relationship with future NFL performance than a variety of physical ability tests administered during the NFL Combine." If you're good in college, in other words, there's a pretty decent shot you'll be good at the next level.
Tuesday night, after watching Lin beat the Raptors at the buzzer and reading these silly and self-serving quotes from a former Cornell player who guarded Lin a few years back, I reviewed the Harvard grad’s advanced stats from his senior year (2009-2010). As an avid college hoops fan who was personally invested in the career of Crimson coach Tommy Amamker, I was familiar with Lin at the time, but I had forgotten just how dominant he really was.
Via Ken Pomeroy’s site ($ required), Lin was on the floor for 79 percent of Harvard's available minutes and used almost 27 percent of the team’s offensive possessions. Even carrying this heavy load, Lin registered a blistering 62 percent true shooting percentage (46th nationally) and a 30.5 percent assist rate (66th nationally). He got to the free line a ton, too. If he had a fault, it was that he turned it over too much—21 percent, definitely below average for elite point guards—but he balanced out those takeaways by recording one steal for every 20 of Harvard’s defensive possessions, the 16th highest rate in the country. Yes, Lin posted these numbers against weaker competition than guards at major conference schools: the Crimson's strength of schedule was 239th out of 345 Division 1 teams in 2010. But Lin didn't fool around.
We can take it one step further by comparing Lin to a few other combo guards that excelled for big programs that season:
Jeremy Lin: 79% min, 27% usage, 62% true shooting, 30% assist rate, 21% TO rate, 68% FT rate
Evan Turner: 74% min, 34% usage, 58% true shooting, 37% assist rate, 21% TO rate, 39% FT rate
John Wall: 83% min, 27% usage, 56% true shooting, 34% assist rate, 24% TO rate, 53% FT rate
Landry Fields: 90% min, 31% usage, 56% true shooting, 19% assist rate, 13% TO, 50% FT rate
Jon Scheyer: 91% min, 23% usage, 57% true shooting, 25% assist, 11% TO rate, 41% FT rate
I didn't include rebounding numbers, as the bigger Turner laps the rest of the field. And that makes sense: he was probably the most prolific player in America that year. (Sorry, Harangody.) But Lin still stacks up, and was a decidedly more efficient player than Wall (the top pick in the NBA draft) even given a comparable usage rate. Plus, he came to play in big games, notching 25 on 10 shots against nearby Boston College and 30 points, nine rebounds, and two big dunks in a six-point loss to then-14th ranked Connecticut.
I’m not blaming NBA general managers for their early skepticism. Chris Dudley was the last Ivy Leaguer to play an NBA game, way back in 2003, and Lin’s stint in Golden State last year—in which he shot just 39 percent from the floor—proved he needed time to adjust to the tempo of the pro game. What’s clear, as unofficial Lin biographer Pablo Torre wrote this week, is that “evaluators of basketball talent, in particular, failed to see the whole picture.” It’s hard to say whether or not race played a factor. But the evidence of Lin’s playmaking ability was staring NBA execs in the face, if they cared to look.
RSS Feed
