The Minnesota Timberwolves GM might hint that the lottery to determine the NBA draft is fixed, but the league has worked hard to have the system incorporate the mathematics of chance, improve lesser teams and still include the element of luck.
Julio Cortez / Associated PressDavid Kahn, left, general manager of the Minnesota Timberwolves, Kevin O'Connor, center, general manager of the Utah Jazz, and Nick Gilbert, 14, right, the son of Cleveland Cavaliers owner Dan Gilbert, stand on the stage representing the final three teams during the 2011 NBA basketball draft lottery.
As the NBA's worst team in 2011, the Minnesota Timberwolves had decent odds of scoring the top pick in the league's player lottery Tuesday night and launching a franchise turnaround. At 25 percent, or one in four, the T-wolves' chances were better, in fact, than any other team, including the second-worst Cavaliers.
But when league vice president Lou DiSabatino popped four numbered ping pong balls at random from the NBA's lottery hopper, the Cavaliers came up winners. The Cleveland team had had only a 2.8 percent shot at getting the top pick on that draw. That led T-wolves General Manager David Kahn to hint to reporters, perhaps jokingly, that the NBA's lottery system was rigged to favor teams that were sentimental favorites.
"This league has a habit - and I am just going to say habit - of producing some pretty incredible story lines" on draft day, a smiling Kahn said, citing the presence on the lottery podium of Cavaliers owner Dan Gilbert's teenage son Nick, who has a serious nervous system disorder. Kahn also mentioned that last year's draft winner, the Washington Wizards, was represented by the widow of the team's owner.
"As soon as the 14-year-old [Nick Gilbert] came up there, it was lights out," Kahn said of his team's chances. "We were done. This was not happening for us." (Kahn later said he didn't mean to imply the draft was fixed, but expected the NBA would fine him for his remarks.)
Simple statistics, however, seem to be the real source of the T-wolves' missed opportunity to land a top hoops prospect. While the Cavs' odds were long, they weren't zero.
Mark Duncan / Associated PressThen-Cleveland Cavaliers general manager Jim Paxson holds the ping pong balls that won Cleveland the top pick, and LeBron James, in the 2003 NBA Draft.
"There's a difference between improbable and impossible," said Matthew Leingang, a clinical mathematics professor at New York University who has blogged about the statistical probabilities of winning the NBA lottery. "If you have a six-sided die and five of the sides are painted red and one is painted green, you would certainly expect a red face to come up. But that doesn't mean a green face is not going to. The most likely answer is red, but it's not the certain answer."
The NBA lottery process is a study in the arcane mathematics of chance, and of the league's attempts to boost ailing teams without completely squashinig the role of luck.
The NBA has set up a system that gives teams with the fewest wins during the previous season a better chance than more successful franchises at landing the top high school or college player. The league has tinkered with the process over the decades, trying to improve its fairness and reduce the opportunities for tampering.
In the early years of the draft, from 1966 to 1984, a simple coin flip between the East and West conference's two worst teams determined who chose first. Each had a 50 percent chance. The loser got the second pick, with subsequent choices awarded to the remaining teams based on the reverse order of their won-loss standings, from last to first. The idea was to build parity by giving poor-performing teams the best shot at good young players.
As former Plain Dealer sports writer Burt Graeff pointed out in a 2003 story, the coin-flip method enabled the lowly Milwaukee Bucks to nab UCLA standout Lew Alcinodor in 1969. Alcindor, the future Kareem Abdul-Jabbar, led the Bucks to the NBA championship two years later.
The coin-flip system's relatively high odds that a bad record would land a good player led some critics to complain that teams were deliberately losing games to position themselves in the draft. The Houston Rockets, who dropped nine of their last 10 games in the 1984 season, were the target of some of that speculation.
Sensitive to the charges, the NBA shifted to a lottery draft in 1985, expanding the pool of low-winning teams competing for the first pick from two to seven, and drawing the winner from envelopes spun in a barrel.
Ron Frehm / Associated PressPatrick Ewing.
The method reduced the odds of winning the top pick from 50 percent to about 17 percent, but it wasn't immune to controversy. Stories abound - which the NBA dismisses as ridiculous - that the envelope holding the New York Knicks' card was bent or marked somehow. That supposedly allowed league commissioner David Stern to recognize and pick it, sending future superstar Patrick Ewing to the team with the nation's biggest TV market.
The NBA switched the lottery process again in November 1993 to the current ping pong ball method. The system slightly bumped up the chances of the worst team picking the first player, to 25 percent, while lowering the odds for the team with the best record to 0.5 percent.
Here's how the process works:
At the NBA's headquarters in Seacaucus, N.J., 14 ping pong balls, numbered 1 to 14, are put in a pneumatic hopper like those used for Mega Millions and other state lottery drawings. A league executive, watched by reporters and security personnel, opens a valve, allowing four balls to pop out.
The four balls have 1,001 possible number combinations. To make the calculations easier, the league rounds the number to 1,000, tossing out the "11-12-13-14" sequence.
Before the drawing, the NBA randomly assigns a portion of those remaining 1,000 ping pong number combinations to each of the 14 non-playoff teams participating in the draft. It's like giving each team a bushel of pre-selected Ohio Lottery tickets.
The number of combinations that each team gets is "weighted," based on their won-loss record. The worst team - which in 2011 was the Timberwolves - gets 250 of the 1,000 possible ping pong numbers, or a 25 percent chance of winning the lottery. The next-worst team - the 2011 Cavs - holds 199 (19.9 percent odds), the third team has 156 (15.6 percent), and so on, down to 5 (0.5 percent) for the non-playoff team with the best record.
Only the first three draft picks are made with the ping-pong method. The rest are assigned based on the remaining teams' win-loss records.
Teams can trade for picks, and, in fact, the 2011 Cavs held the Los Angeles Clippers' eighth pick in addition to their own second one. The Clips' pick had 28 of the 1,000 possible ping pong numbers, and it was one of those sequences - 14-13-7-8 - that scored the top pick for Cleveland. The odds of that happening were 2.8 percent, considerably longer than the T-wolves' one-in-four chances.
Tony Dejak / Associated PressCleveland Cavaliers' LeBron James, the first pick in the NBA draft, holds up his jersey during a 2003 news conference.
Sportswriters have noted that under the ping-pong system, NBA teams with the worst record have won the top draft pick only twice. (One of those was the 2003 Cavs, who were tied for worst and used their pick to land LeBron James.)
That seems to suggest that something's wrong with the system, since the most odds-favored, NBA-weighted team hasn't often gotten the top pick. Meanwhile, prior to the Cavs' 2011 win, three NBA teams with odds of 5 percent or less had scored the winning ping pong balls.
But as probability experts and sports statistics bloggers have pointed out, the probabilities of those outcomes - of worst teams not getting the top pick, and of better teams getting it - aren't that far out of line with expectations.
One statistics blogger calculated in 2008 that during the current ping pong era, the worst team should win the draft between three and four times, so winning twice isn't much off pace - about 19.5 percent, compared to the expected 21.5 to 25 percent.
It's also important to note that the relatively few NBA lotteries that have been held - what statisticians call a small sample size - make it hazardous to draw major conclusions about whether the process is tainted. Lots of results over a long period of time are what give numbers-crunchers the confidence to declare a pattern.
"If you flip a coin ten times and you get heads seven times, you still can't say the coin is fixed," said Leingang. "If you flipped it 1,000 times and it came up heads 700 times, you probably could conclude that."
Likewise, a couple of decades' worth of once-a-year NBA lottery results "is not enough of a sample to conclude anything," said Leingang. "Just because a team is supposed to get the No. 1 pick 25 percent of the time doesn't mean that, in 20 years, you would expect exactly five times where the worst team got the best pick.
"We may have to go a couple of hundred times before we get to conclude the NBA is fixing the lottery."