Tuesday, April 24, 2012

Parity in the NBA: The Plight of the Underdog

The NHL is in the midst of wrapping up the first round of its 2012 playoffs, in which the West's top seed has already been eliminated and the first and second seeds in the East will each need to win a seventh game to advance. The Panthers are on the brink of their first playoff series win since 1996, and the Coyotes just won their first playoff series since...ever.

Meanwhile, the NBA's Charlotte Bobcats are careening toward the worst single-season winning percentage in NBA history, and the Indiana Pacers are the only top-10 team in the NBA that could be regarded as a mild surprise.

Yes, it's another lesson in parity, and specifically the NBA's lack thereof. Why, exactly, does the NBA have so little competitive balance? It turns out that championship teams reliably exhibit some of the exact same qualifications, and that we can measure them before the playoffs begin--and in some cases, before the season even takes place.

The NBA's historical lack of playoff parody is well-established; the higher-seeded teams reliably make deeper runs in the playoffs.

To wit: Only three eight-seeds in NBA history have ever knocked off a one-seed in the first round: Denver topped Seattle in 1994, the Knicks upset Miami in 1999, and the Warriors beat the Mavericks in 2007. Only the Warriors won a seven-game series; both Denver's and New York's wins came when the first round was still a best-of-five format.

Even more shockingly, the 1996 Houston Rockets, a sixth seed, are the only seed outside of the top three to win a single championship since the 1983-1984 season. In the past 28 years, the eventual NBA champion has been a first seed 15 times, a second seed five times, a third seed four times, and a sixth seed once.

Most astoundingly of all, the last 28 titles belong to just eight teams: The Lakers (eight titles in that time frame), Bulls (6), Spurs (4), Celtics (3), Pistons (3), Rockets (2), Heat (1), and Mavericks (1).

So statistically speaking, out of the last 28 NBA championships, 96 percent have gone to top-three seeds and 64 percent have been won by either the Lakers, Bulls, or Spurs.

Why is it so difficult for lesser regular-season teams to win in the playoffs? And an even more troubling question: Why is it so difficult for bad teams to reverse their fortunes and earn more championships?

The answer to the first question is that in the NBA, more than in any other sport, talent tends to prevail over the course of a seven-game series. The sheer number of possessions in an NBA game that result in points tends to propel the more talented team to a victory.

Consider the NHL or baseball. One fluky goal, or one mistake by a pitcher that turns into a home run, can very well be the difference between a win and a loss. In football, a single interception or a poor ball spot can cost a team an extremely valuable scoring opportunity.

In the NBA, one missed shot, bad turnover, or incorrect call rarely means the difference between winning and losing. If that shot, turnover, or call comes at the end of a close game, it can seem to exert undue influence on the outcome, but remember that an enormous number of other previous plays have already taken place to produce precisely the situation the two teams find themselves in.

That means that a less talented team needs to count on a large number of fluky plays going in its direction, over and over again over the course of a game, for several games over the course of a series. The greater the talent discrepancy, the luckier the worse team needs to be. Add it all up, and it's no wonder that the eight-over-one upset has happened just three times in NBA history.

That only answers part of our question, though. We haven't yet figured out why we see the same teams earning those championships year after year. Why is it so difficult for new teams to break through?

Part of the answer lies in the fact that the NBA is an extremely individualistic league, in that individual players can create outcomes for their teams more than they can in other leagues. We see similar effects when an ace starting pitcher wins a disproportionate amount of games for an otherwise struggling baseball team, or when an NHL goaltender carries a less-talented team to a lot of low-scoring wins. In basketball, every player is a potential ace starting pitcher.

Again examining the last 28 years, we see the same names pop up in the NBA Finals MVP column repeatedly. In fact, six players have accounted for 64 percent of all Finals MVP awards in that time frame: Michael Jordan (six Finals MVPs), Tim Duncan (3), Shaquille O'Neal (3), Larry Bird (2), Hakeem Olajuwon (2), and Kobe Bryant (2).

Also on the list are Kareem Abdul-Jabbar, Magic Johnson, James Worthy, Joe Dumars, Isiah Thomas, Chauncey Billups, Dwyane Wade, Tony Parker, Paul Pierce, and Dirk Nowitzki.

Jordan, Bird, Olajuwon, Abdul-Jabbar, Johnson, Worthy, Dumars, and Thomas are all Hall of Famers. Duncan, O'Neal, Bryant, are surefire first ballot selections, as is probably Dirk Nowitzki. Parker will likely make the Hall of Fame because of his international standing. Chauncey Billups is the only member of that group unlikely to be elected to the Hall of Fame.

So it takes not only star power but Hall of Fame-caliber players to win an NBA championship. Indeed, many of those players were at once time or another considered the single best player in the NBA.

And there's the rub. These players are so few and far between, that of course it's difficult for a team to find them. But surely some stars develop late, allowing teams to draft diamonds in the rough that can turn around a franchise's fortune?

Unfortunately not. Of that list of Finals MVPs, Abdul-Jabbar, Johnson, Worthy, Olajuwon, O'Neal, and Duncan were all selected first overall in their respective drafts. Jordan was famously drafted second overall after Sam Bowie in 1984, and Thomas was drafted second overall in 1981. Chauncey Billups was the third overall selection in 1997.

That means that of the last 28 NBA championship squads, 19 featured a Finals MVP drafted either first, second, or third overall.

Do we observe the same phenomena today? If so, the best teams should typically feature the best players, and those players should have largely been chosen very high in the draft.

To measure that effect, we can look at the top ten players in ESPN's John Hollinger's PER statistic, which measures a player's per-minute productivity. Since, generally speaking, coaches will allot more minutes to their most efficient players, this will help us figure out whether the NBA's best players belong to the NBA's best teams, and whether they were drafted disproportionately high.

The table below lists the top ten players in terms of PER, whether their team will make the playoffs, whether their team currently holds a top-three seed in its respective conference, and where each player was drafted.

Player Team Playoffs? Top 3?       Draft Pick     
LeBron James Miami Yes Yes 1
Chris Paul LAC Yes No 4
Dwyane Wade Miami Yes Yes 5
Kevin Durant OKC Yes Yes 2
Kevin Love Minnesota No No 5
Dwight Howard Orlando Yes No 1
Manu Ginobili San Antonio Yes No 57
Derrick Rose Chicago Yes Yes 1
Andrew Bynum LAL Yes Yes 10
Blake Griffin LAC Yes No 1

The results speak for themselves. Four of the top ten players were selected first overall, and only Bynum and Ginobili were selected outside of the top five. It's worth noting in this case that Ginobili was an international draft pick and Bynum was selected at the age of 18, which may help to explain why they fell in the draft relative to comparable top performers. Ginobili also plays only 23 minutes per game for the Spurs.

It's also worth noting that in addition to Ginobili, the Spurs feature former first overall pick Tim Duncan, ranked 13th in PER. In addition to Bynum, the Lakers also feature former 13th overall pick Kobe Bryant, ranked 18th in PER.

Only Kevin Love's Timberwolves will miss the playoffs, and the Heat, Thunder, and Bulls--all featuring top-two draft picks in the top 10 in PER--are the most likely favorites to win the NBA title this year.

Much is made of the disastrous middle ground in the NBA, occupied by low-seeded playoff teams. Those teams have historically no shot at all of winning an NBA championship, but also an extremely low probability of selecting the kind of player who could put them in a position to earn a higher seed, and therefore a better chance at a championship.

This research shows that to be very much the case. Overwhelmingly, the best regular season teams feature extremely high draft picks, and that regular season success repeatedly translates to postseason success.

It's no wonder that below-average NBA teams take it upon themselves to tank in hopes of a better draft pick (just ask the Warriors what, exactly, is wrong with David Lee's groin). A handful of extra ping pong balls in that year's draft lottery is far more crucial to a team's long-term well-being than a couple of ultimately meaningless wins at the end of the regular season.

The news is even more troubling, though, for teams like this year's Pacers, owners of the NBA's fifth best record. The Pacers, currently holding onto the third seed in the Eastern Conference, feature no top-10 picks on their roster.

In the past 28 years, no such third-seeded team has ever won an NBA Championship. Ever. In fact, last year's Mavericks were the first team in that time frame to win as a three-seed or lower without featuring a top-3 pick (Dirk Nowitzki was chosen ninth overall in 1998).

So what are the Pacers to do? To win a championship, they'd have to buck decades' worth of history that tells us they can't. Of course, they've also almost definitely put themselves out of the running for a draft pick that would allow them to select the kind of player that could propel them over the top.

Of course, at least the Pacers are winning. The Rockets and Bucks, for example, will miss the playoffs but still likely end up with just a low lottery pick, barring the intervention of a little luck with the ping pong balls.

That's not to say that every high lottery pick is a ticket to success. Ask the Wizards how they're doing with  John Wall right now. It's also not to suggest that a superstar former high draft pick is an automatic ticket to the conference finals. Carmelo Anthony and the Knicks are currently finding that out the hard way.

But it does suggest that certain criteria are more or less required for a team to be considered championship-caliber material: namely, a top-3 seed and an excellent former high draft pick. That leaves us, at the beginning of each playoff season, with just a small handful of teams that have any real chance to bring home a title.

And that is the great travesty of the NBA. Bad teams are terrible. Good teams often just aren't good enough, and never will be. And mediocre teams find themselves in purgatory, unable to move very far in either direction.

On the bright side, if you're not a fan of the Bulls, Heat, Spurs, Thunder, or Lakers, you've got a little extra free time to kill while the NBA playoffs are on. Then again, you probably knew that. After all, it's been that way for years.

1 comment:

  1. An addendum and semi-summary: I think that there are three major factors that contribute to the state of affairs I've revealed above.

    First, the ability of one player to take over a game plays a huge role. I think that's the most likely explanation for the frequency with which championship teams feature transcendent players. Matt Kemp and Clayton Kershaw posted historically awesome seasons for the Los Angeles Dodgers last year...and the team still finished 11.5 games back in the NL West. Meanwhile, Kevin Love is the only top-10 PER player who isn't going to make the playoffs. Basketball is the least team-oriented of the four major sports.

    Second, as I mentioned in the post, the number of scoring opportunities in a game and the length of a series plays a big role. The more scoring plays over the more games, the more often we're going to see talent win out. A lesser NFL team is much more likely to outscore a better opponent in the first half than it is to outscore that same opponent in a full game. NBA games feature a ton of scoring opportunities, and a seven-game series means a lesser team will rarely see enough bounces go its way to make up for a significant talent discrepancy.

    Finally, I think there's something to be said for the difference between skill and athleticism and how easy it is to measure each. The NBA skews more toward athleticism than any other league, and it's much easier to quantify someone's height and leaping ability than it is to measure his or her "basketball IQ" or "field/court/ice vision" or whatever. That makes it easier to determine which players have a real chance at greatness, and those players are consequently much more likely to be chosen higher in the draft. Players who don't stack up physically have much lower ceilings in the NBA than they do in other sports, I think.

    Add it all up and you've got rare franchise players who are, for the most part, pretty easily identifiable and therefore are usually going at the very top of the draft. Those players exert undue influence on the outcome of each game they play in, because individual ability plays such a huge role. And as we increase the sample size over more and more scoring plays and more and more playoff games, we see the outcomes regress, predictably, toward the teams with the most talented individual player(s).