By David Roher
Consistency is funny. We usually consider it a good thing to have, and we remember prolonged, steady excellence for a long time. Joe DiMaggio’s 56-game hit streak. Cal Ripken’s 2,632 consecutive games. Oscar Robertson’s 1962-season-average triple-double. Takeru Kobayashi’s winning the Nathan’s Hot Dog Eating Contest for six years in a row.
But there’s nothing about consistency that is inherently good. It just means that there’s a continued production of results that are all similar to each other, which says absolutely nothing about the quality of those results. There are plenty of things that are consistently terrible, like three-week-old sushi and the Los Angeles Clippers. On the blog, we’ve been pretty consistently ignoring anything about basketball. That’s another bad thing, and something I’ve decided to change with this series. Where better to start than the New Jersey Nets?
The motivation for the series was the fantastically dreadful manner in which the Nets opened their 2009-10 season, losing their first eighteen games. As they proved in game #19, however, they’re not a bad enough team to lose every single time they step on the floor. In fact, according to John Hollinger’s current Power Rankings, they’re not even the worst team in the league. Even during their 18 Games That Shook the World, the Nets’ points scored and allowed dictated that they should have won 2 or 3 (through their Pythagorean Expectation, a method of estimating record based on runs/points/goals).
Luck, or lack thereof, was probably the greatest cause of their streak. But it got me thinking: was there something about the Nets, other than simply their inability to play quality basketball, that enabled them to provide us with such a shining beacon of ineptitude? Perhaps, I thought, it had to do with their consistency.
Let’s say a team scores 80 and allows 130 points per game, which would likely make them the worst team in history, worse than the Nets could ever dream of being. Their Pythagorean winning percentage would be just .001. But couldn’t the way in which they achieved those averages give them a drastically different result? For example, this team never scores less than 75 or more than 85. However, it allows 70 or 190 points per game, depending on whether it was home or away (this team really loves its fans), which means they win every single home game, giving them a winning percentage of .500. Nothing in this example would ever happen to that degree in real life, but hopefully through the exaggeration you can see the possibility of it happening in reality to a lesser effect.
Maybe the Nets, by staying close to their atrocious averages, were more likely to go on a winless streak. To address that question properly, however, we’ll have to dig a lot deeper into the consistency question. I’ve compiled, through the use of our own Dan Yamins’ aggregated game-by-game results, a consistency index for every team since the start of the NBA in 1950. It’s simply a measure of the standard deviation of each team’s game-by-game point totals. The analysis of the data will come in the next installments (including adjustments for differences in average point totals, games per season, etc.) but just to give you a sneak peek, here are some of the results:
Most Consistent Offenses since 1950 (Team, SD of Points Scored per Game)
1999 Miami Heat, 6.74 (Lockout year)
1986 Detroit Pistons, 8.09
1954 Milwaukee Hawks, 8.40 (Shorter season)
1975 Philadelphia 76ers, 8.40
1994 Minnesota Timberwolves, 8.62
Least Consistent Offenses
1950 Sheboygan Red Skins, 15.75 (Shorter season)
1951 Fort Wayne Pistons, 14.99 (Ditto)
1984 Denver Nuggets, 14.78
2007 Washington Wizards, 14.72
1991 Houston Rockets, 14.58
Most Consistent Defenses
1999 Sacramento Kings 7.69 (Lockout year)
2001 Miami Heat, 8.68
2002 Miami Heat, 8.79
1964 St. Louis Hawks, 8.81
2007 Atlanta Hawks, 8.83
Least Consistent Defenses
2008 Toronto Raptors, 15.53
1984 Denver Nuggers, 15.52
1993 Philadelphia 76ers, 15.50
1950 Denver Nuggets, 15.42 (Shorter season)
1951 Fort Wayne Pistons, 14.97
Over the next few posts, I’ll be looking at these numbers further, determining both their causes and possible effects. Stay tuned.