We are now two weeks into the 2011-12 NBA campaign, and if there's been one defining characteristic of the season thus far, it's been blowouts. Any given night, some team is getting run by 20-plus points. Often, these are good teams. On Friday, it was the Portland Trail Blazers losing by 25 points in Phoenix. A night later, the Chicago Bulls suffered a 15-point loss to the Atlanta Hawks that wasn't as close as the score indicated. Meanwhile, the Philadelphia 76ers blew out the Toronto Raptors in a 97-62 final that was every bit that lopsided.
During the Portland game, I tried to coin a name for these kinds of losses where a team has no energy and can't hit anything from the field: the Lockout Loss, one caused by this year's compacted schedule, even if it's not necessarily at the end of a lengthy stretch of games. The Lockout Loss is a great explanation for the number of blowouts we've seen.
To try to come up with a statistical definition for lockout losses, I went to my schedule-adjusted ratings, which are calculated on a game-by-game basis. My first cut was to look at games where a team's outcome--the margin, adjusted for location and quality of opposition--was 20 points worse than its usual performance. Through Saturday, I found eight such games, including the aforementioned Portland loss. Over an equivalent period at the start of last season, there were four games that were so far out of character for teams. There were seven in 2009-10, but just three in 2008-09.
There isn't any particular pattern to Lockout Losses. Half of them have come as the second game of a back-to-back, but the other half featured fairly typical rest--one day off three times, and two days off for the Los Angeles Clippers before a 25-point loss to the San Antonio Spurs.
(The complete list of Lockout Losses: ATL 106, @NJN 70 (12/27); @SAS 115, LAC 90 (12/28); NYK 114, @SAC 92 (12/31); @CHI 104, MEM 64 (1/1); @MIA 118, IND 83 (1/4); @PHX102, POR 77 (1/6); NYK 103, @DET 80 (1/7).)
We'll see whether this pattern continues. One odd aspect of the statistical definition I'm using for a Lockout Loss is that they become more common as the season goes on, since a single outlying loss is no longer such a big factor in a team's overall rating. By the end of last season, there were 70 total losses that met the criteria--including seven by the Atlanta Hawks alone. Charlotte, Miami and Portland had four apiece, while the Boston, Golden State and Houston never played 20 points worse than their usual level. The spread of teams at both extremes indicates that nights like this don't really tell us much about a team's overall quality.
Of course, there have been many blowout losses that don't qualify as Lockout Losses because the team in question has played so poorly overall. The evidence that something unusual is happening this season is even more overwhelming when we look strictly at lopsided games, regardless of location or competition. Of the 119 games played through Saturday, 23--nearly one in five--was decided by at least 20 points. That's way up from the same period for recent seasons. The chart below shows the standard deviation of scores, as well as the number of 20-point games.
Year SD 20+%
2009 7.1 .092
2010 7.9 .160
2011 8.2 .160
2012 8.5 .193
The last two years have featured an identical number of blowouts through the first 119 games--19 apiece, or four fewer than this season. That's a difference of about one extra blowout every three or four days. The beginning of the 2008-09 campaign was remarkably competitive, with just 11 margins of 20-plus points through about the first two weeks.
As you might expect, lopsided outcomes are more likely early in the season, when teams are still trying to integrate newcomers and install new systems. Here are the final numbers in the same categories for the last three years.
Year SD 20+%
2009 7.4 .144
2010 7.5 .137
2011 7.6 .122
As it turned out, 2008-09 ended up featuring more blowouts than the other two seasons, which indicates how little such a small sample tells us. The standard deviation of scoring margins ended up being remarkably consistent. Still, based on what we've seen so far, it looks likely we'll end up with a standard deviation somewhere around eight points, a noticeable change from the recent past.
What's surprising about less competitive play post-lockout is that it did not apparently happen in 1998-99. In fact, the exact opposite trend was apparent--the start of that lockout-shortened season was unusually competitive. Here's how it scored, both over the first two weeks and for the full season, on the same methods.
Year SD 20+%
early 6.2 .076
full 7.2 .103
Such a counterintuitive finding makes it more difficult to construct a logical narrative around the numbers we've seen so far. Maybe they're just a fluke, or maybe they're an indication that this year's shortened season will be very different from 1998-99. For now, it's too early to tell.
There's another interesting trend in the early going that the numbers confirm. Whether it's because of the rapid travel, the additional importance of fan support as a motivational factor or whatever explanation, home-court advantage has been more important early this season than ever before. The league home-court advantage is 4.9 points per game and home teams are winning 64.7 percent of the time. Typically, teams win about 60 percent of their games at home and enjoy an advantage of about three points, so the difference is noteworthy. This trend too is worth watching the rest of the shortened schedule.
Get your copy of the most comprehensive guide to the new NBA season. Pro Basketball Prospectus 2011-12 is now available in .PDF and paperback format.
Kevin Pelton is an author of Basketball Prospectus.
You can contact Kevin by clicking here or click here to see Kevin's other articles.