Last Year’s Warriors Weren’t The Best Ever, But This Year’s Might Be


There was plenty of “greatest of all time” speculation swirling around the Golden State Warriors at this time last year, as they tore through the Western Conference after breaking the 1995-96 Chicago Bulls’ record for regular-season victories. As it turns out, though, the Warriors weren’t even the greatest team of the 2015-16 season, since they lost to the Cleveland Cavaliers in the NBA Finals. I suppose we all learned our collective lesson, because there hasn’t been anywhere near as much GOAT chatter this time around.

The irony, of course, is that this version of the Warriors might actually be the best NBA team ever. Although they “only” won 67 games during the regular season, the 2016-17 Warriors had a better schedule-adjusted point differential than they did in their 73-win season, and then they rattled off what is so far the most dominant postseason of any team in recent memory (including the fabled 2001 L.A. Lakers).

And if they can take care of Cleveland without much trouble in Game 1 of the NBA Finals, they’ll officially become the GOAT — at least, according to the Elo ratings, our pet system for judging an NBA team’s strength at any given moment.

As my boss wrote on Saturday, the LeBron James Cavaliers have had a tendency to make Elo look silly. But by now, Elo has learned from its mistakes and is (mostly) giving Cleveland its due. The main reason our prediction system is still giving the Cavs a minuscule 10 percent chance of defending their championship, then, is that the Warriors are just so ridiculously dominant.

Golden State’s Elo rating is currently 1850.5, which ranks second in NBA history to the peak rating of the ’96 Bulls (1853.1) — which was set after Chicago took a 3-0 lead over the Seattle SuperSonics in the NBA Finals. (The Bulls’ Elo would dip to a final rating of 1823 after they lost two of the next three games to the Sonics.)

TEAM SEASON DATE GAME NO. DURING PLAYOFFS? ELO
1 Bulls 1995-96 6/9/96 97 1853.1
2 Warriors 2016-17 5/22/17 94 1850.5
3 Warriors 2015-16 12/11/15 24 1838.6
4 Warriors 2014-15 6/16/15 103 1822.3
5 Celtics 1985-86 5/29/86 96 1815.7
6 Bulls 1996-97 2/4/97 47 1811.3
7 Spurs 2015-16 3/19/16 69 1800.1
8 Lakers 2008-09 6/14/09 105 1790.0
9 Bulls 1997-98 6/10/98 101 1788.3
10 Pistons 1988-89 6/13/89 99 1788.1
The best peak Elo ratings in NBA history

Source: ESPN, Basketball-Reference.com

To surpass those Bulls and set a new peak-Elo record of 1853.2, the Warriors would need to prevail by eight or more on Thursday night. That’s far from an unreasonable goal; Golden State has the superior rating and is at home, so Elo actually considers them nine-point favorites for Game 1. (The Las Vegas betting line for Game 1 has the Warriors winning by seven.) According to research by Wayne Winston and Jeff Sagarin, the probability of a nine-point NBA favorite winning by at least eight is about 54 percent.

But even if Cleveland plays the Warriors close in Game 1, a new Elo record is inevitable as long as Golden State keeps winning. A string of one-point victories in Games 1 through 3 would be enough to push the Warriors past the Bulls’ mark with at least a game to spare in the series. But any loss could be a major setback for the record-breaking bid. Even a three-point loss in Game 1 would need to be followed by four straight four-point wins to pull Golden State ahead of Chicago; a stray double-digit loss would require four 10-point wins in a row. And two losses in the series might prove too much for the Warriors to overcome unless they also rack up offsetting blowout wins, particularly if the Cavs’ wins are spaced apart in the series (Elo gives more weight to more recent games).

Then again, it’s also possible that the Warriors could set an all-time peak Elo record after Game 1 and then the Cavs could rally back to win the series. In that scenario, Golden State’s season would parallel that of the 2007 New England Patriots, who set the NFL’s peak Elo record when they rattled off 18 straight wins before losing the Super Bowl. We’re guessing the Warriors would probably rather have the title.