Will Cavs, Rockets and Raps keep slumping after a weird January?

— -- A quick look at the NBA's January standings shows they don't look much like the standings at the start of the month.

While the Dallas Mavericks, Minnesota Timberwolves, Philadelphia 76ers and Washington Wizards have surged up the standings, three top teams as of New Year's Day have slumped. The Cleveland Cavaliers, Houston Rockets and Toronto Raptors combined to go 73-26 during the 2016 portion of their schedule but were just a game above .500 (25-24) during January.

How common are such swoons? Was LeBron right to be worried about the Cavs?

According to data mined from NBA.com/Stats, Cleveland's decline in winning percentage from the first three months of the season to January was the fourth-largest over the five full seasons since the lockout-shortened 2011-12 campaign:

Though declines by the Raptors (minus-.217) and Rockets (minus-.155) weren't nearly so unusual in and of themselves, they contributed to a January that was less predictable from recent results than any in recent memory.

The correlation between a team's win percentage as of Dec. 31 and in January was just .395, much lower than past seasons. And while net ratings were far more stable (.580 correlation between January and the previous three months), that was still the lowest over the past five seasons.

Though it's nice to quantify how unpredictable -- or downright weird -- January was in the NBA, the real interest comes in determining how much it tells us about teams going forward. Using this data, the best way to predict a team's winning percentage from February through April over the past five seasons (with a .785 correlation) has used their net rating (margin of victory per 100 possessions), with net rating through Dec. 31 weighted about five parts to four parts for January net rating.

That means each game in January tells us relatively more than each game beforehand, but since there are about twice as many games played before the calendar turns, the pre-January sample is still more predictive as a whole.

Applying that formula to how teams have performed this season before and during January yields the following predictions for winning percentage from February onward:

Now, I wouldn't suggest these projections are better than those generated by FiveThirtyEight's CARM-Elo method or ESPN's Basketball Power Index. They don't account for schedule strength or preseason projections. Still, they provide an interesting way to balance recent streaks with season-long performance.

This method suggests that concerns shouldn't be too high in Houston and Toronto. The Raptors have a better projection than the Boston Celtics, who have passed them for second in the East, while the Rockets rank third in the West going forward.

However, this method does suggest real cause for worry on the part of the Cavs, who were so bad in January as to produce the East's fourth-best projection. I'd take the over, but LeBron James might not be overreacting to how Cleveland is playing.