The statistical accounting of basketball has gone through some considerable refinement over the past decade or so, but some basics have remained largely unchanged. Points are almost always indicative of a team’s offensive and defensive prowess: Once you adjust for pace, a team that scores a lot of them usually has a good offense and a team that allows relatively few has a good defense. Rebounds offer the same nearly ironclad conclusions: A good rebounding team gets more shots, scores more, and prevents its opponent from scoring.

Assists, however, are another story. In theory, teams with lots of assists excel at moving the ball. The Russell Celtics, the Showtime Lakers, and the championship-era Knicks are the classic examples; so too are the current Suns. Is the converse true? Do teams with few assists run stodgy, ineffective offenses full of selfish players?

You’d be surprised. The Dallas Mavericks rank near the bottom of the league in team assists this year at 17.7 per game. Only the Knicks, at 17.6 rank lower. The Knicks have a good excuse – they’re implementing a complicated new offensive system led by a new coach with almost an entire roster of new players. But the Mavericks, who host the Knicks next Monday, have been a top offensive team for years, and still score at frightening pace despite the absence of former assist master Steve Nash. This season, the Mavs rank second in the NBA in Offensive Efficiency with 113.6 points per 100 possessions.

The flaw may lie in how assists are allocated. Consider the following sequence from the Maverick’s 102-93 win over Golden State last week.

Point guard Devin Harris brought the ball up to the top of the free throw circle and passed to center Erick Dampier in the low post. Dampier faked toward the baseline, then passed the ball to swingman Josh Howard at the elbow. Howard started to drive the baseline, drawing Dampier’s defender to him. Howard returned the ball to Dampier, who, seeing the Golden State help defense close on him, passed on the contested layup and fired the ball to the opposite corner, where Jerry Stackhouse had spotted up to shoot a three-pointer. Harris’s defender rotated to Stackhouse, who moved the ball back to Harris near the top of the three point circle. Harris faked left and drove between his man and the help defenders to score an easy floater in the lane.

The sequence was a superb microcosm of the Dallas offense. Howard, Stackhouse, and Dampier each passed on a makeable shot and made good passes to teammates, which ultimately exhausted the Warrior defense and resulted in a high percentage shot and two points. Officially, no assist was given on the play because Harris seemingly created his own shot.

By contrast, a few minutes later, Golden State had the ball when a deflection caused a play to break down. Guard Jason Richardson looked to take the shot late in the shot clock but was smothered by Dallas defenders. He tossed a desperation pass to forward Troy Murphy, who nailed a 15-footer as the shot clock expired. Richardson was awarded an assist.

In theory, instances in which an assist isn’t really indicative of good ball movement and instances in which good ball movement isn’t rewarded with an assist cancel each other out. But the Mavericks stat line suggests otherwise. Not only are the Mavs near the bottom of the league in assists, but their average of 17.7 per game ranks them among the worst all time. Only the miserable 1976-77 Nets rank lower at 17.34. Those Nets were in the first year after trading Julius Erving to the 76ers for cash, and they were an awful team. Their Offensive Efficiency was 92.4, last in the league, and their 22-60 record was the league’s worst by eight games. Other than the fact that they play with that round orange thing, they couldn’t be more dissimilar to the current Mavericks.

The Mavericks offense generates few assists because it is geared to space the floor to create isolations for their premier offensive threats, forward Dirk Nowitzki and guard Jason Terry. Both are nearly unstoppable one-on-one players and the Dallas offensive spacing is designed to thwart teams who try to double those guys. In response to a double, the Mavs can move the ball quickly to the open man, often behind the arc. As a team, the Mavericks nail 39.1% of their threes, good for fourth in the league. But if the player fakes the three and dribbles to the lane before hoisting a shot, often no official assist is given.

This points out the flaw in the way assists are given. They can mean that a team moves the ball very well (as is certainly the case with Nash and his Suns), but their absence isn’t a sign of a stagnant, ineffective offense. Last season’s Supersonics and Pacers also finished in the bottom 10 all time fewest assists, and Seattle was second in the league in Offensive Efficiency at 112.5.

If we are to extrapolate a team’s strengths and weaknesses from statistics, then the next wave of basketball metrics will need to improve ball movement measurements. This isn’t hockey – a pass then a score doesn’t necessarily imply causation. Also, some teams move the ball purposefully while others play hot potato until somebody heaves a shot to be the 24-second clock. Separating and quantifying these scenarios will enable us to get a much clearer picture of offensive quality. The way assists are currently allocated lets a lot of good offense go unacknowledged.