NBA Four Quarters (02/17/17)

By Sean Pyritz

We're Going Streaking

Over the weekend, the Miami Heat concluded an improbable 13-game winning streak in Philadelphia – a fitting end to one of the strangest stretches in NBA history. Prior to the streak, the Heat sat on a tank-tastic 11-30 record. Injuries to Justice Winslow and Josh Richardson mixed with swirling Goran Dragic trade rumors doomed Miami's season to the bottom of the lottery. However, this scrappy group played inspired, hard-fought basketball for three weeks to dig themselves out of its grave. The dynamic backcourt duo of Dragic and Dion Waiters combined to average 43.4 points, 8.5 rebounds, 11.5 assists per game; Hassan Whiteside averaged a double-double and patrolled the paint on defense; and James and Tyler Johnson brought a strong jack-of-all trades punch off the bench.

From a historical perspective, the Heat's streak is just the 67th single-season winning streak of at least 13 games in the shot clock era (since 1954). Their 0.268 winning percentage entering the streak is the third-lowest overall and the lowest by far for streaks coming so late in the season. On average, the teams that won at least 13 games finished the season with 60 wins. The 2008 Portland Trail Blazers, who also won 13 games in a row, finished at only 41 wins and missed the playoffs. If the Heat miss the playoffs this year, they would become only the second team with a 13-plus game winning streak to count lottery balls. Just two games back of Detroit for the 8th seed at the All-Star Break and with Josh Richardson's return impending, the Heat are in position to make a run for the playoffs, an impossibility a month ago and insurmountable at the beginning of the season after losing both Dwyane Wade and Chris Bosh. Whether or not the organization wants to make that playoff run is an entirely different and more sinister discussion for another time.

Skeptical Statistics

Over the past few weeks we've analyzed the nuances of the possession from definition to estimation. We can now begin investigating the implications of the possession on analytics at-large, starting with one of its poster children: offensive rating. Regular readers are very familiar with offensive rating, or simply points scored per 100 possessions. The formula for figuring out offensive rating is simply: (Points/Possessions) * 100. I've been tracking games manually over this time to provide a sample of games to compare statistics against the major information sites Basketball-Reference (B-R) and (NBA). Keep in mind that all the games I've tracked have been close games, so it is not necessarily a representative sample of all NBA games.

As it stands now, my sample of ten games leaves 20 offensive ratings to examine. The average offensive ratings for the 20 teams from B-R, NBA, and my own tracking are 112.81, 109.61, and 112.78, respectively. Last week we hinted that B-R's more accurate method of estimating possessions made it the more reliable site for possession based statistics. Now we have more concrete proof of the case for B-R over NBA on the analytics front, at least when it comes to team-based statistics. The closeness of B-R to my own tracking is a bit misleading though. If you take the absolute difference in offensive rating between B-R and myself for each team, you get an average deviation of 1.5 points per 100 possessions.

The largest individual difference comes from the Minnesota Timberwolves. In its January 30 matchup versus Orlando, B-R lists the Wolves' offensive rating as 108.98, while I had it at only 105.05. While I did have a different possession count, there were other circumstances surrounding points scored that we will expound upon at a later time. For now, we have established the advantage of B-R when it comes to overall team analytics.

The Boston Celtics took one of the most heartbreaking losses of the season in Chicago on Thursday night. The hearts of Celtics fans must be secretly beaming due to another stupendous performance from Kelly Olynyk off the bench. In the four games Boston played over the past week, Olynyk has averaged 16.8 points, 7.0 rebounds, 3.0 assists, and 56.3% shooting from three-point range on 4.0 attempts. Always an elite offensive center due to his playmaking and shooting abilities, Olynyk has continued to grow in his fourth season, boosting his finishing abilities well above career norms. In fact, Olynyk is among the league leaders finishing around the basket. Even if he never develops the mindset to be a better defensive player, Olynyk is already a deadly third big with a 6th Man of the Year Award certainly on his career trajectory radar. Until then I hope he accepts our 6th Man of the Week honors.

A Tale of Two Defenses

The San Antonio Spurs have the best defense in the league. The Denver Nuggets have the worst defense in the league. Nearly ten points per 100 possessions separate the two teams. This past week both teams visited Madison Square Garden and, predictably, the results were vastly different. Below, the Knicks performance against each defense is detailed. Half stands for half-court possessions, which are determined by my judgment (more on that in the future). Other is everything that doesn't fall into a strictly half-court possession like transition. The biggest difference between the two defenses is their half-court defense. The Knicks spent nearly the same proportion of both games in the half-court, yet scored at a rate 30 points per 100 possessions worse against San Antonio.

The first place to start when analyzing this result is the fact that New York changed lineups when facing the Spurs – starting big to match the Spurs size. By playing with two traditional bigs on the floor, the Spurs can force other teams to do the same and inherently hurt their own offense by shrinking their spacing, particularly in the half-court, which is exactly what happened here. Bolstering its half-court defense is the personnel advantage of the multi-time Defensive Player of the Year, Kawhi Leonard; and he didn't even start on Carmelo Anthony – that assignment went to Danny Green. Denver was forced to play Darrell Arthur on Melo for large stretches of the game. The Spurs bigs are experienced and capable of concentrating on and anchoring a defense for longer stretches than the young guns in Denver, Nikola Jokic in particular.

When breaking down the Knicks performance in the half-court, a few things jump out. First, the Spurs held the Knicks to just 50 % effective field goal percentage. On the other hand, the Nuggets allowed an eFG% of 67 %. A major factor in these numbers was pull-up mid range jumpers. The Knicks shot 8/9 on pull-up two-pointers against Denver, but just 4/15 against San Antonio. The Spurs corralled and harassed Melo much better, as to be expected with the personnel advantage. Melo was just 4/12 in the half-court against the Green/Leonard duo, including 1/6 on midrange pull-ups. Denver got smoked by Melo in the half-court, allowing him to shoot 10/14 which featured a four for four showing on catch-and-shoot three-pointers.

Ultimately, these numbers point to a lack of discipline and focus on the part of the Nuggets defense, allowing far too much penetration in the half-court against what is an average Knicks offense. The Spurs are very disciplined and make a point of running opponents off the three-point line corralling them into the midrange without introducing too much help and forcing scramble situations. Denver relies upon its offense to overpower teams, a successful strategy against New York. San Antonio typically doesn't need its offense to show up to win (it usually does anyway) but the Spurs could not overcome a 36.3 % shooting night to win at MSG.