Stat Central: Welcome To The New NBA, Part Two — Death To Overlays

The Houston Rockets relentless pace highlights the reasons we need defensive rating. (Photo Credit: Michael Dunlap, HoopsHabit.com)

Check out part one, the season’s first Stat Central column, on the evolution of the three point shot.

During a typical NBA broadcast I roll my eyes, audibly sigh, sarcastically mock, or sometimes just yell vehemently at my screen out of sheer frustration for the ignorance coming from the screen. There are exceptions — a few local broadcast teams and crews are a treat to catch on league pass — and I readily admit there is a level of snobbishness and lack of compassion on my part here that is a serious knock against my character. But when people that get paid to analyze basketball are not particularly good at it, I get a little agitated.

One of the more frustrating offenses is typically a fault of the broadcast crew, and not necessarily the guys calling the game. That offense: those all too frequent statistical overlays that display points or opposing points per game along with field goal or opposing percentages in an effort to highlight a teams’ offensive or defensive prowess. There was a day and age where points per game and field goal percentage were the most accurate markers of defensive and offensive success we had at our disposal. That time is long gone and we have much more adequate markers, namely offensive and defensive rating.

Who Cares How Many Points You Scored?

These “ratings” are calculated estimates of how many points a team allows or scores per possession. So, if a team has a 11o offensive rating, they averaged 110 point per 100 possessions. And if a team has a defensive rating of 90, they hold teams to 90 point per 100 possessions. These number are more effective than per game numbers, then, because every NBA basketball game has a different number of possessions in it. If the teams play at a slower pace, their will be less possessions, and obviously, less points scored; while teams playing at a faster pace would get much more shot ups and have more possessions. Just because the teams playing at a faster pace scored more points does not necessarily mean they were better on offense, and likewise, the teams playing slower are not better defensively because they held the opposing team to less points. Instead the teams that scored more or allowed less points relative to the amount of possessions played should be considered the better offense and defense.

Here are some more tangible examples. The Houston Rockets allowed 102.5 points per game last season, ranking 28th in the NBA. Despite that low ranking Houston actually rode the defensive prowess of Omer Asik to an average NBA defense last year. The Rockets earned a 103.5 defensive rating which ranked 16th in the NBA — a much more passable rank than the 28th number mentioned earlier. See, the Rockets played at an incredibly fast pace last season, actually earning a league high 98.64 pace rating (all ratings via NBA.com) — pace rating is the calculated estimate of a team’s average number of possessions per game. Since Houston played so fast, their games typically had a high number of possessions and as a result were high scoring for both teams. So Houston gave up a lot of points per game, but were actually league average when came to their actual points per possession. Their points allowed per game ranked them in the bottom of the league, but their defensive rating painted a clearer picture of their average defense.

The D’Antoni era Suns were another group that appeared worse on defense than they truly were because of their frenetic tempo. For the 2007-08 they played at a trailblazing 99.05 pace (4th) and relented the 25th lowest points allowed per game mark (105 points). They, like Houston, were actually at the middle of the pack defensively though, ranking 16th in defensive rating with a 104.9 figure.

Not taking pace into account can make a slower paced team look better on defense as well. Take last year’s Brooklyn Nets as an example. They Nets played relatively slow last season, only earning a 91.23 pace rating (28th). They only allowed 95.1 points per game — ranked 6th, defensively — but only earned a 103.6 defensive rating. That number ranked 19th and Brooklyn actually relented more points per possession than Houston despite holding teams to 7.4 less points per game. Brooklyn played extremely slow a gave up a low number of points because of it, while Houston played at a rapid pace and allowed plenty of points as a result. But the two teams were actually close to identical concerning how they fared defensively

There is one more example I want to look at that highlights offensive rating being just as effective. The Memphis Grizzlies only scored 93.4 points per game last season — tied for 26th in basketball. However, Memphis also played the second slowest pace in basketball (91.15) which heavily factored into their low per game totals. When accounting for pace the Grizzlies actually fare in a much more average manner, earning an offensive rating of 101.7 which ranked 18th in the NBA. Memphis was not a great offensive team by any stretch last year, but they were nowhere as bad as their points per game ranking would indicate and were in fact quite average as far as scoring the basketball.

The Pace Effect

Before we get to field goal percentage, I do want to use this topic as an opportunity to see if teams that play at faster speeds are actually better on offense — and conversely do teams that play slow play better D. And upon very surface level analysis of last year’s numbers, the answer is…”kind of”. Only two of the teams that ranked top ten in pace last season finished with an offensive rating below 103 — the Suns and Milwaukee Bucks (remember the good times Phoenix). So in general playing faster last year did seem to boost a team’s offense, except in cases where the offense was generally deplorable on all fronts.

Defensively, four of the NBA’s top defense — Indiana (1st), Memphis (2nd), Chicago (5th), and Miami (7th, all in defensive rating) — were bottom ten in pace. However, slower pace did not automatically result in good defense, either. The Pelicans played the slowest pace in basketball but only managed a 107.6 defensive rating — the third worst figure in basketball. Similarly the Detroit Pistons and Toronto Raptors were also bottom ten in pace, but allowed over 104.5 points per 100 possession — a horrible defensive rating placement.

Their are examples of fast paced teams playing good defense — particularly, the Spurs ranked 6th in pace and 3rd in defensive rating — and the Heat, New York Knicks, and Nets are all examples of bottom ten pace teams earning top ten offensive ratings. But their is a precedent of faster paced teams having an easier time on offense, and slower paced team having more success on defense. It makes sense. Playing fast opens up the court more for ball penetration and open shots; it is easier to grab good post position off of a break; and it is also easier to take advantage of cross matches and slow retreating defenders in transition. But playing fast does not condemn a team to horrid defense and playing slow does not confine them from being an offensive powerhouse (I mean the Miami Heat for God’s sake!!!!). It does require great coaching and player discipline, though, to succeed on both ends while playing at either extreme.

It’s not the percentage of shots you make; it’s where you’re taking them from

Understanding how field goal percentage became one of the accepted measures of offensive efficiency is not difficult. If you make a higher percentage of the shots you take, you would figure that means you are ‘better’ on offense. This thinking is an issue for two reasons. 1) Threes are worth more than twos and 2) if you are fouled on a missed shot attempt it is not recorded as a field goal. Take for example, last year’s Boston Celtics team. They ranked 6th in the league in field goal percentage, shooting 46.5 percent from the field. Despite their high tendency to make shots, Boston was actually below average in terms of actual points per possession. Their low scoring numbers, despite a good field goal percentage, was a product of their scarcity from the free throw line and from three. The Celtics made only 17 free throws per 100 possessions — 20th in the NBA — and ranked 24th in made threes per 100 possessions (making 6.4).  The Celtics were good when it came to the basic “take shot, make shot”, but not enough of those shots came from three and they did not earn many ‘free’ points at the line.

The shortcomings of field goal percentage as a measuring stick show up on the defensive end as well. The San Antonio Spurs and the Chicago Bulls were two of the best defensive teams in basketball last season — ranking 3rd and 5th in defensive rating. However, these two teams only ranked eighth (Chicago, 44.2) and ninth (San Antonio, 44.3) in field goal percentage. These two teams, though, had the concept of preventing the three ball the most deeply embedded into their defensive schemes. Chicago ranked 2nd in threes allowed per 100 possession, allowing only 6 threes; while San Antonio ranked 3rd right behind Chicago only relenting 6.5 threes per 100 possession. San Antonio was also great at defending without fouling, as well. The Spurs only relented 15.5 free throws per 100 possession, which ranked 3rd in the NBA. So while these teams field goal percentages barely cracked the top ten, their defensive ratings better reflected their defense as a whole and acknowledged their ability to take away particularly affecting shots.

And that is the advantage offensive and defensive rating as statistics. It may not delve very deep into the reasons a particular offense or defense is effective, but it is the best measurement of just how effective that offense or defense is.