Quick Stat Hits: The Defence - What Happened?

There was a great piece this morning by Zack Lowe looking at the Raptors' defensive issues of late. I thought I'd take a look at exactly which bits of Toronto's defence have been struggling, and what impact that has been having.

First, the difference between the first 14 games and the most recent 20 games is pretty stark. In the first 14 games, the defence gave up a paltry 97.8 points per 100 possessions. That would be good for second best defence in the league right now. In the 20 games since, 109.1, good for third worst in the league. So, where are those 11.7 additional points coming from?

First, we have to look at who they've played. In the first 14 games, their opponent's average Offensive Rating (ORTG) was 102.8. Since, it has been 104.1. So, applying that difference to a league average Defensive Rating (DRTG), that means they've actually performed as a 98.6 DRTG team in the first 14 games, and a 108.6 DRTG team in the last 20.

Next, we'll look at some of the strengths of the defence from the first 14 games (on a per 100 possession basis).

Opp FT%: 67.4%
Opp TOV: 18.2
Opp FG%: 45.3
Opp 3PT%: 35.0
Opp 3PA: 22.0
Opp BLKA: 4.9
Opp PFD: 22.8

So, the opposition was missing free throws, turning the ball over a lot, shooting OK, hitting 3's at an average clip (but not taking too many of them), getting blocked at the rim a good deal and drawing fouls on about 1/5 of their possessions, respectively.

What has happened to those numbers since?

Opp FT%: 77.4%
Opp TOV: 14.0
Opp FG%: 46.8
Opp 3PT%: 36.2
Opp 3PA: 23.7
Opp BLKA: 3.9
Opp PFD: 22.1

So. First and foremost, the opposition is hitting their free throws now, and at an above average clip. The league average FT% is 75.2%. Normalizing the FT% from the first and second data sets to league average changes the defensive numbers from 99.6 and 108.6 to 100.5 and 108.1. So now we have to account for 7.6 points of defensive slippage.

The biggest noticeable difference is in the turnover rate. We can adjust for opposition here as well. The opponents in the first 14 games typically turn the ball over 15.6 times per 100 possessions. Last 20 games, 14.9. So 0.7 turnovers are based on the opposition. The rest of the difference in turnover rate is due to a difference in the defence. That means 3.5 turnovers (each turnover is worth about 1 point) have been lost in the latest 20 games.

Those are the big two. Simple regression to the mean in terms of opposition free throw percentage has cost the Raptors about 2.5 points (or a drop from second in the league to fourth). The Raptors defence has stopped forcing turnovers, costing them about 3.5 more points (a drop from fourth down to 18th). The rest of the change has been:

a) a tougher schedule, and

b) a combination of a couple more 3-point attempts allowed at a similar clip and a slightly higher overall shooting percentage

These two things are harder to nail down in terms of the cause, besides "defensive breakdowns".  Although it should be noted the Raptors are giving up two more fast break points per game, which probably leads to that higher FG%.

So what to do about it? How were the Raptors forcing turnovers earlier but aren't now?

To get an idea, let's look at some individual Raptors' steal numbers as a proxy for turnovers forced.

Player | STL rate first 14 | STL rate last 20

Kyle Lowry: 1.7; 2.6
DeMar DeRozan: 2.3; 0
Terrence Ross: 1.3; 1.2
Greivis Vasquez: 1.7; 1.0
James Johnson: 2.8; 1.6
Lou Williams: 3.4; 1.7

Lowry has improved and Ross has held steady, but otherwise the primary guard and wing players have dropped way down in terms of generating steals.  Losing DeRozan hurts as well, with Vasquez and Williams taking a lot of his minutes, and both are generating far less steals lately than he was.

It seems the primary cause of the drop in turnovers forced is a lack of aggression by the Raptors' smaller players in the passing lanes compared to earlier in the season. Perhaps an increased minute and offensive load for all of them caused by the absence of DeRozan is the reason behind this.

With DeMar returning soon, it will be interesting to monitor the steal rates of the wing players on the team. If those go back up with his return, we should see the team return to an above average defence, even if a more reasonable opponent FT% keeps them from looking like an elite defence as they seemed to be earlier in the year.  Throw in his free throw generation as a way to suppress fast break points and force half court sets from the other team, and perhaps his impact is felt more on the defensive end than we thought - just not in as direct a way as some others.