"My rule was I wouldn't recruit a kid if he had grass in front of his house.
That's not my world. My world was a cracked sidewalk." —Al McGuire

Friday, January 04, 2008

Providence Numbers Recap

In our Providence Preview, we said to watch for Offensive Rebounding Percentage, the Turnover Rate that Providence was able to force, and how well Marquette did on eFG% because Providence was weak in that aspect. Marquette ended up with a decisive advantage in every category, which led to the dominant performance.

Offensive Rebounding Percentage (OR%)
As a reminder, OR% is the percent of total available offensive rebounds that a team is able to grab. The equation is Team OR / (Team OR + Opponent's DR). In other words, every defensive rebound by an opponent is a potential offensive rebound too.

Both Providence and Marquette were amongst the top 15 schools in the nation at OR% (40.9% and 42.2%, respectively). For the game, Marquette managed 48.1% (well above season average) and Providence was held to 27.9%. Advantage Marquette.

Turnover Rate
Providence was averaging a forced turnover about one in every four possesions (24.3%). However, for the game, Marquette only had a turnover rate of 19.4%, which is also below the team's season average of 20%. Advantage Marquette.

Effective Field Goal Percentage (eFG%)
Finally, we said to watch for Providence's eFG% defense, which was in the bottom third of all D1 schools, primarily because opponents were shooting very well from three (38. 4%). Marquette is very strong at eFG%, and managed to show it against PC with an eFG% of 61.7. A good chunk of this high result came from shooting 43.5% from downtown. Advantage Marquette.

Finally, we said to look for the performance of 6'11" C Randall Hanke. He played 10 minutes and finished with 2 points. Can't win them all...

On the individual player side of things, Cracked Sidewalks is unveiling a new feature for the rest of the season.

Individual Results on a Per-Game Basis

Just a quick explanation of the various columns.

Possessions are the view of the amount of time that the player involved ended the action on the floor. A turnover is a possession, a Field Goal Attempt is a possession, and an Offensive Rebound cancels a possession. The actual equation is FGA + TO - OR + (0.475*FTA).

The next four columns are statistics from Dean Oliver's Basketball on Paper. Offensive Rating is similar as is found on Pomeroy's site for Marquette, except we are showing it on a per-game basis. Higher Offensive Ratings are better. The Individual points produced is "the number of points a player produces through scoring possessions, accounting for three point shots and how well he does at the foul line" (Oliver, 152). Basically, since basketball is a team game, it's a view of how many points can be attributed to that player.

The Defensive Rating is taken from the team's Defensive Efficiency (93.3 against Providence), and then accounts for each player's role in the offense (through blocks, steals, Defensive Rebounds, Fouls). Lower Defensive Ratings are better. Finally, Net Points Added are a view of whether or not the player had a net positive or net negative contribution to the game.

This information is literally only available here at Cracked Sidewalks.

How to read the table

Dominic James had ~14 possessions in the offense and delivered an Offensive Rating (ORtg) of 181, which was well above the team Offensive Rating of 128. He contributed 19.8 points on an individual basis. His defensive rating (DRtg) was 97.3, slightly worse than the team's defensive efficiency. However, James' net impact on the game was 9.4 points of margin, so almost half of Dominic James' points contributed positively. Dominic James had an outstanding game.

On the other hand, McNeal had a net neutral impact on the game. McNeal accounted for 8 possessions and had an ORtg of 109.0 (below the team's rating). McNeal contributed 9.5 individual points, despite the box score showing him ending up with four points, mostly because he had 9 assists that resulted in scores for other players. McNeal's DRtg of 90.4 was better than team's Defensive Efficiency of 93.3. However, at the end of the day, his net impact of 0.8 was about neutral (albiet slightly positive). Of course, McNeal is still dealing with the flu.

We hope that you enjoy this additional information going forward. On to West Virginia, where we have been given a 49% chance of winning the game.

edit: Corrected mis-leading language associated with Defensive Rating. McNeal's defensive rating of 90.4 was better than the team's average of 93.3.


Anonymous said...

No personal offense, but I don't buy the individual stats you show, particularly on the defensive end.

They simply do not capture enmough of what happens on the court.

To say that Jerel had a below team average performance defensively is totally innacurate. While he only had a few steals (stats), he deflected at least 7-10 balls being brought up the court. That not only rattles the opponents offense, it sets a tone and pace for the defense that others catch on to.

Again an interesting tool, but wildly inaccutrate in too many places to put much faith in.

Anonymous said...

How is defensive rating calcuated exactly? I understand that it is based on the team rating, but what "role in the offense" stats factor in?

Also, Crean praised McNeal's defensive effort after the game. Why the difference between the rating and his impression?

Anonymous said...

I agree - McNiel had their gaurds in tears when they were bringing the ball up the court - a huge factor in the defensive game.

Just because he was not hauling down rebounds or blocking shots, does not mean he did not take part in the D...

Rob Lowe said...

Thanks for pointing this out. I guess that it is misleading as written. For Defensive Rating, the lower the rating the better. (For ORtg, the higher the better) So for McNeal to have lower DRtg than the team average is actually a good thing. McNeal had one of the best Defensive Ratings on the team.

In addition, I agree that the Defensive Rating statistic has limitations. Traditional box scores do not capture the additional impact associated with things such as deflections, or more specifically, when a player guards an opponent and the opponent hits/misses. And until box scores do capture this information (which folks at APBR are pushing for), we are stuck with an approximation.

The calc for DRtg is actually based on the team Defensive Efficiency, and then modified based on the role that player contributes for things like blocks, steals, defensive rebounds, fouls, %Min played.

I do trust these ratings. For instance, last year McNeal had the best Defensive Rating for the entire team at 89, and that was over the course of the whole season.

Rob Lowe said...

Thanks for the feedback, folks. Blog entry has been updated.

Anonymous said...

I'm trying to get a handle on this...

So hypothetically, if we had a whole team of DJ's last night, we would have won by close to 50? (9.4 x 5)

And if it was made up of only Cubes, we would have lost by 10 or so?

Rob Lowe said...

So hypothetically, if we had a whole team of DJ's last night, we would have won by close to 50? (9.4 x 5)

And if it was made up of only Cubes, we would have lost by 10 or so?
Basketball is still a team game, so not everything can be broken down on a per-player basis.

But yes, that's a good way of looking at it. You could attribute about a third of the win margin last night to James' efforts.

For example, the three leaders in net points for the entire season are James (61.2), McNeal (59.1), and Hayward (44.7). No surprise there, right?

Kevin Buckley said...

Slightly OT, I did think UProv did a little crying last night.

Watching Acker, Cubes, and McNeal dancing cheek to cheek with them every time they took the ball up court? Wow. Glad they're on our side.