I’ve always wanted to do a study on the behavior of the league as a whole. When I started following the PBA a year ago, I literally had NO idea what the base is for most of the advanced statistics in the league. Most of my analysis over the past year have been dependent on just averages of the current season. If the average was 95 at that time, I’d be saying that’s the average. The NBA is totally different; there are already certain baselines that are accepted – league average points per possession is about 1.06 (one), league average effective field goal percentage is 50 percent, turnover rate is 13, etc.
Fortunately, we can now start doing a preliminary study on this (now that I have two conferences with two years of data). Over the next few years, I’ll continue adding more data to give us an even clearer picture. For now, we’ll have to settle with this. I’ll also add a couple of my thoughts on what these numbers may mean.
The Philippine Cup
As you can see, there’s a developing story that the Philippine Cup league average is somewhere between 96 and 100 (it’s actually near 97.5). That’s a very important discovery (why it is will be discussed more later).
The Commissioner’s Cup
Note: The truncated 2014 Commissioner’s Cup is due to the scheduling for FIBA.
The Commissioner’s Cup is a harder conference to crack. Last year, the mark hovered right around 95 while this year hovers right around 100. That’s big gap and it doesn’t give us any clues as to what the baseline is for the Commissioner’s Cup. Probably a big reason for this is the quality of imports that enter the league.
Note: The size of the bubble is PER.
The bubble chart of ORTG versus USG & PER (table 2.1) shows that the 2014 Commissioner’s Cup seems to feature a better set of offensive players. This is because, comparatively, there are more big blue circles than there are big red circles. Additionally, there are more blue circles in the upper right quadrant. What this means is that the quality of imports (offensively, at least) is better in the 2014 edition of the Commissioner’s Cup (versus the 2013 edition).
Much of a team’s success in an import-laden conference hinges on the performance of the said import, his cohesion with the team (2013 featured three teams that changed their import more than twice [Globalport, Petron, Talk ‘N Text] while 2014 features just one team [Ginebra]) and the domino effect on the performance of the other locals that in totality, probably,made the 2014 Commissioner’s Cup a much better offensive conference (or worse defensive conference) than last year.
What else can we take from the data?
Looking deeper into some other trends
As you can see, trends that are the easiest to spot are the consistency of the offensive rebounding percentage and the free throw rate.
Over the past two seasons, the offensive rebounding percentage has hovered right around 30 percent while the free throw rate has hovered right around 22 percent. This is important because it establishes a baseline from which all teams (and upon further study, all players) can be compared to. If, say, we know that the average offensive rebounding percentage is 30 percent, then a team that rebounds more (or less) than that is considered above (or below) average even if the season has just started.
Moreover, it puts a bigger emphasis on players that are good defensive rebounders. Because rebounds have a ceiling (i.e. there is only one rebound to go around every missed shot), there’s a certain supply/demand relationship caused by the marginal value of a rebound. Since the PBA has (hypothetically) better offensive rebounders (than, say, the NBA), then the demand on good defensive rebounders increases. This is in stark contrast to the NBA (where the league average ORB% is usually set around 23~25 percent) where good offensive rebounders are more valuable. I don’t know how much the 5~7 percent increase in ORB% tips the scale to the favor of defensive rebounders, but the point is defensive rebounding is a much bigger necessity here in the Philippines than it is in the NBA.
There isn’t a clear baseline for three-point shooting and free throw shooting. The best we can do right now is make a guesstimate – in this case, I’d say the baseline three-point percentage is about 30.5 percent and the baseline free throw percentage is about 68.5 percent. How is this important?
Besides knowing which players are below/just/above average in these categories (thereby allowing team managers and coaches to strategize appropriately), it also allows us to create a hypothesis on the value of these “tendencies” (if combined with historical data on points per possession). An average three-point sh0oter will, on average, yield 91.5 points per 100 possession, a mark that’s six points below league average (or more, depending on the quality of the import). Similarly, an average free throw shooter will yield about 137 points per 100 possession (a mark that’s about 40 points above league average).
Compare this to the NBA (where league average three-point shooting is set at about 36 percent, league average free throw shooting is set at 75 percent and points per 100 possession is set around 106) — the value of a three-point shot decreases (and inversely, by supply and demand, the value of three-point shooters are increased). The value of free throws is also down, but not by much, which means players who can draw and make a lot of free throws are about as valuable in the PBA as they are in the NBA.
Lastly, as we continue trudging along this quest to complete the database for the PBA, I’m also on my own personal quest to find out the PBA (and UAAP’s) own version of the “efficiency” threshold. The idea, with credits going to Evan Zamir, is that there is a certain frontier that when reached, allows us to classify these specific players as the most efficient scorers in the league.
It is supposedly a relatively easy task to achieve. You just take a certain group of players, plot them along a scatter plot and then look at that boundary that only a few people can break. For the NBA, with the help of Basketball Reference, it’s a simple (albeit not easy) thing to do. The PBA and the UAAP, with the dearth of data available, is a lot more difficult. With more data, we can finally have a working hypothesis on what that frontier is for the PBA and the UAAP and then use it as a baseline for evaluating efficiency of scoring.
Here’s the PBA’s scatter plot, so far:
From the visual, you can point out that there’s a clear boundary in the plot swarm. Because my data visualization skills is not yet as good as I want it to be, I can’t quite show you guys the frontier line. But it’s there.
There’s actually a group of players that are just within/over that line namely: Josh Dollard (59% TS, 36% USG), Pennisi (76% TS, 14% USG), Wesley Witherspoon (63% TS, 28% USG), Leon Rodgers (51% TS, 37% USG) and Eric Dawson (51% TS, 36% USG), just to name a few.Over the coming years, we’ll be able to fill this table with more and more points to give us a clearer view of what the frontier really and consecutively providing us a clearer look at the averages in Philippine basketball.