So when I set about writing this article, I knew that I had to quantify offensive plays in a certain way. Here's the list of questions that I asked myself, and an explanation as to why I asked them.
The most obvious way to do that was by initially deciding whether the play was a fast break or not. Fast Break plays are almost always unrehearsed and haphazard, so there's no sense in lumping them together with halfcourt plays. However, it does make sense to see what percentage of plays were fast break plays, so you can more accurately quantify the pace that the team played at.
Secondly, it made sense to determine what play the team was running each time they brought the ball down the court. For sanity's sake, I only recorded the play that the team originally intended to run, because things that happen after that can often be disorganized and random. I'll admit that I kept my classification of each play as simple as possible. Luckily for me, the Thunder often run really simple plays, but for the occasional play that required more than two passes for the ball to get into scoring position, I simply titled it "Three or More Passes". Some of them I recognized as HORNS, but I don't have an intimate knowledge of every type of play that the Thunder might call, so I decided to lump them into a single category. I also recognize that a simple play might be a variation of HORNS or another complicated play, but I'm mostly looking for the ultimate intent or goal of the play. Remember, the goal here is to quantify what the offense actually does, not exactly how Coach Brooks originally tried to lay things out.
Moving on, the next thing to determine was who the ballhandler was. For this category, I didn't use the player who brought the ball up from the backcourt, because a lot of the time that player would immediately pass the ball off to a pre-determined player. Rather, I used the first player who had to make an actual decision in regards to the play. Basically, they set up the play, decided whether to pass or shoot, and acted on it. So really, the ball might pass through one or two players before it got to the actual ballhandler that I'm speaking of. It's not the purest sense of the word, but it gives you a better idea of who's really setting up the offense.
The next category concerned who the play was actually intended for. Basically, this is the player who was expected to score if the original play was to work correctly. I understand that this could be a judgement call on several occasions, but I did my utmost to identify the intent of the play based on my years of experience in watching this team.
We move into the territory of the finite when considering how many players were moving on the initial play. This one's hard to screw up, as I basically count how many players had moved from their original positions once the original play had run its' course.
The actual shot taker was simple to figure out as well. This is the player with whom the play ended. They either shot the ball, were fouled, or turned the ball over in some fashion.
The type of shot taken category covers all six ways that an offensive possession can end. It can be a two pointer, a three pointer, an And 1, a foul, a loss of the ball out of bounds, or a turnover.
Whether the shot was made determines whether the shot attempt, if taken, went in.
Whether the play was a turnover determines whether the play was a turnover. (Yo dawg!)
Defensively, I tested for three very simple categories. The first was whether a trap was attempted throughout the entire defensive possession. To qualify as a trap, two players had to converge on another in an attempt to steal the ball. Switching did not qualify as a trap. The second category was whether a trap resulted in a forced turnover for the opposing team. If a turnover was forced but a trap was not attempted, then it doesn't count in this category. The third category was whether the attempted trap resulted in a easy score for the other team because someone wasn't defending their man. In order for this category to be counted, the opposing player had to be open as a result of the trap.
First of all, I'd like to acknowledge that the data in my study might be flawed because of a limited sample size. However, I cannot further pursue the study because seeing Reggie Jackson or Steven Adams play with the starters is rare, and because it took me weeks to compile this data.
Secondly, I'd like to acknowledge that the data might be skewed by the type of opponent faced. For example, the Steven Adams lineup saw a bunch of offensive success because they faced the defenseless Warriors and Clippers, so there wasn't much need to run complicated offensive sets. I will bear this in mind as I make my conclusions, but I do think that general trends will still be visible despite the type of opponent.
Thirdly, I'd like to acknowledge that a lot of the data in my study is categorized based on my judgement. Since I categorized these plays weeks apart, it's entirely possible that I might have judged a play one way on one day, and another way on another day. However, there's no way I can account for this human error other than attempting to be as fair and impartial as possible.
Fourthly, I'd like to acknowledge that the data may contain errors. I'm probably mildly dyslexic, and failed several math classes through my years in school. Thus I painstakingly used rudimentary methods to calculate this data and it took me days, while it would have probably taken J.A. Sherman about 5 minutes. So if you notice any inconsistencies....you're probably right. I invite you to make a more accurate data table with the data that I've provided in the link at the top of this page. Still, I have taken care to look over the statistics, and the trends I'm trying to show are more important than the exactitude of the numbers.