Offense Wins Games... Volume I(Measuring Offense)

Almost two months ago, I've analyzed defensive capability of every Euroleague team and formed a defensive ranking. Yet, it isn't all about defense in basketball. We should also talk about offensive side of the game. Offense has always been more appealing side of the game. Most of basketball fans tend to root for more offensive teams. So, I decided to analyze offensive capability of the Euroleague teams, as I've done with defensive capability.

Let's start with the same question; How can we measure offensive capability? As I've mentioned before, "points allowed" numbers are just not enough for us. Even possession adjusted points allowed numbers are not good enough, unfortunately. We need to combine all offensive stats to create an offensive ranking. Therefore, we need to bring all of them to common base, as I've done with defensive stats.

Now, let's analyze offensive stats. First of all, what can happen in a standard offense. Attacking team can score the ball, miss the shoot or turn the ball over. Other than that, offense could be interrupted by a defensive foul and also if team can grab the offensive rebound after a missed shoot, they can charge again. In every possible scenario, either attacking team can get what they want or opponent's defense carries the day.  Let's assume that attacking team scored in an offense. It means that they've shot well. To measure shooting ability, I've used the F-Shoot stats -which is a stat that we've come up with- instead of ordinary shooting percentages or true-shooting since F-Shoot is more detailed and team oriented stat than others. Also a good shoot may/may not be created by an assist which means we should also take assists into consideration for a good offense. What if the attacking team misses the shoot? There are two options actually; either they grab the offensive rebound and set one more offense or opponent team grabs the defensive rebound. It's very rewarding to get the offensive rebound actually. First of all, it prevents your opponents to organize a fast-break offense, but it's not important for this analysis since we are just looking into offensive actions. Still, an offensive rebound is very important for offensive organizations, because, if you have a chance to organize a second chance offense, statistically you would shoot with higher shooting percentage for a fact. It means, while every shoot attempts are weighted with potential score of an standard offense -FGT- [1] and team's field goal percentage, every second chance shoot attempts will be weighted with FGT and second chance percentage which is fairly higher than the field goal percentage.

These all scenarios above are the pretty ones. However, there are some unpleasant ones as well. For example, attacking team could miss the shoot and couldn't get the offensive rebound.Then, since they've lost thier chance to score they should be penalized. Moreover, attacking team could turn the ball over. If a team commits a turnover , they should be punished for it too.Because, turnover means that team refused the chance of scoring number of points as much as their potential score per offense. It should be their penalty. One other scenario is, an offense could be interrupted by a defensive foul.It's the most tricky part actually, because it's more complicated to weight such positions. We need to think it through. If the opponent didn't commit a foul, the attacking team would attack with their field goal percentage and score as many as their FGT. So, opponent's foul eliminates that possibility. Instead of that, team can score as many points as their free throw percentage -if they manage to get to the free throw line- or they can organize another offense. At this very moment, I needed one more item to organize them; frequency of getting to the free throw line. This input solved all my problems, I was able to determine potential score of the attacking team after they get fouled thanks to line frequency.

Other than these, I also used a new approach in this analysis which I didn't use in defensive capability analysis. I didn't use it in TDR, because I've recently found it as a matter of fact. I wanted to make a difference between a team which is eliminated in the group phase and a team, let's say, reached to the playoffs. Because, it would be harder to keep playing with high offensive efficiency as tournament progressed.For instance, if a team is eliminated in the Top 16, they should have competed with the teams in approximately 67% percent of the league. On the other hand, a team who reached to the Final Four would be competed with the teams in 17% of the league. So, there should be a difference between these two teams. I used this approach to make the ranking smoother and happily it served for my purpose.

As defensive capability analysis, I analyzed the last year first. Let's look at the offensive rankings of the Euroleague Season 2011-12.

Let's look at the Top 3 of the list. Even though 4 out of first 5 are Final Four teams, our winner is a team who couldn't reach to the playoffs. Real Madrid was one of Top 3 teams for almost every input I've evaluated. Impressive... And Panathinaikos; even though they aren't at the Top 2 for any category, they've got pretty high points from every category. As for CSKA, turnovers was their biggest enemy.They have terrific shooting and asist numbers way above the league's standards.Yet,they look like an ordinary team in terms of turnovers. 

In the second part, I'll show the historic offenses of the Euroleague's history.

_____________________________________________________________________________

[1] Field Goal Tendency. For more info, look at the annotation of the Care To Defend post.

Care to Defend? Volume I (Measuring Defense)

It has always been a question in basketball talks; Defense or Offense? Is it possible to say that one is more important than the other ? Well, there could be an analysis about that later, but in this analysis,I'll focus on the defensive side of the game for now.

How can we measure defensive capability? Are "points allowed" numbers enough to measure it? Despite the fact that these numbers can tell us lots of things about defensive capability, it's still not enough. Say, Team X allows 60 points per game while Team Y allows 70 points per game.Yet, Team Y still could be better defensive team than Team X. If Team Y scored 100 points and Team X scored 70 points per game, then Team Y has better margin between Points Allowed and Points Scored than Team X. Now, here comes the one of the greatest devices of the modern statistics, possession. Don't worry,I won't say that I've just adjusted these numbers with possessions. Even though it's also a good approach to understand defensive capability, it's not enough too. Also, you can find these numbers in any site, no need for me to analyze it. How about some defensive statistics like steal,block or defensive rebound? Likewise,they aren't enough to understand defensive capability by themselves.

We just need to combine all defensive inputs to measure defensive capability. But again, they are different stats from each other, it won't be easy to combine them all. Thus, we need to bring them together to one,common base. I chose to value every defensive act with points. Let me explain. If a team didn't allow its opponent to score a three-pointer,that team deserves +3 points for its defensive effort. Let's say they stole the ball. Then, they would stopped their opponent to organize an offense. Yet, it is still difficult to assign any point to this effort. So let's think backward, if the team didn't steal the ball , opponent would attack as they attacked normally in the game. It means they would attack with their average field goal percentage in the game. Yet, would they go for three-pointer or send the ball to the rim for two-pointer ? It will be decided by their field goal tendency rate[1]. So if we multiplied opponent's field goal tendency rate for that particular game with opponent's average field goal rate -again,for that particular game- we could estimate that how many points the opponent can get from a standard offense. Let's assume that Team Y attacks with standard mean of 2,5 points. If Team X can steal the ball from Team Y, then they will be awarded with +2,5 points for their defensive effort. I applied same logic to every other defensive act.Thereby, carrying all the defensive inputs to the same base became possible. I made the analysis for every season since 2000-01. In this part, I'll show some some statistics from last season. Here are the best four defensive teams of 2011-12 Season.

Coincidentally -or is it?-, Final Four teams were the best defenders last season. It surprised everyone that Olympiacos won the championship, yet they were in fact one of the best defensive teams last season. Sure, it  surprised everyone that they beat Barcelona and CSKA Moscow who look unbeatable -especially CSKA with AK47-. On the other hand, they were pretty good on the defensive side of the game. Let's look at how they improve their defensive rates.

Even though Olympiacos didn't start the season in a great shape, after Top 16 phase they speeded up their defense. It should have helped them to reach their dreams. It was kind of a fairy tale what they did in the Istanbul,indeed. However, it wasn't surprising at all that they reached to Final Four. Now, let's look at the worst four defensive teams.

Three of the teams above eliminated in the group phase. As for the only team didn't, which was Zalgiris Kaunas, they couldn't reach the playoffs. It's not sane to say defense is the only key to succeed. Yet, it's obvious that it's a huge part of it. If you could be a solid defensive team, chances for you to achieve something in the Euroleague become higher.

In the second part, I'm going to show the best defensive teams in the Euroleague history.

_____________________________________________________________________________

[1] Field Goal Tendency Rate is used to define teams' tendency to choose 3-pointers or 2-pointers. If a team use 3-pointers more often, then their potential score in one  possession would be higher. Because if they made the shoot,then they will get 3 points for a possession instead of 2 points. However, probably they would shoot with lower field goal percentage,so their potential score for one possession will multiplied with lower number. If you don't have couple of snipers, then going for lot's of 3-pointer would most probably hurt your points of standart mean for a possession.