Stats & JVG: How efficient were we really last season?

Following a very interesting and straight forward post by Tyler Durden (The Orlando Trade is Looking Like a Steal), I was so looking forward to see if we could demonstrate his theory with numbers, and came up with surprising results (I actually agree with the ideas he presented, although the conclusions are a bit rash .


Calculations are based on the very ideas he mentioned already: regardless of how many possessions you have, once a team has possession of the ball, it can:
- Turn it over
- Score a basket
- Miss a shot

A team offensive efficiency can thus be measured with the following formula: Successful possessions/Total possessions

This can be estimated as:

(Field Shot Made + Free throw Made/2 ) / (Turnovers+ Field Shot Attempted + Free Throws Attempted/2 )

The most arbitrary measure is the free throw figures divided by 2 for estimated possessions because of basket + FT+ Foul situations, technicals... We could argue here.
The basic result is nonetheless independent of the pace of the game each teams applies.

Here are the results:

1 Sacramento Kings 51.10%
2 Seattle SuperSonics 50.45%
3 Dallas Mavericks 50.06%
...
5 Minnesota Timberwolves 47.82%
...
8 Los Angeles Lakers 47.03%
...
10 Houston Rockets 47.02%
...
16 Orlando Magic 46.09%
...
20 San Antonio Spurs 45.86%
...
22 Detroit Pistons 45.26%
...
29 Chicago Bulls 43.64%

Surprising isn't it? Rockets are 10th over teams like San Antonio and Detroit and the main reason it is not relatively better than Dallas, Wolves or Lakers is because of its abysmal total of turn overs. It is incredible: 300 more TO than a high octane offense like Dallas!
Please note the Magic's rank last year and consider their very good TO totals (26th). This very good news for us as we now have three of their starters.

How come this ranking is so different than the actual end of season ranking? Defense.

The ideal way to evaluate the defensive efficiency of a team is to calculate the Offensive Efficiency of their opponents while they meet. Alas, I do not have the numbers per team for this, if anybody knows how to get them, I would sure like to benefit from their knowledge.

If this was available, the defensive efficiency would be the inverse of their opponent offensive efficiency. This would highlight directly if a team is a tough match-up to the other's style of play.

As I don't have those numbers, I hope to have the next best available measure:

I assume that our opponents have the same total of possession we have per game: As soon as you did something with the ball, score, miss, loose it, your opponent has it. This is a problem as the numbers of possessions per quarter are not always even and the slower the pace of the game is, the more this is an issue. On a season average, this should even out somewhat.
The same caution must be taken for the free throws as mentioned previously.

Once your opponent has the ball, defense comes in play: You can steal the ball, or make him miss and re-take possession: defensive rebound.

Important: I do not consider blocks here as it would duplicate numbers: Either you block the ball and gain possession, in this case you can score, miss or loose it and your opponent has a new try afterwards => this is already in my totals.
Or you block the ball, but do not gain possession, still counts as one possession for the opponent. This logic can be argued too, but blocks do not necessarily mean possession gain (I really do not want to be-little blocking, it does have an important psychological impact that can also go both ways). Ideally: if a block is followed by a possession gain it should count as steal too, in this case, I would be safe.
Charges and offensive fouls (good defense) are much more problematic and that is an issue I cannot get out of the way. As mentioned previously the best thing would be the opponent's stats for offense.

We end up with: Opponent Total Possessions - Defensive rebounds - Steals / Opponent Total Possessions, the lower the score, the better the team's defense.

Again, not satisfactory, but let's look at the results, Defensive Efficiency:

1 San Antonio Spurs 61.38%
2 Houston Rockets 61.72%
3 Minnesota Timberwolves 62.04%
4 Indiana Pacers 62.33%
5 New Jersey Nets 62.37%
6 Detroit Pistons 63.06%
...
13 Los Angeles Lakers 64.05%
...
20 Chicago Bulls 64.37%
...
22 Dallas Mavericks 64.99%
...
25 Seattle SuperSonics 65.40%
...
29 Orlando Magic 67.20%

My biggest surprise here was the wolves' standing, apart from that, this pretty much what is expected in retrospect. I did expect Detroit higher. Maybe using the preferred method would adjust these standings accordingly. This is also where we could understand Orlando's deceptive season: Orlando had all the fire power it needed, it just lacked the defense to go with it. Most of the time you are hearing that: 'Tracy had no other star besides him', maybe we should reconsider...

The defensive efficiency is probably misleading, some assumptions can be hotly debated and it only gives a rough estimate, but what I like, is that these number challenges a number of assertions like:
'JFG offense is lacking', 'Mac Grady had no one to share the offensive load'. They also illustrate that Houston has under-achieved last season, probably due to the intangibles that make statistics so un-reliable for predictions .

In the end, I think these intangibles like experience, abnegation, execution and smart play at the right time, at the right place is what made Detroit champions and Rockets' early exit.

Rockets management seems to be on the right course: combine JVG's & Houston's defensive qualities with Orlando's low turnover and offensive actors.

Look forward to the season and feddback of other number crunching posters...

I am trying to get 2003-2004 seasons' stats to apply the same logic and compare the Rudy / JVG 'effect'.

 

Well, after additionnal search, I found out that ESPN has opponents stats, I'll have to do the numbers again...

Previous results were done with SI BTW.

 

Montevideo:

Attempts are taken into account in the total possession: one attempt (Field Goal attempt) is considered as one possession.

 

I apologise for this unwanted effect

I thought that this topic was not confined to: did we steal anything from the Magics?

I really appreciated your post, except that it was a bit harsh...

 

Well, I agree, a higher number of hot babes in the stands probably increases the efficiency, but it does increase it for both teams. In the end, as both teams give the max, it probably exposes who thebest tean is!

This is why it is better to compare your efficiency versus the one of your adversary.

As a follow up support to moniker's post:
It is indeed the case, this is why ratio of possessions are taken into account, this allows a fair comparison betweem high tempo teams' versus slow half court teams.

Wichever teams manages to impose its style will probably be the most efficient for the game.

 

Tyler, it is good thing you mention this!

I hesitated a long time before taking into account made free throws instead of total FT attempts! Afterall, a Ft is not only a free throw but also a foul for the opponent: you gain twice.

The reason why i did include the FT% in the equation is the following: being efficient behind the free throw line should be considered as part of being efficient offensively.

If you can't convert your free throw, you probably should avoid getting to the line to have a potent offense, confer the hack a shaq theory... unless you get all your opponents players out of the game...

I had a hard time objecting against that...

 

Well, FG percentage is part of the ratio.

And this ration accounts for TO, steals and so, while at the same time leveraging for paceo f the game.

I am reworking the ration with my previous idea ( your opponents efficiency as an estimate your defenisive efficiency)

I am also doing the maths for the previous year too, in order to see if Rudy / Gundy did have different impact.

It will be interesting to see the 'yao' effect too.

 

'ration' is supposed to be 'ratio'.

Too bad I cannot edit my post.

 

Tough sell here, as expected:

If you take into account you can play a style of game (cut inside, go to the basket) that generate a lot free throws, it is probably because your offense is more efficient for you if you do so: you'd better be a good FT shooter.

No coach would like to set plays that exposes his players on their most glaring weekness.

Especially at the end of games.

If you need two points from anyone, you'll probably draw a play to have your point guard fouled instead of your 70% Ft shooting center.

That is offensive efficiency as a direct consequence of a coach's decision.