clock menu more-arrow no yes mobile

Filed under:

The Downbeat #1125: The Worthless Player Proving them Wrong Edition

A Bleacher Report slideshow (I know), Questions and thoughts from national writers about the next year's Jazz team, and a look at True Shooting Percentage.

This guy approached a TS% of 60% last year. For you kids at home, that's great.
This guy approached a TS% of 60% last year. For you kids at home, that's great.
Jerome Miron-USA TODAY Sports

ESPN is starting a series of "Burning Questions" articles (Insider). Here's what they asked about the Jazz in their Northwest Division set:

Which one of the Utah Jazz's "Core Four" makes the biggest jump this year?

One of the reasons I've [author Amin Elhassan] been bullish on the Jazz's future is the cadre of young talents known as the Core Four: Alec Burks, Derrick Favors, Gordon Hayward, Enes Kanter. They represent varied skill sets and different positions, and have the opportunity to grow together ...

This is, of course, exactly why many of us Jazz fans are so excited for the team this year.

However, I'm not sure their question is really what I'd call the most important one. I don't think the who is so important. Nor is the biggest jump, i.e. greatest year-over-year improvement.

No, I think the two questions that matter most are: (a) How high can the Jazz's best player reach, and (b) How many of them can give consistent great-to-elite production?

So for whatever reason, I was perusing the internet and ended up at an old Bleacher Report slideshow.

I know.

But bear with me.

It is titled "The 50 most worthless players in the NBA." I was curious about how many Jazzmen would be on the list (Biedrins, of course, seemed guaranteed), but it turns out the article is from January, 2012. This is right at the beginning of the lockout season. And so the list became much more interesting to me, because we all know what has happened over the two seasons since. Here are a few players that made the list in 2012:

  • DeMarre Carroll (this was when he was barely signed for the Nuggets). Since he became a decent hustle guy off the bench.
  • Steve Novak. He's since become an ace three-point shooter.
  • Kosta Koufos. He, of course, became an effective player and the starting C for a 57-win team.
  • Jamal Tinsley. Our best PG last year ... and actually moderately effective despite his amazing FG%.
  • Nikola Pekovic. Just signed a $60 contract that seems totally reasonable.

There were also a lot of usual suspects (Eddie Curry, Luke Walton, Darko). But the inclusion of the five guys above shows something really fun about sports: despite all the stats, all the attempts to predict, all the things you believe you know ...

The truth is you don't really know. Unexpected things always happen.

The Basketball Jones recently put up a post about the most interesting players on each team in the Northwest division. And by interesting, they mean as players and according to what they may or may not do on the basketball court next year.

So before looking, I tried to predict who it would be (yes, I am teaching predicting as a reading strategy to my 5th graders right now). The most interesting new player has to be Trey Burke, right? Unless you're going that guy and pick Biedrins just to get attention. For returning player? I debated and debated. Finally I had to guess Favors.

Well, the Burke prediction was right. Even if the article is pretty down on him (a recurring theme, thanks to his poor shooting in Orlando). But the returning player? Well, the article basically punted. Favors AND Kanter AND Hayward AND a mention of Alec Burks. Sorry, TBJ, but trying to construct an "all of the above" option for an opinion piece dooms 5th graders to a lousy Writing Assessment score. You gotta show some guts, make a choice, and run with it like you mean it.

So I'm gonna make a choice: Enes Kanter is the most interesting guy to me. The amount of improvement he made from rookie to 2nd year was insane. If that trend continues ...

Well, there's many reasons why I think the Jazz will end up with 35-40 wins this year. Enes is reason 1-A.

Speaking of my crazy 35-40 wins prediction ... SBN's Tom Ziller mentioned, just as an aside into another piece about paying Larry Sanders lots of money, that "Center and power forward remain the most important defensive positions in the league." It's a bit of conventional wisdom I generally agree with—that these two spots certainly seem to have the biggest impact on the team's defense.

Which is why I am so excited to see what Favors and Kanter can do. We've seen strong work from them over and over, in their limited minutes. We've seen flashes of greatness. If they can pull off this kind of play consistently, well, the Jazz will have the beginnings of something great. Something—in the words of the great Olvia—"Most Wonderful!"

Charley Rosen, now writing for HoopsHype, and the NBA media's resident curmudgeon went on a little rant about analytics. One of the strinking things to me was how Rosen showed he actually doesn't understand what advanced analytics are trying to do. Case in point:

... the ability to block shots is not always a reliable measure of a superior defender.

I don't think there's a single believer in analytics who thinks blocked shots is a universally reliable way to identify superior defense.

Anyway, Rosen's article and misunderstandings reminded me that there has been a request for us at SLC Dunk to go through and do a sort of Advanced Stats for Dummies. So I'm going to start dedicating a portion of the DB for it. Today's topic:

True Shooting Percentage (TS%)

What it tries to measure: scoring efficiency

What values are good: Around 50% is meh. 53% is a threshold for a somewhat decent scorer. About 56% and above show an effective scorer. 60% and above is elite. 65% and above is beyond elite and starting to look like an all-time great year.

How it is calculated: (points)/(FGA + (.44 x FTA))

Why it is calculated this way: There's several issues regarding scoring efficiency:

  1. Not every made basket is worth the same amount of points.
  2. Not every shot has the same likelihood of going in.
  3. Wildly different shot distributions can result in the same amount of points (e.g. 3-6 on layups = 2-6 from three = 6-12 on free throws = 6 points scored in 6 attempts to score.
  4. Because different types of scorers tend to put up different types of shots, it is helpful to find a metric that results in a consistent and easily compared value to a guy's scoring efficiency, regardless of the type of shots he takes.
So to handle these issues, the TS% uses:
  1. Points in the numerator: total points doesn't care how they were scored, just that they were scored.
  2. FGA in the denominator: this lets 3-6 from two feet be the same as 2-6 from three ... as they should be because both result in 6 points.
  3. 0.44 x FTA in the denominator. This coefficient adjusts the FTA to account for the (a) that two FTA's generally represent a single attempt to score and (b) number of FTA's that tend to be and-1's (i.e. do not represent a unique attempt to score)

Strenghts to TS%: It gives you a straight, easy, consistent, reliable number to compare scoring efficiency. It doesn't include any event that wasn't an attempt to score. It doesn't care whether a shot is pretty, nor how the points are scored. It simply tells us how efficiently a player scores in his opportunities to do so.

Weaknesses to TS%: It tells you nothing about how a player scores. It is also fairly variable: Just because a guy put up 66% one year doesn't mean it will repeat. In fact, it will more likely NOT repeat.

Yucca's take: TS% is my favorite way to talk about scoring efficiency. It's easy, it incorporates everything, and it doesn't muddy the results with additional events that weren't attempts to score. And though it can vary a lot from year-to-year, good players tend to stay in the good area (56%-60%). This also makes it helpful to look at TS% over several years and see what a player's general range seems to be. So as long as you are looking for a reasonable range rather than a specific magic number, you will do just fine with TS%.