APBRmetrics

gabefarkas · Joined: 31 Dec 2004 Posts: 1313 Location: Durham, NC

I caught the following graphic on SportsCenter this morning:

Anchor Steve Levy took about 10-15 seconds to explain what TS% was (in layman's terms), and then said something like "Plus/Minus? I didn't even know they had that in the NBA. That's what you get when ESPN takes it to the next level!"

So, does this mean we've arrived, so to speak?

Mike G · Posted: Tue Dec 09, 2008 8:34 am Post subject:

Is the decimal in the right place? Or is it easier for the layman to grasp "Equivalent baskets per 100 attempts"?

To me, it makes sense to think of 'baskets per attempt'. Also, I fear it looks too much like a 'quarterback rating', and as such may be rejected by the general public.

B-R.com uses a decimal fraction (like .675) to describe TS%, but a 100* multiplier for other things (Stl%, etc).

Knickerblogger uses the real number for W% (with a lead zero), multiplies by 100 for shooting%'s.

Yahoo's box scores have FG% and such in .465 form; but on the player pages, they go to 46.5 . No one seems to multiply W% by 100.

At NBA.com, they have Kobe's FG% at .468, Rebounds (per game, I guess) at 5.30, APG at 4.4, SPG at 1.58, ... It's a total mess.

Hollinger's (ESPN) player pages show Wade's .582 TS% (on TV it's 58.2 ?) and his Usg of 32.7 . Which one is a 'percent'?

Maybe, since we got TS% defined and named, we can also promote consistency in the depiction of a decimal fraction.
_________________
`
36% of all statistics are wrong

gabefarkas · Joined: 31 Dec 2004 Posts: 1313 Location: Durham, NC

I think you're over-thinking it.

In my Clark Kent job, we use the form "XX.X%" to display percentages all the time. I think that's the more common way, to be honest. If you go into Excel and apply the percent format to a bunch of numbers that all have a leading decimal point, Excel converts them to that format automatically.

Mike G · Posted: Tue Dec 09, 2008 10:01 am Post subject:

Well, consistency is generally good, and I don't have any trouble reading either
52.5%
.525
But simplicity is good, and the 2nd option takes up less room in a column. More columns can fit on a screen.

If you drop the % sign, you lose the functionality in your spreadsheet:
What are the chances that a 70% FT shooter will make 2 of 2?
Not 70 * 70, but 70/100 * 70/100.
Meanwhile, .7 * .7 = .49, simple as can be.

A guy who made .675 of his shots in last night's game -- did he really make 67.5 of every 100? Or 67.5/100 of his shots*100? It's almost like going back to Roman numerals.

People don't really think about 55 cents as being 55% of a dollar, do they? It's just $.55 . No conversions necessary.
_________________
`
36% of all statistics are wrong

Ryan J. Parker · Joined: 23 Mar 2007 Posts: 711 Location: Raleigh, NC

Nice TV!

Very cool, though. They should have kept the % of course.
_________________
I am a basketball geek.

Mike G · Posted: Tue Dec 09, 2008 11:23 am Post subject:

Maybe I should be more jubilant at seeing TS% on a TV. My appreciation of the stat dates to this example:

John Hollinger · Joined: 14 Feb 2005 Posts: 175

Wow! I hadn't seen that ... that's awesome.

gabefarkas · Joined: 31 Dec 2004 Posts: 1313 Location: Durham, NC

MattB · Joined: 22 Jun 2006 Posts: 38 Location: Lowell

What next?

Pace Calculations?

Don't the have an ESPN4000 or some channel that this can be an every day thing. Get some real experts on there to go over this stuff?

Harold Almonte · Joined: 04 Aug 2006 Posts: 616

Oh, oh! Maybe "true shooting" should be renamed: "easy practical scoring". Royal didn't shot better, but scored better and smarter (1.52 pps vs. 1.16 pps). That's what you get when you take your FT/FGA to the next level. It's like the "easy inside assists list"(in another topic) against the total assists/G, where PGs were benefited from good inside primary scorer teammates to change list's spots.

Of course, the kind of usage, and the level of usage, is what makes a player different than the other. Might the team's off. strategy hold an increase in the Royal's usage? If so (and I don't beleive it), then he will be the better scorer all the time.

What's next? Production predictions from usage-Eff. tradeoff.

basketballvalue · Joined: 07 Mar 2006 Posts: 208

HoopStudies · Posted: Wed Dec 10, 2008 12:40 pm Post subject:

gabefarkas · Joined: 31 Dec 2004 Posts: 1313 Location: Durham, NC

Serhat Ugur (hoopseng) · Posted: Wed Dec 10, 2008 3:59 pm Post subject:

It would be interesting to watch some games with an advanced stats people live commentary. Most of the viewers might get bored but we've been for a long time getting bored with their cliches?

Also, at the pre-game shows, I would do anything to see a scouting report that coaches/staff use.
_________________
http://www.nbastuffer.com

MattB · Joined: 22 Jun 2006 Posts: 38 Location: Lowell

As cliche as it is, a podcast of vcast would be nice...

Get a few of the more involved people and take guests. Even a weekly thing would be interesting.

Who has the time tho...