Total TO or "regular" TO?

[agentkirb]

I'm trying to build a stats program that parses through play by play data and calculates the box scores. I don't need programming help. The issue I have is determining if I want "Total TO" or just TO. According to the play by play data I have and box scores from NBA.com, there is a difference between "Total TO" and regular TO. It seems like regular TO just calculates all of the individual TO numbers and totals them. The "Total TO" category accounts for "Team Turnovers" like shot clock violations. So if I'm interested in the team values, "Total TO" should be what I'm looking for, correct?

The reason I ask too is because I've dealt with a similar situation with "Rebounds" vs "Team Rebounds". And the difference there is they'll give a team a "Team Rebound" if you like block a shot out of bounds. They want the total rebounds to match up with the total number of missed shots. But for statistical purposes if I'm trying to figure out who the best rebounding team is, I probably don't want to account for rebounds that come from blocked or tipped shots.

[J.E.]

So if I'm interested in the team values, "Total TO" should be what I'm looking for, correct?

Yes

The reason I ask too is because I've dealt with a similar situation with "Rebounds" vs "Team Rebounds". And the difference there is they'll give a team a "Team Rebound" if you like block a shot out of bounds. They want the total rebounds to match up with the total number of missed shots. But for statistical purposes if I'm trying to figure out who the best rebounding team is, I probably don't want to account for rebounds that come from blocked or tipped shots.

Team Rebounds in PBP are kind of weird. There are some that you definitely want to ignore, like a team rebound after 1 of 2 FT missed. But I think most of them should be recorded in some way.
Take your blocked shot, whoever gets the team rebound after the block either stays in or gains possession, so it's kind of important who gets it.
If the ball clanks from the rim directly to out of bounds, that's a team rebound. Teams that are good at boxing out should have more of these; don't you want to somehow reward them?

[Mike G]

If you get your shot blocked out of bounds with 3 seconds on the shot clock, is that a Team Offensive Rebound?
You get the ball back with 1 or 2 sec. to shoot, which is hardly as rewarding as an OReb off the rim, which comes with with another 24 sec.

Unless there are players or teams that are really good at boxing out and letting rebounds go out of bounds, it may be that individual reb are proportionately distributed among players and teams just about as they should be, to represent rebounding ability.

And sometimes a team will have the higher % of both OReb and DReb, yet have a lower % of total Reb.

[agentkirb]

Take your blocked shot, whoever gets the team rebound after the block either stays in or gains possession, so it's kind of important who gets it.

I haven't completely looked into this, but I imagine if a single person recovers the ball after a shot was blocked it would theoretically record a rebound for that person.

I think for the purpose of recording the number of possessions, you can look at some formula involving FGAs to calculate that. So team rebounds wouldn't matter.

If the ball clanks from the rim directly to out of bounds, that's a team rebound. Teams that are good at boxing out should have more of these; don't you want to somehow reward them?

See, this is kind of the question I'm asking here. I guess we answered it for turnovers. But for rebounds it's kind of the same thing. I don't want to count rebounds that come off of blocked shots out of bounds if people normally don't count them. If it's normal for people to count them, then I want to also do that. But if I have some kind of weird rebound forumla... I want to be able to count on the fact that my numbers are valid. So I either have to not include team rebounds or include them (or include just some of them). I'll have different numbers either way so I need to get this as right as possible.