What's new
By:

Game Thread Jan 23, 2024 06:00PM MT: Jazz at Pelicans

Added to Calendar: 01-23-24

MT Steve said:
That'd just be for unassisted attempts though, regardless of make or miss.
Click to expand...
I was unclear there using the word shot. I meant made field goals.

AST% = assisted buckets divided by the sum of assisted and unassisted buckets.
 
SoberasHotRod said:
I'm not quite following your logic, so maybe there is something lost in translation.
Click to expand...
Try using total assists or potential assists instead.

ORtg also has issues as while I do believe it correlates with assists, it also correlates with offensive rebounds and most importantly general FG%. Probably correlates some with steals as well.

EDIT: Oh yeah, turnovers as well. Forgot to mention.
 
HermanG said:
I was unclear there using the word shot. I meant made field goals.

AST% = assisted buckets divided by the sum of assisted and unassisted buckets.
Click to expand...
Ah, that makes sense.

So essentially, teams with elite scoring in isolation would be punished by this metric whereas teams that are very bad in isolation would be rewarded, even though both teams could be equally competent with moving the ball.
 
HermanG said:
I was unclear there using the word shot. I meant made field goals.

AST% = assisted buckets divided by the sum of assisted and unassisted buckets.
Click to expand...

You're not wrong, but you're describing it in a confusing way. It's the number of assists by a player, divided by the total number of field goals made by the team while said player was on the floor.
 
Fosto said:
You're not wrong, but you're describing it in a confusing way. It's the number of assists by a player, divided by the total number of field goals made by the team while said player was on the floor.
Click to expand...
That's for an individual player though, the chart posted was for teams.
 
SoberasHotRod said:
Hmmmm.... Here are Passes/Possession by EFG% and Potential Assists/Possession by EFG%

View attachment 15843
View attachment 15844
Click to expand...
Makes sense though, doesn't it?

With the top chart, the longer a possession goes, the less likely it is to result in a score, but longer possessions will also have more passes.

With the bottom one, I think the potential assists on made baskets would be on significant better looks on average than potential assists on missed baskets. I'd imagine potential assists on missed baskets would be plentiful in teams that aren't good at moving the ball to generate good looks.
 
MT Steve said:
Makes sense though, doesn't it?

With the top chart, the longer a possession goes, the less likely it is to result in a score, but longer possessions will also have more passes.

With the bottom one, I think the potential assists on made baskets would be on significant better looks on average than potential assists on missed baskets. I'd imagine potential assists on missed baskets would be plentiful in teams that aren't good at moving the ball to generate good looks.
Click to expand...

I can see what you are saying on the passes/EFG% data, but not the potential assist data.

A potential assist, as I understand it, is any pass that leads to a FG attempt. I would have assumed that FG attempts from a pass have a much higher chance of being made than a FG attempt not from a pass.
 
Nerds.

JK, I'm enjoying the discussion.

Sent from my CPH2451 using Tapatalk
 
MT Steve said:
Ah, that makes sense.

So essentially, teams with elite scoring in isolation would be punished by this metric whereas teams that are very bad in isolation would be rewarded, even though both teams could be equally competent with moving the ball.
Click to expand...
Exactly my point. I need to hire you to translate me for others. Whats your going rate?
 
SoberasHotRod said:
I can see what you are saying on the passes/EFG% data, but not the potential assist data.

A potential assist, as I understand it, is any pass that leads to a FG attempt. I would have assumed that FG attempts from a pass have a much higher chance of being made than a FG attempt not from a pass.
Click to expand...
I was thinking of things like players shooting the ball right after receiving it on not great looks, which would still be a potential assist but not a high percentage shot.
 
SoberasHotRod said:
OK, I think I figured out a way to get what I was looking for. I averaged the FG% for all of the teams assisted vs unassisted and there is a huge difference. The average FG% for an assisted FG is 62.7%. The average FG% for an unassisted FG is 37.3%. So something isn't adding up to me for why AST% isn't correlating better with OFFRTG? I made the same plot for AST% vs FG% EFG% and it was similarly uncorrelated.
Click to expand...

FWIW I'm pretty sure I misunderstood this stat from NBA.com as well. It isn't the FG%, but % of made fgs that are assisted or unassisted. That's why if you add them up it equals exactly 100%.

Does NBA.COM have anywhere that gives their calculations for each of their stats, or a longer explanation than just the glossery?
 
OK, I'm giving up on the FG% on assisted attempts vs FG% on unassisted attempts. NBA.com at least doesn't keep track of assisted vs unassisted attempts, and I can't figure out a way to get this data. I always thought that potential assists were all of the passes that led to a potential assist, but it isn't. A potential assist is a pass that leads to a shot after one or less dribbles. An assist is a pass that leads directly to a made basket. So potential assists include when someone shoots free throws, but assists do not (I actually figured out a way to separate out the potential free throw assists), but more importantly an assist can include any number of dribbles, but a potential assist is only one.

One nice thing is I did find the stat glossary, so at least I can look up what everything means now: https://www.nba.com/stats/help/glossary.
 
In leu of not being able to find the unassisted vs assisted efg%, I think the below is probably the best way to look/think about it. Here is the efg% based on the number of dribbles before a shot:

1706017954716.png
 
SoberasHotRod said:
In leu of not being able to find the unassisted vs assisted efg%, I think the below is probably the best way to look/think about it. Here is the efg% based on the number of dribbles before a shot:

View attachment 15847
Click to expand...
That is pretty good imo, but the biggest problem is probably the fact that putbacks bloat that 0 dribble number.

I appreaciate the effort though. It would have been interesting to get some kind of estimate how assists impact efficiency. But since teams analytics departments use paywall next gen stats like Second Spectrum and you kinda often hear coaches quote their team assist totals... I think its fairly safe to say they care about that number.
 
You must log in or register to reply here.

Share this page
Back
Top