| View previous topic :: View next topic |
| Author |
Message |
jmcfaul13
Joined: 12 Jan 2011
Posts: 5
|
 Posted: Sat Jan 15, 2011 12:18 pm Post subject: Unrecorded Useful Stats |
 |
|
I've been thinking about creating a site that records stats no one else does, but that would still be considered useful. An example-Recording the average time into the shot clock a team shoots each game. I suspect that teams that get into their sets quicker or shoot the ball sooner in the shot clock have a higher fg%, but of course, no one records that information.
Are there any stats you would like to see that no one records? |
|
| Back to top |
|
 |
Crow
Joined: 20 Jan 2009
Posts: 817
|
| Posted: Sat Jan 15, 2011 3:14 pm Post subject: |
|
|
There are a few sites that do some of the same, but maybe not that prominently, so probably would want to know what they are doing if you want to focus on the truly uncounted.
Recording the average time into the shot clock a team shoots each game would be good, though a rough estimate could be found now from what 82 games reports and the distribution is useful information too.
Some other ideas:
Second chance points has been mentioned before, preferably at player and team level for the season.
Same with fast break points for the season if you think it can be fairly done with a second cutoff for scoring shorter than 10 seconds (that data is available at 82 games). Maybe use 5 or 6 seconds?
If you wanted to pursue little things. maybe shots and FG% within the last 1-4 seconds of a quarter to get at player behavior in that circumstance.
Shots from more than 28-30 feet.
Airballs based on clock stop and no individual rebound? Not needed, might be considered cruel but it could be interesting to see how often the pros do it in their games.
A frequency showing how often a player "uses" the possession 2, 3, 4, etc times in row?
Possession scoring streaks? Maybe same for assists and turnovers to round things out. |
|
| Back to top |
|
|
bchaikin
Joined: 27 Jan 2005
Posts: 689
Location: cleveland, ohio
|
| Posted: Sat Jan 15, 2011 3:34 pm Post subject: |
|
|
Are there any stats you would like to see that no one records?
non-steal forced turnovers. any time a player commits a turnover that is not credited as a steal to a defender, credit a defender for good defense with a non-steal forced turnover....
surely there will be turnovers with no obvious defender deserving credit. but you count enough of these as credited to defenders and we'll get a good idea of who is making an impact on defense, similar to how we count blocks or charges drawn...
right now milwaukee is best in the league with 9.0 non-ST forced TO/100poss, the lakers the worst with just 6.7. but i wouldn't know which players on the bucks to give the credit to... |
|
| Back to top |
|
|
jmcfaul13
Joined: 12 Jan 2011
Posts: 5
|
| Posted: Sat Jan 15, 2011 6:26 pm Post subject: |
|
|
| Crow wrote: | There are a few sites that do some of the same, but maybe not that prominently, so probably would want to know what they are doing if you want to focus on the truly uncounted.
Recording the average time into the shot clock a team shoots each game would be good, though a rough estimate could be found now from what 82 games reports and the distribution is useful information too.
Some other ideas:
Second chance points has been mentioned before, preferably at player and team level for the season.
Same with fast break points for the season if you think it can be fairly done with a second cutoff for scoring shorter than 10 seconds (that data is available at 82 games). Maybe use 5 or 6 seconds?
If you wanted to pursue little things. maybe shots and FG% within the last 1-4 seconds of a quarter to get at player behavior in that circumstance.
Shots from more than 28-30 feet.
Airballs based on clock stop and no individual rebound? Not needed, might be considered cruel but it could be interesting to see how often the pros do it in their games.
A frequency showing how often a player "uses" the possession 2, 3, 4, etc times in row?
Possession scoring streaks? Maybe same for assists and turnovers to round things out. |
Hmm, that's interesting. I checked out 82games, seems they do season stats for shot clock usage: http://www.82games.com/0910/0910WAS3.HTM
I don't see any stats for individual games however, so I suppose it's still an option. I was thinking more of a layout like espns.. where you can see game logs as well.
I like your idea for second chance points, I'm just suprised no one does that already?
I was also thinking about recording how long each player has possession of the ball, does that sound ridiculous or interesting? |
|
| Back to top |
|
|
EvanZ
Joined: 22 Nov 2010
Posts: 292
|
| Posted: Sat Jan 15, 2011 6:46 pm Post subject: |
|
|
| bchaikin wrote: | Are there any stats you would like to see that no one records?
non-steal forced turnovers. any time a player commits a turnover that is not credited as a steal to a defender, credit a defender for good defense with a non-steal forced turnover....
surely there will be turnovers with no obvious defender deserving credit. but you count enough of these as credited to defenders and we'll get a good idea of who is making an impact on defense, similar to how we count blocks or charges drawn...
right now milwaukee is best in the league with 9.0 non-ST forced TO/100poss, the lakers the worst with just 6.7. but i wouldn't know which players on the bucks to give the credit to... |
My PBP code records these as "team turnovers", which I split credit among teammates. With a little bit more effort I could keep track of counterparts for these.
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir |
|
| Back to top |
|
|
EvanZ
Joined: 22 Nov 2010
Posts: 292
|
|
| Back to top |
|
|
jmcfaul13
Joined: 12 Jan 2011
Posts: 5
|
| Posted: Sat Jan 15, 2011 6:52 pm Post subject: |
|
|
| EvanZ wrote: | | jmcfaul13 wrote: |
I was also thinking about recording how long each player has possession of the ball, does that sound ridiculous or interesting? |
How would you get this? |
Recording the game and watching it...very tedious...but I feel like most stats are based off the play by plays that everyone has access to... so if you want to do something new, you'll have to record it yourself. |
|
| Back to top |
|
|
Crow
Joined: 20 Jan 2009
Posts: 817
|
| Posted: Sat Jan 15, 2011 7:28 pm Post subject: |
|
|
| On second chance points, STATS LLC is doing it and providing it to paying customers (newspapers) but no one has offered it free to the public as a full database, just the occasional team level second chance citation in newspaper articles. |
|
| Back to top |
|
|
greyberger
Joined: 27 Sep 2010
Posts: 51
|
| Posted: Sun Jan 16, 2011 12:27 am Post subject: |
|
|
| Quote: | | Recording the game and watching it...very tedious...but I feel like most stats are based off the play by plays that everyone has access to... so if you want to do something new, you'll have to record it yourself. |
If you're going to invest a lot of time in a project like this, it's good that you're making sure the results will be worth looking at.
Something that would be cool to see (and I don't think redundant with anybody else's project or even paid services like Synergy) would be pass charts. This would be something like the score sheets demonstrated at the beginning of BoP except coded.
The key idea is that every half-court pass is recorded along with other key moments in the possession. In BoP, this means dribbling that advances the ball or occurs in the post but you could code less or more information (say ball-screens) and still have some interesting results. If you include a timestamp or the result of the possession the pass-chart can be reconciled with play-by-play information.
It would be useful or at least interesting information to see, even it it were just for one team. |
|
| Back to top |
|
|
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
|
| Posted: Sun Jan 16, 2011 9:01 am Post subject: |
|
|
Do we know what fraction of a player's missed FGA are offensively rebounded?
If not, this might be pretty good to know. A .400 shooter whose misses are recaptured .40 of the time is better than a .450 shooter with a .20 recovery rate.
_________________
`
36% of all statistics are wrong |
|
| Back to top |
|
|
EvanZ
Joined: 22 Nov 2010
Posts: 292
|
| Posted: Sun Jan 16, 2011 9:23 am Post subject: |
|
|
| Mike G wrote: | Do we know what fraction of a player's missed FGA are offensively rebounded?
If not, this might be pretty good to know. A .400 shooter whose misses are recaptured .40 of the time is better than a .450 shooter with a .20 recovery rate. |
I would think these data would be skewed towards bigs who put back their own misses.
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir |
|
| Back to top |
|
|
Mike G
Joined: 14 Jan 2005
Posts: 3604
Location: Hendersonville, NC
|
| Posted: Sun Jan 16, 2011 10:28 am Post subject: |
|
|
Also skewed toward penetrators who break down the defensive rebounding wall. And midsize guys who follow their misses.
About half of the top eFG% in the league are primarily outside shooters.
There's a big difference in effectiveness of shots with/without offensive rebounders in position.
_________________
`
36% of all statistics are wrong |
|
| Back to top |
|
|
EvanZ
Joined: 22 Nov 2010
Posts: 292
|
| Posted: Sun Jan 16, 2011 10:35 am Post subject: |
|
|
| Mike G wrote: | Also skewed toward penetrators who break down the defensive rebounding wall. And midsize guys who follow their misses.
About half of the top eFG% in the league are primarily outside shooters.
There's a big difference in effectiveness of shots with/without offensive rebounders in position. |
Yeah, I agree this could make valuation better. For example, in ezPM I treat every missed shot as worth the same, and if the ball is rebounded by the offense, the same credit for that. If it's true that some shots are "more or less reboundable", than I could adjust those credits. In theory...
(and thinking about it, if the average ORR is 30%, it might be 20% in some cases and 40% in other cases, which would obviously be important to know)
This could be one of the things that shows up on APM that none of our models capture (right?).
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir |
|
| Back to top |
|
|
Crow
Joined: 20 Jan 2009
Posts: 817
|
| Posted: Sun Jan 16, 2011 12:45 pm Post subject: |
|
|
| Recording how long each player has possession of the ball and how it moves could be interesting. Though labor intensive, you could select a manageable cut and glean some insights. Maybe pick from the top 6 teams and a few games where they play each other. And maybe even just focus on time when all 5 starters or at least 4 are present on offense to cut down on the work and target understanding of the most normal / set offensive patterns. What these guys do probably will generally describe what other teams and lineups do or should do and it would be most relevant for the playoff match-ups to come. |
|
| Back to top |
|
|
ScoutingMachine
Joined: 02 Nov 2010
Posts: 7
|
| Posted: Mon Jan 17, 2011 6:16 pm Post subject: |
|
|
Another idea...
You can do something regarding shot selection and offensive rebounds of any given player. Something like these:
What is the average ORB% for players that shoot more than 50% of their shots from three?
What is the average ORB% for slashers who take more of their shots at the rim?
What players standout in this situation? For example, Jared Dudley is a guy that shoots lots of 3s but manages to have a high ORB%.
Last edited by ScoutingMachine on Mon Jan 17, 2011 6:26 pm; edited 1 time in total |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
