EPTNBA Season 2017-2018 - Page 50
| Forum Index > Sports |
|
Twinkle Toes
United States3605 Posts
| ||
|
JimmiC
Canada22817 Posts
| ||
|
JimmyJRaynor
Canada17708 Posts
+ Show Spoiler + Milwaukee doesn't look like world beaters but they are 10-4 since hte Bledsoe trade. i imagine they'll keep getting better as Bledsoe fits in better with the team. | ||
|
Jerubaal
United States7684 Posts
| ||
|
JimmyJRaynor
Canada17708 Posts
https://streamable.com/dm70u Once the Spurs get Anderson back I can't see Boston beating them in a 7 game series. If the Spurs manage to have both Leonard and Anderson integrated into the team and in game shape i think the Spurs could sweep Boston. George Hill is turning out to be a really expensive back up point guard. He played 18 minutes tonight. NBAMath.com takes a microscope to his sharp decline in performance. https://nbamath.com/swishes-with-sampson-knicks-flex-offense-mavericks-improvement-and-eric-bledsoes-fit-in-milwaukee/ | ||
|
andrewlt
United States7702 Posts
| ||
|
Twinkle Toes
United States3605 Posts
| ||
|
Twinkle Toes
United States3605 Posts
Win Shares Usage Rate PER BPM OBPM | ||
|
JimmiC
Canada22817 Posts
| ||
|
JimmyJRaynor
Canada17708 Posts
looking back at something i said recently... On November 27 2017 01:01 JimmyJRaynor wrote: On the Raptors, Pascal Siakam is outplaying Serge Ibaka. Jacob Poetl is outplaying Jonas Valanciunas. Problem is Ibaka and JV combine to make over $40M/year while Poetl and Siakam make 10% of that. If this keeps up Casey has got some tough decisions to make. a tip of my hat to Casey for sitting Miles, JV, and Ibaka while playing OG, Poetl and Siakam. i guess when you have a coach that has been with the team for many years and knows he won't get fired you get a guy willing to make gutsy decisions that some boxscore readers would call "dumb". On December 09 2017 21:21 Twinkle Toes wrote: Hey JJ, care to explain a few terms: Usage Rate here is a great look at how usage rate is ESTIMATED and how it is better to manually count possessions and chances rather than estimate them. https://squared2020.com/2017/12/02/usage-and-efficiency/ as usual, Justin Jacobs nails it. one key to understanding usage is that "chances" are not the exact same thing as "possessions". If an offensive rebound occurs the # of "chances" will be greater than the # of "possessions". | ||
|
Jerubaal
United States7684 Posts
On December 09 2017 21:21 Twinkle Toes wrote: Hey JJ, care to explain a few terms: Win Shares Usage Rate PER BPM OBPM I'm not sure if these were rhetorical or genuine curiosity; they are rather simple terms. I'll try, however, to prove a point that these things can be explained in actually helpful ways. I've taken this information straight from the BasketBall Reference, but I've paraphrased and added my own commentary. -Win Shares attempts to credit a player with the number of wins he was responsible for. It seems to me to be a purely offensive stat based around points scored. The win shares calculation is the quotient (that's like product, except for division. I had to look up that particular term. xD) of two calculations. The first calculation, marginal offense, considers total player points, league points per possession and offensive possessions (by that player I think). So we're seeing how many points a player scores per possession while considering the league average. The second number is the league points per game multiplied by the quotient of team pace and league pace. So, in a nutshell, Win shares is a player's points per possession (less a factor of the team and league PPP) while adjusting for team and league pace. -Usage Rate is a pretty simple calculation that estimates the percentage of a team's plays a particular player is involved in. That seems like a terrible explanation but that's the way it's worded. It's easy to understand though. It compares a players's FGA, FTA, minutes played, and turnovers to the team's overall values. -PER or player efficiency rating is a player per minute rating that is basically a hodgepodge of values derived from counting stats. It considers: FG, FGA, assists, FT, DRB%, ORB, blocks and personal fouls. It also factors in team assists and team FGA. I think most people think that PER is an ok beginning for advanced stats, but it is very inelegant and surely flawed. I think the problem is that it's not very elegant but is also very broad, so you can't use it as an all encompassing stat but also can't be used for any specific circumstance. It's like a big, white sheetcake. You can use it for any occasion, but why would you want to? -I actually hadn't heard of BPM or OBM before this. It's an interesting concept. It's attempting to make a box score stat in the image of Plus-Minus. Most of the terms are pretty vanilla though: Raw BPM = a*ReMPG + b*ORB% + c  RB% + d*STL% + e*BLK% + f*AST% - g*USG%*TO% + h*USG%*(1-TO%)*[2*(TS% - TmTS%) + i*AST% + j*(3PAr - Lg3PAr) - k] + l*sqrt(AST%*TRB%) It's then put through the regression wringer, which is as far as my math skills will go. Well, I hope that wasn't too much copy paste. I don't think you have to be a math whiz to understand the concepts and understand the purpose of each stat. The limitations of the stats make them interesting, I think. But don't take my word for it! Ba dun dun. | ||
|
Ace
United States16096 Posts
| ||
|
JimmyJRaynor
Canada17708 Posts
| ||
|
JimmyJRaynor
Canada17708 Posts
Kyle Anderson (left MCL sprain), Pau Gasol (rest), Manu Ginobili (rest), Kawhi Leonard (right quadriceps tendinopathy) and Tony Parker (return from injury management) are out for Saturday's game at Phoenix. Danny Green (tightness, left groin) is questionable. On December 10 2017 03:36 Jerubaal wrote: -Usage Rate is a pretty simple calculation that estimates the percentage of a team's plays a particular player is involved in. That seems like a terrible explanation but that's the way it's worded. It's easy to understand though. It compares a players's FGA, FTA, minutes played, and turnovers to the team's overall values. usage rate can be miscalculated when the person uses estimated possessions and "chances" rather than programatically stepping through the play-by-play data and accurately counting the exact # of possessions and "chances". you can get the play by play data from NBA.com this inaccuracy due to "estimating" rather than "counting" also occurs with Defense Rating and Offense Rating. 10 seconds after a game is over and you hear a commentator quote a player's Defensive Rating.. chances are it was calculated using an estimate of possessions based on boxscore data. long after the game is over and someone quotes a player's Defense Rating its important to do some digging and determine whether it came from actual counted possessions or an estimate. On December 09 2017 12:07 Twinkle Toes wrote: I wonder who's gonna break the DJ trade first, Charnuke, Wojbomb, Broussard (lol), or someone else the only big trade last night was the Yankees getting Stanton. | ||
|
JimmiC
Canada22817 Posts
| ||
|
JimmyJRaynor
Canada17708 Posts
Embiid is not playing tonight  | ||
|
Twinkle Toes
United States3605 Posts
On December 10 2017 03:36 Jerubaal wrote: I'm not sure if these were rhetorical or genuine curiosity; they are rather simple terms. I'll try, however, to prove a point that these things can be explained in actually helpful ways. I've taken this information straight from the BasketBall Reference, but I've paraphrased and added my own commentary. -Win Shares attempts to credit a player with the number of wins he was responsible for. It seems to me to be a purely offensive stat based around points scored. The win shares calculation is the quotient (that's like product, except for division. I had to look up that particular term. xD) of two calculations. The first calculation, marginal offense, considers total player points, league points per possession and offensive possessions (by that player I think). So we're seeing how many points a player scores per possession while considering the league average. The second number is the league points per game multiplied by the quotient of team pace and league pace. So, in a nutshell, Win shares is a player's points per possession (less a factor of the team and league PPP) while adjusting for team and league pace. -Usage Rate is a pretty simple calculation that estimates the percentage of a team's plays a particular player is involved in. That seems like a terrible explanation but that's the way it's worded. It's easy to understand though. It compares a players's FGA, FTA, minutes played, and turnovers to the team's overall values. -PER or player efficiency rating is a player per minute rating that is basically a hodgepodge of values derived from counting stats. It considers: FG, FGA, assists, FT, DRB%, ORB, blocks and personal fouls. It also factors in team assists and team FGA. I think most people think that PER is an ok beginning for advanced stats, but it is very inelegant and surely flawed. I think the problem is that it's not very elegant but is also very broad, so you can't use it as an all encompassing stat but also can't be used for any specific circumstance. It's like a big, white sheetcake. You can use it for any occasion, but why would you want to? -I actually hadn't heard of BPM or OBM before this. It's an interesting concept. It's attempting to make a box score stat in the image of Plus-Minus. Most of the terms are pretty vanilla though: Raw BPM = a*ReMPG + b*ORB% + c RB% + d*STL% + e*BLK% + f*AST% - g*USG%*TO% + h*USG%*(1-TO%)*[2*(TS% - TmTS%) + i*AST% + j*(3PAr - Lg3PAr) - k] + l*sqrt(AST%*TRB%) It's then put through the regression wringer, which is as far as my math skills will go. Well, I hope that wasn't too much copy paste. I don't think you have to be a math whiz to understand the concepts and understand the purpose of each stat. The limitations of the stats make them interesting, I think. But don't take my word for it! Ba dun dun. Thanks a lot, I appreciate it. | ||
|
Twinkle Toes
United States3605 Posts
On December 10 2017 03:50 Ace wrote: It wasn't genuine curiosity. Just a dig at JJ so he would explain it and they'd have an excuse to argue with him for 4 more pages and trash the thread up. I admit a small part of it was making JJ brag again about his uncle and watching from the sidelines a couple of hours before the game starts, but it was a genuinely honest question. There seem to be so many formulas and applications, I wonder if there is really a standard metric for it. | ||
|
JimmyJRaynor
Canada17708 Posts
| ||
|
Twinkle Toes
United States3605 Posts
| ||
| ||
