RealGM

04.17 Rockets @ Lakers: Gasol wins 5.4 points for Lakers



04.01 Magic @ Rockets: Asik wins 10.9 points for Rockets



03.24 Spurs @ Rockets: Harden wins 8.7 points for Rockets



03.20 Jazz @ Rockets: Lin wins 7.9 points for Rockets



03.15 Timberwolves @ Rockets： Harden wins 9.1 points for Rockets



03.09 Rockets @ Suns： Dudley wins 6.4 points for Suns



03.08 Rockets @ Warriors: Parsons wins 10.6 points for Rockets



03.01 Rockets @ Magic: Beverley wins 6.5 points for Rockets



02.20 Thunder @ Rockets: Harden wins 20.2 points for Rockets



02.12 Rockets @ Warriors： Parsons wins 6.6 points for Rockets



02.10 Rockets @ Kings: Hayes wins 5.1 points for Kings



02.08 Blazers @ Rockets： Harden wins 14.1 points for Rockets



02.06 Rockets @ Heat : Wade wins 8.4 points for Heat



02.05 Warriors @ Rockets: Lin wins 8.1 points for Rockets



02.02 Bobcats @ Rockets: Patterson wins 7.9 points for Rockets



01.30 Rockets @ Nuggets: Gallinari wins 6.3 points for Nuggets



WPTS means win points. The last column "Final WPTS" is the sum of the 11 WPTS before. The sum of "Final WPTS" of all one team players is equal to the match points difference.

Instructions of the 11 WPTS before "Final WPTS" are as below:
"Shot WPTS" is calculated by using PTS, FGA and FTA. Compare the PTS, FGA and FTA of one player with the other players of two teams. For example, one player shoots 9 times and scores 12 points. The other players of two teams shoot 100 times and score 100 points. This player "Shot WPTS" is 12-9*100/100=3.

In "RB WPTS" :
One offensive rebound points= “points per possession” * “defensive rebound probability”.
In above formula: “defensive rebound probability”=“two teams defensive rebounds” / ”two teams total rebounds”.
One defensive rebound points= “points per possession” * “offensive rebound probability”.
In above formula: “offensive rebound probability”=“two teams offensive rebounds” / ”two teams total rebounds”.
For example in one match two teams get 30 offensive rebounds with 70 defensive rebounds and score 200 points with 200 offensive times. In this game:
One offensive rebound points=(200/200)*[70/(30+70)]=0.7
One defensive rebound points=(200/200)*[30/(30+70)]=0.3
"RB TIME" before "Shot WPTS" is to amplify the minutes played of centers and power forwards and reduce the minutes played of guards and small forwards. So the rebound performances are judged on the player position differently.

"AST WPTS", "STL WPTS" and "BLK WPTS" are calculated by using the number of assists, steals, blocks and minutes played　respectively.
"TOV Lost WPTS" is the losing points due to turnovers and calculated by using the number of turnovers.
"MP Cost WPTS" is to make up for turnovers according to minutes played.
"Shot Cost WPTS" is to make up for turnovers according to shooting times.

"DEF WPTS" means defense win points. If the offensive efficiency of one team is better than the other team, "DEF WPTS" is positive. Otherwise "DEF WPTS" is negative. One team total "DEF WPTS" can only be distributed according to minutes played for lack of opponents’ points and shooting times on each player’s defense.The team total "DEF WPTS" is close to "Shot WPTS".

Sum up shooting times and assist times of every player. Punish the players who are under the average level of team by "OFFEN WPTS". The penalty points "OFFEN WPTS" are distributed to the players whose "PTS/MP" exceeds the average level of team.
"DEV WPTS" is to eliminate deviation. "DEV WPTS" is proportional to minutes played.

Neat stuff. Where'd you get it?

moofs wrote:Neat stuff. Where'd you get it?

its original

Very interesting stuff, Yi!

Looking forward to seeing the comparisons between games as the season closes out (especially when comparing games against the same opponent).

Looks like it has some good potential for stats geeks.

Can you post links to a spreadsheet with the calculations, like a google spreadsheet? (or is it secret?) I'd like to understand it more.

You said there are 11 Win Point factors. Are they equally weighted (e.g. is Reb WPTS worth the same as Blk WPTS)?

Mr. E wrote:Very interesting stuff, Yi!

Looking forward to seeing the comparisons between games as the season closes out (especially when comparing games against the same opponent).

THANKS

madbucky wrote:Looks like it has some good potential for stats geeks.

Can you post links to a spreadsheet with the calculations, like a google spreadsheet? (or is it secret?) I'd like to understand it more.

You said there are 11 Win Point factors. Are they equally weighted (e.g. is Reb WPTS worth the same as Blk WPTS)?

In "RB WPTS" :
One offensive rebound points= “points per possession” * “defensive rebound probability”.
In above formula: “defensive rebound probability”=“two teams defensive rebounds” / ”two teams total rebounds”.
One defensive rebound points= “points per possession” * “offensive rebound probability”.
In above formula: “offensive rebound probability”=“two teams offensive rebounds” / ”two teams total rebounds”.

02.10 Rockets @ Kings updated

Defeated by the Chuckwagon

02.20 Thunder @ Rockets updated

03.01 Rockets @ Magic updated

Really neat stuff Yi. Thanks for keeping everything in one thread as well.

I'll be honest, I haven't had the time to really look at the tables and identify what the numbers really meant. I certainly hope to do that one day.

TMU wrote:Really neat stuff Yi. Thanks for keeping everything in one thread as well.

I'll be honest, I haven't had the time to really look at the tables and identify what the numbers really meant. I certainly hope to do that one day.

thanks

03.15 Timberwolves @ Rockets updated

03.20 Jazz @ Rockets updated

03.24 Spurs @ Rockets updated

04.01 Magic @ Rockets updated

04.17 Rockets @ Lakers updated