In wake of a recent thread in general discussion, I wondered if there was a statistical way to find stat stuffing players. For a first run I just plotted points vs. win shares (for the top 103 players in points). Points is the biggest stat stuffing complaint and the idea is to find the players whose points don't lead to wins (or at least win shares), 2013-2014 season.
The chart:
I then found the player's projected win shares (based on points and the graph's trendline) and looked at the players who most under or over achieved their win share projection. Perhaps unsurprisingly, high scorers on bad teams top the list. The top stat (or points anyways) stuffer: Tony Wroten!
Leading Stuffers - (most underachieved projection)
1. Tony Wroten -121.52%
2. Josh Smith -83.08%
3. Michael Carter -77.62%
4. Victor Oladipo -75.81%
5. Evan Turner -74.785
6. Dion Waiters -70.5%
7. Brandon Knight -58.17%
8. Jamal Crawford -57.77%
9. Nick Young -55.71%
10. Jeff Green -55.06%
Leading Anti-Stuffers (most overachieved projection)
1. Joakim Noah +139.98%
2. Robin Lopez +139.87%
3. Chris Paul +105.47%
4. Andre Drummond +86.96%
5. Terrence Jones +80.1%
6. Jonas Valanciunas +66.77
7. Marcin Gortat +58.74%
8. Mike Dunleavy +58.14%
9. Nicolas Batum +54.39%
10. Jose Calderon +54.34%
What do you think? Let me know if I committed any mathematical heresies. If I wanted to do all "all box score stats" comparison, what should I use instead of just points?
Kevin Love BTW, +20.9% over his projection (31st best of the 103).
Attempt to find "stat stuffers"
Moderator: Doctor MJ
Attempt to find "stat stuffers"
-
- Head Coach
- Posts: 7,237
- And1: 1,883
- Joined: Feb 28, 2005
Re: Attempt to find "stat stuffers"
-
- Sophomore
- Posts: 156
- And1: 76
- Joined: May 23, 2012
Re: Attempt to find "stat stuffers"
WinShares includes stats, so it's no perfect for this analysis. Better to use RAPM, which is completely stat-free. Here's 14-year RAPM http://stats-for-the-nba.appspot.com/ratings/14y.html Sample size is an issue with that stat, so using more years leads to more accurate results
You could compare this to points, NBA Efficiency, PER etc.
You could compare this to points, NBA Efficiency, PER etc.
Re: Attempt to find "stat stuffers"
- RSCD3_
- RealGM
- Posts: 13,870
- And1: 7,278
- Joined: Oct 05, 2013
Re: Attempt to find "stat stuffers"
Jasen777 wrote:In wake of a recent thread in general discussion, I wondered if there was a statistical way to find stat stuffing players. For a first run I just plotted points vs. win shares (for the top 103 players in points). Points is the biggest stat stuffing complaint and the idea is to find the players whose points don't lead to wins (or at least win shares), 2013-2014 season.
The chart:
I then found the player's projected win shares (based on points and the graph's trendline) and looked at the players who most under or over achieved their win share projection. Perhaps unsurprisingly, high scorers on bad teams top the list. The top stat (or points anyways) stuffer: Tony Wroten!
Leading Stuffers - (most underachieved projection)
1. Tony Wroten -121.52%
2. Josh Smith -83.08%
3. Michael Carter -77.62%
4. Victor Oladipo -75.81%
5. Evan Turner -74.785
6. Dion Waiters -70.5%
7. Brandon Knight -58.17%
8. Jamal Crawford -57.77%
9. Nick Young -55.71%
10. Jeff Green -55.06%
Leading Anti-Stuffers (most overachieved projection)
1. Joakim Noah +139.98%
2. Robin Lopez +139.87%
3. Chris Paul +105.47%
4. Andre Drummond +86.96%
5. Terrence Jones +80.1%
6. Jonas Valanciunas +66.77
7. Marcin Gortat +58.74%
8. Mike Dunleavy +58.14%
9. Nicolas Batum +54.39%
10. Jose Calderon +54.34%
What do you think? Let me know if I committed any mathematical heresies. If I wanted to do all "all box score stats" comparison, what should I use instead of just points?
Kevin Love BTW, +20.9% over his projection (31st best of the 103).
What about LeBron and KD?
Sent from my iPhone using RealGM Forums
I came here to do two things: get lost and slice **** up & I'm all out of directions.
Butler removing rearview mirror in his car as a symbol to never look back
Butler removing rearview mirror in his car as a symbol to never look back
Peja Stojakovic wrote:Jimmy butler, with no regard for human life
Re: Attempt to find
-
- Head Coach
- Posts: 7,237
- And1: 1,883
- Joined: Feb 28, 2005
Re: Attempt to find
RSCD3_ wrote:What about LeBron and KD?
LeBron is + 28.33% (24th) and KD is + 20.09 (33rd)
Re: Attempt to find "stat stuffers"
-
- Senior Mod
- Posts: 50,792
- And1: 19,488
- Joined: Mar 10, 2005
- Location: Cali
Re: Attempt to find "stat stuffers"
Like the idea, but yeah, have to use +/- stats on this. Win Shares make Reggie Miller look like a stat stuffer, and he's the absolute opposite of it.
I think RAPM vs points and RAPM vs PER would be good things to try.
I think RAPM vs points and RAPM vs PER would be good things to try.
Getting ready for the RealGM 100 on the PC Board
Come join the WNBA Board if you're a fan!
Come join the WNBA Board if you're a fan!
Return to Statistical Analysis