A brand new Draft stat!

[Plowright]

So some of you are aware I have started a website recently with a few people on here from BCP. We are pleased to announce we have developed our very own NBA Draft stat!!!

This is called the Draft Franchise Rater (DFR): This stat tells us which team makes the best decisions with their first round picks, regardless of position. It must be emphasized that the DFR does NOT rate which teams have drafted the best players! The DFR is a way to measure how good a team’s scouting department and decision-making is come draft time.

Check it out now!!! Also it is our official website launch today so make sure you check out the rest of the site. We have our own NBA Lottery Bigboard and prospect analysis on Dieng and Burke so far

http://thelotterymafia.com/franchise-draft-rater/

[CampNightmare8]

Interesting, I will definetely be taking a look.

[Chef]

So some of you are aware I have started a website recently with a few people on here from BCP. We are pleased to announce we have developed our very own NBA Draft stat!!!

This is called the Draft Franchise Rater (DFR): This stat tells us which team makes the best decisions with their first round picks, regardless of position. It must be emphasized that the DFR does NOT rate which teams have drafted the best players! The DFR is a way to measure how good a team’s scouting department and decision-making is come draft time.

Check it out now!!! Also it is our official website launch today so make sure you check out the rest of the site. We have our own NBA Lottery Bigboard and prospect analysis on Dieng and Burke so far

http://thelotterymafia.com/franchise-draft-rater/

nice work. to make up for the brandon wright anomaly, perhaps calculating in raw playing time too. in other words, balance PER with actual per game value in terms of usage. another nice piece would be to factor in salary. wright is probably an ok value but when you factor in his first 4 years being paid as the 8th overall pick, his production sucked. chandler parsons on the other hand is fantastic. it may actually prove to better a better gauge to only use rookie deals as opposed to extension or FA deals since cap value is so important now.

[Plowright]

nice work. to make up for the brandon wright anomaly, perhaps calculating in raw playing time too. in other words, balance PER with actual per game value in terms of usage. another nice piece would be to factor in salary. wright is probably an ok value but when you factor in his first 4 years being paid as the 8th overall pick, his production sucked. chandler parsons on the other hand is fantastic. it may actually prove to better a better gauge to only use rookie deals as opposed to extension or FA deals since cap value is so important now.

Yep we're looking into including a "minutes played" "games played" etc. only issue is the time it takes to crunch all the numbers, I will update the thread when we have done. Thanks for the advice Chef!

[bing!]

Two words... Win Shares.

[Chef]

Two words... Win Shares.

i was going to add that but i thought that it would too heavily favor teams that win too much. not sure i have a full understanding of how the stat is tabulated.

[bing!]

i was going to add that but i thought that it would too heavily favor teams that win too much. not sure i have a full understanding of how the stat is tabulated.

I know, and I don't really see a problem with that, seeing as the players' overall impact reflects on the W/L ratio of the team they're playing for. That is, a 20/5/5 guy on the worst team in the league is arguably a worse player than a 20/5/5 guy on the best team (despite an identical output), and their win shares would accurately indicate that. Although, if a guy is really good, it shows. Case in point, Kevin Love's impact during the 11/12 season despite the Wolves' poor W/L...

1. LeBron James-MIA - 14.5
2. Chris Paul-LAC - 12.7
3. Kevin Durant-OKC - 12.2
4. Kevin Love-MIN - 10.0

[Chef]

I know, and I don't really see a problem with that, seeing as the players' overall impact reflects on the W/L ratio of the team they're playing for. That is, a 20/5/5 guy on the worst team in the league is arguably a worse player than a 20/5/5 guy on the best team (despite an identical output), and their win shares would accurately indicate that. Although, if a guy is really good, it shows. Case in point, Kevin Love's impact during the 11/12 season despite the Wolves' poor W/L...

1. LeBron James-MIA - 14.5
2. Chris Paul-LAC - 12.7
3. Kevin Durant-OKC - 12.2
4. Kevin Love-MIN - 10.0

but is the inverse true (this is where i may not fully understand the stat)? ie does danny green or mario chalmers have a high win-share because they are on really good teams? if you substituted, say, kirk hinrich for chalmers what would happen to each player's win shares?

[SWedd523]

I know, and I don't really see a problem with that, seeing as the players' overall impact reflects on the W/L ratio of the team they're playing for.

This doesn't really take into account the quality of the team the player is on.

Kemba's Win Shares this year was 4.8 That's lower than a lot of clearly worse players who benefitted by being on better teams

http://www.basketball-reference.com/..._advanced.html

[bing!]

This doesn't really take into account the quality of the team the player is on.

Kemba's Win Shares this year was 4.8 That's lower than a lot of clearly worse players who benefitted by being on better teams

http://www.basketball-reference.com/..._advanced.html

Yeah, but he's also "better" than Steve Nash & Rudy Gay. Even though both are arguably better players than Kemba, they were pretty much duds for their teams this season (especially Nash, whilst Gay went to a worse team and managed to produce less, even though he should have shined there as an only star), and their WSs reflect that.

Comparing him to other PGs who have a higher WS than him, and bearing in mind that there isn't much difference between, let's say, 4.8 and 5.4 of Jeremy Lin (that's half a win), you can see that they are rated, to a degree, accordingly. It takes into account the efficiency, usage rate, and the impact one has on winning a game. It's hardly perfect, I know, but it's the closest thing to it.

One can't quantify skill, but seeing as this game is all about winning and not individual performance, one is better off with a statistic that gauges the impact on wins when determining who is the "better" player during any given season.