Player Rating Discussion

[Alex English]

I think everyone more or less agrees that the player rating system is off a little. I know Ian plans to properly address rating changes after the off-season but I think it's pointless to discuss specific rating changes if we are all working on a slightly different scale of what a rating means.

So hopefully this can start a discussion so the whole league can be on the same page.

This is how I value the ratings:
90+: Franchise players only, a player has to prove they can perform at an elite level while making their team successful.
85-89: All-Star potential players, these are the next best players who are top 5-10 at their position
80-84: Good starters, these player are competent and hold their own but don't dominate
77-79: Bad starters/good bench player, these are your typical 6th man players
74-76: Average bench players, they're decent players who make solid back ups
70-73: Lower end role players, they only fill out your rotation
65-69: 12th men and players who only just make the league, you sign them just to fill out your bench

I'd roughly break the distribution down as follows:
90+: approx 5-10 players (just a few for the whole league)
85-89: approx 20-30 players (at most 1 per team)
80-84: approx 90-110 players (3-4 per team)
77-79: approx 100-120 players (3-4 per team)
73-76: approx 90-110 players (3-4 per team)
70-72: approx 60-80 players (2-3 per team)
65-69: approx 50-70 players (about 2 per team)

Total player pool: approx 450-480 players (at least enough for 15 per team)

_______________________________


That's my personal opinion on the issue. What are your thoughts?

[Danny Longley]

I'd say that this is pretty close to how I see things, nitpick on the 85-89 range wording to something like "All-Star/Borderline All-Stars" because "Potential All-Star" seems to imply something different.

[Deleted]

bump, we need this badly.

[Deleted]

This looks good but it might be too abstract...waybe we need definitions for what constitutes a "All-Star potential players, these are the next best players who are top 5-10 at their position."

[Alex English]

Apr 7, 2014 19:24:47 GMT Vlade Divac said:

This looks good but it might be too abstract...waybe we need definitions for what constitutes a "All-Star potential players, these are the next best players who are top 5-10 at their position."

Well I guess to go back to the Bosh situation. He has made 9 straight All-Star teams. He was a near franchise player in Toronto who has had his stats affected heavily by how good his teammates are. His skills have arguably improved since joining the Heat though as he now has a consistent 3 point shot.

Compared to Smith and Gay who haven't made any All-Star teams. But they are among the best players who haven't. Which makes sense to put their ratings around 85.

[Deleted]

Apr 7, 2014 19:35:16 GMT Alex English said:

Apr 7, 2014 19:24:47 GMT Vlade Divac said:

This looks good but it might be too abstract...waybe we need definitions for what constitutes a "All-Star potential players, these are the next best players who are top 5-10 at their position."

Well I guess to go back to the Bosh situation. He has made 9 straight All-Star teams. He was a near franchise player in Toronto who has had his stats affected heavily by how good his teammates are. His skills have arguably improved since joining the Heat though as he now has a consistent 3 point shot.

Compared to Smith and Gay who haven't made any All-Star teams. But they are among the best players who haven't. Which makes sense to put their ratings around 85.

Kobe made the All-Star Team this year without even playing and I believe Yao Ming did the same thing....who cares? Boy he must have sucked in Toronto if his skills are better in Miami. You can credit the skills of the players around you but he gets a lot of open jump shots because of Lebron and Wade which positively affect his stats.

I am for it if you want to put all three of them at around an 85.

[Ian Noble]

I like it as a general guide, but I think it's important to still use our opinions.

[Deleted]

Apr 7, 2014 20:07:24 GMT Ian Noble said:

I like it as a general guide, but I think it's important to still use our opinions.

I think we need to say something like "if you don't average 15 points a game, you can't be over an 85" because we are getting absurd about certain things.

[Tracy McGrady]

I don't agree with the last post because it's all about impact and presence on a team. Dennis Rodman would not have averaged 15+ points per game but he would definitely be rated 85+ just due to his presence.

[Walt Frazier]

I also don't like Vlade's last post but Tracy, yes.

We should really consider in the end, just before giving our final rating, what kind of overall impact a player has on his team.

Is a player a "Smart" player, or a "knucklehead" (or something along those lines). Are his stats empty? Does his team seem to struggle no matter how well he plays? If he leaves, does his team seem to get better even if the return in a trade seems to be less?

There are really a TON of factors, I agree with where Ian is going. So, when I give my rating after looking at all the stuff I look at, I try to just also put a general "what impact does this player have" spin on everything. I believe a player can have a great impact even on a bad team, if most of the rest of his team is bad. And a good team can have a few bad players whose stats are inflated. No specific examples at this time, but just some food for thought.

[Jeremiah Hill]

You are drunk if you think Kyrie Irving is better than John Wall. Take Kobe and Rose out of the franchise players and put Kevin Love and Blake Griffin in there.

[Chris Mullin]

I think part of the problem is for some players their ratings are only getting voted on by say 5 GMs and another players rating is being voted on by say 10 GMs, but at the end of the day both players are having their ratings changed one way or the other. I dont really think its fair to increase or decrease a player's rating based on a potentially very small sample size of GMs opinions.

Im not sure what the solution would be to help fix this part to the problem, but I feel like there should be some type of minimum number of votes required before a player's rating is increased or decreased.

[Bob McAdoo]

I'm a data guy. Is there a stat like PER a majority could agree on?

Create an algorithm and eliminate all this sports bar bullshit about why my opinion is better than yours. Rerun the process in October and February so people know when rating changes might be coming and PER is visible so there wouldn't be any surprises. Only PER that change by more than 10% (minimum absolute = 1.5) would be implemented.


Ok, now everybody $#!+ all over that idea.

[Jeremiah Hill]

When I consider who I think is a franchise player I don't think about their overall. For example I think that Kyrie Irving is a franchise player, however he's not playing like one YET. Now I would change that term to Superstar. Also allstar level is what I'm going to consider, as there are always snubs etc.

But I went ahead and took the time to just do a count, this is based off of the roster page without any of the changes.

Superstars 90+: 11
All-Star 85-89: 22 (this includes Paul Pierce still)
Great Starters 81-84: 47
Starter Caliber 79-80: 46
Good Bench Guys 77-78: 66
3pt Specialist type guys 75-76: 72
Prospects/End of the Bench 69-74: 138

There are 450 roster spots in the league and these rated players make up 402 of the players on rosters. The rest are lower than that. As it stands right now a lot of the 80 plus players are clustered onto the top 8-9 teams, this is due to the tanktastic nature of the league right now and I feel it will balance out when next year the draft isn't one of the best since the LBJ draft. And the competitiveness will go up. So many teams are tanking this year because it seems like even if you are the 10-15 spots you can get a pretty good player.

As we can see our top tier guys kind of match up with what Alex was saying in regards to how the guys should be distributed.
When it comes to the OVRs of certain guys there are always going to be dudes like for instance DeAndre Jordan who are just so astonishingly good at their role even if it is just athleticism rebounding and defense that they warrant a better rating. I said he should be an 85, do I think that DeAndre is an allstar level talent? While I doubt with the way the west is currently structured he'll ever make a team in real life I do think that he provides that type of value to a team.

The makeup of the league is distributed mostly pretty good. A few guys need shuffling around, there are quite a few egregious ratings that Walt addressed. So I think we'll see a few guys go up to the top and a few guys go up into the 79-80 range.

[Deleted]

Apr 7, 2014 21:14:07 GMT Tracy McGrady said:

I don't agree with the last post because it's all about impact and presence on a team. Dennis Rodman would not have averaged 15+ points per game but he would definitely be rated 85+ just due to his presence.

Yeah I agree with this as well, I merely used it as an example, but there has to be SOME KIND of tangible threshold to measure players against.

[Deleted]

Apr 7, 2014 22:33:36 GMT Bob McAdoo said:

I'm a data guy. Is there a stat like PER a majority could agree on?

Create an algorithm and eliminate all this sports bar bullshit about why my opinion is better than yours. Rerun the process in October and February so people know when rating changes might be coming and PER is visible so there wouldn't be any surprises. Only PER that change by more than 10% (minimum absolute = 1.5) would be implemented.


Ok, now everybody $#!+ all over that idea.

You aren't a data guy or you wouldn't stand for Chris Bosh being an 88.

[Jeremiah Hill]

Apr 7, 2014 22:33:36 GMT Bob McAdoo said:

I'm a data guy. Is there a stat like PER a majority could agree on?

Create an algorithm and eliminate all this sports bar bullshit about why my opinion is better than yours. Rerun the process in October and February so people know when rating changes might be coming and PER is visible so there wouldn't be any surprises. Only PER that change by more than 10% (minimum absolute = 1.5) would be implemented.


Ok, now everybody $#!+ all over that idea.

The only thing with PER is that it overvalues some statistics. For example Brandon Wright is practically Jesus when using PER. I would not be against taking a stat like PER, then somehow averaging that out with say, EWA (take into account games missed) and a couple of the other stats. The human and bar element to the ratings needs to be there however. For example with PER Klay Thompson is like 20 spots below Tristan Thompson, I don't know anyone who would take TT over Klay.


This is especially true because OVR takes into account how athletic someone is. That's why the "bar talk" is necessary. Otherwise why even log in. Maybe instead of that, we put certain dates in to reevaluate certain players. Like for example, 2 months into the real season we stop and look at Rookies, then at the Allstar Break we can evaluate what players are vastly outperforming their ratings. Then around the end of the season we look to weed out the old guys who are still rated too high and look to make a list of guys to take out of the game.

[Bob McAdoo]

Apr 7, 2014 22:46:59 GMT Jeremiah Hill said:

...

But I went ahead and took the time to just do a count, this is based off of the roster page without any of the changes.

Superstars 90+: 11
All-Star 85-89: 22 (this includes Paul Pierce still)
Great Starters 81-84: 47
Starter Caliber 79-80: 46
Good Bench Guys 77-78: 66
3pt Specialist type guys 75-76: 72
Prospects/End of the Bench 69-74: 138

Mr. Hill, thanks for looking at some numbers, that's a step in the right direction. 

Using the term "All-Star" is problematic because the idiot fans vote on that - see Kobe Bryant, etc. We need a better category label.

[Bob McAdoo]

Apr 7, 2014 22:56:57 GMT Vlade Divac said:

Apr 7, 2014 22:33:36 GMT Bob McAdoo said:

I'm a data guy. Is there a stat like PER a majority could agree on?

.

You aren't a data guy or you wouldn't stand for Chris Bosh being an 88.

Please read what I actually wrote, I never once voiced an opinion on Bosh because that wasn't the topic of the thread.

[Deleted]

Apr 7, 2014 23:22:08 GMT Bob McAdoo said:

Apr 7, 2014 22:56:57 GMT Vlade Divac said:

You aren't a data guy or you wouldn't stand for Chris Bosh being an 88.

Please read what I actually wrote, I never once voiced an opinion on Bosh because that wasn't the topic of the thread.

I am still waiting in the other thread for you to respond, if you aren't too big of a bitch to do it.