Ratings for All Players NEW VERSION 2.0

[Fifedawg]

Point taken. Thinking about it, the base should be 3 pointers made per 36 minutes played and the adjustment will be made with shooting %. Not sure who holds the record for that though..

As for adjusting for the shorter line, I'll leave that up to you guys. I won't adjust because I think it fits if someone wants do teams for that era. They should shoot at a rate and % as if there where a shorter line. Now if you wanna to mix and match teams like say play a team from that era against a more recent team, than I would recommend adjusting the 3 pt rating.

[townknave]

Cool, that sounds like a more reasonable approach, though as always the details of what formula you use will be important.

I would also advise you to keep in mind players' Shot off Dribble and Shot in Traffic ratings. If a guy has very high ratings in those areas they will affect 3 pt shooting, so it might be a good idea to tweak their 3pt rating down a bit to compensate or they will end up being too effective.

Are you also figuring in the percentage of a player's 3 pt attempts that are assisted?

[Fifedawg]

Are you also figuring in the percentage of a player's 3 pt attempts that are assisted?

No. That stat is not listed for alot of the older players.

[Fifedawg]

One more tendency to work on and have to re-do Three Point Rating. Hopefully I can't knock that out tomorrow. We may be looking at an early release

Also i'm looking for a lilttle help with the Touches Tendency. This particular tendency is really helpful with the realism as far as the key players getting the ball, but I wanted to know the severity of it. I remember reading somewhere else in the forum where they talked about that, but I can't find it now. What I mean by severity, say if you give a player a 100 touches tendency rating, how often would he shoot the ball if the computer controlled this particular player. Would he jack it up say 30-40 times or does it depend on the touches his teammates have as well?

[NBaller_76]

Hi Fifedawg,

First of all thanks for your work. I've been into ratings since NBA Live 97, and have collected and analysed NBA players' stats since then. But lately I do not have much time to continue doing so. Thus, I really appreciate what you're doing.

Now with regards to Touches Tendency. From what I've read, it seems to be actually the frequency that the CPU uses a player to shoot. It seems like the Primacy rating back in NBA Live 2001, 2003, 2004 etc. Thus, what I used to do is either get the number of field goal attempts per 36 or 40 minutes (or 48 minutes or per game), or else to be more practical (but less accurate), use the field goals made (including 3s). You can also opt to use the Points per game (or 36,40 or 48 minutes). Take the maximum (for example Kobe's 2005-06 season and give that a rating of 99, then find the minimum and give it a 25, and work the ratings in between.

Obviously this is what I've used to do. You can have a different opinion, but you asked for help and I'm trying to give it to you here Please do make sure it is like the primacy rating because I am not sure on it, but if it is, I can guarantee you that it should work fine as I just told you

Enjoy and please release soon

Thanks in advance

[townknave]

Best way to calculate touches would probably be USG%. It's available on basketball-reference.com at last back to the '80s. You can also approximate it by calculating FGA+FTA/2+TO.

With very high touches ratings players tend to get a ton of shots, like 30+ over a 48 minute game. I would give the biggest superstars around 80, otherwise stars tend to get too many shots and role players too few.

[slimm44]

What I've done for touches is to look at the FGA per 40 minutes of each player and having the mininum rating a 40 and the maximum a 70. The scale looks like this

FGA Rating
25 70
23.2 68
21.6 65.5
20 63
18.2 60
16.6 57.5
15 55
13.2 52.5
11.6 50
10 48
8.2 45
6.6 42.5
5 40

I've watched CPU vs. CPU playoffs like this and got very realistic shot attempts by all teams.

[townknave]

I really think you need to also factor in FTA... Missed FGA's where the shooter was fouled don't count as FGAs so Slimm44's approach will underrate guys who get to the line a lot.

Use USG%, it's literally perfect.

[slimm44]

I really think you need to also factor in FTA... Missed FGA's where the shooter was fouled don't count as FGAs so Slimm44's approach will underrate guys who get to the line a lot.

Use USG%, it's literally perfect.

Is this a theory or something that you've seen implemented in the game? I tried the usage % way and it didn't work for me. The rating scale that I posted did work. I'm just wondering if you're opinion is based on theory or on application.

[townknave]

It's based on common sense. Your anecdotal affirmation that something "worked" doesn't mean much because I doubt you did careful, empirical analysis of the data results you obtained over a large sample, you just saw what you saw and went with your gut feelings.

You probably weren't looking closely enough to see if guys who shoot a lot of FTA's are really getting the proper number of touches (and FTA's) in the game.

[slimm44]

It's based on common sense. Your anecdotal affirmation that something "worked" doesn't mean much because I doubt you did careful, empirical analysis of the data results you obtained over a large sample, you just saw what you saw and went with your gut feelings.

You probably weren't looking closely enough to see if guys who shoot a lot of FTA's are really getting the proper number of touches (and FTA's) in the game.

In other words, that's your guess. Here are the Minutes, FGA and FTA attempts from one of the series' in the playoffs I watched. I've tweaked the touches tendencies a little bit since then, but they're pretty close to the same. I still had the 40-70 range but didn't use a specific formula, just kind of eye-balled it. It was pretty close, but since I tightened them up they're a little better. Let me know what you think.

Lakers Vs. Thunder 4 Games
----------------MPG---FGA----FTA (This Season)
Durant--------33.5---28.5-----4.50 (20.6 and 8.9)
Westbrook---33.5---13.5----7.75 (17.7 and 8.5)
Green---------32.7---12------4.25 (13.3 and 5.3)
Harden--------18.8---7.5-----1.25 (7.0 and 3.6)
Ibaka----------20.8---5.0-----2.00 (7.3 and 2.5)
Nrstic----------27.5---8.25----3.25 (7.3 and 1.7)
Thefalosha----25.2---4.25----1.50 (5.0 and 1.8)
Collison--------6.0----.75-----1.00 (3.2 and .6)
Maynor--------12.5---3.5---1.25 (4.6 and .6)
Aldrich--------13.2---2.75-----2.00 (.8 and .4)

----------------MPG---FGA----FTA (This Season)
Bryant--------32.5---21.75--7.5 (20.4 and 8.4)
Gasol---------31.3---12.75--5.5 (14.8 and 5.4)
Fisher--------28.8---8.25---1.5 (6.7 and 1.5)
Odom--------25.5---7.75---4.25 (10.8 and 3.6)
Artest--------27.5---7.25---2.00 (8.0 and 1.2)
Bynum-------24.5---5.25---3.25 (No Sample)
Brown--------11.5---4.50---1.50 (8.3 and 1.4, AVG 19MPG, Rotation Error on Me
(If I had him at 19MPG, Stats should be very
(close
Blake---------16.3---3.50---.50 (4.5 and .4)
Barnes-------14.0---3.25---.50 (6.2 and 1.8)
Ratliff-------13.0---.750----.50 (.8 and .0)

I will post more after you comment on these. Durant and Westbrook were the biggest aberrations but still weren't so far off they can't be considered unrealistic, IMO. Durant's FTA were low because I needed to raise his drive tendency. I had him settling for mid-range jumpers way too much.

[Huthutchuck]

I just have to say excellent work!!! This gives the opportunity to create any player or team that ever was. We may not agree with every number, and no two people ever will, but alot of this work is spot on. I'm in awe, thanks for the contribution to the community.

[Montherlant]

Great work Fifedawg. Thank you so much !

[JustJamie820]

Too cool! I really appreciate the effort you put into this. 62 ratings for 16,617 player-seasons is a monstrous task, but you pulled it off...and quite well, too.

[gOry]

big big thx for all of your hard work bro...

[mortiis]

wow...great work :O

[JaoSming]

thanks for releasing it man, this is going to be really useful for a lot of future mods

[Fifedawg]

Thanks everyone for the kind words. Version 1.1 will be out shortly. I had to fix an error on three point shot rating, so version 1.1 will contain that fix. Working on fix to Shot Average and Shot Ratings as well.

I have already started working on making some old great teams and started with 1969-1970 New York Knicks. Very tough defensive team with lots of good shooters.

[LHead2]

Thanks everyone for the kind words. Version 1.1 will be out shortly. I had to fix an error on three point shot rating, so version 1.1 will contain that fix. Working on fix to Shot Average and Shot Ratings as well.

I have already started working on making some old great teams and started with 1969-1970 New York Knicks. Very tough defensive team with lots of good shooters.

Well, I'd be happy to incorporate your ratings (with some adjustments, maybe) into my Greatest Teams mod.

[Fifedawg]

3 pt Rating been Fixed. Version 1.1 is at the top of the page.

Well, I'd be happy to incorporate your ratings (with some adjustments, maybe) into my Greatest Teams mod.

Do what you gotta do LHead2, let me know how the Mod is coming along.

[JustJamie820]

I've noticed a very minor error...players who appeared in both the ABA and NBA in a single season do not have their NBA stats included. This doesn't affect a lot of player-seasons; my own spreadsheet says it affects 42 of them over a nine-year span, and only a few of those are really all that notable...but someone looking to make an NBA season patch from that time frame might be interested in having them. Let me know if you need to know who all is missing.

I still can't fault the effort you put into this, though. This is still amazing.

[Fifedawg]

I've noticed a very minor error...players who appeared in both the ABA and NBA in a single season do not have their NBA stats included. This doesn't affect a lot of player-seasons; my own spreadsheet says it affects 42 of them over a nine-year span, and only a few of those are really all that notable...but someone looking to make an NBA season patch from that time frame might be interested in having them. Let me know if you need to know who all is missing.

I know Charlie Scott is one of them. In 1971-72 he played 73 games with the ABA Virginia Squires and only 6 NBA games with the Phoenix Suns. In the spreadsheet, he's listed with his total for both the Squires and Suns. He averaged 34.6 in the ABA that year and 18.8 with the Suns and thus his ratings are sky high. You're right that's gotta be fixed. No disrespect to the ABA, but I only want NBA stats listed. If you do have a list of all the players that played in the ABA/NBA in the same season, please post that if you can.

[TonGunner]

Hey man, thanks... great work!

How you calculate the block atribute? Because I see Kareem with 67 block shot ability and dont understand that.

And how you calculate the tendencies? I'm kind new to rosters edit and having so much problems with tendencies. lol

Thanks again, I love it... and sorry for my english, I'm from Brazil!

[JustJamie820]

Charlie Scott is exactly who I was going for with the "notable" part.

Here's the list:
1967-68: Jim Burns, Jim Caldwell, Reggie Harding
1968-69: Rod Knowles, George Lehmann, Jay Miller, Ken Wilburn
1969-70: Rich Niemann, Tom Workman
1970-71: Rich Johnson, Dennis Stewart, Al Tucker
1971-72: Shaler Halimon, Jim McDaniels, Steve Mix, Mike Price, Bob Quick, Charlie Scott, Gary Zeller
1972-73: Roger Brown (1972-1980), Mike Davis, Gary Gregor, Gus Johnson, Erwin Mueller, Luther Rackley, John Trapp
1973-74: Larry Cannon, Travis Grant, Mo Layton, Joe Reaves, Rich Rinaldi, Chuck Terry
1974-75: Don Adams, Terry Driscoll, Bernie Fryer, Lamar Green, Stan Love
1975-76: Don Adams, Roger Brown (1972-1980), Mike D'Antoni, Jim McDaniels, John Roche

That should be all 42 player-seasons. Like I said, there's not a terribly large amount of notable ones (though you'll recognize some of the names). Only McDaniels and Scott in 71-72 have seasons in both leagues with at least 5 win shares combined, and most of those came in the ABA for both.

Hope this helps.

[jenroe962002]

This will be useful to all who will make old rosters or mods
Great work fifedawg