Stats to Ratings Conversion, Part I: Physical

[OldFoolStyle]

These are formulas I have developed over the years for the NBALive series, and most of them actualy improve the game.
Consider these suggestions, and if anyone has developed other formulas please share them.

I usualy create my players in stages, so that is how I'll be posting this: part I: physical, part II: offense, Part III: defense, Part IV: finnishing touches. And please pardon how I write these out, I've long forgotten my mathmatical notation.

For each attribute I will be giving two examples, the players for this will be Michael Jordan from 1988-89 and Karl Malone from 1996-97.

Strength: (weight-100) - height.
This is one of the ratings that EA most screws up, too many guards in the 80s, too many forwards in the 70s.

Using Michael Jordan we have 6'6" and 198 lbs. 198 - 100 = 98, 98 - 66 = 32. So Mike has a "raw" rating of 32, and a "Live" rating of 61.

Karl Malone is 6'9 and 256 lbs. 256 - 100 = 156, 156 - 69 = 87. So Karl is raw 87, Live 93.

Note: for player that are 5'10", 5'11", 6'10", 6'11", etc, you need to round to the next closest height, that would be 59, 60, 69, 70, respectively.

Base Quickness: Probably better if I just use the examples.
I use the # 99 at a fulcrum of 6'6", being Jordan's list height we'll start with that. We then divide his weight in pounds by 10, 198 ~ 10 = 19.8, rounded up thats 20. we subtract the weight equasion(20) from the Height equasion(99), 99 - 20 = 79/89.

For Karl we subtract 2 from 99 for each inch over 6'6", being 6'9" we get 93. His weight goes from 256 to 25.6 then rounded to 26. 93 - 26 = 67/83.

Hopefully my inability to write math notation didn't make this seem harder than it atually is.

Note: For the first number I subtract 2 for each inch the player is over 6'6", or subtract 2 for each inch under 6'6". or just use this guide: 6'0" = 111, 6'1" = 109, 6'2" = 107, 6'3" = 105, 6'4" = 103, 6'5" = 101, 6'6" = 99, 6'7" = 97, 6'8" = 95, 6'9" = 93, 6'10" = 91, 6'11" = 89, 7'0" = 87.

Note2: This rating will be modified later.

Alternative: Consider it racist if you like, but I subtract 10 for light-skinned players.

Alternative 2: If you find the game too fast, move the fulcrum from 6'6" to 6'0".

Speed: Base quickness minus half of the years pro.

For Jordan we had a Base of 79, as a four year pro we subtract 2. 79 - 2 = 77/88.

Malone was a 11 year pro, so we subtract 5.5 or 6. 67 - 6 = 61/80.

Freethrow: Freethrow% X 100 = Live rating.

Jordan shot .850, so his final rating would be 70/85.

Malone shot .755, so his rating would be 54/76.

Note: If you are familliar with most NBALive editors, then you should already know what I mean by "raw" ratings and "Live" ratings.

Hardiness: % of games played = Live rating.

Jordan played in 81 games or 99%, so his rating is 98/99.

Malone played all 82 games or 100%, as the highest rating is 99/99 that is what he gets.

Otherwise we would divide the number of games the player was in by .82.

Note: If the games played is 41 or under, list him as 0/50. Negative ratings sometimes have unpredictable results in the game.

Alternative: Games played = raw rating. Jordan = 81/90, Malone = 82/91.

Endurance/Fatigue: Minutes per game x 2 + 10.

Michael averaged 40.2 mpg. 40.2 x 2 = 80.4, 80.4 + 10 = 90/95.

Karl averaged 36.6 mpg. 36.6 X 2 = 73.2, 73.2 + 10 = 83/91.

Alternative: Mpg X 2.

Primacy: (Field Goal attempts + Freethrow attempts + Assists) divided by Minutes.

Jordan had 1795 fga, 793 fta, 650 asst, in 3255 minutes. 1795 + 793 + 650 = 3238, 3238 divided by 3255 = 99/99.

Malone had 1571 fga, 690 fta, 368 asst, in 2998 minutes. 1571 +690 + 368 = 2629. 2629 divided by 2998 = 88/94.

Thats it for now, next I'll fumble through an explanation of how I get my offensive ratings.

[DANet]

A couple of nights ago I was checking out your Map of the dunking genome thread because I was creating a player and wanted to set his dunk package. I thought to myself at the time "this guy has some good stuff, I wonder where this guy went?"...then two days later you appear .

Again, I'm thinking "this guy has some good stuff"...

[madskillionaire]

this is really cool. this will definitely help me for editing my real attributes in the game

[madskillionaire]

oh 1 question: how did u get the "live" rating for the weight?

[OldFoolStyle]

No Live rating for the weight, i just use their actual weight(or listed weight)in pounds. Malone's is the same as it is in the game, Jordan's I copied off an old Lindy's pro basketball season prevue. I haven't tried it in metric yet, so if that's what your using you'll have to toy with it. The height is just their list height with the ' removed, example 6'6" = 66, 6'7" = 67, etc.

I dissapeared for a while to do another project, which I'm still kinda doing, but will drop in a little more often now.

[Andrew]

Some good work there, though I disagree with a couple of things.

Strength

This isn't really a rating that can be calculated from statistics, and depends on the individual player. It's been documented that Michael Jordan was one of the strongest guards in the NBA - this greatly helped his post game. Thus, a rating of 61 is somewhat low.

Also consider Dennis Rodman, who was not as tall or heavy as other power forwards, but was still very strong.

Apart from that, nice work. Though it's important to remember there are exceptions to the rule, meaning some ratings must be "personalised".

[elwudl]

Well man, great work again!


But I have to agree to Andrew, all things can't be changed that way, despite your method is very fine for the common average NBA-player!

[OldFoolStyle]

At 198 Jordan is heavyer then most guards, then after his retirement he was 215 which is huge for a guard. 61 is only bad compared to the ratings EA gave them. Using this method Jordan, Majerle, Magic, and Drexler are untouchable strength wise, and still heve the speed and quickness to blow past all but the smallest guards.

Yes Rodman would be a little under rated, though I remember him always struggling with Sir Charles emence girth. Rodman also gets a huge bonus to his quickness(i'll explain later), which definately makes up for this. You could always use this as a starting point and then add/subtract 10 based on reputation, I personaly just like to remove oppinion from my work.

[Andrew]

I personaly just like to remove oppinion from my work.

I understand that, it's just that there are exceptions to the rule. In my experience, basing all player ratings on a formula takes away from some of the individuality of the players.

[OldFoolStyle]

The formulas don't realy detract too much from individuality, as it only takes one digit from any stat or figure to change the rating. All it realy does is measure all players by the same yardstick.
This was mostly a reaction to some of the old Live 95 and 96 patches that put overall rating ahead of realism. I'm sure you remember all rookies having better ratings than the all-stars, or Jordan having enough strength to post up Shaq, and countless other errors of judgment.

That being said, I have noticed my Strength ratings are the only physical attributes that have no variables. Speed considers age, Quickness considers defensive prowess, and Jumping considers age and era. I still contend that you can come within 10-15 points of how strong a player is through body mass calculation.

I recommend looking at some of my finnished players, or even trying them out in game to see if the ends justify the means. For the way I play these ratings work, for others they might not.

BTW: All of the players listed in my Classic Teams dumps on the general 2003 board were created using these formulas.

[Andrew]

Please don't get me wrong. I admire the work you have done, and coming up with all these formulas (I'm including the Offense and Defense threads as well) is no mean feat. It's some great work, and I'm glad to see people like yourself contributing to the community this way.

I just contend that there are exceptions, and adjustments would have to be made to certain players. For example, not all players lose their speed or quickness at the same rate as they age, so adjustments would have to be made accordingly, wherever appropriate.

But like I said, I only disagree with bits and pieces, because I'm a fan of "personalising" ratings (though I've always preferred to use the free throw percentage calculation - FT%=Rating in Live). For the most part, I do like the formulas.

[OldFoolStyle]

I'm also going to concede that there's a huge difference between, say, 250 pounds of pure 100% prime beef(young shawn kemp for example) and 250 pounds of donuts(shawn kemp now, for example). Its when it gets to compairing the 250 pounds of donuts to only 220 pounds of beef that I start getting standoffish, and need to set a boundery.

Looking at some of my previous replies, I seem a little bit snippier then I intended. Thats just my American-public-school education rearing up(oops, are we still alowed to make those jokes, i hope ashcroft isn't gonna put me on some terrorist list for that).

[Andrew]

I understand, it's human nature to defend our work. I apologise if my posts came off as suggesting you were wrong, or that your work wasn't good.

[elwudl]

I edited the Nets and the Spurs with all formulars from part 1-3 and made little adjustments in speed and quickness too make the game not too quick. And well man it's fucking great! I made a exhibition game with me the Nets vs. the Spurs and lost in SA by 12. Great stuff man!

But a question:
what formula do you use for jumping and dunking? and for endurance? maybe I only wasn't able to read, maybe I'm a fool

[OldFoolStyle]

Endurance is already up there, I think i called it fatigue though. Jumping and dunking will be up today or tomorrow.

[OldFoolStyle]

yep, i found it, i forgot ot highlite it.

[b]Endurance/Fatigue:[/b] Minutes per game x 2 + 10.

Michael averaged 40.2 mpg. 40.2 x 2 = 80.4, 80.4 + 10 = 90/95.

Karl averaged 36.6 mpg. 36.6 X 2 = 73.2, 73.2 + 10 = 83/91.

[elwudl]

Oh sorry!

I think when you do it with your formula the endurance-rating is too good! I used the formula: mpg x 2! and i find it okay! maybe it's not a big difference to yours but you could try!

[OldFoolStyle]

I've toyed with endurance both ways, the + 10 does keep starters in a bit much unless they get in foul trouble. Without it though, the key bench players cycle out too fast. Most teams will play a 9-10 man roster, and I've even had games where the computer played all 12.

[Santiago Roza (Kq)]

Well calculating the endurance and hardiness ratings based on minutes or games played might be good for Shaq and Kobe, but what about the reserves? Do they have less endurance or hardiness because they play less than the stars (not because they get tired or injured, but because they're reserves...)?.

[OldFoolStyle]

Depends on the player and team. it's based on minutes and games played, a good sixth man that played in 80 games and averaged 20+ minutes will be right up there with the starters. The 14 year vet that only played 20 games and averaged 2-3 minutes will have problems. Most teams will see players in each extreme, and many players right in the middle.