Overall 2.1

[billy]

Hey guys,

I know ya'll have been frustrated with the way the overall ratings are so out of whack with how useful a player is in the sim. I present to you, overall 2.1. In conjunction with the previously announced league-wide evenly applied ratings nerf, which was designed to lower the ridiculously high scoring games we saw, and also make attributes harder to get, this new overall has been formulated to allow people to better understand how useful a player is in the sim than the old formula.

I haven't updated the overalls on the ratings page itself, but the following spreadsheet shows the new overalls for all the players on each team. It's also updated in the league file. This new overall display rating doesn't effect the sim (or substitutions).

data["players"][i]["ratings"][0]["ovr"] = (data["players"][i]["ratings"][0]["hgt"]*5 +data["players"][i]["ratings"][0]["stre"]*3.7 +
data["players"][i]["ratings"][0]["spd"]*5 +data["players"][i]["ratings"][0]["jmp"]*4.7 +data["players"][i]["ratings"][0]["endu"] +
data["players"][i]["ratings"][0]["ins"] +data["players"][i]["ratings"][0]["dnk"]*1.4 +data["players"][i]["ratings"][0]["ft"] +
data["players"][i]["ratings"][0]["fg"] +data["players"][i]["ratings"][0]["tp"] +data["players"][i]["ratings"][0]["oiq"]*2 +
data["players"][i]["ratings"][0]["diq"]*5 +data["players"][i]["ratings"][0]["drb"]*2.4 +data["players"][i]["ratings"][0]["pss"] +
data["players"][i]["ratings"][0]["reb"]*2)/30

[html]

[billy]

For anyone interested in the code for the script itself, so you could run it:

import simplejson as json

jsonFile = open("39.json", "r+")
data = json.load(jsonFile)

seed = 100

for i in range(10000):
try:
shootingLowPost = (data["players"][i]["ratings"][0]["hgt"]*2 + data["players"][i]["ratings"][0]["stre"]*.6 +
data["players"][i]["ratings"][0]["spd"]*.2 + data["players"][i]["ratings"][0]["ins"] + data["players"][i]["ratings"][0]["oiq"]*.2)
dribbling = (data["players"][i]["ratings"][0]["drb"]+data["players"][i]["ratings"][0]["spd"])
passing = (data["players"][i]["ratings"][0]["drb"]*.4 + data["players"][i]["ratings"][0]["pss"] + data["players"][i]["ratings"][0]["oiq"]*.5)
shootingThreePointer = (data["players"][i]["ratings"][0]["oiq"]*.1+data["players"][i]["ratings"][0]["tp"])
rebounding = (data["players"][i]["ratings"][0]["hgt"]*2+data["players"][i]["ratings"][0]["stre"]*.1+data["players"][i]["ratings"][0]["jmp"]*.1
+data["players"][i]["ratings"][0]["reb"]*2+data["players"][i]["ratings"][0]["oiq"]*.5+data["players"][i]["ratings"][0]["diq"]*.5)
defenseInterior = (data["players"][i]["ratings"][0]["hgt"]*2.5+data["players"][i]["ratings"][0]["stre"]+data["players"][i]["ratings"][0]["spd"]*.5
+data["players"][i]["ratings"][0]["jmp"]*.5+data["players"][i]["ratings"][0]["diq"]*2)
defensePerimeter = (data["players"][i]["ratings"][0]["hgt"]*.5+data["players"][i]["ratings"][0]["stre"]*.5+data["players"][i]["ratings"][0]["spd"]*2
+data["players"][i]["ratings"][0]["jmp"]*.5+data["players"][i]["ratings"][0]["diq"])
athleticism = (data["players"][i]["ratings"][0]["stre"]+data["players"][i]["ratings"][0]["spd"]+data["players"][i]["ratings"][0]["jmp"]+
data["players"][i]["ratings"][0]["hgt"]*.75)

#print("old" + str(data["players"][i]["ratings"][0]["ovr"]))
data["players"][i]["ratings"][0]["ovr"] = (data["players"][i]["ratings"][0]["hgt"]*5 +data["players"][i]["ratings"][0]["stre"]*3.7 +
data["players"][i]["ratings"][0]["spd"]*5 +data["players"][i]["ratings"][0]["jmp"]*4.7 +data["players"][i]["ratings"][0]["endu"] +
data["players"][i]["ratings"][0]["ins"] +data["players"][i]["ratings"][0]["dnk"]*1.4 +data["players"][i]["ratings"][0]["ft"] +
data["players"][i]["ratings"][0]["fg"] +data["players"][i]["ratings"][0]["tp"] +data["players"][i]["ratings"][0]["oiq"]*2 +
data["players"][i]["ratings"][0]["diq"]*5 +data["players"][i]["ratings"][0]["drb"]*2.4 +data["players"][i]["ratings"][0]["pss"] +
data["players"][i]["ratings"][0]["reb"]*2)/37.2
#print("new" + str(data["players"][i]["ratings"][0]["ovr"]))

data["players"][i]["ratings"][0]["skills"] = []

if (shootingLowPost > (seed*2 + seed*.6 + seed*.2 + seed + seed*.2)*.61):
data["players"][i]["ratings"][0]["skills"].append("Po") #works
if (dribbling > ((seed+seed)*.68)):
data["players"][i]["ratings"][0]["skills"].append("B") #works
if (passing > (seed*.4 + seed + seed*.5)*.63):
data["players"][i]["ratings"][0]["skills"].append("Ps") #works
if (shootingThreePointer > (seed*.1 + seed)*.59):
data["players"][i]["ratings"][0]["skills"].append("3") #works
if (rebounding > (seed*2 + seed*.1 + seed*.1 + seed*2 + seed*.5 + seed*.5)*.61):
data["players"][i]["ratings"][0]["skills"].append("R") #works
if (defenseInterior > (seed*2.5 + seed + seed*.5 + seed*.5 + seed*2)*.57):
data["players"][i]["ratings"][0]["skills"].append("Di") #works
if (defensePerimeter > (seed*.5 + seed*.5 + seed*2 + seed*.5 + seed)*.61):
data["players"][i]["ratings"][0]["skills"].append("Dp") #works
if (athleticism > (seed + seed + seed + seed*.75)*.63):
data["players"][i]["ratings"][0]["skills"].append("A") #works

print(data["players"][i]["ratings"][0]["skills"])
except:
continue

with open('39new.json', 'w') as outfile:
json.dump(data, outfile, indent=2)

[billy]

Also more data on the new attributes that are so handy.

Totals in the league now:

Po:39
B: 16
Ps: 20
Tp: 37
R: 63
Di: 46
Dp: 14
A: 30

So it takes someone being pretty elite in an area to get an attribute. The amount of these will be kept somewhat steady, unless the talent pool in the league itself changes significantly.

[Allan Houston]

Awesome work.

But I still maintain that Overalls as a concept aren't desirable. NBA GMs don't get to see players' "overall" values.

[billy]

Jan 18, 2019 21:32:16 GMT -5 Allan Houston said:

Awesome work.

But I still maintain that Overalls as a concept aren't desirable. NBA GMs don't get to see players' "overall" values.

Overall 3.0 is just a black box that when you open it up its your own reflection

[Scott Pilgrim]

Jan 18, 2019 21:32:16 GMT -5 Allan Houston said:

Awesome work.

But I still maintain that Overalls as a concept aren't desirable. NBA GMs don't get to see players' "overall" values.

We don't have head coaches or scouts that we hire to evaluate our talent.

[Brian Scalabrine]

Jan 18, 2019 21:21:34 GMT -5 billy said:

Also more data on the new attributes that are so handy.

Totals in the league now:

Po:39
B: 16
Ps: 20
Tp: 37
R: 63
Di: 46
Dp: 14
A: 30

So it takes someone being pretty elite in an area to get an attribute. The amount of these will be kept somewhat steady, unless the talent pool in the league itself changes significantly.

Dp is so damn coveted

[Steve Jobs]

Jan 18, 2019 21:32:16 GMT -5 Allan Houston said:

Awesome work.

But I still maintain that Overalls as a concept aren't desirable. NBA GMs don't get to see players' "overall" values.

That’d be ideal, but unfortunately, overalls DO play into how the sim functions (and/or is directly tied to the individual ratings that DO control how the sim functions), so it’s not realistic to expect our sim league to work without them, even if they shouldn’t be (and actually never HAVE been) the most important part of how successful teams are built.

[Scott Pilgrim]

Finally, little symbols you see next to a player's name like 3ABDiDpPoPsR represent the key skills a player has. This is designed so that you can just glance at a player and easily take in that information. It has no impact on gameplay. A player who barely misses getting a symbol will be about the same as a player who barely gets it.

To see what a symbol means, hover your mouse over it or consult this list:

3: Three Point Shooter
A: Athlete
B: Ball Handler
Di: Interior Defender
Dp: Perimeter Defender
Po: Post Scorer
Ps: Passer
R: Rebounder
basketball-gm.com/manual/

[Stuff The Magic Dragon]

Jan 18, 2019 22:21:19 GMT -5 Scott Pilgrim said:

Finally, little symbols you see next to a player's name like 3ABDiDpPoPsR represent the key skills a player has. This is designed so that you can just glance at a player and easily take in that information. It has no impact on gameplay. A player who barely misses getting a symbol will be about the same as a player who barely gets it.

To see what a symbol means, hover your mouse over it or consult this list:

3: Three Point Shooter
A: Athlete
B: Ball Handler
Di: Interior Defender
Dp: Perimeter Defender
Po: Post Scorer
Ps: Passer
R: Rebounder
basketball-gm.com/manual/

I don't have a mouse

[billy]

Jan 18, 2019 22:21:19 GMT -5 Scott Pilgrim said:

Finally, little symbols you see next to a player's name like 3ABDiDpPoPsR represent the key skills a player has. This is designed so that you can just glance at a player and easily take in that information. It has no impact on gameplay. A player who barely misses getting a symbol will be about the same as a player who barely gets it.

To see what a symbol means, hover your mouse over it or consult this list:

3: Three Point Shooter
A: Athlete
B: Ball Handler
Di: Interior Defender
Dp: Perimeter Defender
Po: Post Scorer
Ps: Passer
R: Rebounder
basketball-gm.com/manual/

This is a lie though, they do effect gameplay.

[Scott Pilgrim]

Jan 19, 2019 0:20:24 GMT -5 billy said:

Jan 18, 2019 22:21:19 GMT -5 Scott Pilgrim said:

This is a lie though, they do effect gameplay.

www.reddit.com/r/BasketballGM/comments/7b4rfn/a_detailed_analysis_of_the_effects_of_tags_xpost/

have you seen this post on the bbgm reddit, I haven't been able to understand it really

[billy]

Jan 19, 2019 0:37:07 GMT -5 Scott Pilgrim said:

Jan 19, 2019 0:20:24 GMT -5 billy said:

This is a lie though, they do effect gameplay.

www.reddit.com/r/BasketballGM/comments/7b4rfn/a_detailed_analysis_of_the_effects_of_tags_xpost/

have you seen this post on the bbgm reddit, I haven't been able to understand it really

I've said something similar to this before and people don't believe me. The synergy is important, and having the attribute means a lot, but being just slightly under it is still basically the same thing.

[Jackie Kong]

Jan 19, 2019 0:43:02 GMT -5 billy said:

Jan 19, 2019 0:37:07 GMT -5 Scott Pilgrim said:

www.reddit.com/r/BasketballGM/comments/7b4rfn/a_detailed_analysis_of_the_effects_of_tags_xpost/

have you seen this post on the bbgm reddit, I haven't been able to understand it really

I've said something similar to this before and people don't believe me. The synergy is important, and having the attribute means a lot, but being just slightly under it is still basically the same thing.

I don't think it is that they don't believe but that they don't know how to recognize when they are slightly under for an attribute.

[billy]

Jan 19, 2019 0:45:21 GMT -5 Jackie Kong said:

Jan 19, 2019 0:43:02 GMT -5 billy said:

I've said something similar to this before and people don't believe me. The synergy is important, and having the attribute means a lot, but being just slightly under it is still basically the same thing.

I don't think it is that they don't believe but that they don't know how to recognize when they are slightly under for an attribute.

It'd probably be overkill but I could create a 2ndary list of Attributes that show if a player has 75% of an attribute

[billy]

Overall 2.1 announced.

Changelog:

Updated the formula (and the first post in the thread) to have more weights, that more accurately depict the effect that each "Rating" has on one of the "skills" in the sim.

I put an upper limit of 5 on any of the ratings weights in overall (note in reality each point of height gives 13 points to skills -- in comparison Tp gives 1 point to skills). This should be a lot more accurate as to how much each person is contributing to the team. However, each category isn't made equally so it isn't an exact representation. For example, would you rather have a guy that gives 100 to mid range or 100 to three points? Also, the way that points in offense and defense actually contribute to creating the results that we see isn't clear to anyone (except the developer). However, this is by far the best representation of overall player contribution you're gonna find. League file uploaded under 39bnew.

<iframe src="https://docs.google.com/spreadsheets/d/e/2PACX-1vQ2JIhJ3iJbLVgEyzkiPfu2z4gk0TvMpBC3jyCaiEKMe7-wJadMb1c6OF8iSAMPrnEGAwWMirFuFOJw/pubhtml" width="900" height="500">

[html]

[Steve Jobs]

Jan 19, 2019 1:02:53 GMT -5 billy said:

Jan 19, 2019 0:45:21 GMT -5 Jackie Kong said:

I don't think it is that they don't believe but that they don't know how to recognize when they are slightly under for an attribute.

It'd probably be overkill but I could create a 2ndary list of Attributes that show if a player has 75% of an attribute

That’d be dope.

[billy]

added.

[Kareem Abdul-Jabbar]

Is there more to come for the final current? Or are we done here? ^^

[billy]

Jan 19, 2019 19:58:53 GMT -5 Kareem Abdul-Jabbar said:

Is there more to come for the final current? Or are we done here? ^^

It's probably still weighted a little too much toward defense, but I think it's okay. I'm not sure we'll ever get a 1-1 rankings system, in real life players that we see as perfect we don't actually see as having multiple flaws as if they were built in a sim engine. Curry or Harden is a "Flawless" player amirite? But if you build a "Flawless" player in the sim they're gonna have a lot of skills Curry and Harden don't. Like Interior defense, or great rebounding. So it's always gonna be difficult to try to make these numbers match up with the idea of rankings that everyone likes to carry in their back pocket, but I think this is better than before.

[Kareem Abdul-Jabbar]

Jan 19, 2019 20:26:28 GMT -5 billy said:

Jan 19, 2019 19:58:53 GMT -5 Kareem Abdul-Jabbar said:

Is there more to come for the final current? Or are we done here? ^^

It's probably still weighted a little too much toward defense, but I think it's okay. I'm not sure we'll ever get a 1-1 rankings system, in real life players that we see as perfect we don't actually see as having multiple flaws as if they were built in a sim engine. Curry or Harden is a "Flawless" player amirite? But if you build a "Flawless" player in the sim they're gonna have a lot of skills Curry and Harden don't. Like Interior defense, or great rebounding. So it's always gonna be difficult to try to make these numbers match up with the idea of rankings that everyone likes to carry in their back pocket, but I think this is better than before.

Thanks Commish. ^^

[Shaq O'Neal]

Just a question on the ratings.In the league file when i go into edit player and change the team of the player,the rating goes down .Any reason why this happens?

[billy]

Jan 25, 2019 23:01:56 GMT -5 Shaq O'Neal said:

Just a question on the ratings.In the league file when i go into edit player and change the team of the player,the rating goes down .Any reason why this happens?

Because that’s the overall calculated by the sim engine, and this one is calculated by me. You also notice they lose their .9 attributes as well.

[Shaq O'Neal]

Jan 26, 2019 8:40:52 GMT -5 billy said:

Jan 25, 2019 23:01:56 GMT -5 Shaq O'Neal said:

Just a question on the ratings.In the league file when i go into edit player and change the team of the player,the rating goes down .Any reason why this happens?

Because that’s the overall calculated by the sim engine, and this one is calculated by me. You also notice they lose their .9 attributes as well.

Oh okay cool got it!

[billy]

new ratings tab at top to show the calculated overall