CSL x NBA (please read and comment)

[Marcos_Beck]

Guys,

Most of you know our league's shooting ratings are well below modern basketball standards.
People constantly whine as players are shooting very bad since the DDS22 transition, with many failing to crack acceptable level of accuracy, like shooting 37-39% from the field with a small number of threes attempted.

NBA has reached a level of offense that is barely entertaining anymore, as defenses are almost meaningless and players only shoot at the rim or from deep most of the time. This is not interesting at all, however our current CSL standards are really low and could use some changes.

I've done a few studies I'd like to share with you, it has been allowed by admins and we can have discussions that could improve our league a lot using that. There are a few concerns about how to do things, like one of them is making sure no players gets nerfed at all or overpower the engine and breaks the game, but those are workable things I believe.

Here are some numbers comparing CSL and NBA shooting (for a the whole league and one full regular season)



And a study that shows the average for every shooting rating in CSL



And a study that puts CSL ratings ranked into percentiles (if you have any questions let me know below)



With all that in mind I've suggested some ratings improvements we used to test stuff out and see how we could improve the league, and here's the result for 3 tests with improved ratings, as well as new percentile distributon with those ratings increases in mind (keep in mind it is not definite or perfect, but I think it's important to share those)




I'll share more stuff in the next post. Some concerns and objectives I had when making this study and something we can use as guidance to buff ratings if it ever happens.

[Marcos_Beck]

1- Don't nerf players: when you change ratings some players will be better, some will look the same, so there might be a feeling they got worse. But no player will be nerfed, no player will look worse than what he already is.

2- Don't break the engine: we gotta be very careful when buffing top players as they are already pretty strong in comparison to those who really need buffs. If we buff top players too much they could end up unstoppable. The solution could be scaled buffs to buff everyone in general but help players who currently suck more

3- Engine values versatility: this engine gives players with better scoring versatility more shots. The goal then is to help raise volume by making some players useful in more aspects of the game (but they will still look bad if they are bad at something, just a bit more useful). Don't expect a bad midrange shooter to all of a sudden look like DeMar DeRozan out there.

4- Inside game is OP right now: most of the buffs would be naturally directed to FGRA and FGMID, which are two areas that needs a lot of help. But right now inside game in general is very strong, some because of the volume of shots post oriented players get in comparison with perimeter oriented ones, but also because the league has not learned to use 3pt shooting as well as we could. We gotta make sure to not overbuff players in ranges from 0-10 feet from the basket, even more because it will make more styles of play and strategies viable and that's more fun for the league overall.

5- Spacing the floor helps opening the lane: with increased FGRA there could be more incentive for teams to reach the rim, therefore space the floor better. Shooters would be more valuable and important, mid range also as it takes away players from the post and helps reaching the rim. This has been mentioned in Kyrus' PB22 default ratings studies and tested and confirmed once again here. Putting all your players in the post is not a good idea, as it has clogged the lane and produced a terrible 0-3 to 3-10 ratio in the league overall.

[Silogical]

Its a really complicated issue, i think the only way to do this is tiny changes to slowly get it where you want it. The 1st two changes for me would be

1. Increase FG-MID accross the league. use a formula so its balanced
2. draft team creates players with ATB in the 39 range with high SCR/PAS (id be interested if we have a single player like this in the league and who it is)

these two changes will bring balance, right now the league is just who can dominate inside the best.

[Marcos_Beck]

Its a really complicated issue, i think the only way to do this is tiny changes to slowly get it where you want it. The 1st two changes for me would be

1. Increase FG-MID accross the league. use a formula so its balanced
2. draft team creates players with ATB in the 39 range with high SCR/PAS (id be interested if we have a single player like this in the league and who it is)

these two changes will bring balance, right now the league is just who can dominate inside the best.

I have a feeling I'd love to try but putting out together more data would be a helluva effort I'm not willing to do lol
I think 3pt shooting is very low overall because most guard / wings have very high pass preference. Because they are players who tend to get more touches compared to big men overall, those players have less available points to be distribuited into ball actions, and since you want to make balanced players, there's not much to work with and you can't for example go 15 pull up and 15 catch and shoot on a high level guard.. And you would even have to have some drive / kick, drive / shoot and post up as well to increase volume overall (that's what engine values).

Big men tend to have much lower pass % and don't need to spread out in multiple areas so they can stack post up and drive / shoot as much as possible, every touch they get is a 0-10 ft shot basically.....

I mean how do you get Lonzo Ball (88 pass %) to shoot like 7 3s a game without sacrificing everything else? Impossible. Might be impossible to shoot 7 3s a game even with all pull up.

[Stockton12]

ive read it ..but im not the sharpest tool in the shed so ill stay out of it

[Silogical]

Big men tend to have much lower pass % and don't need to spread out in multiple areas so they can stack post up and drive / shoot as much as possible, every touch they get is a 0-10 ft shot basically.....

I mean how do you get Lonzo Ball (88 pass %) to shoot like 7 3s a game without sacrificing everything else? Impossible. Might be impossible to shoot 7 3s a game even with all pull up.

yeah i dont think we try to convert current players into 3 pt shooters. We just start creating some high volume shooters who arnt complete trash on defense.

The league basically has high volume/high efficiency/trash defense and high efficiency/low volume/good defense.

Where are the high volume/med efficiency/med-good defense? high volume shooters will compete for shots with bigmen which will lower their volume which will increase their efficiency.

[blackice]

This is extremely complicated and must have taken a very long time to write Marcos, so kudos on that front. I think efficient offense is hard to come by, particularly on the wings which puts a high emphasis on players (stars) who play inside. Ayton is a good example of this, or guards that primarily score inside. On the other end, it's hard to find true two way players who can really shoot it & defend at a high level. I've struggled to find a balance and I think guys that are pure 3&D like (for example) Herbert Jones or Dikembe Dixson are extremely valuable. My issue is moreso finding the caproom to bring these role players back on good contracts as they tend to get targeted and overpaid whereas the true superstars like Zion are max worthy from day 1 so it's hard to find a middle ground. You have to essentially hit on a superstar like Nikola Jovic on a rookie deal and then go all in and find value in your surrounding players. This is why true rebuilds really allow teams to go from bottom to top quickly whereas teams like my Raptors are sort of in a grey area (which I'm fine with as I think our defense will carry us), but that is my perspective. You and Silos seems to have a better grasp of this then I do, and I would love to hear more input so I can learn and am certainly open to new ideas.

[Marcos_Beck]

Big men tend to have much lower pass % and don't need to spread out in multiple areas so they can stack post up and drive / shoot as much as possible, every touch they get is a 0-10 ft shot basically.....

I mean how do you get Lonzo Ball (88 pass %) to shoot like 7 3s a game without sacrificing everything else? Impossible. Might be impossible to shoot 7 3s a game even with all pull up.

yeah i dont think we try to convert current players into 3 pt shooters. We just start creating some high volume shooters who arnt complete trash on defense.

The league basically has high volume/high efficiency/trash defense and high efficiency/low volume/good defense.

Where are the high volume/med efficiency/med-good defense? high volume shooters will compete for shots with bigmen which will lower their volume which will increase their efficiency.

You are one of the guys with highest engine knowledge here and all your points makes A LOT of sense overall.
Just one thing I want to point out, anything we do I believe we gotta do it for present and future. Not only future classes as teams can tank to get the enchanced future draftees and in 5 years they will just dominate the league as draft top players will be SO MUCH better than current CSL players, who will likely look like trash overall.
I mean we have around 450 players in our rosters. If we don't adjust current players as well, in 5 years we will have 250 players from future draft classes occupying roster spots in CSL. That's basically the whole class. They'll be so much stronger than current CSL players it will be unfair, with the exception of a few players like Jovic for example.

[Silogical]

You are one of the guys with highest engine knowledge here and all your points makes A LOT of sense overall.
Just one thing I want to point out, anything we do I believe we gotta do it for present and future. Not only future classes as teams can tank to get the enchanced future draftees and in 5 years they will just dominate the league as draft top players will be SO MUCH better than current CSL players, who will likely look like trash overall.
I mean we have around 450 players in our rosters. If we don't adjust current players as well, in 5 years we will have 250 players from future draft classes occupying roster spots in CSL. That's basically the whole class. They'll be so much stronger than current CSL players it will be unfair, with the exception of a few players like Jovic for example.

It is very possible to create different players that are balanced and not OP.

It would be very easy to create a high volume shooter with around 54TS% and reasonable defense. You can create a guy like this with top 5 pick strength, you can create a guy like this with top 10 pick strength and so on.

same goes if you want guards to shoot better from 0-3 feet. You can start giving guards better FG_RA and lower other atributes to balance.

The league is shaped by the players we create and we have been creating the same players over and over. We dont have this efficiency because of the engine we have it because that is the players we've been creating.

IMO we need to put more trust in the draft team and give them the ability to create players with preferences outside the maximum. So they can create even more unique players to shape the league like we feel it should be shaped. It would be very easy to create a player with 20 drive/shoot that isnt over powered.

[andrei]

My question here is: what are we trying to achieve and why?

Any change you do there will be players who shoot better and there will be players who shoot worse.

I personally really don't like the NBA right now where every other week somebody breaks either team or individual offensive record. If anything, I'd prefer to go back to 90s basketball.

When I look at the stats - 70% inside? Would you shoot 70% inside if you had a Mutombo, Hakeem, Robinson, Mourning, Ewing, Shaq, Rodman, Oakley, Davis and Davis, Garnett, Duncan etc waiting for you out there every night? The players were good and they were allowed to be physical inside on defense - now you shoot 70% because nobody can touch you. And players are brought up to pretty much just learn how to shoot and ignore defense because why bother if you can't do anything?

My point is I would not want to mimic current NBA

[Marcos_Beck]

100% agree with you Andrei
That’s why in our tests with improved ratings I didn’t want to go all the way to 57 TS%
55 TS% as a league average should be more than enough to have better offense overall (with players shooting over 40% from the field again) and defenses still highly valuable.

In my tests I think I went too high with inside scoring, mid range scoring was nice and 3pt shooting could use some slightly boosts as I agree with silo, we have very few good 3pt shooters who can play and not be a net negative defensively, so boosting 3pt shooting a bit even if our ratings are above nba average would be a nice thing because we shoot like half the volume.
And also we could improve FT shooting overall since we are like 5% below NBA’s historical numbers of around 77%

[drumr]

I think you should study the newest engine because I believe these were all of his goals in addition to making defense actually work. I have read 24 was a lot easier to make different eras of basketball.

[Rizzo]

Thanks for all the work Marcos. I have been out of the game for a while in terms of draft creation and overall testing of the engines so I trust your data. If changes need to be made, there are a lot of brilliant minds here to make those accordingly.

[wms02a]

If we are going to change engines lets wait and see how that goes first?

[Marcos_Beck]

If we are going to change engines lets wait and see how that goes first?

DDS 25 should be available November 2024 and the changes suggested here would only happen next offseason at least, which should be around the time. The thing is, DDS games are always full of bugs for a while until Gary fix those in future updates so it might take a while to use PB25 after all.

[KW]

If we are going to change engines lets wait and see how that goes first?

the idea of switching to a newer engine is new information. Marcos has been putting in a *ton* of research and testing in regards to this league's issues over the last couple seasons. I've provided some input here and there and run a few test sims, but Marcos is doing a lot of the heavy lifting here and I generally agree with all the changes he's proposing... if we plan to stay with PB22 long term. If we're moving to PB25 when it comes out (presumably in time for next season), then yes it would make sense to test these changes on that engine rather than this one.

The main issue with this league's rating setup is that the low to middle tier of the interior scoring ratings (FG_RA, FG_ITP, RA_Rate, DunkRate) are far too low, and have been since the PB3 to PB19 transition. It's why guards and wings have a really tough time getting to the 0-3 range, and force up tons of bricks in the 3-10 range, and unless your bigs are on the higher side for their position, they may often have those problems too. In addition, we'll need to balance it out with small jumps in jump shooting on the low end to middle class of shooters so that teams will be comfortable instructing players to shoot from range, since this engine doesn't seem to punish a lack of spacing nearly enough.

Speaking as a member of the draft team, you'll notice that our drafts have (for the most part) followed league percentiles with a few outliers. We are not going to suddenly put in entire drafts of players with a massive jump in those ratings, as that would upset the balance of the league. We plan to keep drafts in a similar range (or marginal increases at most) so that global changes are easy to make when they do happen.

As for Andrei's concern, nobody trying to mimic 2024 NBA basketball with defensive essentially being legislated out of the game and every team scoring 120 a night. However there is a middle ground between that and the inefficient 90s-00s ball we have in the engine right now. Some of that is the engine, but some of it is the current makeup of ratings, ball actions, and floor ranges in this league that have been issues since PB19. These are things that can be corrected with a rating update. There is a 5-7 TS% difference between the CSL and the modern NBA. If we can bridge about 2/3 of that gap, and get our league average to the 56 TS% range, I think I'd be very happy with that given the engine's limitations.

Between our commissioners, Marcos, and whatever input I'm able to give between now and then, I'm confident we'll eventually work out a system that works for everyone, isn't going to upset the balance of the league, and isn't going to be too drastic in any direction.

[mgtr81]

Since the very first scouting report I got when I joined the league years ago, I already said shooting ratings, specially inside shooting ratings, were very low. Among the answers I got were that I was living in the past, I didn't know how the engine works, etc. Nice to see some years later it seems that our shooting ratings were actually low.

[Silogical]

Speaking as a member of the draft team, you'll notice that our drafts have (for the most part) followed league percentiles with a few outliers. We are not going to suddenly put in entire drafts of players with a massive jump in those ratings, as that would upset the balance of the league. We plan to keep drafts in a similar range (or marginal increases at most) so that global changes are easy to make when they do happen.

When global changes happen the lower the rating the bigger the buff. So for example a 30 MID would go to a 35 MID and a 45 MID would go to a 46 MID. Global changes are balanced in that regard.

So if the draft team buffed certain areas slightly and nerfed certain areas slightly to balance the player a good global change would not upset the balance of the league (although the problem right now is the balance of the league is upset).

Also the biggest buff needed isnt an efficiency buff its a 3 pt volume buff. 3 pt efficiency is already too strong, so we wouldnt have to worry about the admins making a high volume 54TS% 3pt shooter a high volume 58TS% 3pt shooter.

Im also against global changes when we can fix whatever we want to fix through player creation over time, but i would help with a global change if the admins wanted help. Im not even sure what global changes anyone even wants besides FG_MID and 3-10 feet. Global changes are never fun.

Since the very first scouting report I got when I joined the league years ago, I already said shooting ratings, specially inside shooting ratings, were very low. Among the answers I got were that I was living in the past, I didn't know how the engine works, etc. Nice to see some years later it seems that our shooting ratings were actually low.

completely agree with this, and just because we didnt do it the yr we shouldve, and we didnt do it 5 yrs after the year we shouldve, doesnt mean we shouldnt start doing it now.

[KW]

Silo, we're at least partially on the same page. Yes any increases that would be made should largely benefit the low end to middle class. The top guys (for most not all ratings) do not need fixing. And they should be simple enough that a single excel formula can convert a rating.

Such as FGMID_NEW = if(FGMID_OLD<45, roundup(2/3 * FGMID_OLD + 15, 0), FGMID_OLD)

would do nothing to values 45 and up, but would adjust the scale from 0-44 to 15-44 and bump every player up but more on the low end.
25->32
31->36
34->38
38->41
42->43
And so on

That's just an example I threw together in 5 minutes but the kind of adjustments we should be making. not something that's going to make a mid range savant like Jovic or Culver even more dominant

[Silogical]

Yeah that's what id push for if we did go with a global change. We can continue to fine tune the formula until its perfect before implementing. The admins dont typically like sharing their formula so we have to just trust them.

However even if we put hours into a formula some players will still come out ahead and some behind, thats why i dont like global changes.

Changes in creation is fair. Is our FG_MID low and our DEF is high? start giving players better FG_MID and lower DEF on average. 3 pt volume low? create a few more medium efficiency high volume 3 pt shooters, and fewer low volume/high eff 3 pt shooters.