PSA: A guide for fictional draft class makers

[Barncore]

DRAFT CLASS PHILOSOPHY

Creating players via draft class and watching them go to work within 2k17's awesome gameplay might be the most satisfying thing you'll experience within the game. Also, having multiple fictional draft classes at your disposal to enhance your MyLeague/MyGM experience is imperative, but sometimes it's hard to trust somebody else's class. Well, this post intends to positively affect the collective creative we have here at OS, by offering some perspective on what exactly we're trying to achieve when using custom made draft classes over generated ones.

Who am i to speak up on the matter? Well, just a passionate basketball fan just like you. But also, I've been following the NBA religiously since i started playing NBA Live 2001. I've been extensively analyzing NBA drafts extensively since 2006. I've seen all the trends and patterns in my time. Also, I was among the first people to make fictional draft classes back in 2011 (I'm Nick at NLSC).
My aim here is to set some key universal principles/objectives for what makes a great fictional DC, in order to provide inspiration for the many talented DC makers we have here

Fictional draft classes - What are we trying to achieve? And what should we be mindful of?

1. Essentially we're trying to create players that have a specific identity. That's the ultimate purpose. I.e. Their tendencies match up with their attributes, which match up with their animations, which match up with their badges, etc etc. They come across as their own unqiue entity on the court. Suggestion: Create players with purpose in mind. Imagine the type of player you want to create BEFORE you start, write it out, THEN do the edits.

2. Have "the NBA of tomorrow" in mind when you create draft classes. Think about the overall balance of the league. What positions are in vogue and what positions are lacking at the moment? Do the opposite of that - balance it out. Think about the direction the league is heading (hint: less iso ball, less post ups, more zone defense, more floor spacing, more team ball, versatile forwards more favoured, poor foot speed less favoured, pick&roll defense ability more favoured). Also, think about what kinds of stars/identities the NBA universe hasn't seen yet. Suggestion: i don't think i've seen a 6'5" PF for a while. Nor a post-up PG. Where all da 7 foot shooters at? Get it?

3. Edit your players with both gameplay and simulation in mind. Tendencies and attributes are critically important for stat simulation. But try really hard to make their animations reflect the kind of player they are (in terms of attributes/tendencies/body), so they are fun to use in gameplay, and so that they feel right. It needs to match up. The marriage of player type to animations is the key to a great fictional draft class.

4. Compatibility. Edit your player attributes/tendencies with the official 2k roster in mind, so that they are universally compatible (unless intended otherwise obviously). Suggestion: Have a look at the attributes/tendencies of NBA players on the 2k roster with specialist abilities to get an idea of where the benchmark is for certain things. Look at guys like Dwight, Rubio, Korver, Tyreke, Draymond, Kobe, Toney Douglas lel, etc. Get a feel for things if you haven't already

5. Names. Be sure to use player names that are in the 2k audio bank already, so then the commentator says their name when they receive the ball, and the PA announcer says their name when they score. This is pretty easy to do, as you can browse them yourself. Also, be sure to make the player name match up with their chosen nationality (goes without saying). Try to come up with unique/fresh sounding names. Think about how in the NBA, a lot of star players have unique first names that eventually become synonymous with their identity. Kobe, Kawhi, LeBron, Kyrie, Kemba, Klay, etc. Aim for that with half of your star players. (Same goes for last names. Harden, Lillard, DeRozan. You've probably never heard the word before seeing their name before.) If it's a future star, let the player have at least one unique first or last name. Some stars can have boring names too (Dwight Howard lol), that's fine, as long as it's something that can become synonymous with their identity.
Suggestion: Stuck for names? Try this fake name generator. It's pretty cool because it comes up with a name, location, birthdate, favourite colour lol, and a bunch of numbers that could be used somehow. And you can set the gender, nationality, and age range before you click generate.

6. Different colleges have different cultures/identities, and therefore recruit/produce different kinds of players. It's not hugely important, but it's something to look out for. Here's my observations so far: Arizona is athleticism/conditioning focused, great athletes come from there. Michigan State tends to be high on intangibles. Duke tends be about polished skillsets, particularly offensively. Gonzaga and Creighton are full of white role players. Wisconsin too. Kentucky for 5-star level talents, dominant physical traits or transcendent skill levels. North Carolina and Kansas for 4-star recruits. Marquette for mentally tough guys. Connecticut and Georgetown for fundamentally sound wingers. Washington for raw athletes. UCLA for "flavour of the month" type prospects. Small schools for underrated/undersized guards. Louisville for 3pt shooting in guards and defense in bigs. Florida for shooting. Anyway you get the point. I'm sure others can add to this.
You don't have to stick to this by any means, any player could go to any college. This is just a rough guide.

7. Particular tendencies to be mindful of:
--Shot (this affects their FGA / PPG a lot in sim. A player could have high offensive attributes but if his Shot and Touches tendency are low he will score under 10 ppg)
--Touches (so does this, and also determines how often plays are run for them in gameplay. Very important!)
--Roll Vs Pop (100 is roll, 0 is pop. Important for non-guards as it determines whether they roll to the basket or fade to the 3pt line after setting screen. Balance as you see fit)
--Foul (determines foul trouble probability!)
--Block shot, pass interception, on-ball steal (probably has just as much, if not more, affect on block and steal rates in simulation than the actual attributes do)
--Dish to open man (this is the most problematic tendency with generated classes, as all generated non-PG's tend to get really low numbers for this (e.g. between 0 and 15) which results in low APG numbers that ALL tend to be 0.9apg or lower. This eventually creates misbalance in the league after 5+ seasons of simming where only PG's are the ones with high assist numbers and everyone else has below 1apg, which ruins suspension of disbelief. So this must be paid attention to. My rule of thumb with this is that it should be over 20 no matter what. Even Hassan Whiteside has a 26 for this. Most standard non-playmaking bigs can be around 20-40. A versatile big/forward that is more orientated towards ball movement would have between 50 and 65. Most average SG/SFs will be around 50. Most PG's will be above 70. PG's who are specialist playmakers (i.e. Rondo, Rubio) are 90+. You get the idea.

7B. The tendencies listed above are of particular importance, but i urge draft class makers not to neglect the rest of the player tendencies overall. Be it fictional or real. They are SO important to both gameplay and simulated stats. Classes are not at reliable quality without edited tendencies. Bottom line. Suggestion: Edit tendencies before editing attributes.

8. Play types. One of the awesome features of NBA2K is that plays are run depending on a combination of the available plays in the team's playbook, the type of "playtypes" a player has under his name, and the touches tendency. In other words, the playtype determines what role(s) the player has within the offensive scheme! To me that is heaps of fun to play with. Make sure the play types match up with the player abilities: Suggestion: Edit playtypes last! After attributes and tendencies etc, then you have a better idea of how a coach would use the player

9. Potential & peak. For those that don't know, the potential rating is basically the ceiling of where their overall rating could end up in their peak. So when you're setting somebody's potential to 85, you're saying he's gonna develop into an 85 overall player. Injuries can sometimes derail somebody reaching their full potential, keep that in mind. Peak starts & end are super important. Think of it this way: 'Peak start' defines how quickly a player will break out. If you set it late, he will gradually progress slowly. If you set it early, he'll go BANG BANG and make big strides in his first few years. When you set peak starts & ends, take into account the kind of player they are. As a general rule of thumb, i generally make "players who rely on their athleticism" have a peak end around 30 or 31 (think Dwight). Remember, the peak end is when their ratings START to decline. So really when you set the peak end at 31 you're saying that from 31-34 he will slowly go down in ratings. For players who use their skills more than their body (like shooters, playmakers), i make their peak end later than others. I.e. say a 3pt specialist doesn't start declining until 33. For the transcendent stars, i let their peak end be late too.
For peak starts, i think about the kind of player they are, and their age. As a general rule of thumb, centers and point guards take longer to develop than wings and forwards do, for some reason. So i might have a SF/SG peak start at 23 or 24. But i might have a PG or C be around 25 or 26. However, it also depends on what age they're drafted. I wouldn't give a freshman PG a peak start of 26, cos that's 7 years of development which seems too long to me. I'd probably go 24 or 25 in that case. But i might give a senior PG a peak start of 26. Same goes for centers. My general rule of thumb for most players is to develop until 25 and then start declining at around 31. I vary it based on body type, position, their age when drafted, and potential.

9B. Sensical trajectory: Age vs current ability vs potential ability. Think about what class of college they're in, and let that decide how you rate them. E.g. Most freshman should have lesser current ability than seniors, unless the freshman is transcendentally talented. And most seniors will probably have less upside than freshman (although there are definitely exceptions to the rule). It probably doesn't make much sense that a freshman would be rated 78 overall or something and then have a potential rating of 79. If a 19 yr old is that good already then that means he is a transcendental prospect that is ahead of the development curve. It would be silly to think he wouldn't develop at all in his next 4 years.
Also, usually there's a reason why seniors stuck in college so long, and it's usually because their upside was considered limited by NBA standards. NBA prospects generally stay in college in order to develop their game and prove their worth to the NBA. So think about giving your seniors less athleticism or height, but more polish & mental ability, i.e. more readiness. More badges, perhaps. For sophomores and juniors, anything goes. No rules. It varies so much

10. Looks & gear. Not really my field, but my general rule of thumb here is just to give everybody their own unique look as much as possible. The less generic the better. The more varied identities the better. Suggestion: Stop giving everybody cornrows lol. Nobody has that look anymore.

As we know, we can't edit players in MyGM mode, which makes downloadable fictional DC's an essential part of the NBA landscape experience, because if we were to edit every single DC ourselves then we wouldn't have time for the actual game! Luckily we have a great community here and we have accumulated a decent amount of passionate fictional DC makers.

Please speak up if you have anything to add, or change/discuss, and i'll add it to the list.

It's early days, so much potential for awesome DC's ahead! Just make sure you create them for PS4, not Xbox

Cheers,
Barn

[Barncore]

Another quick note on potential, this is what Leftos said about potential a year ago in regards to 2k16. I assume it's the same or similar for 2k17.

"Potential is supposed to represent an approximation of the ceiling of a player, be it a future ceiling or a past ceiling. He can reach it by peak start, maybe a few years after that, maybe even surpass it by a couple of points. Injuries, lack of playing time, bad training staff, can all hinder that progress.

Potential is adjusted every year. Before peak age, after offseason progression, we look at whether the player is still on track to hit his potential by the time he hits his peak. If he didn't progress enough, we adjust potential down to reflect that. If he progressed to the maximum capacity, his potential stays the same. If a player hits his potential early, there's a probability that he'll go a point or over, and potential will be adjusted up as well. By the time a player hits peak start his potential should be the same as his overall. For players that regress (between peak start and peak end in a very minor way and then in a major way after peak end), potential reflects the max historical overall the player had.

So, potential is always at least equal to the overall rating."

[pierralban]

Hey Barnsey (well you'll always stay Nick to me, i started making draft classes thanks to your Nictional Draft Classes), that's a great post, might want to go back myself on FDCs, i'm on PS4 now.

If you set the peak start to early, won't the player ever reach his full potential ?
Like if i have Player A being 19 with a potential of 90, and i set his peak start at 23, i'm not sure he'll have enough time to develop into a 90 player. What's your take on that ?

[Barncore]

Quote: Originally Posted by pierralban

Hey Barnsey (well you'll always stay Nick to me, i started making draft classes thanks to your Nictional Draft Classes), that's a great post, might want to go back myself on FDCs, i'm on PS4 now. If you set the peak start to early, won't the player ever reach his full potential ? Like if i have Player A being 19 with a potential of 90, and i set his peak start at 23, i'm not sure he'll have enough time to develop into a 90 player. What's your take on that ?

I remember you! Good to see ya

I don't think that's how it works. From what i can tell, a player will get to his potetial as best he can. Say for example a player is rated 70 and need to get to 85 in 3 seasons, then he will jump up b 6-7 pts each season to get there.

I could be wrong though, i haven't tested 2k17 yet, i'm just assuming it's the same as 2k16

[CWSapp757]

@pierralban After test that I have done last year (and a few this year), the answer to your question depends on the progression slider. This year, if you keep the slider at 50, they will not jump very much in overall. Therefore if that Player A you were referring to has an overall of a 70, he will more than likely not reach that potential by the age of 23 unless you make his overall closer to his potential or up the progression slider. I think we all need to test that slider some more and figure out the best place for it.

Also, great post Barnsey. I agree with everything except using names in the 2k database. For me it limits the different names you can use too much. After three or four classes it becomes an issue for me. Of course, I usually play in silence since I'm generally watching a movie or TV while playing so it doesn't bother me nearly as much. Just a personal preference of mine.

[YogurtProducer]

This is honestly perfect. It would be amazing if this community could manage to
make 80 fictional draft classes (or more) that are at least useable and not broken.

I plan on trying at some point soon to pump out some that might be bare but at least playable. We need some real clarification on the peak start/end thing as I would like to take advantage of creating busts and more variables in play.

When you talk about considering the future of the NBA, what kind of potential levels are you talking about? Like if you were to make say; 10 draft classes. How many guys would you make 90+OVR, 80+, high 70s, etc.

Also how high would you make each players base OVR? I was considering most 1st rounders should be 69+OVR and then we could utilize the peak start feature and different potential ratings to get better variance. For the rest of the 2nd rounders I was thinking 65+ because how often does 2k even rate a guy below 65? I think Bruno Caboclo is the lowest on the game other than some random FAs.

I know it's a lot to ask for guys but even if people have a base template available for multiple years it would help a ton.

[MarvellousOne]

Great post Barnsley, very good insight. I think this year is going to be another fun year for making draft classes, I'm already getting a jump on my DC for this year. Loving the fact we can edit the class throughout you ML season.
Hoping to make much better classes this year for ps4 community.

[capt3031]

Quote: Originally Posted by MarvellousOne

Great post Barnsley, very good insight. I think this year is going to be another fun year for making draft classes, I'm already getting a jump on my DC for this year. Loving the fact we can edit the class throughout you ML season. Hoping to make much better classes this year for ps4 community.

Yeah a couple things I notice real quick was that 2k gives u an option to randomize a class and not automatically do it which bother me a lot last year and say if u upload one class in the beginning of the year u can change it throughout the year which is kinda cool 2