Dynamic Difficulty

[dukebeatsuncagain]

NBA2k games, in my opinion have always not had proper team-to-team variation in skill, especially when scaling up the difficulty. Games like FIFA and MLB: The Show have this down. Playing bad teams feels completely different from playing good and elite teams, but even the bad teams can surprise you every now and then if you don't execute.

I always felt this wasn't exactly the case for NBA2k. Obviously this all depends on the player, but for me personally, on Pro I could torch the 96 Bulls with the Bobcats. That should never happen. On HOF, the Bobcats could torch me as the 96 Bulls. Also should never happen. Maaaaayyybe once in a bunch of games, but we are talking about the greatest team of all time in that example.

To make a long story short, I am going to start assigning a difficulty level based on the team I am playing's overall team rating. Throughout a season, injuries, fatigue, and trades cause a teams overall to fluctuate, so when you play a team in November, it could have a drastically different feel when you play them again in March depending on who they have on the court that night.

Alright. I have nailed down what I am going to run with for the next couple of months. I did a lot of thinking, at work...naturally, and decided to go in a relatively different path than what I was originally thinking. This is what I'm working with...

New Association w/ post-trade deadline roster
-Full seasons + games
-Simulation Game Style

Rules:

Regular Season Dynamic Difficulty Power Ranking

#1-5 - Hall Of Fame
#6-15 - Superstar
#16-24 - All Star
#25-30 - Pro

Regular Season Dynamic Difficulty Seeding

#1 - Hall Of Fame
#2-4 - Superstar
#5-7 - All Star
#8 - Pro

Online Integration:

Any game on the schedule can be replaced by a quick-match game online. After a win or loss, trial and error simulate in the appropriate outcome.

This can break up the monotony of an 82 game season. I personally go through online spurts, but in the end it all comes down to offline. This is just a little something to add some spice to the season, and make those online matches count towards something.

Simulation Of Games:

User has the ability to simulate Regular Season games if the team they are playing is -2 difficulty in the current week's power rankings.

I never know when I should sim, this in a way solves that. Really, the only way you are even allowed to sim games is if you earn it. Getting to the top of the league means faster progression through the year. It is a risk and reward decision however. If you sim a game against a Pro team, you might lose. You have a much greater chance of winning if you played it out.


So that's the idea. Not re-inventing the wheel, but for me it definitely makes a difference. When I have Milwaukee come into my building it feels insanely different than when I travel to Miami. To beat the bad teams, you have to play to your strengths, execute, and get the win. But when you play an elite team, you will have to play damn near perfect, along with a couple good calls, to get the win.

[cch99]

Re: Dynamic Difficulty

This is actually a really good idea. I struggle with finding that spot of getting a good difficulty level, where i usually beat teams I should and lose to teams I should. I might use this to a degree but my problem is if I played a 90+ team on HOF with the 96 Bulls I would still get smoked all the time. I think each person would have to find their sweet spot of difficulty. So if you get good games between even teams on all pro for example, then play teams better than you on Superstar, and teams worse than you on Pro.

I think this is a really difficult thing to program because you have to almost program it into the game that teams will take nights off against poor teams so you have upsets now and again.

I also think they must build something into the sim engine to get bad teams to win sometimes but they end up winning more than they should and you end up seeing some teams that are really good having mediocre records and some poor teams make the playoffs. One of the small things that has always bugged me about the 2k series is that when the sim engine ends up giving a bad team a good record, that's fine but if i have a bad team I will often beat the "better team" because they aren't really good it's just the sim engine somehow gets them to keep winning.

It might be something where you go by record also to account for teams playing above there level. So lets say my even teams level is all-pro. I am the Sixers Playing Milwaukee which would normally be All-Pro since they are both equally bad. But the sim engine has Milwaukee playing Winning Basketball, I then play them on Superstar to account for them outplaying their ratings in my current Association.

Just trying to jump onto this and add. I really like this basic concept.

[dukebeatsuncagain]

Re: Dynamic Difficulty

Not being able to find a good difficulty is exactly what spawned this idea for me. So frustrating bouncing around.

All your points are great. I dumb'ed it down to the ORV intervals to keep it as confusion-free and 'black and white' as I could, but did have another idea. Kind of what you were saying with record based, I was thinking about the Power Rankings as well as using the records. So like you were saying, teams who are hot and playing above their OVR rating, at that time, will perform up to how they have been.

As far as elite teams still smoking you on HOF, when you are playing with an equally elite team, that is still something I am trying to get a better hold on. I have some ideas, but still trying to hammer something out.

Keep and/all suggestions and ideas coming.

[cch99]

Re: Dynamic Difficulty

Going by Power Ranking is a great idea so there is no guessing That way you can do a set amount for example, anyone say 8 spots are more above you you jump up a level, 8 spots or more below and you jump down.

I'm not sure the best number to do, it might have to be more or less. If you have a team like Miami and are Ranked #1 if it's only 8 spots you will end up playing almost everyone down a level and probably blowing teams out all the time. Even at #20 you would still be playing 2 teams down a level and you aren't even a playoff team.

You have to kind of figure out where that cutoff is that teams are noticeably better and worse. You might just have to do it by personal feel.

For me personally i played a game last night, Portland vs LA Lakers in Portland a random game in my association on All-Star which is my even level. LA was 0-5 and Portland was 2-3 but there Ratings were LA #19 and Portland #3. That's not Power Rankings it was there Pre-Game Ranking but for sake of argument let's say that it was PR. I ended up winning by 3 and the Lakers missed a buzzer beater and I had to fight back from 7 down in the 4th to get the win. I felt like, and this is a consistent feel when i play random games, that the level kept it artificially close. If it was a one time thing I would have thought it was great, LA played better than expected and kept it close but it's a normal thing where on All-Star teams that I should be better than consistently keep it close.

Which brings up another thought, what about Home and Away. You could make the numbers in the Power Rankings based on a Home or Away game. Maybe the difference is on a sliding scale for home and away. Using the Above Lakers example. If Home was +/-8 and away was +/-16(playoff difference), In Portland I would have played the game on Pro but in LA on All-Star.

I know it's getting more complicated but anyone could take what they wanted and throw out what they didn't to keep it simpler.

[dukebeatsuncagain]

Re: Dynamic Difficulty

The home and away difference, I think, would be a big enough of an in game difference, I don't think you would need to go that far in depth. I do find winning road games to be a little harder, especially on the higher difficulties.

Maybe Pro should come into play for the very bottom of the barrel. You have the ability to really give it to em'. I was also thinking that since generally playing on HOF is damn near impossible, maybe only have 1 or 2 teams have that level attached to them, at the top of the PR, the bottom 5-8 be at Pro, then the rest split up between All-Star and Superstar.

This could also carry into the playoffs and seeding.

#1 - HOF
#2-4 - Superstar
#5-7 - All-Star
#8 - Pro

Gives huge incentive to get the #1 seed because then the highest difficulty you would have to play is Superstar, except in the Finals. And being the #8 seed spells doom. Squeak by in the regular season and run into a brick wall in the 1st round.

Another thing to point out is that I play with Real Player FG% setting for shooting. I think this is the only way to be able to also keep this realistic. On lower levels the timing window is so big it is too easy, and on the higher levels you can't buy a bucket. With Real FG%, you get a good steady(ish) shooting night to night.

[Chocolate Chip]

Re: Dynamic Difficulty

what sliders are u using?
SIM
Default
Custom

the cpu get a hidden boost get of +5 to +40 to their attribute from all star to HOF.

which is why

On HOF, the Bobcats could torch me as the 96 Bulls

.

[bo.jangles344]

Re: Dynamic Difficulty

On all-star in my career which is the lowest possible difficulty in the mode, every opposing player has a +5 to all their attributes. This makes 60s into 70s and 70s to 80s and so on. That's why Jeff Green lit me up for 25 and Chris Johnson had 18.

This, to me, is a cheap way to make the game harder. I'd much rather play blacktop than career because of this. At least I know I'm going to get a variance of skill level online.

[stublerne]

Re: Dynamic Difficulty

I struggle with finding that spot of getting a good difficulty level

[dukebeatsuncagain]

Re: Dynamic Difficulty

On all-star in my career which is the lowest possible difficulty in the mode, every opposing player has a +5 to all their attributes. This makes 60s into 70s and 70s to 80s and so on. That's why Jeff Green lit me up for 25 and Chris Johnson had 18.

This, to me, is a cheap way to make the game harder. I'd much rather play blacktop than career because of this. At least I know I'm going to get a variance of skill level online.

I'm with you there. I hate that that is their approach to the difficulty, but it it what it is. I have had such a hard time finding the difficulty sweet spot, this is an effort to end all that. Hopefully.

So far so good.

As far as other settings I am using...

-Simulation game style
-Zero defense assist

[dukebeatsuncagain]

Re: Dynamic Difficulty

This really all comes down to player skill/preference. For me, I can beat the CPU on HOF using realistic play, just not very often. Out of 6 test games on HOF I went 2-4. Considering that, I only made the top 5 teams in the power ranking.

[Coach2K]

Re: Dynamic Difficulty

NBA2k games, in my opinion have always not had proper team-to-team variation in skill, especially when scaling up the difficulty. Games like FIFA and MLB: The Show have this down. Playing bad teams feels completely different from playing good and elite teams, but even the bad teams can surprise you every now and then if you don't execute.

I always felt this wasn't exactly the case for NBA2k. Obviously this all depends on the player, but for me personally, on Pro I could torch the 96 Bulls with the Bobcats. That should never happen. On HOF, the Bobcats could torch me as the 96 Bulls. Also should never happen. Maaaaayyybe once in a bunch of games, but we are talking about the greatest team of all time in that example.

To make a long story short, I am going to start assigning a difficulty level based on the team I am playing's overall team rating. Throughout a season, injuries, fatigue, and trades cause a teams overall to fluctuate, so when you play a team in November, it could have a drastically different feel when you play them again in March depending on who they have on the court that night.

Here is my 'chart thingy' showing my basic idea. Nothing too fancy or technical. Pretty straight forward...

Hall Of Fame - 90s
Superstar - 80s
All-Star - 70s
Pro - 60s

I usually play at the minimum of All-Star, so I made that the 'lowest' group. You never see a team drop into the 60s for overall, but if a bad team is demolished by injuries, or if they can't re-sign anybody...it could happen. And if it does, I will beat them by 30.

So that's the idea. Not re-inventing the wheel, but for me it definitely makes a difference. When I have Milwaukee come into my building it feels insanely different than when I travel to Miami. To beat the bad teams, you have to play to your strengths, execute, and get the win. But when you play an elite team, you will have to play damn near perfect, along with a couple good calls, to get the win.

I am not sure if it's because I haven't had enough practice yet but in past years, it seemed easier to to dominate lower rated teams. In 2K14, I am not as sure yet. So I can't say it's a fact.

On the surface this seems like a good idea and I thought that it was pretty creative but if you are going to do something like this, I would probably take a slightly different view.

First off, in the real NBA, bad teams beat the good ones all of the time. The reason is that the bad NBA teams don't know how to bring the effort required to beat teams night in. But bring in the elite teams and suddenly they bring the energy and focus necessary to win. It's not everyday of course, but if they do, the next night, they face a bad team and often lose.

Let's take the Jazz this year. They are currently 19-35. Yet they have beat Miami, OKC, Houston and Phoenix as an example.

So, as odd as it might be, if you are going to do it, I'd almost suggest that you keep the difficulty high on the low teams to represent the effort they would bring when you come into town and lower it on the really good teams.

I believe that the game takes into account home and away and wouldn't adjust difficulty for that.

If you are going to rate teams, I would rate them a little differently than the power rankings. I would take into account the number of 90+, 80+, and 70+ rated guys on the team like in the chart below and I'd lump them into categories of how good you would have to be to win with them like I did with this chart to begin the year for my hall of fame challenge not necessarily go down to rookie or pro levels. This chart was for a different purpose.



I think it's a creative idea and maybe workable if that's the way you want to handle it.

[dukebeatsuncagain]

Re: Dynamic Difficulty

Alright. I have nailed down what I am going to run with for the next couple of months. I did a lot of thinking, at work...naturally, and decided to go in a relatively different path than what I was originally thinking. This is what I'm working with...

New Association w/ post-trade deadline roster
-Full seasons + games
-Simulation Game Style

Rules:

Regular Season Dynamic Difficulty Power Ranking

#1-5 - Hall Of Fame
#6-15 - Superstar
#16-24 - All Star
#25-30 - Pro

Regular Season Dynamic Difficulty Seeding

#1 - Hall Of Fame
#2-4 - Superstar
#5-7 - All Star
#8 - Pro

Online Integration:

Any game on the schedule can be replaced by a quick-match game online. After a win or loss, trial and error simulate in the appropriate outcome.

This can break up the monotony of an 82 game season. I personally go through online spurts, but in the end it all comes down to offline. This is just a little something to add some spice to the season, and make those online matches count towards something.

Simulation Of Games:

User has the ability to simulate Regular Season games if the team they are playing is -2 difficulty in the current week's power rankings.

I never know when I should sim, this in a way solves that. Really, the only way you are even allowed to sim games is if you earn it. Getting to the top of the league means faster progression through the year. It is a risk and reward decision however. If you sim a game against a Pro team, you might lose. You have a much greater chance of winning if you played it out.


I have a week off of work starting at 5 pm, and I am going to go all in on a new Association with the New York Knicks using these guidelines. Pumped!

[yjdykyjen]

Re: Dynamic Difficulty

This is actually a really good idea.

[yjdykyjen]

Re: Dynamic Difficulty

This is actually a really good idea.

[fyukfyuken]

Re: Dynamic Difficulty

This is actually a really good idea.