OSUPiper's CH2K8 CPUvCPU Experience

[OSUPiper]

Quote: Originally Posted by Brock

Just a suggestion, I think if you wanted role players you could just drop tendencies of non offensive players by like 10-15. 3pt specialists, drop their close/med tendencies.

Probably so. I know from past experience that you can create a Garbage guy with strong rebounding and low shot tendencies. That dude only eats what he can kill by himself.


I think you can create a 3pt guy just by lowering his Drive tendency. That would be interesting to try out, just to see if the game would really have him floating around on the perimeter on offense.


For lockdown defenders, I am wondering if simply moving points from offensive awr to def awr would do it. I find that Off Awr affects shot frequency more than the Coach's Tempo. This is why Power 5 teams never run at as slow a pace as small schools, even high tempo small schools.

[Brock]

Quote: Originally Posted by OSUPiper

Probably so. I know from past experience that you can create a Garbage guy with strong rebounding and low shot tendencies. That dude only eats what he can kill by himself. I think you can create a 3pt guy just by lowering his Drive tendency. That would be interesting to try out, just to see if the game would really have him floating around on the perimeter on offense. For lockdown defenders, I am wondering if simply moving points from offensive awr to def awr would do it. I find that Off Awr affects shot frequency more than the Coach's Tempo. This is why Power 5 teams never run at as slow a pace as small schools, even high tempo small schools.

From what I've seen, 2k actually tried to make one-role types of players, but still fall into the "good all around but not great at one thing". If you looked at a player and they sided more on the def side and high def awareness you could just boost their on ball def.

I've seen this with a few hsaa. One I just recently edited has an extremely high on ball def (at 97) and his def skills are above average, I'll admit not great, and I think his offense was average/solid.

[OSUPiper]

Slight changes made to sliders (see Original Post).

[OSUPiper]

Just watched an interesting matchup with my smallish but very talented Duke team against Rutgers, whose Euro big man might go first in the draft. He was averaging about 15 FGAs p/gm in only about 25 mins.


He got into foul trouble against my athletic forwards, and we played great D at the post, but he ended up taking 9 shots. When he got position, he was unstoppable, so I liked that. He just should have been more of a threat.


He is a skilled passer, and he did give the ball up for good outside looks, but there were times when he had my PF 1v1 and kicked out.


As a result, I have substantially increased CLOSE tendency. I will watch the effect and advise in this thread. My fear is that I already had too many FGAs underneath from the PG, SG, and SF spots (generally, second chance shots) that I felt should be passed back out. My original settings achieved that result, but I think now that I am hamstringing post-centric teams.


If anyone has seen similar issues, please post. It will be hard to replicate this issue since good offensive big men are rare, but I plan to watch this change play out. If it works, I will post suggested changes to the sliders, probably with additional editing to rosters.


ONE LAST NOTE: In this same game, my PF (who starts at C) attacked the basket and scored above his season average. He has a better mid-range game than most PFs and Cs. So, this may be an isolated issue, but I want to test my theory that the Shaq-range post players are underserved at this time. (I realize that the Rutgers C would not play like Shaq, but his ratings suggest that his range is pretty much inside 10 feet.)

[jww]

i have never been able to replicate a consistent low post presence in the game. when i have increased close tendency, it seems to mostly increase backdoor cuts. the only way i have been able to increase low post is to create a playbook that emphasizes low post plays. that combined with a high close tendency player attribute seems to result in higher low post presence.
i have a question about your sliders. i think that they make the game look great, have run many games in quick game. shooting percentages are consistently low, and i am wondering if these even out over a season when in legacy mode. i am using dartmouth, and i have simmed a season, and want to achieve in game stats that resemble season stats before starting a career legacy.
great work on these sliders though, i think they get as many things right as is possible with this game engine.

[OSUPiper]

jww:


Thanks for the input and kind words.


First, I agree with you on the low post issue. I see exactly the same thing: a greater increase in backdoor attacks over the expected increase in post offense. I have been able to emphasize a low post game on teams where the offense has to go through the PF/C, i.e., when the post player has higher shot tendencies and wings have lower shot tendencies. My theory is that this is a by-product of who gets the most 'touches' in the offense.


NBA2K has a slider for this which is very useful in fine-tuning the issue. Unfortunately, CH2k8 preceded that slider. So far, however, I like the result of bumping the CLOSE slider.


Next, I cannot comment on the sliders in Quick Game, but I can say that FG% will be lower early in a season, and it tends to creep up during the season. I believe this has to do with player progression generally favoring offensive ratings. For instance, not only does player awareness increase, but so do at least five separate ratings all linked directly to FG%. Add in Vertical and Speed (almost neutral in application between offense and defense) and Confidence, and the impact is skewed toward offensive stats.


The second issue re: FG% is player ratings. I routinely find (at small schools) low-rated players that tend to shoot 3s despite a 53 3Pt rating. That is ridiculous. That guy does not get a scholarship to play Division 1... D1 small schools often have effective 3 point shooters. Small schools lose ground on big schools on (1) quality plus size/height and (2) basketball IQ.


As an example, a 3pt sharp shooter at Kansas is going to be a better shooter than his counterpart at Brown. The kid from KU might win a game of HORSE 7 or 8 out of 10 times even. But the advantage the KU kid has is that he can rise up and knock down that 3 with a 6'7" wing flying toward him, or that KU-Guy knows how to space himself on the floor, or even that KU has three other guys on the floor that the defense is worried about so this guy gets more open looks.


CH2k8 generates a Freshman that's not just a D overall but a 'D' across the board. Some players are consistent across the board; some great players are really good across the board, and some are great at just a few things. But an undersized kid with average quickness who can stroke the 3 can find a scholly at a small school, while a kid that is below average across the board just is not getting that free ride anywhere, in my opinion.


I polished my sliders through the last 10 seasons at Duke, although I coached at schools at each level. My purpose in doing so is that within 5-6 years of most dynasties, I end up at a major program. Thus, it is most important for sliders to be fair and fun at that level. Having said that, aside from the issues above, I have found that the sliders translate well to each level against similar competition. Moreover, I am happy with the individual stats and the challenge of placing guys on the All-Conference and AA teams.


I am always looking for new ideas to try, with the goal to reduce player edits.

[OSUPiper]

So, in watching and analizing the post-offense issues, described above, I noticed that SGs and SFs take more shots than they should, even accounting for tendencies. (This is partly a game engine issue, as has been discussed in other threads.) Further, it appears that these are usually 3pt attempts.

My theory is that these ‘extra’ attempts are the result of passes back out from the post (e.g., cross pass or kick out from a double team). The effect is FGAs migrating artificially from the post to the wing.

Now, cross passes need to happen; it is a very normal and effective play for open 3s. It just happens too easily and too often with my sliders. I believe this is because of my Help Defense and Help Recovery settings.

I settled in on the present sliders for Help D and Recovery to simulate defensive movement, create strategic give and take on defensive pressure and Zone vs. Man, and reward good ball movement (a la San Antonio Spurs). I am hoping that, by reducing the frequency of Help D and boosting Help Recovery slightly, the lane will clear out more. This will (1) create more opportunities for dribble-drives and (2) create more one-on-one opportunities in the post. A post scorer should elect to take it to the hoop when help defense is late getting there in most circumstances.

Preliminary results are positive. For one thing, the game opens up even more, which means more animations, more varied play and playmakers, and frankly, more fun to watch. My team literally had a ton more personality. I also saw several pull-up jumpers at mid range, which has been rare with any slider set I have seen.

I watched a team with zero presence in the post that relied on their PG and SG for about 1/3 of their shots and over half their scoring. The SG averaged 10+ FGAs p/gm and 3.5 3pFGAs p/gm. He ended with 11 FGAs and 5 3pFGAs. The PG got into foul trouble and was ineffective, so no help there.


Interestingly, the C took 11 shots... I think this is a by-product of offensive rebounds and putbacks, but it is an issue to watch. Hopefully, though, this will trend in the same way, which would let me drop the CLOSE tendency and reduce the wings cutting to the goal so often. This particular Center is a nice player, former 5-star juco PF playing out of position on a decent but undersized Power Conference team. It is not unrealistic that he had these stats, but his tendencies do not support those stats night-in, night-out. MID tendency may be slightly high.



As expected, FG% dropped. This is because there are more contested shots on the perimeter and more shots in traffic (because a player commits to driving the lane before help arrives). I will not mess with FG% sliders until this gameplay issue is worked out.

For testing, I now have HELP D = 31, and HELP REC = 45.

[OSUPiper]

Also, I should add that I made the following changes for testing purposes, relative to the Help D / Help Recovery sliders:


Close tend 16
Mid tend 62
3pt tend 19
Drive 64

These changes are for testing only; I have not changed the slider set on Page 1 at this time, although so far things have improved. I want to make this clear that these changes, despite seeming slight, may have a pronounced ripple effect on shot distribution and FG%. As such, I DO NOT recommend taking the Page 1 set and just adding these changes as a long term legacy solution.