Stats

[Fenix]

Was anywhere mentioned that this game will a have a new statistics calculation engine? I want to have Rodman type of rebounding numbers (99 offensive/defensive rebounding), I want a 35 points per from the ultimate scorers (99 for offensive awareness), 12 assists per game from Kidd type of PGs in a run-and-gun systems. Plus, it sucked this year that you couldn't view players per game stats in the 'view player' screen. And career numbers! Why is this so hard to put into the game?

[AlwaysWhat,NeverWhy]

I want to have Rodman type of rebounding numbers (99 offensive/defensive rebounding)

Given the fact that few if no active players can actually post Rodamn-esqu rebounding numbers, I think that's not a good idea.

I want a 35 points per from the ultimate scorers

Other than AI and Kobe , few can currently go above 30ppg. The 35ppg plateau is unreachable, with MJ and Wilt both out of teh league.

12 assists per game from Kidd type of PGs in a run-and-gun systems.

I agree there. Having the assists leader average a paltry 7.5 sucked in Live 2005....

[stockton12]

maybe they through stockton was the only one who could do that in assits:D . i do beleive it needs bumping up

[BOSS]

And they should trim down how may rebounds an SF gets. McGrady gets like 10 a game in my dynasty

And my bad to be off topic, they should add a like season high stats and stuff like that.

[beau_boy04]

I agree McGrady gets about 10 rpg and so does Paul Pierce. Al Harrington averaged 20/10 in the Hawks and that's not very smart because he hasn't been able to do so in real life but the game thinks he should averaged 20/10 because he's the main guy in the Hawks and everybody else in there pretty much sucks.

[Andrew]

My understanding of the stats engine is that the highest ratings will result in simulated statistics around the league leading number from the previous season. Thus, Steve Nash with his ratings should average at least around 12 assists in NBA Live 06.

[beau_boy04]

Well Jason Kidd will averaged at least 10 apg for sure with a healthy Jefferson, Vince on the wing, a much experienced Kristic, and if Rahim is in fact joinin them, I see Jason beating Nash out in the assits stats.

[Andrew]

Assuming the stats engine will in fact allow players to average up to around 13 assists per game, it should also be a possibility in NBA Live 06.

[Fenix]

Given the fact that few if no active players can actually post Rodamn-esqu rebounding numbers, I think that's not a good idea.

I think it would be possible of Reggie Evans to post 15+ rebounds per game, given enough minutes. Look at Kevin Garnett (39 mpg). He is a defensive rebounder extraordinaire (10.5 rpg), but not at the same level on the offensive end (3.0 rpg). Lets give him 95 rating for his defensive and around 70 for his offensive rebounding. 95 for the offensive rebounds would be Ben Wallace (3.9 rpg - 37 mpg). If you now edited KG and gave him 99 rating on the both ends, he ould be capable of grabbing around 11 defensive rebounds+4 offensive -> 15 rpg. A stronger, bigger player could grab 16-17 rpg with the same ratings.

Other than AI and Kobe , few can currently go above 30ppg. The 35ppg plateau is unreachable, with MJ and Wilt both out of teh league.

So you wouldn't give them 99 for their offensive awareness. A seven footer with a body of a young Shaq, scoring mentality of Wilt and skills of Olajuwon would be capable of giving you 35+ on a mediocre team.

And off coure - 6th man! I want Jamison type 6th men, who can give you 15 off the bench.

[beau_boy04]

Last season, on my Nba Live Dynasty Mode, K Korver averaged 12 ppg and became the 6th Man of the Year! however I didn't see B Gordon averaging his 14-15 ppg, so on and so forth.

[BrotherJRB]

The simulation stats engine is bad....no calculations for bench players and minutes....for instance a guy like Stromile Swift who averaged(I didn't look these up only guessing) 9ppg 5reb and 1 block in 26min would start and play 38 min and average the same everything in NBA Live. They used to have a thing called Dstats which determined simulated stats....which provided the ceiling. However in recent years I think it changed but the same problem still exists of players being "capped" by something predetermined.

I certainly hope more variables occur in the future of all sports games. Those spectacular seasons are often judged by these changes in stats.

I think Iverson could easily average 40 a game if Philly and the NBA adopted the run and gun style of old....which will make a comeback mark my words....on Doug Moe's perm full court basketball will reign again

[Kemp]

Hmm... the stats don't seems to be realistic enough.

[Andrew]

They used to have a thing called Dstats which determined simulated stats....which provided the ceiling. However in recent years I think it changed but the same problem still exists of players being "capped" by something predetermined.

The problem with Dstats was that players wouldn't show much statistical decline as they aged, if any at all. However, they did have the advantage of giving players rather realistic numbers.

The current system relies on maximum numbers, to put it simply. For example, a player with high offensive awareness and passing ratings is capable of averaging a high number of assists, up to a certain point (generally the NBA high for the previous year). With Nash setting the bar fairly high this year, I believe players with high ratings should be able to get around 13 assists tops using the current system.

That system is not without problems, there's not enough variables to handle the bench scoring at this point and sometimes the scoring is too evenly distributed. But in a way, it's better than the rigid nature of Dstats.

[BrotherJRB]

Andrew but with the database editors I could edit the leagues Dstat field and get real stats in year 2010 in my dynasty!!

I am kind of suprised that sports games have such a hard time simulating stats....shouldn't this be one of the easiest things to do, considering it is all mathmatics? I always imagined this would be the easiest thing to do(think of all the text based sports games guys used to make in their spare time back in the day). Baffled

[Andrew]

True, that's the advantage Dstats offered. But they also caused problems with aging players, Shaq could be 40 years old and 60 something overall, yet he's still putting up the same numbers he did in his prime, under the Dstats system. I think some sort of combination of the two systems might work. Dstats would be fine if they could decline with a player's abilities.

[beau_boy04]

Yeah a combo of the two - ratings and DSTATS is the best solution and also taking the age in consideration. The engine system should look for the last season DSTATS and should try to match/increase those numbers if the player has a 1 in DESTINY or maybe 2. Destiny and age should work together. The older the player the lower the destiny, but then we had a player like Karl Malone who consistently averaged 20/10 over the course of his career - that's why we can't depend solely on the ratngs because his ratings would decrease over the year and the engine would look for the DSTATS and match his last season numbers (if possible) if there aren't any other great offensive superstar in the same team.

I hope I made myself clear enough.

[beau_boy04]

There's something wrong Andrew! you said the simulation system is based upon last season leaders and that's why we can't get Nash to averaged over 10 apg in Live 05. Then why in the hell nobody can't averaged the 28 ppg Tracy McGrady averaged ? in my last season in franchise mode Kevin Garnett led the league in points achieving 27 ppg but before that season the leaders were averaging under 25 ppg which is very funky. Hopefully in Live 06 we'll see more of the Iverson 30.7 ppg and Kobe 27.6 followed by LBJ 27.2 from this past season.

Thanks.

[Mark.]

I think the scoring in the league leaders was too low overall, I mean sometimes playerrs averaging 19.5ppg are in there......

[fgrep15]

One problem with rebounds was that rebounding ability was affected by jumping ability, which should not happen. Swingmen who could jump and in rela life are good rebounders would have very high rebounder numbers.

Rebounding stats should be affected by only the rebound ratings, that's why they're there. Stackhouse is 6'6, and is athletic, yet is a worse rebounder than Iverson.

Also I hate how a guy with 99 assists, and 99 offensive awareness would still struggle to get 10 APG. Anyone with 95+ in assists should be getting at least 10 APG.

Then you give a guy 99 primacy and 99 offensive awareness, and he's only averaging 20-22 PPG? Does that even makes sense?

Also one very important thing that is needed is having players go to the line. No one goes the to line, the league leader in FTA/G averages like 5 FT's per game . Iverson averages 30 PPG mainly because he get's to the line 10 times a game in addition to taking a lot of shots, not because he takes 31 shots a game, makes 42%, and goes to the line 5 times.

Foul Drawing rating - http://www.nbaliveforums.com/ftopic23193.php

The ratings shouldn't be based on what happened the previous year, their's no constant standard like that in the NBA. Even if it was, I put Kobe at 99 on primacy and he has like an 87 OAwareness, and he wasn't averaging up to 23 PPG? How does that happen? A guy with a 99 primacy no matter what should average at least 25 shots PER 40 minutes [should be more like 26-28]. That means Iverson wouldn't even have a 99 primacy, his would be like 95. Jordan when he was scoring 37 PPG was averaging something like 28 shots, and he shot close to 50% from the field and went to the line about 12 times a game.

[Metsis]

About DSTATS... The biggest problem of the system was that you drafted a great scoring center that was automatic when you played him and in simulation he got like 11 points and 11 boards as he had the defensive center DSTATS, when you averaged 25/10 with him while playing. I had this happen in Live 2000 or 2001 can't remember which one.

Another thing about DSTATS was in Live 2003, you could have teams that have highly skilled starting fives that average 110 points per game, just the five guys and the bench guys would participate too. So you could have a team that would average 140 points per game... Not since showtime... And a starting five that would average 40 rebounds and 35 assists per game...

It was just bad...

Stats are not easy in basketball... It is extremely hard to predict how someone plays day in day out... A guy could go on a shooting spree and get 40 points in one game and then run cold for the next 4 games and average about 8 points in those... How do you figure this into it all??? Then there's the fact that playing against some guys is just plain harder than against others... So you could get 25 points against a bad defensive player and 10 against a great defensive player...

The primacy system is a step in the right direction, but it's way too far off... We should as the coach/gm and everything in between be able to assign guys into roles... So you could decide what stat setting the guy would use when simulating... You could have the scorer, the rebounder, the role players of various things, the playmakers, the defensive stopper etc. Pretty much in the line with the new Super-star stuff. And when you assign a player like Antawn Jamison to a Defensive role, the ratings would then decide how well would the guy do in the role and that would decide what sort of stats he gets from game to game... Then player happiness could be effected by the players role too... So a guy in a wrong role would get unhappy and maybe demand trade or something and that would too effect his daily stats... Then the team chemistry would move the stats too... Then there's still the daily form of the player...

Let's list them here, how I see would get a good set of stats:

1. The role of the player (and how well he fits into it)

2. Against who you are going against

3. How happy is the player

4. How good is the team chemistry

5. Does the player have a good or a bad day

The different super star packages could cover the first item on the menu, but what with the players that don't have access to any super-star stuff...

I've always thought that you should have more control over what are the players roles during simulation... If I want T-Mac to be my main offensive weapon above Tim Duncan (if they were on the same team) then I as a coach should be able to assign them so.

I had Stojakovic and T-Mac in a dynasty and it always just hurt my eye when T-Mac shot 8 times in a game or Stojko put up 9 shots... It's just unrealistic... Those guys should be shooting 15-20 shots per game atleast. No matter who they play against... The defenders effect should be more like does he make 12 of 20 or 6 of 20 shots... But the fact is the guys that have the green light to put up shots, do so, no matter who's defending them. It's their job to put the points on the boards, that's why they get 14 million dollars a year... So they should put up shots if it was their job.

What makes the stats calculations even more difficult, is the fact the most of the stats are related to other stuff... Like if you can't score, you don't get assists, but you get more rebounds. If the scoring is hot the rebounds go down. So a team that is great defensively should get more rebounds on defensive end as they make the other team miss more shots. But a team with a bad defensive ethic, should get less rebounds as the opponents make shots at a higher rate...

The only way for the AI to truly simulate a game stats correctly is doing it play-by-play... One posession at a time. But it would take a very long time to do that... And the machines we have probably wouldn't be up for it.

The stats are incredibly hard to balance and when you add predicting the future into the mix, it only gets worse... They'll get there... Maybe they will some day get it right, but Basket ball is never pure math... It's a game of passion.

[Andrew]

There's something wrong Andrew! you said the simulation system is based upon last season leaders and that's why we can't get Nash to averaged over 10 apg in Live 05. Then why in the hell nobody can't averaged the 28 ppg Tracy McGrady averaged ? in my last season in franchise mode Kevin Garnett led the league in points achieving 27 ppg but before that season the leaders were averaging under 25 ppg which is very funky. Hopefully in Live 06 we'll see more of the Iverson 30.7 ppg and Kobe 27.6 followed by LBJ 27.2 from this past season.

I said that's the way it was designed. I didn't say it worked perfectly everytime, I think we're all aware there's still some glitches in the system.