Thursday, October 9, 2008

Simulated statistics, Part I

There's been some grumblings about 2k's sim stats being even more messed up than previous seasons. Let's sim an association and see what we get.

LEAGUE LEADER STATISTICS

PPG
Lebron 29.4
Kevin Martin 28.6
Iverson 27.6
Joe Johnson 27.4
Kobe 27.1
Wade 26.9
Gay 26.8
McGrady 26.2
Durant 26.0
Melo 25.6
Dirk 25.2

NBA.com scoring leaders

In real life only 5 players averaged 25 ppg. Here we've got 11.
In real life 31 players averaged 20+ppg which is exactly how many there were in this sim, 31.

So the scoring problem seems to be too many players are in the upper 20s.

Offensive Boards
Howard 3.4
Duncan 3.4
Garnett 3.2

NBA.com Rebounds

In real life Tyson Chandler led with 4.1 and traditionally the league leader is usually around 4. The shooting percentages are reportedly too high so this plays a factor here. Big Ben averaged 3.2 in real life which was only good for 10th. In 2k9 only 3 players averaged that much.

Rebounds
KG 11.3
Howard 11.0
Duncan 10.5
Okafor 10.2

Those are the only players in double figures. It's probably worth noting that I've heard people get even fewer than this. I wonder if the edits I've made have anything to do with it, maybe by the time I'm finished there will be more

Assists
Calderon 10.1
Deron 10.1
Kidd 9.8
Nash 8.7
Ford 8.3
LeBron 7.9
Iverson 7.9
Hinrich 7.8
Wade 7.7
A. Miller 7.7
Felton 7.6
Roy 7.5
Harris 7.2
Billups 7.2
Udrih 7.1
Sessions 7.1
Parker 6.9
Ellis 6.9

NBA.com assists leaders

11 players in real life averaged 7 assists, here we've got 16. The difference isn't too bad and you can attribute it to the higher shooting percentages and scoring.

Steals
Kidd 2.4
LeBron 2.3
Artest 2.2

NBA.com steals leaders

These I don't have a problem with, they see fairly accurate for what seems to be the first time. Paul I believe suffered some injuries so he didn't qualify for any leaderboards if you're wondering why he's absent from the two categories he led the league in last year.

Blocks
Howard 3.0
Duncan 2.7
Kirilenko 2.6
E. Thomas 2.4
Darko 2.4
Yao 2.3

NBA.com block stats

I don't really have a gripe here, Camby was the only player with 3 bpg last year (3.6) and there's only one with 3 here.

FG%
Brand 553
Amare 530
Howard 529
Melo 527
Curry 524
West 520
Kaman 519
Deng 517
Bosh 513
Gasol 511
Yi 510
KG 507
Granger 507
Aldridge 503
J. O'Neal 500
T-Mac 500
RJ 499
Peja 498
LeBron 497
Wade 495
Yao 495
Dirk 494

NBA.com FG% leaders


Finally the category people are pissed about. I don't necessarily have a problem with the bigs having high FG%, Howard and Amare obviously should be towards the top, but some of these guys are notorious jump shooters even for big men. Jermaine O'Neal hasn't posted a FG% this high in years. T-Mac takes mostly jumpers and shouldn't be anything close to 50%. Ditto for Peja, Dirk, and YI JIANLIAN.

Of the top 30 in FG%, only 5 averaged a three made per game: Calderon, Deron, Nash, Marion, and Mike Miller. The first three are excellent shooting PGs who also have great shot selection. Marion gets most of his points on dunks and layups. Mike Miller is perhaps the league's most efficient shooter.

The rest of the PG/SG/SF in the top 30 barely take threes: Josh Childress, Ronnie Brewer, Monta Ellis, and Grant Hill. All are slashers who get most of their points in trasition.

None of these players are heavily guarded players like Melo or T-Mac.

3pt%

NBA.com stats

In real life Kapono and Nash were the only players who shot higher than 45%.

In the game?

Kapono 557
J-Rich 531
Herrmann 494
Cook 488
Peja 486
Barry 482
Vujacic 479
Korver 476
Granger 476
Evans 476
Lewis 474
Atkins 473
Quinn 471
Brooks 467
Gay 464
Pietrus 463
Hayes 462
J.R. Smith 459
Blake 459
Smush 457
S. Jackson 457
Chalmer 452
Gibson 452
Carroll 452
Finley 451

25 players in the game shot 45%. Again, TWO shot that high in real life. Ladies and gentlemen, I shit you not.

63 players had a 3pt% higher than 40%. In real life only 33 players had a 3p% higher than 40%


THREES MADE
13 players had 200+ threes made. You know how many hit that many in real life? Try THREE.

Peja 280
Allen 272
S. Jackson 265
Lewis 254
Gay 249

In real life, J-Rich led the league last year with 243. Five players in a 2k9 simulation trump that.


STATISTICS BROUGHT TO YOU BY THE 2K INSIDER! The other one.

8 comments:

Anonymous said...

Pretty interesting breakdown you have there. It might be a good idea to run a few simulations of the "living rosters" in comparison with your own. The 3pt% is definitely ridiculous, but not really surprising for a basketball video game. I also have to question the tendencies that players have to shoot 3s, because I have a feeling there may be too many attempted.

Funny thing is, some bigs might not have a high enough shooting percentage, but I don't think across the board it will be a huge issue. I do wonder if the ones you posted are the leaders, because if so I'd have to ask where Shaq is at, he definitely should not be shooting lower than 50%. Howard shot .599 last year (career .567) and Amare shot .590 (career .540).

I think that once you get the 3pt% sorted out, some of perimeter player percentages will go down too. Some players though are fine like Wade who has shot pretty close to 50% many times in his career (career .481, his % went down to .469 last year with his injuries) and Lebron. These are usually more of the players who drive to the basket more often than settling for jumpshots.

lkasprzak said...

O man u r good at this :)

Kev said...

good stuff! keep it up

Benjamin said...

Rashidi,

instead of comparing stats to last year only, maybe do a 3 year average.

for example, how many guys scored 25ppg+ per season on average last 3 years (when rule changes were pretty established).

this is Benicio10 by the way, if you haven't figured it out.

Rashidi said...

the problem with 3 yr average is it would heavily penalize the younger players and favor some declining vets.

Basketball is great in that the stats are much easier to predict than a sport like baseball, where there are a cajillion sub stats.

Benjamin said...

i meant to look at 3 yr average for ...

the # of guys with 10APG at the end of a season,

or the # of guys with 25PPG +

etc.

cuz each year it varies.

Rashidi said...

Ah. Well, it does to a certain extent but it's usually more dependent on the players than it is anything about the league. There was a period where J-Kidd was the only great passer in the league, when Stockton retired in 2003. Then Nash broke out on the scene and there were two. Last year Paul, Deron, and Calderon all broke out and we have 5 great passing PGs in the NBA when 5 years ago we only had 1.

The NBA went for awhile without a 30 ppg scorers too. Then LeBron was drafted, and that's pretty much history. Of course, the rule changes did play a role here, I don't think Tony Parker or Monta Ellis could shoot 50% from the field 10 years ago. Calderon/Nash yes because they have Stockton-esque shot selection, Mike Miller yes because he's akin to Jeff Hornacek.

I'm starting to build some theories to test based on what people are posting though.

tyrthomas said...

Hey Rashidi quick question. When 2K updates people's dunk and move packages throughout the season will your updates reflect those changes as well?