Who would you have? Carmelo Anthony or Andrei Kirilenko?

Looking a little deeper at the Nuggets, they have outscored the opponent more often with Carmelo on the floor than with him off. So, blowouts are probably having an effect.



As far as point differential vs win/loss, I do think that point differential is a better indicator of how good a team is over the long haul. Wins/losses often are skewed by luck at the end of games.

San Antonio last season had, by far, the best point-differential for instance.

Early this season, a lot of people were impressed by Milwauke's nice win/loss record even though they were actually scoring less than opponents per 48 minutes. Well, as the season progressed, the wins/losses started to conform to the point differential and now they're a .500 team.

 

I see what you're saying but don't understand this. Wins/losses is the best indicator out there of how good a team is...

...had Milwaukee kept winning (or getting lucky, as you say), then they would have had a positive point differential.

 

Head to head matchups can be informative, but TWO GAMES? Anthony outplayed Kirilenko in one of those games, and Kirilenko had better numbers across the board in the other.

I'm sorry, but that too me doesn't say a lot.

PER tells us more about a player's offensive ability than opponent PER tells us about his defensive abilities. On offense, you get credit for your numbers. On defense, you're responsible for not only your opponent's numbers but the whole team. This is even more the case for a guy like Kirilenko. Is it really necessary for me to elaborate on this point?

Scoring ability is PART of the game. Magic wasn't good simply because he had the ability to score but decided not to. He was good because of everything he did on the court put together.

Who's better: Ben Wallace during the Piston's championship season or Antoine Walker when he was scoring 20-25 ppg?

I think you're overrating scoring ability. There's more to the game than that for a player. Just because the object for the team is to outscore the opponent, that doesn't mean the object for the individual player is to outscore an individual on the other end. I'm sure I'm just not understanding you correctly, because I know you can't actually believe that.

And I'll just say again, Anthony would help some teams more than Kirilenko because of his scoring. But I think more teams in this league could probably use a guy like Kirilenko rather than Carmelo.

Ok, so what makes Duncan better than Iverson, if Iverson is scoring more? Isn't it because Duncan does other things better (rebounding, team defense)? So why credit Duncan for those things, and not credit Kirilenko?

Is Ginobili better than Duncan this year? On offense, certainly. PER doesn't capture defensive ability, and Duncan is the pillar of San Antonio's defense.

PER shows that Duncan has been a top 5 player in the league for years. If you only went by scoring average, that wouldn't be the case.

I agree with this. That's why I said in my first post in the thread that Anthony would be a better choice for a team desperate for a go-to scorer. For most teams, though, I think Kirilenko would be better.

Ok. I think he's a very good team defender from what I've seen. He may not be an intidimidator in the class of Dikembe/Hakeem (who is?), but he does cover more ground.

 

Not necessarily. If they improbably won games throughout the season in the same fashion, they're point differential would have stayed the same (negative). Of course, if they won EVERY game for the rest of the season, then the point-differential would have probably become positive.

Consider this: Team A and Team B over the course of 20 games

Team A wins 12 games by an average of 1 point. They lose 8 games by an average of 10 points. So,
W-L: 12-8, point-diff: -3.4

Team B wins 10 games by an average of 10 points. They lose 10 games by average of 2 points. So,
W-L: 10-10, point-diff: +4.0

Now, which team would you rather face: Team A or Team B?

Going by win-loss, Team A was better. But Team B has a significantly higher point differential.

What I'm saying is that Team B is probably the better team and they were just unlucky in some games while Team A was lucky in some of theirs. I consider point-differential a more instructive statistic evaluating the two team's performance than win-loss record.

 

I disagree. Team A was better. It is a fact. It is also why teams make playoffs based on record and not point differential.

Again, I see your point, and how point differential is a useful stat.

But if point differential was truly a better indicator of the better team, then we should rank teams differently, etc.

As your example demonstrated, there are many mathematically issues with looking at just point differential.

Over an 82 games season, however, the only luck left over in wins/losses is injuries, or lack there-of.

 

You call it luck. I call it clutch.

 

If Team A beats Team B 3 out od 4 times by a average of 2 pts
but the one win by Team B is by 25

Which one is the better team?

Rocket River

 

Off of that, I'd probably say team B, actually. Winning 3 games out of 4 by an average of only 2 points doesn't really say much.

I know the cliche that whoever wins a series is the better team, but if the team that wins is only winning by a point or two each time, that really doesn't prove much to me.

Of course, I wouldn't endorse giving the series win to the team with the better point differential. That would take away from the drama of the game.

It's probably not a popular thing to say, but the best team doesn't alway win. That's just the nature of sports.

 

It's a bit of both, probably.

 

It says that his masterful defense can't stop Carmelo's offense, which is why his blocks and steals are nice but not as valuable.

I thought PER counted both offense and defense, and why is PER on offense any less misleading due to a whole remainder of a team being involved? Your PER is helped if you're surrounded by deadeye jumpshooters and you get more assists.

And he was much better at scoring and rebounding and assisting than Kirilenko can ever hope to be. Your point?

Well, Wallace was an elite rebounder, again, something that Kirilenko has NEVER come close to being. There's a bigger difference between the 3-4 extra rebounds per game that Wallace would give you than Walker than the 3 more rebounds that Kirilenko would get you over Anthony. The difference between great and good is WIDER than the difference between good and average. Once again, these things aren't fungible. If they were basketball would be a simple linear equation.

But to answer your question - it's quite possible that Walker in his heyday WAS a much better player than Wallace. I'd definitely take Antoine in his best seasons (22 ppg, 9 rpg, 5apg) over Wallace in 2004 (9 ppg, 12rpg, 3bpg).

I know it's a lot more rational than believing the object is to outsteal or outblock.

You're arguing "fit" over ability. I'm just arguing ability. In the abstract, Carmelo gives you more.

I think Iverson is great - this year you could defintely say he's better. I love Iverson. But anyway, Duncan as a scorer is far more competent than Kirilenko can ever be. He's a number 1 guy that you can focus an offense around. Anthony is at that tier also. Kirilenko is not and never will be barring a dramatic change.

Yeah but I wouldn't do that, and I'm not doing it now. Duncan is a primary option on offense that you have to double team. So is Anthony. Kirilenko? No.

Fine, but like I said with scoring, while there doesn't seem like there's that much numerical difference between his 3 blocks a night and Hakeem or Dikembe's 4 blocks a night - in reality it's completely different.

Like I said, these stats aren't always fungible. Being great at one thing changes the game in the way that being good at multiple things does not.

 

Like I said, Kirilenko is a great team defender. I don't think Carmelo scoring well in a few games against the Jazz is proof that he's not a good defender. He's not like Bruce Bowen or Ron Artest where all his focus is geared towards stopping his man.

Firstly, PER doesn't cover defense very well. It only takes into account steals, blocks, and defensive boards. Much more to defense than that. Does PER cover offense perfectly? Of course not, but it's much better for offense than defense.

What I said is that Opponent PER isn't as good an indication of a player's defense as PER is of a player's offense. The reason is that a player's defensive responsibilities isn't solely guarding his man.

If you want proof, just look at the PER leaders -- they are almost always players recognized for their skills. But if you looked the Opponent PER of players, they are all over the map. Particularly for players who are more team defenders. To take just one example, Ben Wallace has a much higher Opponent PER than Yao, but he's clearly a more valuable defensive player.

My point is simply the obvious observation that a player should be measured by everything he does on the court, not just his scoring ability.

I'm glad we agree that there's more to basketball than scoring.

Yes Wallace is a better rebounder than Kirilenko. But Kirilenko is a better offensive player than Wallace. Altogether, I'd say that Kirilenko is roughly as valuable to a team as Wallace might be (it might depend to some extent on the existing setup of the team, of course).

It would probably depend on the team (as usual), but in most cases I'd probably take Wallace. Walker was/is a frighteningly inefficient scorer.

The object for a player is to help his team win, period. Not to "out"-anything his opponent.

When I consider the question "who's the better player", I ask myself which player I'd rather have if it's unknown who the rest of the team will be. So skill is part of it, but another crucial part is the ability to play well with others (call it "chemistry", "teamwork", or whatever).

Ok, so if I understand you correctly, the primary thing you look at when evaluating a player is his scoring ability on offense and his ability to demand double teams, open up opportunities for teammates, etc? That's very important, and I'll admit that the ability to draw double teams could very well be undervalued in PER.

But you could also argue that Kirilenko does things on defense that aren't credited enough in PER either.

Perhaps, but I also feel that if you have a player that's good at many different things, that also adds something extra ("more than the some of the parts"). And Kirilenko isn't merely "good" as a team defender; I consider him great. He's highly disruptive on the defensive end, because he'll get deflections and bother players all over the court.

As is usually the case with these lengthy internet conversations, I don't think our views are as far apart as they may seem. Carmelo is a greater scorer. Kirilenko is more versatile. Carmelo's skills will be a better fit with some teams, Kirilenko's skills will be a better fit with other teams. Scoring ability is very important, but so are other aspects of the game. We agree on these parameters. The rest are just details.

 

Kirilenko is the better all around player. He plays great defense, is a good rebounder, and is slowly developing an offensive game...

However, he's not the kind of guy you can build a franchise around. Mello will make clutch shots and has the ability to grind out victories by himself if necessary. He has the intangibles necessary to become a franchise player. Kirilenko would be an amazing role player on any team, but he can't be the focal point.

Case in point being the debate over whether to pick Dwight Howard or Emeka Okafor. Okafor is the more polished player who all around might even be better than Howard. But you cant make him a franchise player. Howard has the upside and is slowly developing a game that makes him more conducive towards being a franchise player. (although he's not really close to the caliber franchise player as Carmello Anthony)

 

I agree since AK47 can block like crazy!... and he's only 6'9 like Ben Wallace..... thats one of the only white dudes in the nba with hops..... LOL bu Carmelo can shoot a good percentage and make game winning shots

 

so winning 3 out of 4 games does not make you the better teams
head up

interesting
I guess the playoffs are pretty meaning less

Rocket River

 

Durvasa you are really digging deep into some obscure stats to try and prove your point. Since you love to quote 82games so much - I believe I will take Melo the most clutch player in the league over AK who can t even crack the top 40.

 

If you win 3 out of 4 by a basket, and you get blown out in the other. No, I don't think that means you are the better team.

Look at it this way: suppose they continued to play 40 more times. Based on those first 4 games, who would probably win a majority of those 40 games? I'd pick Team B, even though they only won one of the first four games. It might seem counter-intuitive at first, but when you consider that very close games are often largely decided by good fortune (the ball happening to bounce a certain way on a few possessions), whereas blowouts indicate one team was clearly better than the other, it makes sense.

I don't see why the playoffs are meaningless. If you play a game of chess with a friend, that isn't meaningless even though whoever wins isn't necessarily the better chess player. That's just how games are. Is the NCAA tournament meaningless even though you have upsets where the better teams often get knocked off?

 

That +/- and PER is an obscure stat to you doesn't make them invalid. The information they convey isn't particularly complicated (PER sums up the positive statistical contributions and subtracts the negative contributions).

What does the second part have to do with me using information published by 82games?

 

Melo hits another game winner!!!!!!!!

I would even take him over lebron now.

 

I don't know if I'd take Melo over Bron, but this is amazing:

http://sports.yahoo.com/nba/recap;_ylt=ArfNIw490AZ7Q6BzmXdhWoy8vLYF?gid=2006040607

 

I find it hard to believe that Jordan only managed 11 game winning shots in his career...