Who would you have? Carmelo Anthony or Andrei Kirilenko?

Both players have very good talent, yet they are vastly different types of players.
Andrei Kirilenko is very good on defence
Anthony is very good on a offense

 

Well, he hasn't said any stupid things this year (that I've heard about). Maybe he's finally growing up.

And yes, I've been very impressed with him lately.

I think Melo has the potential to be a very good defensive player too. He has the physical gifts, and defense can be taught.

 

Melo's actually been a really good citizen this year. He has quite a few charity organizations set-up throughout the Denver area and is really doing a lot for the community. I'm proud to say I'm a Melo fan now. Before there was always a little bit of a social bias. With phenoms like LeBron, it's easy to forget that Melo was only 19 his rookie year.

Personally, I don't think he'll ever be a great defender, but I'm fine with that. He's not a weakness on defense right now either. I'd like to see him take less risks on D, but all in all, I think he's a better defender than many think.

I just wish we could get him a shooter to help him.

 

Well, I'm not ready to be a Melo fan yet. But if he keeps up this stretch of good behavior, maybe I'll forget the dumb things he did during his first couple of years. Oh, and one other thing: he needs to improve his 3s, or else stop jacking them up.

The defensive potential is there, he just needs to want to do it.

Hah, everybody wants a shooter.

 

A blocked shot takes away 2 points from the other team, and actually since the shot only has a 50% chance of going in (that' being generous), let's say it takes away one point.

So Kirilenko's 3 blocks = 3 points per night. Anthony's 1 block = one point per night.

THat doesn't make up for the 12 point differential between them, every single night - which will likely grow as Anthony gets better. And I don't see Kirilenko's defense doing too much on Anthony.

Kirilenko is a role player, Anthony is a star. If you switched both players around, Denver would be worse and have to rely on Andre Miller and Marcus Camby to score points - sure they'd have an extra weakside shot blocker, but who gives a sh-t? Eddie Griffin is a good weakside blocker, and he sucks. Jawan Oldham was a good weakside blokcer. Nice ability to have, but by no means critical, and Andre Miller scoring , well that's not where you want to be; Utah would be better and their crappy offense would be energized. Defensively if they played hard they'd be fine.

Eddie Curry and Ben Wallace? That's a totally inappropriate comparison - and you know it. Curry is not a 27 (and potential 30) ppg scorer and never will be. HE's not even a 20 ppg scorer. He's a freaking 12 ppg scorer. He is not even in Anthony's ballpark as a scorer. For that matter, Kirilenko is NOWHERE CLOSE to Wallace as a defensive player. Wallace is one of the great rebounders of his generation. Kirilenko has never broken double digits in a season.

YOu guys don't realize, points per game are not fungible. Two 15 ppg scorers do not = the value of Allen Iverson or Kobe Bryant - those guys bring so much more with just the ability to put up that much. The gap between a star player that you have to focus your gameplan around when you play against him and a second tier player like Kirilenko cannot be made up with a few blocked shots or steals.

This is not Eddie Curry vs. Ben Wallace. It is Rodney McCray vs. Reggie Miller. I'll take Reggie Miller.

 

Agreed. I think Kirilenko would fit in with basically any team, since he doesn't need plays run for him to be effective. He moves very well without the ball and gets a lot of offensive rebounds. Defensively, he's obviously much better than Carmelo.

If the team already has a couple go to scorers, Kirilenko would definitely be a better choice. Maybe if you're the Bulls or the Bobcats you'd rather have Carmelo.

 

Well, you'd need to also factor in that on offense, Kirilenko is not using up nearly as many possessions on his own shot attempts and is in fact getting more assists, so saying there's a 12-point differential offensively is misleading. Plus, Kirilenko is obviously a much better rebounder.

Kirilenko is a star, and I'd argue he's more valuable to his team than Carmelo. Just compare their +/- (impact on team per 100 possessions) and PER the last 3 seasons:

Code:

05/06
		MPG	+/-	PER
Kirilenko	37.7	+12.3	20.8
Anthony		37.0	-2.0	22.5

04/05
		MPG	+/-	PER
Kirilenko	32.9	+13.6	24.4
Anthony		34.8	+1.5	16.7

03/04
		MPG	+/-	PER
Kirilenko	37.1	+12.1	22.6
Anthony		36.5	-1.0	17.6

Kirilenko has had an off year in terms of individual statistics (scoring efficiency is way down), so his PER is less than Carmelo's for the first time. But he's had a major impact on his team (shown by +/-) every year, and is in fact always among the league leaders in this regard. But these numbers show that, overall, the Nuggets have performed just as well with Carmelo on or off the court. This season, the Nuggets have played a little better (relative to the opponent) when Carmelo is sitting on the bench!

I think you're overrating the value of scoring and underrating the blocks and steals -- based on individual numbers, their production is still pretty close even though Kirilenko is having an off year statistically and Anthony is having a career year. But beyond looking beyond the box score numbers, it seems that Kirilenko has a significantly greater impact for his team. I'm sure it's due, in part, to the Jazz having a weak bench, but the difference is so large (and consistent through the years) that I have a hard time explaining it away with that reason alone.

 

You're making the same mistake.

By the numbers ONLY, a guy who can get 8 rebounds (Kirilenko) vs. who gets 5 rebounds (Anthony) = a guy who is roughly 1.6x better at rebounding than the other.

By the Numbers ONLY, a guy who scores 27 ppg vs. a guy who scores 16 ppg is roughly 1.6x better at scoring than the other.

by the numbers ONLY, a guy who gets 1.6 steals vs. a guy who gets 1 steal is 1.6x better at stealing than the other.

So, by the numbers ONLY it seems the gap between Anthony and Kirilenko is roughly matched in certain categories, right? right.

But in reality, the gap between a 27 ppg scorer and a 15 ppgs scorer is MUCH wider than a 5 or 8 rebound a game guy or a 1.6 vs 1.0 steals guy.

27 ppg is the type of thing you build a team around, or your gameplan around if you're the other side. 8 rebounds a game is nice, but honestly who cares? Mo Taylor could get 8 rebounds a game if he played 40 a night. That doesn't make him a great rebounder or even put him up close to the top rebounders in the league.

Now, Kirilenko is an excellent shotblocker. Bully for him. Do opposing teams build their game plan around neutralizing his shotblocking ability or his 5 assisting ability? I sure wouldn't.

Finally, I see that Carmelo is hitting 27 a game head to head against Kirilenko - his defense doesn't seem to be paying dividends there. Again, steals and blocks are nice, but they don't tell the whole story on defense.

If you looked at steals and blocks ONLY, you'd be under the impression that Eddie Griffin was a good defensive player. I'm not saying Kirilenko's bad on defense at all - it's just that there's more to the story - such as the fact that Anthony is holding opposing players to a LOWER PER than Kirilenko this year.

One guy's a star, the other guy's a role player.

BTW, I don't know why you're using data from 3 seasons ago. Anthony's improvement has been dramatic over that span - I don't really care how he played at age 19 in 2003. The fact is that right now today he's a better player, and at age 21 - he's only going to get better and the gap is going to widen.

 

You want to compare two players based on a sample of a few games?

I'm not looking at steals, blocks, rebounds "ONLY". I'm looking at all the numbers, puts together (that's what PER does).

Altogether, Carmelo's numbers are a little better this year (but not by much). Considering that Kirilenko is having a down year and Carmelo is having a career year, it's not a given that Carmelo will continue to outperform Kirilenko by individual numbers in the future.

I don't put too much value on opponent PER as a defensive measure. There's too many cross matchup during an NBA game, and Kirilenko is clearly a superior team defender.

You seem to be saying that 27 ppg over 15 ppg makes Carmelo better, end of story. I think there's much more to the game than scoring points, and if you look at everything together in a logical way, Kirilenko appears to be pretty close, if not better.

The past has relevance, if you want to predict trends. Based on the past, we can expect Kirilenko to play better in the future, which could very well put him statistically on the same level or ahead of Carmelo.

But I notice you didnt address the +/- of the two players. If Carmelo's 27 ppg is so much more valuable, why is this impact not actually reflected in the numbers?

---

Carmelo is younger, so I admit he has more scope for improvement in his game. Kirilenko is probably reaching his prime right now.

I don't want to overstate either player's case. Like I said before, I think it depends on the needs of the team. Some teams need a go-to scorer more than anything, and Carmelo might be more useful for those purposes. But I think there are more teams that need a player who does a bit of everything, and Kirilenko is perfect for that. I think Kirilenko is the kind of player who'd help any team. Carmelo might help a few teams significantly, but have little to no impact on many other teams.

BTW, I checked Opponent PER (though I still maintain it's not a very useful measure). Kirilenko's Opponent PER is better than Anthony's this season, so I don't know why you said the opposite.

Code:

		Own PER		Opp PER		Net
Kirilenko	22.0		12.7		+9.3
Anthony		23.6		14.5		+9.1

So, even by this measure, Kirilenko is ahead.

 

You want to exclude head to head competition?

Thanks, I know what PER is. And I also know that Carmelo's is not only higher, but those of his opponents is lower.

So do I. And both the numbers and the intuitive evidence stacks in favor of Carmelo.

By this logic, Carmelo has played worse in the past. So we can expect him to regress.........

If you want to look at trends - Carmelo is improving - Kirilenko has plateaued and is battling injuries. That's the trend I'm concerned with. He also can't shoot from the outside and is not big enough to put up lots of points inside. That trend doesn't point to him being anything more than he is now: a second fiddle.

Wait a second, I thought PER was most important, now you're switching gears on me to +/-? And raw plus minus at that and not adjusted +/-?

Anyway it is reflected in the numbers: Denver is running away the division, and the superior palyer who allegedly does more for his team is once again counting lottery balls.

Diversity is nice, excellence is better.

 

Of course, I put more weight on performane over the course of a season, versus peformance in a few select games.

From 82games:

Code:

		Own PER		Opp PER		Net
Kirilenko	22.0		12.7		+9.3
Anthony		23.6		14.5		+9.1

We have different intuition, so I'm not going to argue that part.

And, yes, the individual numbers are in Carmelo's favor this season (slightly). But I also give extra credit for Kirilenko for better versatility (I think the value of this versatility is "greater than the sum of the parts", so to speak).

That's not what I said. We can expect Kirilenko to play better. I agree Carmelo will improve as well, since he's young.

The injuries is a fair point. Certainly, a team needs to consider that. As far as scoring, Anthony is undisputably better in that regard.

Did I ever say one should be looked at in exclusion of the other? Both are relevant.

Adjusted +/- is hard to come by, but there are some who have done it.

Kirilenko's adjusted PER in 02/03 and 03/04 was excellent, according to this analysis. Only Garnett and McGrady were better in those years.

Unfortunately, I don't know of any recent data publically available, but it would appear that the raw +/- isn't a fluke.

Riiight ... because team success is only dependent on the quality of play of one player.

Besides, if you compare how their respective teams perform while they are on the court, it's pretty much the same:

Code:

		Team	Opp	Net
Kirilenko	106.5	105.9	+0.6
Anthony		107.4	106.4	+0.6

The difference is that when Kirilenko goes out, the Jazz struggle. When Anthony goes out, the Nuggets still usually outperform the opponent. What does that say about their value to their respective teams?

And you define "excellence" by scoring ability, apparently, while I prefer to look at the player's entire skill set.

 

I thought rookies were signed for like 5 yrs under the old CBA

Rocket River

 

They can give him a max extension this summer. But i don't buy this Melo is only playing hard for a contract stuff. In truth he's been playing extremely well for about a year and a half now, since George Karl came on board.

 

Excuse me for a second.

HAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHA

I don't care what the numbers say. The Nuggets look like CRAP when Melo is out of the game. Their offensive production grinds to a halt pretty quickly. Normally, I'm a fan of the numbers telling the story, but I can guarantee to you that this is one case that the numbers are VERY misleading.

Bingo. The contract talk is one of the dumbest comments I've seen for Melo's improved play.

 

You'd exclude it totally? Odd, I generally see the actual games as important. But why bother with games when we have stats.



From 82games:

Code:

		Own PER		Opp PER		Net
Kirilenko	22.0		12.7		+9.3
Anthony		23.6		14.5		+9.1

Do you have a link? Because 82games says different, in fact it says the opposite - that ANTHONY has a higher net rating

http://www.82games.com/0506/rolandratings0506.htm

.

It says something about their back-ups - which is why the adjusted +/- would be valuable to have as you noted.

The object of the game is to outscore your opponent, not to out diversify him.

 

I don't understand what this means. You want to look at the performance in select games during the season, whereas I look at the performance for all the games in the season. I'm not the one "excluding it totally."

I don't believe there's much relevance in focusing on per game averages for only 2 games. That isn't much of a sample.

And I never said stats substitute for watching games either. Both are important (you can't digest EVERYTHING that happens in a game without statistics, as any coach will tell you). But I do trust stats for an entire season more than an impression based on a couple games.

http://www.82games.com/0506/rolandratings0506.htm

Those numbers are a couple of days out of date, which explains the discepency. But even from that page, Kirilenko's Opponent PER is 12.6 and Anthony's is 14.3.

Anthony
Kirilenko

Maybe we're looking at different numbers. Am I missing something?

Anyways, as I said, I don't think Opponent PER is a useful guage of defensive ability. Particularly for a guy like Kirilenko who has a much greater responsibility as a team defender.

So would you say that Iverson has been a better player through his career than Tim Duncan? Do you think that Domique Wilkins was better than Magic Johnson?

Yes, the object of the game is for a team to outscore the opponent. That doesn't mean the more a player scores, the better that player is. There are a lot of things that go into helping your team score and limiting the team's ability to score. We can agree on this, right?

 

http://www.82games.com/0506/05DEN12D.HTM

This shows that the Nuggets are in fact a little better on offense when Carmelo is on the floor. The eFG% is much better, which might be what you are observing in the games you've seen. However, turnovers go down significantly when Carmelo is out (maybe they aren't fast-breaking as much). Keep in mind that these numbers are looking at offense per possession. It's quite possible that the Nuggets play at a much faster pace with Melo in the game, and so when we watch the game they would be seemingly scoring at a MUCH faster rate.

On the flip side, defensively the Nuggets have been better with Anthony on the bench over the course of the season. So, overall, the net difference is that they have been better when Anthony is sitting.

I'm not going to go so far as to say that this means Anthony doesn't have value for the Nuggets. I still think he and Camby are their most valuable players. If he was on the inactive list, current bench players would have to see more time playing against starters and it would probably have a net negative impact. Also, Anthony seems to elevate his play in crucial game moments, and that isn't shown in these numbers.

I'm not going to go so far as to say that this is a "holy grail" type stat. But almost all the great, franchise players usually have a strong, positive +/-, so when a player like Anthony who puts up pretty good numbers has a negative +/-, you have to take notice and wonder why.

Code:

[b][i]Top Ten Rated Players (by PER)[/i][/b]

[b]Player		PER	+/-[/b]
Dwyane Wade	28.4	+17.1
LeBron James	28	+17
Dirk Nowitzki	28	+8.3
Kobe Bryant	27.4	+12.3
Elton Brand	27.3	+4.8
Kevin Garnett	26.9	+9.1
Allen Iverson	26	+8.3
Yao Ming	25.7	+8.2
Paul Pierce	24.1	+1.9
Gilbert Arenas 	23.8	+4.1

 

I'd be hard pressed to come up with a better way of comparing two players than having them play against each other. Anthony's offense beats Kirilenko's defense head to head it seems. This is not surprising - a great offensive player will beat a great defensive player more often than not, for the simple reason that a man running forward has an inherent ability to get to a spot more quickly than a man running backwards.

I can play this game too - PER doesn't effectively gauge what a dominant scorer like Anthony means to an offense by forcing teams to focus on him.

Why do you guys feel the need to resort to phony analogies like this? Yes, Dominique was a better scorer than Magic, but Magic was no slouch either - he regularly put up 20-25 ppg throughout the bulk of his career, and regularly put up big scoring games and shouldered the load whenever he had to. Kirilenko is merely adequate as a scorer - do you really think he's just holding back because he's got Kareem and Worthy running next to him on the break? He doesn't have the ability to "turn it on" and become a primary scorer. He's simply not that good.

Same with Duncan and Iverson. (by your numbers, btw, Ginobili is better than Duncan this year -- is he really?) Sure, Iverson scores more points, but Duncan, this year excepted, has historically been a 20+ scorer and a primary option - again, something that Kirilenko will never be.

It goes a long way. A great scorer is unstoppable generally if he's on his game no matter who is out there with him. A great defensive player doesn't impact the game much if the other four guys are lapsing defensively. It's like the Rockets when McGrady and Yao were out. Sure,they could play defense as hard as they wanted. But without a scorer they were sunk as soon as they went out there. You could put Kirilenko out there and I don't think the results would change much. YOu put Anthony out there and they have a fighting chance.

Finally, I don't think Kirilenko is really THAT great on defense. He blocks a lot of shots (which seem to come in a help role), but is not a Dikembe/Hakeem style intimidator who patrols the lane and alters a team's game plan. And his rebounding, again, is all right but really not that much to write home about. It's better than Anthony, but by no means is he an elite rebounder.

 

Melo plays 37 mpg. The Nuggets are 9 games over .500.

So, either the Nuggets are winning these games in the 11 minutes Carmelo isn't on the floor (which I somehow doubt), or the stat is missing something in its analysis. Like what? I don't know how it's calculated. But an example would be blow-outs versus close-wins. Maybe the Nuggets tend to get "blown out" in thier losses while winning by less in close games. The fact that Denver is a high scoring team with only a 1.1 ppg positive margin for the team as a whole anyway, I'd imagine this is the case. And it would negatively impact his +/- (I think, again, I don't undertand fully the calculation). But in the end, it doesn't matter how much you win/lose by. You're record changes the same regardless.

This is my problem with some of these stats. They can never completely cover every aspect.

 

I tend to agree with you, and always find it odd when AK47 has a good all around game, that you will immediately see it posted about in this forum. Um, Kobe, Lebron, T-Mac (when healthy) all do the same thing regularly (except the blocks), but are significantly better scores.

Then you have other, lower tier players that put up good all around games as well. Heck, Josh Smith is 6th in the league in blocks. Gerald Wallace is 9th. And he scores more than AK, gets only 1 rebound less a game, steals 1 more per game, and has a fair number fewer assists....but overall, all-around, very similiarly talented to AK, and people don't have an orgasm every time he does well.

AK is a good player to be sure. Above average and occassional all-star type seasons. That's it. Coincidentally, that is about the level Carmelo is currently at, too, but I think this season he should have been an all-star and has seperated himself somewhat from AK.