Exaggeration Time!

You may have heard something about Carmelo Anthony recently.  He appears to have been traded, I think to a big city somewhere.  His first game with the Knicks was last night, which they won in a mighty display of Carmelo’s dominance, by six points over the Bucks.  But that isn’t really what I want to talk about.  Instead, I want to pick on a post at Hoopdata that discussed the game.  Particularly, this part of the post:

*Though Anthony didn’t shoot well on 10 of 25, the rest of the team was 30 of 55 thanks partly to Carmelo taking so much of the opposing defense’s attention.

This is ground that’s been covered recently (here‘s a post by Kevin Pelton that also links to posts by Dave Berri and Nate Silver).  The general public take is that Carmelo is a superstar.  The general (I guess?) stats take is that Carmelo is above average, but not great, due to his average shooting efficiency; he scores so much because he takes a lot of shots, and he isn’t above average at doing many other things.  The perspective offered by Silver and Pelton, and presumably shared by others, is that Carmelo’s inefficiency isn’t too big a deal because he opens up the offense for his teammates.  This is exactly what the Hoopdata post claims happened last night; essentially the defense has to worry about Carmelo scoring from anywhere (even though he apparently doesn’t do that especially well), and so his teammates have an easier time getting open or in general getting easier shots.

Let’s ignore the fact that one game is not enough data to really inform this argument at all.  Let’s ask, how much better did the Knicks really shoot last night now that Carmelo is there drawing all that defense?  We’ll start with the box score.  Only eight players got on the court for the Knicks last night, one of which was obviously Carmelo.  As the quote says, he went 10 for 25, consisting of 1 for 2 on three pointers and 9 for 23 on twos.  This season Carmelo is shooting 33.3% on threes, so 1 for 2 isn’t overly surprising, and 47% on twos, meaning we would expect him to hit 25*.47 = 11.75 shots on average, so he underperformed a bit there.  His effective field goal % (and true shooting, for that matter) were below his season averages, so this wasn’t a great game for Carmelo, although he did chip in ten rebounds and a couple steals.

What about his teammates?  We can use the same reasoning to see how well they shot last night.  Let’s start with the other big Knicks acquisition of the season, Amar’e Stoudemire.  He shot 6 of 13, all two pointers.  With his season average we would expect him to make 6.6 of 13 shots, so he was pretty much on par.  Amar’e did help his true shooting by making all seven of his free throws, but presumably Carmelo didn’t help him get more open at the free throw line; his 7 attempts are also just below his season average.  So not much going on here.  Ronny Turiaf went 3 for 3; we would have expected 2 makes so there’s an extra make here.  Of course, on three shots who can say for sure.  Landry Fields was 2 for 3 on twos and 1 for 3 on threes, both right about on his season averages (we would expect 1.76 and 1.2).  Bill Walker was 0 for 1 and 1 for 3; we would have expected him to maybe make the two and to hit a little over one of the three pointers, so this was below average.  Shawne Williams was 2 for 2 and 1 for 3 compared to expected values of .833 and 1.4, so again not too surprising.  None of Turiaf, Fields, Walker, or Williams get to the line much, so their grand total of 2 free throw attempts doesn’t say a lot either.  In total, these Knicks were pretty much on par for their two point shooting and perhaps just below average on threes, although there aren’t a lot of shots to look at.

That leaves two players.  One is Toney Douglas, who was already on the Knicks.  He had a big shooting night, going 7 for 7 on two and 3 for 5 on threes.  We would have expected 3.3 and 1.8 makes, so he definitely came out ahead.  Then we have Chauncey Billups, who came over from Denver with Carmelo.  He shot 3 for 6 from two and only 1 for 6 from three, compared to expected values of 2.6 and 2.6.  Chauncey’s three point shooting was definitely not helped by Carmelo (or Amar’e for that matter).  So Toney came out ahead on two pointers, but his extra three point makes were canceled out by Chauncey’s misses.  I can’t find a line-up breakdown, but given that Toney is a back-up guard I can’t imagine that he and Carmelo overlapped for more than maybe 20 minutes, so Carmelo probably can’t be responsible for all of his great shooting last night.

Here’s the summary: with the exception of Toney Douglas, everyone shot pretty much as expected on two pointers given that only Douglas, Amar’e, and Billups took more than three of them.  And again with the exception of Douglas, everyone shot a little below average on their three pointers.  That is, everyone on the Knicks shot pretty much how we might have guessed they would have before the game.  The Knicks were playing at home against the crummy Bucks and only won by six, largely because Carmelo, Amar’e, and Billups went 25-26 from the line.  With their season averages we would have expected the trio to make just over 22, so basically half the Knicks’ margin of victory was lucky free throw shooting.  On the other hand, Carmelo had an off game, so the Knicks might have won by more if he shot at his average.  All in all, game 1 of the Carmelo-Knicks experiment didn’t leave me very impressed.  I certainly don’t see any evidence that his teammates did any better because he was there.  But it was only one game; maybe the Knicks will start shooting like crazy.  They were already 8th in the league at both eFG% and offensive rating; if Carmelo boosts their performance like Silver and Pelton expect, they should challenge the top teams the rest of the way (at least on offense).

This entry was posted in Uncategorized and tagged , , . Bookmark the permalink.

10 Responses to Exaggeration Time!

  1. EvanZ says:

    “The Knicks were playing at home against the crummy Bucks and only won by six, largely because Carmelo, Amar’e, and Billups went 25-26 from the line.”

    Alex, the Knick’s ORTG last night was 117.0. The Bucks season DRTG is 102.7 (5th in the NBA this season). You don’t think that’s important to note?

    ” largely because Carmelo, Amar’e, and Billups went 25-26 from the line. ”

    Last I checked, free throws count towards TS% and ORTG. Part of having an efficient offense is getting fouled. Anyone who draws a double team regularly will either get fouled himself often, or “create” fouls for his teammates, since that defender will be out of position, leading to more defensive block calls upon trying to recover (either the defender himself or a teammate).

    • Alex says:

      Carmelo went to the line 7 times, which is below his season per-game average even though he played a few minutes more than normal. As I said in the post, the rest of the Knicks all shot about as many free throws as you’d expect. The sole exception was Chauncey, who did get to the line a ton. Maybe Carmelo’s effect only worked on him for drawing fouls and Toney Douglas for shooting accuracy? In any event, I think it’s kind of ridiculous to claim after one game that the rest of the team shot so well because Carmelo was there. There will be almost 30 games for the effect to show up; maybe we can hold off until then.

      The line you quoted is also referring to the fact that if those three shot free throws as expected, they would have made about 3 less in a 6 point game. Seems important to me; it was not the margin of victory I would expect over a crummy team if Carmelo was vastly improving the Knicks. He certainly didn’t help Chauncey shoot perfectly at the line. If you take away the above-average makes at the line, the Knicks won by about the average home field advantage. Shouldn’t they be better than the Bucks? If Carmelo was supposed to make the Knicks better, it didn’t seem to happen in the way I would have predicted.

      • EvanZ says:

        “Seems important to me; it was not the margin of victory I would expect over a crummy team if Carmelo was vastly improving the Knicks. ”

        Crummy team? Not defensively. Milwaukee is one of the best defensive teams in the league. I mentioned that, but you seemed to be ignoring it.

        • Alex says:

          They have a negative point differential; they’re a crummy team. The Bucks have also given up big offensive games to the Clippers, Philly, and the Pistons; sadly, none of them made any big trades right beforehand for us to attribute the outburst to.

          I didn’t have a double standard in the post; I said we’ll ignore the fact that it’s silly to look at one game. When we do, I’m not seeing convincing evidence that Carmelo was a game-changer. If anyone wants to argue it further I think it’s foolish because it’s just one game. There are two reasons to think the hoopdata claim jumped the gun a bit, the sample size and what actually happened, but I gave them the benefit of the doubt on the first one. I retain the right to use either argument beyond that.

          Something I didn’t specifically point out in my last comment – the one player who did get to the line a lot (Chauncey) is a guy who’s been playing with Carmelo all year. So that would be another strike against the Carmelo campaign – at least for one night.

          • EvanZ says:

            “They have a negative point differential; they’re a crummy team.”

            They don’t have a crummy defense, though. And that is what the NY offense was going up against, and that is what your post was discussing.

          • Alex says:

            On a given night, any team can have a bad defense, or an average defense, even if they’re usually good. One thing the Bucks are not good at is keeping opponents off the free throw line, and that’s where the Knicks won (at least, that’s my impression). Are there numbers on how likely it is for a team to play an average offensive game against a good defense?

            It sounds like you’re strongly on the side of Carmelo helping his teams in ways that don’t show up in his own numbers? Assuming the Heat play up to their top 5 offensive and defensive ratings, the Knicks will drop to below season average on both numbers through three games in the Carmelo Era. I’m not one to jump to conclusions (that’s really the point of the article), but it isn’t going to look good.

      • EvanZ says:

        “In any event, I think it’s kind of ridiculous to claim after one game that the rest of the team shot so well because Carmelo was there. ”

        Wasn’t the point of your post to rebut the pro-Carmelo argument with one game as evidence? I’m not understanding why there is a double standard here.

    • bduran says:

      I think his point about the team shooting 92% from the line not having much to do with Melo is valid. Helping them get there (they took 3 more than average), but not converting.

      It looks like their Ortg was about 114 (114 points on about 100 possessions). However, if you take away four points because you assume season average FT% this goes down to 110. Their season Ortg is 109.9. So they did average against an above average team. Not too shabby.

      Also, it looks like Tony Douglas played most of his minutes with Melo according to the popcornmachine.net gameflow.

  2. Neerav Vyas says:

    “All in all, game 1 of the Carmelo-Knicks experiment didn’t leave me very impressed. I certainly don’t see any evidence that his teammates did any better because he was there.”

    I think Alex made it clear that hoop data was jumping to conclusions and that looking at the data seemed to refute the claim more than supporting it. BUT it’s all based on a small sample size.

    We’ll see how this all pans out (and I really hope that nate silver and the melo fans are right). I would expect the player to watch is landry fields and see how he performs with melo on the floor. If there’s one player who benefits from all the attention on STAT and Melo it should be fields..

  3. EntityAbyss says:

    Well, the knicks lost to the cavs tonight… and Carmelo shot inefficiently and got fouled out, sooo what’s the reason there?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s