Explaining the improved New York Knicks

Last summer I projected the New York Knicks to win 39 games this season.  So far, the Knicks are outstripping my prediction, and are aimed closer to 47 wins.  How have they improved?

Here is the team’s Win Chart to date:


New York Knicks
Win Score
Win Contribution
Player Winning
L Fields
2.3 +3.82 3.7__(-0.5) +0.495 1.152%
R Felton
1.0 +0.79 2.5__1.5 +0.126 .637%
A Stoudamire
4.8 -0.05 1.9__2.0 -0.008 .494%
D Gallinari
2.6 -0.30 1.6__1.9 -0.042 .451%
W Chandler
3.4 +0.21 1.9__1.7 +0.074 .538%
T Douglas
1.6 +0.90 1.5__0.8 +0.083 .655%
R Turiaf
5.0 -0.43 0.6__0.9 -0.026 .429%
B Walker
2.4 +0.26 0.5__0.5 +0.010 .546%
S Williams
4.0 +2.35 0.4__0.0 +0.037 .902%
R Mason
T Mozgov
5.0 -3.78 (-0.1)__1.1 -0.153 (-.138)%
A Randolph
4.6 -5.72 (-0.2)__0.6 -0.094 (-.468)%

MWS Record: 14.3__10.7


For a comparative with last season’s Knicks, here is a link to the team’s 2009-10 Win Chart.  Based upon the above numbers, the following conclusions can be made:

1. Landry Fields is driving the improvement

I don’t know how the New York Knicks were able to find the swingman Landry Fields, but his incredible productivity is driving the team’s success.  Last season the team employed below average Win Contributor’s in his role.  He is well above average, and since he is playing a large chunk of minutes, he has put himself in the MVP conversation with his incredible start.

2. Felton is a big improvement at the point

Last season the Knicks employed Chris Duhon as their main point guard, and he was well below average.  This season those minutes have gone to Raymond Felton, and he has been above average.    Felton’s production is actually less than he provided last season to the Charlotte Bobcats, nevertheless the substitution effect between he and Duhon has been substantial.

3.  The bit players have raised their games

Toney Douglas, Wilson Chandler, and Danillo Gallinari have all raised their games a bit, which cumulatively has had a nice impact on the team.  Bill Walker is playing at about the same level as last season, which was above average.

4. Amare is not the reason

Finally, let me point out my contention that Amare Stoudamire has not been an adequate replacement for the departed David Lee.  Stoudamire’s numbers are way down from what they were with Phoenix, and they are a step down from the numbers Lee was providing the Knicks last season.  Of course, that could be considered good news.  If Stoudamire can get himself back up to his Phoenix levels, and if all else holds, the Knicks ceiling may be even higher than the one they have already hit.

21 Responses to “Explaining the improved New York Knicks”

  1. brgulker Says:

    It’s so frustrating to me as an NBA fan that no one seems to grasp that Amare isn’t the superstar his contract indicates he should be. Clearly, other players are more instrumental than him, but it’s as if the media feels compelled to heap praises on him. Irritating.

  2. Jon Says:

    You guys are delusional, do you watch Knick games? Hollinger has Amare 6th in the entire NBA in wins added, i wonder which one is correct??? Fields is a nice role player with a chance to be a bit more, he rebounds very well, hits open jumpers, and makes the extra pass. But he’s a complementary player, who succeeds largely because the defense is focused on other guys. Amare had a slow first couple of weeks but he’s been insanely dominant since then (man-to-man defense notwithstanding). The whole offense clicks because of him, not because of Fields!

    • tywill33 Says:


      How does Hollinger calculate wins produced? I didn’t even know that he did. I’d love to see his methodology.

      The reason I downgrade Stoudamire, for better or worse, Jon, is that my version of the Win Score metric considers comparative opponent productivity (loosely “defense”) and those numbers bring Stoudamire’s numbers down. If you look at the production half of the Roland Ratings on 82games you will see that even by Hollinger’s PER rating, Stoudamire’s opponents are outproducing him.

      Its like Brandon Jennings performance last night. If you look at his stats, you would say “great game!” But I contend he had a net negative effect on the Bucks chances to win because the Mavericks he guarded significantly outproduced him. If this blog had broader appeal, I am certain Bucks fans would be jumping ugly on me.

      With all that said, your opinion is clearly in the majority, mine and others the minority.

      Thank you for adding some fire to the debate, though

  3. brgulker Says:

    Hollinger has Amare 6th in the entire NBA in wins added, i wonder which one is correct???

    LOL! The irony is so amazingly thick!

  4. Jon Says:

    Lol at your lol brgulker! Hollinger’s stats/ratings are clearly more predictive and better pass the sniff test than wins produced. What was that old Bill James line – a good stat should jive with common sense 75% of the time and surprise me 25% of the time? When your stat (WP) defies both common wisdom and nearly all other advanced metrics a high % of the time, and does a poorer job of predicting the future than those things, odds are you’ve got a bad statistic on your hands. That said, if Ty could provide the link to the 2010 Roland ratings (seem to only be able to find older ones) i’d be interested to see them. Part of it is the fact that the Knicks have no true center so Amare is miscast in some games (such as guarding Brook Lopez) and has poor help. However, i still find it very hard to believe he’s been outproduced in 1-on-1 battles in aggregate, so i’d like to delve into those #s.

    • brgulker Says:

      Hollinger’s stats/ratings are clearly more predictive and better pass the sniff test than wins produced.

      Are you joking about the predictive part? Hollinger’s PER and derivatives do little to nothing to predict team wins. And regarding the sniff test, I don’t really care. I’m interested in rigorous academic critiques more than sniff tests for most topics, including basketball statistics.

      Part of it is the fact that the Knicks have no true center so Amare is miscast in some games (such as guarding Brook Lopez)

      I don’t know if this is as true for Ty’s metric as Dr. Berri’s, but in Berri’s, the positional adjustments between PF and C are quite small. So being “miscast” would be incredibly minor, if at all. Furthermore, it would be fair insofar as Amare is actually playing C, whether suited well for it or not, and him playing C has affects on the team. That’s inescapable.

      and has poor help.

      Ah yes, the bad teammates argument. Have you read the research that generated Wins Produced and WP48?

      What was that old Bill James line – a good stat should jive with common sense 75% of the time and surprise me 25% of the time?

      So, common sense says that common sense should be right 75% of the time. I think they call that a tautology.

  5. Jon Says:

    Hmm i think i’m looking in the right place now and it has Amare at 24.0 and his opponent at 20.3. It also has him with mediocre +- numbers, which is maybe whats driving some of your logic? That seems harsh, since +- numbers have so much noise in them, most people think you need to look at them over a multi-year sample, let alone a 20 game sample. Also seems weird b/c David Lee, who WP loved so much (i’m assuming same for you Ty but correct me if i’m wrong) had a subpar adjusted +- for the entirety of last season, and allowed opponents a 22.9 PER (compared to Amare’s 20.3, assuming i’m looking in the right place). I would say that while Amare is below average defensively, and Lee is god awful, neither is as bad as those numbers makes it seem, the fact they have/had horrible help in post defense contributes somewhat to those numbers as well.

    • tywill33 Says:

      You were absolutely right all along… I realized it right away, but I had some clients on my ass and I didn’t think you’d come back so soon… sorry for the mistake!

      The point remains the same though, his opponent PER, and his adjusted +/- both show that perhaps he isn’t the real reason the Knicks are improved.

      But, as you say, I don’t see many Knick games. But I wouldn’t trust my eyeball assessment anyway. Good comments, you sparked a debate!

  6. Jon Says:

    No I’m not joking. If PER is so terrible at predicting wins than why are Hollinger’s team win predictions so much better than everyone in the WP Smackdown right now? Why does a player’s PER stay more stable when he switches teams than his WP? Why does nobody else “realize” that Marcus Camby is the greatest player of all-time? Wins Produced forces past data to line up with team wins and be correct at the team level, while horribly misallocating the wins among players. Then when you have to predict how teams will do that have had player movement from the prior year, you end up with terrible predictions. Telling me something correlates with wins at the team level when looking backwards means nothing. Tell me you can predict in advance whats going to happen based on a stat, or its worthless. Believe me i work in a forecasting/modeling industry, and if i had a model that produced results (predictive/future-looking) as poor as WP it would be in the scrapheap already.

    • tywill33 Says:

      Its not that tough to predict wins in the NBA. The majority of teams’ records tend to be close to what they were the season prior. That’s why I stopped predicting.

  7. Jon Says:

    Again, if you actually watched Knick games, you’d know that Amare is forced to guard people that he can’t really guard, such as Brook Lopez, because the Knicks lack a true center. Does this mean his value to the team is negated??? If they didn’t have Amare, they’d still have to cover Brook Lopez with a non-center (or even worse Mozgov), and he’d still score at will. However they would lose Amare’s offense. You don’t find it odd that the Suns are struggling badly without Amare? That Steve Nash was devastated about Amare’s loss? That the Knicks are doing so well with him (albeit against an easy schedule – oh wait its all b/c of Landry Fields that’s right). That everyone in the media and associated with the NBA is giving Amare the lions share of the credit? That even most “stats” guys have Amare rated as an elite or at minimum all-star player? But no, you and your WP savants are the only ones who know the truth about what a fraud he is! How stupid everyone else is!!!!!

    • brgulker Says:

      The positional argument is an interesting one. You seem to be implying that Amare isn’t a good C, especially when he has to guard bigger C’s who are good offensively.

      That’s fine, as far as it goes.

      My response is simply this: If Amare isn’t good at defending other good C’s, then it ultimately hurts the team — right? If Amare is surrendering points and rebound at a greater rate than he’s grabbing them, he is being outproduced — right?

      So ultimately, Ty’s point — no matter what positional games we might want to play — stands. Amare is regularly outproduced by the opposition, which has a negative impact on the team’s ability to win basketball games. You seem to be agreeing with this with your comment.

      And in a nutshell, that’s what his numbers Ty is using say — Amare’s overall contribution to wins isn’t as significant as it might appear, because the opponents he plays against tend to outproduce him regularly.

      You don’t find it odd that the Suns are struggling badly without Amare? That Steve Nash was devastated about Amare’s loss?

      Amare is a good player, and the Suns tried to replace him with not so good players. No, I’m not surprised at all they are struggling. They lost a good player and didn’t replace him with a good player. But he’s not a great player, and while he’s certainly contributing, he’s not the primary reason for the Knicks improvement.

      And Nash isn’t devastated. He’s still a phenomenal player. I have no idea what you’re talking about.

      But no, you and your WP savants are the only ones who know the truth about what a fraud he is! How stupid everyone else is!!!!!

      ad hominem for the win?

      • tywill33 Says:


        I should be blogging on the Knicks, not the Bucks! Wow, talk about a spike… I thought my writing had declined (it probably has), but now I think the brunt of the yawns were directed at the Bucks. I’m shocked that more people don’t gravitate toward a 1950s style offensive team!


    • tywill33 Says:


      Come on now! First of all, you are taking this way too personally! If you believe Amare Stoudamire rocks, then he rocks.

      Second though, the argument you are putting forward is illogical. You are making it sound as though Stoudamire ought to be relieved of any defensive responsibility simply because he is guarding players you assume he cannot guard! How do you do that?? Its like arguing that Johnny Bench’s awful years as a third baseman ought to be reevaluated as though he were still playing catcher!! He’s on the field with the specific assignment to carry out third base duties! To whom should we allocate his errors? The guy in Triple A who “should” have been playing third?

      If the Knicks choose to play him at center, they must see some potential benefit in that assignment. Indeed, he probably enjoys offensive advantages over other centers. You cannot accept the offensive benefits and then simply disregard any defensive costs! That’s inconsistent and illogical.


  8. Decoding the Knicks Secret Rebuilding Plan | The Wages of Wins Journal Says:

    […] If you want another take on Landry Fields, see Ty Willihnganz discussion at Courtside Analyst (it really isn’t a different take, just […]

  9. Jon Says:

    Brgulker – As Ty has now said, Amare is not getting outproduced by his competition, not even close. He’s allowing them to put up above-average numbers, but he is clearly outproducing them. Would he be even more valuable if he was a great defensive player – of course. I still think he’s a tremendously valuable offensive player who has some defensive benefits – rebounds ok, blocks shots well. Regarding Nash – he has publicly made many comments about how he was upset the Suns did not match the Knicks offer for Amare, about how Amare was an amazing player/teammate, etc. I didnt say his game was devastated, i meant personally he was devastated knowing the Suns had taken a championship contender and dismantled it. And most WoW analysis suggested that the Suns would be just fine without Amare, that Josh Childress was probably 95% as good, etc. Just go back and read the previews on Dr Berri’s site. As far as who is a more valuable factor in the Knicks resurgence – i will continue to believe that it is clearly Amare. You guys can think its Landry Fields because thats what WP tells you, i’ve realized there’s no convincing you of that point. One of the things that bothers me about the WP bloggers is how the stat is treated as dogma. It tells you how valuable a player has been – with no uncertainty about it, it’s treated as fact by you guys. Here’s an example of a good stat-driven analysis that has a bit of a bigger-view picture as well.

    Because of Fields’ rebounding, scoring efficiency, and ability to guard multiple positions, he’s been a perfect fit with the Knicks and an indispensable part of their rotation, and the +/- numbers reflect that. When Fields in on the court, the Knicks outscore their opponents by an average of 7.13 points per 100 possessions. When he sits, the Knicks get outscored by an average of 5.21 points. That’s a 12.34 net +/-, easily the best mark for any Knick.

    Is Landry Fields as good as Amar’e Stoudemire? Of course not. Should he be in the rookie of the year conversation? Not as long as Blake Griffin continues his campaign against rims. But he’s playing a key role on a good team a few months after being drafted with the 39th pick, and that’s pretty impressive.

    Ty , appreciate the tone of your dialogue. However regarding predicting wins, its just not true that using last year’s totals is hard to beat. Not in a year like this one with so much player movement. Most systems/predictions that i’m tracking are easily beating last year’s win totals (most people realized Cleveland would now suck for example). Wins Produced driven picks are struggling to best that benchmark so far, which again makes me very skeptical about the stat’s ability to allocate wins properly among players.

  10. Jon Says:

    Ty, I’m really not taking it personally. Just got frustrated with some of br’s comments and perhaps responded a bit immaturely. I hope you are watching Knicks-Celtics tonight though!

  11. brgulker Says:


    For what it’s worth, I definitely overreacted. I think I probably misunderstood your Hollinger comment actually, now that I’ve re-read it. I took you to be talking about PER, but now that I look again, it looks like you might have been talking about his prediction metrics, not PER. My mistake if that’s the case.

    I think PER is a pile of dog poo, and I tend to get frustrated when people tout it about as something that has any explanatory power as it relates to team wins. I think the work of Dr Berri and others has completely debunked PER.

    Without making this debate last longer than it needs to, I’ll simply address your comment about “dogma.”

    I think that WP is the best box-score based metric with regard to connecting individual performance to wins. I don’t think it’s perfect. I don’t think it’s a one-size-fits all. I don’t think it tells us everything about the game of basketball.

    I also think that it’s much more useful in a macro type of explanation than a micro type of explanation. To oversimplify (to the point of risking being incorrect), I don’t think WP tells us much about which team will win an individual game. Sure, you’d put your money on the team with “more productive players,” but ultimately, on any given night, anything can happen.

    But I do think that WP, over the course of 82 games or multiple seasons, can tell us a lot about the value of any particular player to any particular team. In other words, Amare Stoudamire might beat David Lee in a game of one-on-one, but personally, I’d rather take my chances with Lee (especially at his price point) as the centerpiece of my front court than Stoudamire — because I think that WP is reliable, and WP favors Lee.

    Further, I and many other WP bloggers treat the metric scientifically in that the idea is constantly being put to the test, and in some cases, refined. In other words, the scientific was and continues to be applied to the analysis, and that’s actually the polar opposite of a dogmatic position.

    I (and plenty others) am open to being wrong — if there’s something better out there, I welcome it. I just haven’t seen it yet.

    That said, take care Jon. I meant no offense, but I certainly did overreact. My apologies.

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 )

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

%d bloggers like this: