Statistical Analysis. Humor. Knicks.

Thursday, October 2, 2014

Knicks Roster Analysis – Point Guards

Hi, I’m Kevin Pelton. At the risk of going all Lionel Hutz on you, you may remember me from such columns as “Page 23″ at Hoopsworld.com and such contests as KnickerBlogger’s 2004 Bloggers Bracket. Over the last couple of months, his KBness and I have shared some e-mails and AIM conversations, and I was flattered when he asked me to do a little guest blogging during his vacation. After giving him some crap about vacationing on the best day of the NBA year, I gladly agreed and offered to give an outsider’s take on the Knicks. I’m basically thinking of this as my chance to do one chapter’s worth of a Pro Basketball Prospectus-style annual.

As KB said in introducing the guest bloggers, I’m a Sonics fan, but I’ve followed the Knicks more closely than the average NBA team the last couple of years. I guess it’s the contrarian in me that makes me feel a certain kinship with a group of guys roundly criticized as underpaid. I championed the Knicks as a playoff team in my preview this year, repeatedly insisting they were better than the Celtics. Lo and behold, I nailed the C’s record and was one game off on the Knicks. Just forget the fact that both teams remade their rosters during the season.

Before we start examining the players in detail, some technical notes about the statistics I’ll be using in the statistical summary:
TS% – true shooting percentage, the best measure of offensive efficiency (PTS/(2*FGA + .88*FTA))
Reb% – percentage of estimated available rebounds grabbed
Pass – 50 * ((AST/MIN)^2)*(AST/TO)

The other measures are all derived from my possession-based rating system, which creates an imaginary team composed of four average players and the player in question. Off and Def are this team’s offense and defense ratings, Win% its winning percentage, and WARP the wins the player is worth over a replacement-level player.

Value is derived from a slightly adjusted WARP formula and uses the Marginal $/Marginal Win concept I’ve adapted to basketball from the late Doug Pappas. I only have this for last season. Salary is the player’s 2004-05 salary (from Hoopshype.com).

Without further ado. ?

Stephon Marbury

Year    MPG   PPG   RPG  APG   TS%  Reb%  Pass   Off   Def  Win%  WARP  Value  Salary
01-02 38.9 20.4 3.2 8.1 .519 4.7 5.12 93.1 91.1 .547 9.7
02-03 40.0 22.3 3.2 8.1 .520 4.6 5.06 93.4 89.9 .592 13.0
03-04 40.2 20.2 3.2 8.9 .519 4.6 7.05 93.0 89.2 .601 13.6 $11.48 $14.63

I spent the summer of 2002 “covering” the Suns for News@Hoopsworld, and the process made me a Marbury fan. That summer, Marbury was feeling the full wrath of the comparison between him and the player he was traded for, Jason Kidd. Marbury was fairly blamed for a foolish DUI, but the blame for the teams’ performance was unjustified, as it usually is. Kidd is a better player, but he’s also been the best point guard in the NBA over the last three years. It wasn’t Marbury’s decision to effectively trade Clifford Robinson for Bo Outlaw, just as Kidd didn’t draft Richard Jefferson or magically heal Kerry Kittles.

Statistically, Marbury is one of the league’s most devastating offensive forces. It’s my belief that players who are good at more than one thing don’t get as much credit for those skills as do one-dimensional players, and Marbury might be exhibit A in that argument. Last year, Marbury posted an identical assist/turnover ratio to Kidd’s and handed out only slightly less assists per minute, but anyone suggesting that they were in the same league in terms of passing would be laughed off the ‘net.

With the Knicks, Marbury drifted slightly more to the true point guard side of things, sacrificing a point per game for an assist per game, a trade-off I imagine Lenny Wilkens was happy to see him make. It’s not inconceivable that Marbury could lead the NBA in assists next season.

The concern is that Marbury gives it all back at the defensive end of the court. Hey, look, here’s a quote that says just that!

“Marbury’s one of the top 10 players on offense,” Wayne Winston, half the brains behind WINVAL, told the Washington Times. “Everybody thinks this guy is a great player. But when he’s on defense, he gives it all back.”

Indeed, per 82games.com, the Knicks were 6.7 points per 100 possessions better on offense with Marbury in the game, 5.6 points per 100 possessions worse on defense.

But is that right? Plus-minus numbers, particularly the adjusted kind WINVAL uses, are valuable, but they’re not the complete story on defense. John Hollinger reported in last year’s Prospectus that the Suns ranked fourth in defending starting point guards, and 82games.com also reports that Marbury held opposing point guards in check.

Marbury’s other big weakness is that sometimes he tries to do too much. The playoffs were the quintessential example of that; the image of Marbury forcing it time and time again in desperation against the Nets will be hard to forget (and not just because I picked the Knicks to pull the upset). Marbury put up 23 shots a game over the last three games of that series. He’s been at his best when paired with a strong power forward along the lines of Kevin Garnett and Amar? Stoudemire — and the Knicks might just have someone like that on their roster.

I think the defense requires a slight downgrade to the numbers I get for Marbury, but he’s still certainly amongst the top five point guards in the NBA and likely amongst its top 20 players. At $14 million-plus next year and for many years to come, he’s somewhat overpaid, but he gives the Knicks a star player they haven’t had since Patrick Ewing, and the price paid for him in the trade with Phoenix was worth it.

Moochie Norris

Year    MPG   PPG   RPG  APG   TS%  Reb%  Pass   Off   Def  Win%  WARP  Value  Salary
01-02 27.4 8.1 3.0 4.9 .471 6.3 4.14 89.7 91.1 .463 2.9
02-03 16.8 4.4 1.9 2.4 .470 6.7 2.32 88.4 89.7 .468 2.0
03-04 12.8 3.5 1.0 1.8 .471 4.5 1.93 87.7 88.4 .454 0.9 $2.528 $3.850

Since I’m only going back three years, Norris’ last good year doesn’t show up. The last three years, Norris has barely been adequate for a backup point guard, and last year he was even worse than that after seeing his passing and rebounding numbers tank. If there’s good news, it’s that Norris did pick up his performance after joining the Knicks in a trade for Clarence Weatherspoon, pushing his field-goal percentage from a dreadful 31.0% to 40.8%.

Most point guards come out better offensively than defensively by my system, which makes sense. With scoring and passing, most of their contributions come on the offensive end of the court. But Norris hasn’t been an efficient scorer in the last three years and has only been a good passer one of those years.

As a price for unloading Weatherspoon’s larger contract, Norris isn’t that bad, but the Knicks shouldn’t feel particularly compelled to play him, and if he’s still in the rotation next fall, that’s not a good sign.

Frank Williams

Year    MPG   PPG   RPG  APG   TS%  Reb%  Pass   Off   Def  Win%  WARP  Value  Salary
02-03 8.0 1.3 0.9 1.6 .393 6.3 4.15 86.1 90.3 .372 -0.1
03-04 12.8 3.9 0.9 2.2 .478 4.3 2.76 88.1 89.5 .432 0.5 $2.216 $0.957

Williams has just recently been discussed here, so I’m not sure entirely how much I have to add for the discussion. Unlike Dave, I wasn’t a particularly big fan of Williams in college. I recall thinking of him as an underachiever (I also abhorred Illinois teammate Brian Cook), and scoffing when people got excited about his summer-league play before his rookie seasons.

After a couple of NBA seasons, however, I have to agree with Dave that the Knicks need to keep Williams and give him more action. Offensively, Williams and Norris were similar players last season, and neither was very good. The first place there’s a difference between the two of them is that while Norris will be 31 this summer and is on the downside of his NBA career, while Williams turned 24 this season and has plenty of room to grow.

The second difference is defense. I hadn’t really investigated Williams’ defense very much before this, but there’s little question statistically that it’s fantastic. Williams’ on-court/off-court comparison is the reverse of Marbury’s – 5.7 points per 100 possessions better on defense (and also 1.4 points per 100 possessions better on offense). Williams’ individual defense also looks great; he limited opposing point guards and shooting guards both to a microscopic 10.1 PER.

Sadly, I’m going to copy Hollinger again by using my similarity scores to assess the future prospects of the Knicks’ youngsters. Williams’ closest age-24 comparable is Jeff McInnis, at the time playing limited minutes as a backup in Washington. It would take a couple of years, but McInnis eventually developed into an adequate starter. The next four names on the list — Morlon Wiley, Anthony Goldwire, Dan Dickau, and someone named Lowes Moore — aren’t as encouraging, but next after that is Scott Skiles and Sam Cassell also lurks in the top ten. So there’s some breakout potential there.

KnickerBlogger correctly points out that there won’t be a ton of minutes for Williams next season if Allan Houston is back, but what about the possibility of just cutting bait on Anfernee Hardaway? Hardaway isn’t a bad player by any stretch of the imagination, but he’s not a part of the Knicks’ future and Williams could be. I think Williams is plenty thick enough to play 20-25 minutes behind Marbury and Houston as a third guard in a three-guard rotation and that would really help the Knicks’ perimeter defense.

Kevin Pelton writes “Page 23″ for Hoopsworld.com on a semi-regular basis. He can be reached at kpelton@hoopsworld.com. Check back Wednesday for his analysis of the Knicks’ shooting guards.