March 30, 2010

Warriors SCORE Board, 3.30.10

Previous Editions:

We have updated the numbers and links at left. (It had been awhile... Coby Karl was still in there. Ewwww.) Now, we update the storied SCORE Board, with one tweak to our statistical recipe., we love you like our electronic brother from another mother, but you just don't update your Simple Rating often enough. As such, we'll now be getting our dash of plus-minus data from the dogs at Basketball Value, who revise their numbers daily. We're going to stick with the unadjusted overall figures, as several Warriors have played too few minutes to merit adjustment. There is noise in these numbers. Hopefully, the other elements of SCORE will outweigh that noise.

Notes: the NUOR (ewwww) reflects efficiency differential, not raw point differential, so the unit of measurement there is "points per 100 possessions", not "points per 48 minutes". Also, the Wins Produced data (we're using Bradford Doolittle's Wins Produced, not the Berri flavor) are oddly inflated this time around... that metric currently seems to think that every NBA player is amazing. We could toss their numbers, but the system makes a good point -- NBA players are amazing at basketball! Plus, we didn't want to have to mess around in Excel more than was necessary. For now, WP3K, you get a reprieve.

To reiterate: the proper way to interpret these results is to have complete faith in their validity and wisdom.

To the Board!

#73: GSW @ LAC 3/28/10

PostThoughts -- Warriors 121, Clippers 103
A ROAD win?! Sweet jumpin' fuck! The Dubs snapped a thirteen-game losing streak outside of Oakland in this one, and posted their first road win against a non-Nets/non-Wolves team since November.

That cold night against Dallas sure didn't seem to stick; the Warriors resumed their layup line here, and the Clips bricked a few too many to keep up. This game was very much in keeping with recent trends, as the Warriors got badly outrebounded (53-38), but scored (53.4 FG%) and passed (2.27 A/TO) with extreme efficiency. Was Monta's absence a positive? Hard to say... his net plus-minus results haven't been bad lately, and he's been around for many of the slickest offensive performances. But at the very least, one can safely say that he wasn't missed here.

Both Turiaf and Tolliver had more than twice as many points as shot attempts, which is the kind of complementary work you've gotta love. And while our man Reggie is still flying far south of his previous efficiency -- 25 points on 22 shots ain't nothin' to write home about -- seven assists against one turnover buys you a lot of goodwill around here. Curry had the kind of quiet, foul-heavy, good-but-not-great game we saw a lot from him in the season's first couple months. Given the crushing number of minutes he's received as a rookie, he's held up admirably.

As odd as it is to say about a team that's 5-30 outside of their building, the Warriors have a chance to grab another couple road wins before the season wraps up. They'll lose in Salt Lake City tomorrow, but Sunday brings a road trip replete with patsies (the Raptors, the Wizards, the Wolves, these Clippers), and the Warriors' final game is against a banged-up Blazers team that will almost certainly be resting up for the postseason. A mini-run to end the year is very possible. On a strategic level, it'd be the exact wrong time to start winning, but it'd be nice for the players to feel a little sunshine on their faces, and it'd be nice to wrap up this Nellie fiasco once and for all.

#72: DAL @ GSW 3/27/10

PostThoughts -- Mavericks 111, Warriors 90
A wild streak of high-quality offense came grinding to a halt here, with eight of nine Warriors shooting below 50% and Reggie and CJ netting a mere 13 combined points off of 28 attempts. On some level, it was just one of those nights... when Rodrigue Beaubois goes 9-for-11 from long distance, all you can do is just tip your hat to him and move on. But this game should serve as a grim reminder of the risks of predicating your attack on jump shooting. (Incidentally, the ninth Warrior did not shoot 50% or better: Ronny didn't attempt a shot in his 19 comeback minutes.)

Anthony Tolliver had an eye-popping 21 rebounds here, including an other-eye-popping eight on the offensive end. It's hard to see much upside in him, but every now and again, he has a game that makes you wonder. In a mere seven minutes, Chris Hunter contributed three rebounds, a block and a team-best +4 plus-minus, extending his strong results in the latter category. Hunter is the only Warrior who's seen the team outscore its opponents during his time on the floor this season; whether anyone in the organization is aware of that is an open question.

This was the first time the Warriors got blown out in Oracle in four months, and you never like to see your dogs go down hard at home. But when you're outstroked, you're outstroked. Hardly the most frustrating loss of the season.

March 27, 2010

The Ultimate Warrior

Last time out, we documented the Dubs' recent transition from a high-scoring, no-D team into, well, a really high-scoring, no-D team. The average Warriors game in the last three weeks has been a 124-122 loss. Several teams haven't scored or given up that many points in a single game all season... the Warriors are living there right now. The league hasn't seen a team put up numbers like those since the freak-show Nuggets of the '80s and early '90s.

Pace isn't the issue here, either. The Warriors continue to play at a league-leading pace, but their pace (100.3 possessions per contest) hasn't actually quickened lately; they're not playing faster than they were earlier in the season. The scores of Warriors games are rising because they're featured unbelievable offensive precision on both ends. The Warriors aren't shooting more often, they're making shots more often... their opponents are doing the same, to say the least.

That this streak has coincided with the rise of Reggie Williams can not be considered a coincidence. Because, make no mistake: to date, Reggie Williams has, on a per-minute basis, been one of the most effective offensive players in the league this season. His simple per-36 minute averages -- 21.2 points, 5.3 rebounds, 3.2 assists -- are nice, but they understate the quality of his contributions by a significant degree. Firstly, Reggie's True Shooting Percentage is a laughable .664... only two guys, Blazers Nic Batum and Jeff Pendergraph, rate ahead of him, and neither scores nearly as often as he does. In fact, his combo of scoring frequency and scoring accuracy is damn near historic... only ten guys have posted a .650+ TS% and a 20.0+ Usage Rate in 300 or more minutes, and six of them are named Barkley, Dantley, Gilmore, McHale, Nash and Stoudemire. Secondly, Reggie's 3.2 assists come at a cost of a mere 1.5 turnovers. The only Warrior who boasts a better assist/turnover ratio than Reggie's 2.07 mark is CJ Watson.

Essentially, Reggie has played like a more efficient Corey Maggette who can pass. And metrics that rely heavily on offensive production have noticed: Reggie's 21.0 PER would place him 22nd in the league (just ahead of Maggette, ironically enough), and his Offensive Rating of 128 (per Basketball-Reference) is higher than that of any player with enough minutes to qualify. The Dubs score 115.8 points per 100 possessions when Reggie's on the floor... only four guys have totals that high in big minutes, and they all play for Phoenix.

It's not like Reggie's just outperforming Anthony Morrow on offense. He's outperforming everybody on offense. No current Warrior -- not Curry, not Monta, not Maggette -- has ever posted across-the-board production this good. Reggie has not just been a good scorer... he's been a big-impact offensive player, a guy who systemically improves his team's offensive ability when on the floor. That's not bad for a call-up, and it suggests he'll be a big asset going forward, as long as he's not the worst defensive player in NBA history.

Sadly, we can not yet rule out that possibility. The Warriors have given up 120.3 points per 100 possessions with Reggie on the floor, an eye-poppingly horrible number... the next-worst mark on the team, Devean's, is miles behind at 113.8. No full-time NBA player has a mark worse than 115.8 (the three worst finishers, as you might imagine, are all Raptors). And thus, for all of his offensive magic-making, Reggie has not made a big impact on the team's ability to compete. In the thirteen games since Reggie signed, the team has been outscored by 4.1 points per 48 with him on the floor and by 6.5 points with him on the bench... the current Warriors are a little better when he plays, but only a little, and still well south of good. Because some of the offensive pixie dust he sprinkles on the Warriors lands on the other team, as well.

When Reggie Williams is on the floor, everybody can make a shot, and nobody can make a stop. Putting Reggie Williams into a Warriors game is like pouring gasoline on a grease fire. He's like the dudes at the end of "Desperado", the guys you call in when there's no other option but to burn the village to the ground. He has shown himself to be the living embodiment of all that the franchise loves... he is the most Warriory Warrior to ever walk the face of the earth. All hail Reggie Williams. All hail the ultimate Warrior.

March 25, 2010

Catchup: Games #65-71

Apologies for the recent radio silence. Look, sometimes fantasy baseball just happens. You've been there, dog... don't hate.

It hasn't been an uneventful time; like a bleary-eyed and recently shot Agent Cooper, we are a bit overwhelmed by the flood of news emanating from Oakland. Don Nelson has tipsily claimed he wouldn't mind coaching summer league. Raja Bell has finally relinquished his inexplicable hold of a roster spot. Stephen Curry has finally broken his "injured Warrior" cherry. Chris Cohan has finally acknowledged an interest in ending our long regional nightmare. And to top it all off, the good guys have played some more basketball games! Since we last checked in, they have

beaten the Raptors 124-112, on a rare night where both Curry and Monta excelled;
lost a 124-121 heartbreaker to the Lakers with a classic mix of heartening and head-scratching play;
topped the Hornets 131-121 behind a trio of career nights from D-League alums;
gotten stomped 147-116 by the Spurs, making their 24th straight loss in San Antonio one to remember;
lost a 123-107 snoozer in Memphis, extending their road losing streak to 13;
been edged 133-131 by the Suns, with a bone-rattling Amar'e dunk summing up the season nicely, and
won a 128-110 semi-laugher against the Grizzlies, making Memphis the first good team to lose to the Dubs twice.

This was an absurd stretch of basketball, with scads of highlights, lowlights and moments of interest, and had we not been so busy assembling a juggernaut of an f-ball squad (Lincecum, Greinke and Johan --we'll talk later), we'd have very much enjoyed putting these games under our usual nerdy microscope. But time keeps on slippin' into the future [synthesizer solo], and we mustn't dally overlong in the past. So let's examine this heptet of games as a block. What did the Warriors do well? What did they do poorly? What trends are worth noting?

March 12, 2010

#64: POR @ GSW 3/11/10

PostThoughts -- Blazers 110, Warriors 105
In some ways this was a typical loss, the kind we saw all too often in December and January: death by smallball, an open-and-shut case of bad coaching. In other ways, though, this broke the mold of Warriors futility we've seen in recent years.

As Kevin Harlan et al helpfully reminded us, the Blazers had not won in Oracle since 2004. Through the good, bad and ugly of the past six seasons, the Warriors had held serve against Portland; the offensive talent and home court advantage were always just enough to keep a decent team at bay. And that looked to reman the case through the first three-and-a-half quarters last night. As it turns out, however, this Warriors team is bound for even lower depths. Oracle is not the salve it once was... the few bright spots are winking out, one by one. This team is fucking AWWWWful.

The key numbers from last night: 43 minutes total for non-perimeter players, a 53-36 rebounding deficit, 2 poor performances from the supposed team leaders, 31 fouls (24 by the starting lineup alone), 6th straight loss, and an attendance of 17,308 and dipping, dipping, dipping.

Ladies and gentlemen, your Golden State Warriors. Pity the collegian that has to wade into this swamp next fall.

March 10, 2010

The Warriors and Truthiness

This weekend saw the largest edition yet of the MIT Sloan Sports Analytic Conference, an annual gathering of the nerds that provides a rare high-profile moment for APBRmetricians. Many of the people whose work we've linked to -- Hollinger, Simmons, Pelton, Paine, Sill -- were in attendance, as were former Dub Avery Johnson and noted Barry son Brent. Not in attendance? Anyone currently affiliated with the Warriors. Some attendees estimated that as many as 16 of the 30 NBA teams (including most of the best teams) sent representatives to the conference; no one saw any sign of an envoy from Oakland.

This is not surprising, but it is nevertheless dispiriting. Given the wealth of geek talent in the Bay Area and the storied successes of Billy Beane's analytic work just down the road, there is no team that should have cottoned to advanced statistics more quickly than the Warriors. Instead, they continue to languish in ignorance; while most intelligent front offices were keeping an eye on Cambridge, Larry Riley and his comrades twiddled their thumbs. And so, on the day that BP's Kevin Pelton announced he was now working as a consultant for the Pacers, the Warriors lost their third game in a row. On the day that the conference itself took place, the Warriors lost their fourth in a row. And on Monday, while the basketball cognoscenti buzzed over the statistical insights of the weekend, the Warriors lost number five.

This goes beyond the little on-court efficiencies that statheadedness can add. For a franchise as rudderless as this one, a refusal to reckon with the realities of the statistical record is a refusal to deal with reality itself. And indeed, we have seen some decision-making and signaling from Oakland that evokes all the intelligence and clearheadedness of the previous presidential administration. Every day, this team drives home the lesson that you can't spell "Warriors" without "W".

Let's take a trip down Memory Lane here in Worrierville and revisit some of the team's more harmful '09-'10 delusions:

March 9, 2010

D-cisions, D-cisions

This summer, whilst sifting through the wreckage of the season, the Golden State braintrust will have to fish or cut bait on several players, by either extending a qualifying offer (if not something more generous) or by letting them walk. The Warriors will clearly work to keep Anthony Morrow in hand, and will at least make an effort to keep C.J. Watson around. More interesting, if less important, will be their decisions on the D-League dudes. To hear the always-brilliant Hoopdata tell it, the Warriors would need to spend $1.25 million to retain Anthony Tolliver and $1.03 million to retain Chris Hunter. Reggie Williams is still on his initial ten-day contract; if the Dubs lock him up for the year, he'd be in line for a similar $1+ M qualifying offer.

Let's screen these scrumptious scrubs. Who should the Warriors retain? Who will they retain?

Chris Hunter
Per 36: 12.1 points (.501 TS%), 7.6 rebounds, 1.5 assists, 1.2 turnovers, 0.5 steals, 1.7 blocks, 6.8 fouls
Positives: Hunter's been able and willing to mix it up underneath, on a team that's sorely needed that -- his shot-blocking numbers are solid. His foul frequency is not necessarily a negative, as he's been explicitly deployed as an enforcer at times; if any Warrior should have a high foul rate, it's Hunter. He's passed pretty well for an inexperienced big, and he gets to the line decently often given how rarely he shoots. He seems like a prince of a guy. And his on-court/off-court stats and his RAPM showing indicate that the Warriors have been more effective on both ends when Hunter has played...
Negatives: minutes too scant and context-dependent to draw any conclusions with confidence. Things we can conclude with confidence: Hunter is neither a frequent nor an efficient scorer, and he's a significantly below-average rebounder for a center. That's not a combo you like to see.
Should They Re-Sign Him? Probably not. The plus-minus data suggests that he may be a bit better than his production, but no bigs with production like this ever rise to the level of "good". It wouldn't be a tragedy to plant him at the very end of the bench... there are worse players in the league. There are also better ways to spend a million bucks.
Will They Re-Sign Him? Doubtful. Expect his bench big spot to go to a vet, for better or (in the case of another Mikki type) for worse.

Anthony Tolliver
Per 36: 12.5 points (.502 TS%), 7.5 rebounds, 2.2 assists, 1.1 turnovers, 0.7 steals, 0.8 blocks, 2.9 fouls
Positives: Tolliver is a stretch four, an extremely valuable type of player in the current NBA. Like Hunter, he seems to be a well-liked and high-character guy; like Hunter, he fares quite well by plus-minus. And he boasts some surprisingly good results in small areas: he's done an impressive job of avoiding fouls for a big man, and his 1.96 assist/turnover ratio is the second-best on the best behind CJ's. Tolliver is a piece that seems to fit the Warriors well.
Negatives: Trouble is, he's really not much of a piece; even with his occasional big games, he posts scoring and rebounding numbers that are essentially identical to Hunter's (though as a four, the rebounding bar for Tolliver is not quite as high). Tolliver would need to be an elite three-point shooter to be really interesting, and with a .333 mark in college, a .371 mark in the D-League and a .338 mark in the NBA, there's no real indication that he is one.
Should They Re-Sign Him? An unenthusiastic yes. Floor-stretching bigs, even bad ones, have their uses.
Will They Re-Sign Him? Yes. And given the risk that he'll get playing time over superior players in Randolph and Wright, it may be a re-signing Warriors fans come to despise.

Reggie Williams
Per 36: 19.4 points (.595 TS%), 4.6 rebounds, 4.6 assists, 1.5 turnovers, 0.6 steals, 0.3 blocks, 1.8 fouls
Positives: THIS GUY'S FUCKING AWESOME!!!!!!!!!!
Negatives: It's only been five games. There's no indication that he can play any defense whatsoever. The Warriors have over eighty other swingmen, all of whom can score.
Should They Re-Sign Him? Without question. All of the quibbles above are relevant: the sample size is tiny, he looks to be a horrible defender, and he plays a position that's already laughably overstocked as it stands. But bad teams need to stockpile assets, and Williams, a guy who has excelled and over-achieved at every level in Curry-like fashion, looks for all the world like an asset. The Warriors should lock him in for next season and worry about the details later.
Will They Re-Sign Him? Probably... he plays the style that the team (over)values. However, we can't rule out the possibility that the front office will idiotically refuse to renounce the rights to Devean George or Raja Bell, forcing Williams back into the minors (and, in short order, onto another NBA roster).

#63: GSW @ NOH 3/8/10

PostThoughts -- Hornets 135, Warriors 131
A blast of a loss. That's not to say these losses don't hurt... it would've been nice to see the guys' effort be rewarded with a win at some point during this road trip. But in the scheme of things, with Wall and Turner looming ever larger, an outcome like this is damn near optimal. Morrow, Devean and REGGIE~! all had comically effective games. Curry did not, and has quietly posted five poor performances in his last six outings... however, the kid's gassed beyond belief and receiving more defensive attention than he ever has, so it's probably nothing to worry about.

Given the all-around hot shooting and decimation of the front line, it's forgivable, but worth noting all the same: the Warriors' total of 23 rebounds was the lowest total by any NBA team in a game this season. And if you're wondering, yes, dem Dubs are again on pace to be the worst-rebounding team in league history, edging Nellie's Warriors of twenty years ago.

Winless though it was, this was probably the Warriors' best-played road trip of the year: four competitive performances against good teams, with the only dud coming in one of the three or four least winnable games of the season. Thursday brings a four-game homestand that looks breezy in comparison, with three beatable opponents (Portland, Toronto, these Hornets) and zero back-to-backs. Even with lottery balls in mind, it'd be nice to see the Warriors take two of these four games. The players and fans deserve a few smiles, even if Cohan and his cohort do not.

March 6, 2010

#62: GSW @ CHA 3/6/10

Charlotte Bobcats
Record: 29-31 (20th) • Point Differential: +0.4 (15th) • Pace: 90.6 (27th)
Off. Efficiency: 103.6 (t-25th) • eFG: 48.8 (t-21st) • TO: 17.3 (30th) • OReb: 25.9 (20th) • FT/FG: 25.8 (t-3rd)
Def. Efficiency: 103.1 (6th) • eFG: 49.5 (t-12th) • TO: 16.9 (3rd) • OReb: 25.4 (8th) • FT/FG: 20.0 (t-3rd)

The Warriors' road record is what one might call "gross" -- at 4-26, they're exactly one game ahead of the Nets. Winning away from home is no easy feat for a young and undermanned team, but if you ever want to be worth a damn, you've got to break through and beat the odds every now and again. Today, against a flawed and slumping Bobcats team, seems as good a time as any. Last time the Dubs came to this building, Jamal Crawford lit 'em up for fifty.

On the other hand, the Warriors' last game against these fuckers was none too pretty. The 'Cats are a game out of eighth, and will be plenty motivated. And they excel at coaxing turnovers and fouls out of their opponents, two things the Warriors have been all too happy to relinquish. Plus, you've got to figure Jack wouldn't mind clowning the Dubs again in front of his new fanbase. The Warriors will need a very big night from someone to pull out a win here.

This game features an interesting X-factor: Curry and Morrow grew up in Charlotte. Morrow got eight shot-free minutes in the Warriors' visit last year, but both will essentially be appearing in their hometown as established NBA players for the first time. Will they excel in front of their loved ones, or buckle under the pressure?

Warrior To Watch: Stephen Curry, who has again wrested this label away from all comers for a bit.
Bobcat To Watch: Tyrus Thomas, a Randolphian enigma who has profited greatly from his recent change of scenery. We can only hope that our skinny shot-blocking goofball will not require the same transition.

March 5, 2010

#61: GSW @ ATL 3/5/10

Atlanta Hawks
Record: 39-21 (6th) • Point Differential: +4.8 (t-5th) • Pace: 90.8 (26th)
Off. Efficiency: 111.3 (4th) • eFG: 50.2 (t-11th) • TO: 13.0 (1st) • OReb: 27.6 (t-6th) • FT/FG: 21.9 (t-20th)
Def. Efficiency: 106.0 (13th) • eFG: 49.8 (16th) • TO: 15.7 (12th) • OReb: 27.0 (19th) • FT/FG: 20.5 (t-6th)

Let's see here... on the road, against a good team that's playing well and sore over losing to the Warriors two weeks ago? Yeah, this'll be a long one. At least we're likely to get some more sweet Reggie Williams action... Maggette may be back, but doesn't figure to play much.

Warrior To Watch: Stephen Curry, who has yet to put up a really effective game in this sans-Monta go-around. With Mike Bibby and Jamal Crawford guarding him, that could easily change tonight.

Hawk To Watch: Joe Smith, just for a nostalgic chuckle.

March 3, 2010

#60: GSW @ ORL 3/3/10

Orlando Magic
Record: 41-20 (3rd) • Point Differential: +5.8 (3rd) • Pace: 92.7 (t-15th)
Off. Efficiency: 109.3 (9th) • eFG: 52.6 (3rd) • TO: 15.3 (12th) • OReb: 23.6 (26th) • FT/FG: 25.0 (t-6th)
Def. Efficiency: 103.1 (4th) • eFG: 47.7 (t-3rd) • TO: 14.1 (26th) • OReb: 22.7 (t-1st) • FT/FG: 20.8 (8th)

As poetic as we've been waxing about the current Warriors lineup, there's just no way to forecast a competitive game here. The Magic are 23-6 at home, and only one of those six losses came at the ends of a sub-.500 team... the Wizards caught them on a Caron Butler buzzer-beater in a wacky game last month. These guys don't wreak much havoc on the offensive boards, but they do everything else very well. And since both Atlanta and Boston are within three games of catching Orlando for the second seed in the East, the Magic still have plenty to play for. The Warriors would need monster shooting nights from both Curry and Morrow to take this one, and even that might not do it. Tonight, the Magnificent Eight are simply outgunned.

Warrior To Watch: Chris Hunter. Someone will need to man up to Dwight Howard, and Ronny just doesn't have the beef to do it. Hunter does, and he's shown a willingness to smack big intimidating dudes. Look for a nice hard first-half foul to get the Warriors fired up.

Magician To Watch: Dwight Howard. He still isn't good at hitting free throws and passing out of double teams... that's an attackable combo. But the Warriors don't have enough bodies to play Hack-A-Howard, so he could have a pretty monster night.

The Magnificent Seven Ride Again?

The few Golden State fans that are masochistic enough to watch every game witnessed an unexpected treat last spring -- some competitive play from an unheralded and undermanned group of guys. Azubuike, Crawford, Kurz, Morrow, Randolph, Turiaf, Watson... these were not names that struck fear in the hearts of men. And yet this heptet, with occasional help from more ballyhooed guest stars like Monta and Biedrins, keyed a three-game winning streak and an inspiring win in Salt Lake City before running out of gas with two games left to go. These guys moved the ball well, they defended gamely, and they functioned well as a unit. They exceeded expectations, and were damn entertaining in doing so. Bay Area media wags dubbed this ragtag bunch The Magnificent Seven.

Something similar seems to be brewing on this road trip. Curry, George, Hunter, Morrow, Tolliver, Turiaf, Watson... it's another punchless-sounding lineup, even with a ROY contender in the mix. Combined, these seven guys make 11.1 million dollars, barely more than Monta Ellis makes by himself. But these seven rate as the most effective court presences on the team. These guys have been successful in their limited time on the court together. And last night, in a season marked by near-historic incompetence on the road, these guys were a few bricks away from toppling a likely playoff team in their building. The most fun stretch of '09-'10 Warriors basketball may be unfolding before our eyes.

In addition to the loveable CJ/Morrow/Ronny trio, this unit has four key traits in common with the Magnificent Seven of yesteryear:

March 2, 2010

#59: GSW @ MIA 3/2/10

Miami Heat
Record: 29-31 (18th) • Point Differential: +0.7 (14th) • Pace: 90.2 (28th)
Off. Efficiency: 105.5 (21st) • eFG: 48.9 (t-20th) • TO: 14.4 (t-6th) • OReb: 26.0 (19th) • FT/FG: 21.8 (22nd)
Def. Efficiency: 104.8 (10th) • eFG: 48.2 (7th) • TO: 16.0 (t-8th) • OReb: 25.1 (6th) • FT/FG: 25.8 (25th)

The Heat have lost four in a row, and have fallen into a virtual tie with the Bobcats for eighth in the conference... this is a must-win game for them. The Warriors are 4-23 on the road. The available Golden State roster tonight consists of Li'l Stephen Curry, a banged-up Ronny Turiaf, a long-dead Devean George, and five players the Warriors have acquired either from the D-League or the waiver wire.

And yet, this game could be interesting. For if you managed to slog through our previous post, you may have noticed something: besides the debuting Reggie Williams, the seven Warriors suiting up tonight are the team's seven most effective players, according to regularized adjusted plus-minus. And the most likely starting lineup tonight -- Curry, CJ, Morrow, Tolliver, Turiaf -- has been monstrously effective, outscoring opponents 147-101 in its 57 minutes of court time to date.

Now, there's some flukiness in those numbers, to be sure... a Curry/CJ/Morrow/Tolliver/Turiaf lineup probably won't average 123 points a game long-term, and especially not against a rock-solid defensive team like the Heat. There's no real reason to expect a win tonight. But the guys that will be playing during this road trip have played very, very well together. Don't be surprised if the Dubs are more competitive on this road trip than they've previously been.

Warrior To Watch: Reggie Williams, who, with Curry, gives the Warriors the leading scorers in college basketball from each of the last three seasons. He'll get plenty of minutes from jump street... can he show enough to earn himself a job somewhere in the league next year?

Heatperson To Watch: why, it's Dwyane Wade, you silly goose!

More Fun With Hoop Numbers!

Early in this blog's e-life, we looked at the regularized adjusted plus-minus numbers calculated by Joe Sill at Hoop Numbers; these are the most-adjusted bad boys you'll find kicking around the Web. Happily, Sill has provided a mid-season update, and now lists RAPMs through the games of February 25th. That encapsulates every Warriors game except for Saturday's survival of the Pistons. So it seems a fine time to take a gander at how dem Dubs are faring by this system. Specifically, we'll be looking at the four-year time-weighted data. More seasons of data help to reduce the system's margin of error... even when looking at rookies, you can rate them more confidently when you know more about the histories of their opponents and teammates.

The question this metric essentially asks: "how effective is this player in the role he's been given?" An RAPM of 1.000 means that the player's presence improves his team's point differential by 1 point per 48 minutes; an RAPM of -1.000 means the player's presence degrades his team's point differential by 1 point per 48 minutes. Those may sound like small effects, but they're not... a net point per 48 equates to about three wins over the course of a season, so in a vacuum, a 1.000 RAPM guy would be about six more wins more valuable than a -1.000 RAPM guy. The vast majority of NBA players have RAPMs between 2.000 and -2.000. In the four-year data, no player rates worse than -4.000... only two players rate better than 4.000, those being Dwight Howard (4.717) and LeBron (a brain-breaking 6.046).

Hoop Numbers helpfully splits each guy's results into offensive RAPM and defensive RAPM. We'll take a look at the offensive side of the ledger first... the "Rank" column lists each guy's overall NBA ranking in the category. We'll confine our scrutiny to the players who are still on the roster, so there'll be no beating of the dead horse that is Mikki Moore. As with any stat, grains of salt are recommended -- sample size issues exist for spot-minute guys like Hunter and Tolliver, in particular.

March 1, 2010

Lineup Lunacies

Want a shorthand way to explain how crazy things have gotten in Oakland? Don't bother looking at the Warriors' record, or rebounding futility, or fouling proclivities, or Monta's ineffectiveness, or the passel of 48-minute nights by individual players. If you want to get a vivid sense of Don Nelson's "Beautiful Mind"-style devolution into basketball madness, just take a gander at the Warriors' list of starting lineups over the last two seasons.

Now, injuries have, to be sure, hampered Nellie's choices. And tweaking lineups is nothing new for the big fella. After taking over for Jim Clemons in Dallas in 1997, Nellie used 35 different lineups in the team's final 66 games... even in '07-'08, an unusually healthy and stable season in Oakland, Nellie used 23 different lineups, one of the league's higher figures that year. He's nothing if not a tinkerer. But there's a difference between tinkering and what we've been seeing for the past sixteen months. The 2008-09 Warriors fielded 47 different quintets to open games, easily breaking the previous NBA record of 42 (recorded in Larry Brown's surreally awful season with the Knicks). Nellie only used one lineup -- Monta/Crawford/Jack/Buike/Beans, if you're curious -- as many as five times, and he used that lineup only five times.

This year, things have gotten even crazier. In their 58 games, the Dubs have already used 34 lineups -- no other team has used more than 25. The Warriors have fielded only one starting lineup more than three times, and eight of the ten starts of that lineup (the grim Curry/Monta/Morrow/Vlad/Mikki quintet) were commissioned by Keith Smart. Which means that -- yes, your math is correct -- Don Nelson has not used a single starting lineup more than three times this season. He's used four different lineups thrice, six different lineups twice, and 24 different lineups only once.

In his 48 games, Nellie has used 34 different starting lineups. In the last 13 games alone, he's used 12 different starting lineups, more than half the teams in the league have used all season. Nellie has used two starting lineups (Monta/Morrow/Buike/Jack/Mikki, Curry/CJ/Cartier/Maggette/Vlad) that share zero players in common. And seventeen different players -- everyone who's played more than 67 minutes this season -- have started at least once under Nellie.

The Warriors' injuries, plus the trade of Stephen Jackson, created unavoidable turmoil... circumstances have necessitated some lineup changes. They have not necessitated this. If Nellie had opened the box and done something crazy like playing the best available player at each position every night, the Warriors would've used a total of only fifteen lineups thus far. There was never a game where Nellie needed to start Coby Karl or Cartier Martin or Devean George or Chris Hunter. There were only two games on Nellie's watch where there was a compelling reason to start either Vlad or Mikki, and there was never a compelling reason to start both.

Starting lineups do not tell the whole story, and smarter use of them wouldn't be any kind of panacea. But you simply can't look at the Warriors's starting fivesomes over the last 140 games and find any semblance of a coaching plan. Simply put, Don Nelson is bored, and has been screwing around to amuse himself. With a new scorer aboard, the madness won't be ending anytime soon.