Arturo avatar

Power Rankings #8: The 50% Solution Extravaganza

(A quick note before we start, rankings are through games of Thursdary, January 23rd)

The season just lumbered past the halfway mark this week and it's really been a fascinating season so far. We seem to have gotten some things right (Lakers, Knicks, Nets, Sixers, and Heat to name a few) and some things wrong (most notably the Pacers and Blazers). That's the way it goes with forecasting.  I keep telling you all that I was going to get some of this wrong.
 
But we're here to talk about the year to date. Let's get started with a chart showing the how the rankings have evolved this season:
 
And just so you can look up the numbers better, here it is as a table:
 
 
Indiana, San Antonio, Miami, the Clippers, and OKC have been the cream of the crop this season, with brief cameo apperances by the Rockets, Blazers, and Warriors. It's interesting that a few of these teams aren't at full strength (OKC, Miami, Clips, Spurs) or are pending a move (Rockets). We know that these teams are all contending; we just don't know what their actual final playoff roster will look like. That won't stop me from speculating, though!  
 
51% of the way into the season should allow us to make some firm conclusions.

Let's get to the rankings:

BoxScore Geeks Power Ranking as of 01/23/14
Rank Change Team Point Margin Sched Rk Sched PM Orank Drank
1 Up 2 Indiana Pacers 6.8 30 1.4 16 1
2 Up 4 Oklahoma City Thunder 6.1 1 -1.5 4 3
3 Down 2 San Antonio Spurs 6.0 13 -0.2 3 8
4 Up 4 Minnesota Timberwolves 4.9 17 0.0 8 6
5 Down 3 Los Angeles Clippers 4.4 4 -0.9 5 7
6 Up 3 Portland Trail Blazers 4.2 15 -0.2 1 18
7 Down 3 Golden State Warriors 4.1 19 0.2 12 4
8 Up 7 Phoenix Suns 3.9 24 0.7 10 11
9 Down 2 Miami Heat 3.8 27 1.2 2 24
10 Up 3 Houston Rockets 2.0 23 0.5 7 12
11 Up 1 Toronto Raptors 1.3 16 0.0 17 5
12 Up 5 Brooklyn Nets 1.2 5 -0.9 14 26
13 Down 8 Denver Nuggets 1.1 22 0.4 15 13
14 Up 2 Dallas Mavericks 1.1 7 -0.5 6 23
15 Down 5 Memphis Grizzlies 0.8 26 0.7 19 16
16 Down 5 Atlanta Hawks 0.7 25 0.7 9 20
17 Up 1 Chicago Bulls -0.6 28 1.3 29 2
18 Down 4 Sacramento Kings -0.6 14 -0.2 13 30
19 Up 2 Washington Wizards -1.0 29 1.3 20 14
20 Down 1 New Orleans Pelicans -1.4 6 -0.8 11 27
21 Up 1 Charlotte Bobcats -1.8 21 0.4 25 10
22 Up 1 Utah Jazz -2.0 2 -1.2 23 28
23 Up 3 Cleveland Cavaliers -3.1 18 0.1 27 19
24   Boston Celtics -3.4 9 -0.4 26 9
25   Detroit Pistons -3.8 11 -0.3 22 15
26 Up 4 Los Angeles Lakers -3.9 3 -1.1 21 21
27   Orlando Magic -4.5 12 -0.3 24 22
28   Milwaukee Bucks -5.4 10 -0.3 30 17
29 Down 9 New York Knicks -5.8 8 -0.4 18 29
30 Down 1 Philadelphia 76ers -7.0 20 0.4 28 25
               

 

As always, the rankings include pace and schedule adjusted offense and defense. ICYMI, all the ratings went up as a visual post. Let's talk highlights with an eye towards how the postseason projects:

  • #1 Indiana Pacers Neutral Site Win:72.7% (Up 2,#16 in offense, #1 in defense): Indiana has been the team of the first half. They've consistently been at or near the top of the charts all season long, and Stephenson and George have provided a fabulous one/two punch for them. Did you know Lance has actually been the better player of the two (0.241 WP48 versus 0.232 WP48)? I would say that match up is going to be key for their title hopes. If I was them, I would much prefer San Antonio or OKC or LAC, who will happily play half court with them, over someone like Portland or even Phoenix. Right now they're favorite for the crown.
  • #2 Oklahoma City Thunder Neutral Site Win:70.7% (Up 4,#4 in offense, #3 in defense): With Chris Paul out and LeBron merely fantastic, KD, the Slim Reaper, has seized the crown as the best player in the game. He has been downright otherworldly in the past month. The elephant in the room is that with Russ out, this Thunder team looks every bit like a LeBron Cavs team circa 2009, right down to the questionable rotations and coaching. Could KD carry this team to the Finals? Yes! Particularly given the health issues of his main rivals in the West. Could they win it all? Not without Russ, I don't think. If Presti thinks Westbrook isn't coming back, he should be looking to make a deal for one more piece for this year, because they can't keep wasting shots at the title. Down that path lies KD leaving OKC like LeBron left Cleveland.
  • #3 San Antonio Spurs Neutral Site Win:70.2% (Down 2,#3 in offense, #8 in defense): Very quietly the Spurs are down three starters, Kawhi, Green, and Splitter (0.244 WP48, 0.173 WP48 and 0.175 WP48) and it's really starting to show. When healthy, this is the best team in the West, but take away their best player (Kawhi) and two other really good players and it gets ugly. That forces Tim, Tony, and Manu to play more. I think there's a very real chance we see them come back to the pack and struggle to get home court in round one, particularly since Popovich won't care. I think their season could very much resemble the Mav's 2011 year. We need to account for their #FullStrength roster come playoff time, not what we are about to see.
  • #4 Minnesota Timberwolves Neutral Site Win:66.8% (Up 4,#8 in offense, #6 in defense): The Wolves have now lost 11 straight games decided by five points or less. There's about a 1 in 2048 chance of this happening if they're a five hundred club and they've been better than that. Even with their gaudy numbers (4th overall, 8th in offense, 6th in defense) the model now thinks this team could have real issue making the playoffs. Of course, math tells us that a regression to the mean should be coming and this team should go on a run but at this point, I won't trust it until I see it (Editor Patrick: I agree. I have been burned before.) This team is weak at the small forward position; perhaps a call to the Wizards (Ariza, Webster) or the Celtics (Wallace) for a trade is in order (Editor Patrick: stop trying to dump your bad Celtics contracts on my team).
  • #5 Los Angeles Clippers Neutral Site Win:65% (Down 3,#5 in offense, #7 in defense): DeAndre Jordan being the best center in basketball (0.312 WP48) takes some getting used to. The Clips continue to be a good, solid and well-coached team that really, really misses their superstar CP3. They're holding it together without him, but they are really in danger of losing homecourt in round one and that's going to be very important in the upcoming Western Conference playoffs deathmatch. Winning a road series at the Rose Garden or Oracle? Yikes.
  • #6 Portland Trail Blazers Neutral Site Win:64.6% (Up 3,#1 in offense, #18 in defense): Given the state of the rest of the West and their level of play, we do take the Blazers seriously. It's going to have to wait for a bit, but I have some clear indicators that the Blazers are one of the smartest teams in the league.  They're also young, talented and deep. Given the injuries to the other top Western teams, there is a realistic chance that they get the one seed. They are also a bad matchup for San Antonio. The big concern for this team is a historical one. Other than Mo Williams and Robin Lopez, this team is composed of players who haven't even been out of Round one. I suspect the WCF remains their ceiling.
  • #7 Golden State Warriors Neutral Site Win:64% (Down 3,#12 in offense, #4 in defense): Curry, Thompson, Iguodala, Lee, and Bogut (0.215, 0.072, 0.260, 0.156, 0.289 WP48) are, in my mind, the best starting five in the NBA right now. As with the Blazers, we now have to consider the likelihood of them stealing a one seed in the West. The difference here is that I don't trust this roster to stay healthy (sure enough, Lee and Bogut are dinged up right now). I think they get home court in Round one. I hope it's not San Antonio in round one.
  • #8 Phoenix Suns Neutral Site Win:63.5% (Up 7,#10 in offense, #11 in defense): Phoenix will not go away. Gerald Green has been a serviceable replacement for Bledsoe so far (0.100 WP48) and Goran Dragic has been an All-Star (0.227 WP48 in a crazy amount of minutes recently). They're well coached and well balanced on both sides of the ball and their top eight in minutes are over 0.095 WP48. Hell, this team just beat down the Pacers by 24 at home. It's crazy to think that they're really too young to do much more than scare someone in round one, but there it is. I love their future though.
  • #9 Miami Heat Neutral Site Win:63% (Down 2,#2 in offense, #24 in defense): Are the Heat done? Let's quote our preseason preview: "The Heat will last as long as LeBron and Wade do...The major problem the Heat face is that their bench is really old. Allen, Battier, Haslem and Andersen are a great set of players to complement a star. The problem is that they're ticking time bombs given their age”. This seems prescient now. Here are a few things to keep in mind: 14-15 wins is all I estimate the Heat need to the rest of the way to get the two seed, and they're about 110 days away from game one of the ECF in Indiana. In the interim their only concerns are keeping Wade (...and Ray...and Birdman...and Battier) healthy and getting Oden ready. I am convinced that their plan is Cole, Wade, LeBron, Bosh, and Oden with spot minutes for Rio, Ray, Battier, and Birdman, and that we won't really start seeing it until the last month of the season. The Road to the belt still goes through South Beach.
  • #10 Houston Rockets Neutral Site Win:56.8% (Up 3,#7 in offense, #12 in defense): Patrick Beverly figured prominently in our preseason projections for this team and his return bodes well for them. The emergence of Terrence Jones (0.205 WP48) as a legit power forward does as well. The fact that he is currently their best per minute player does not. Houston is perhaps the deepest and most talented roster and they still have that Asik trade pending. With the Spurs, Thunder, and Clippers wounded, everything seems poised for them to make a charge (and at 8-2 over their last ten, they're kinda charging). I worry about their focus and their inconsistent play. At the end of the day, I feel like this is a team that needs to get punched in the face a few times before they get serious. Harden and Howard have the talent to play at an elite level but they're not there this year. They have all the pieces but I just don't think this is their year. Morey could turn it around with the right trade but as constructed, they need some seasoning.
  • #11 Toronto Raptors Neutral Site Win:54.5% (Up 1,#17 in offense, #5 in defense): Toronto continues to look like the third best team in the east. Even more impressively, they have done so through adversity and injury recently. Tyler and Salmons are key cogs for this team (0.195 and 0.125 WP48) and they were missed. I really look forward to seeing if Toronto has another trade to boost them some more. With their strong defense and improving offense, Raptor Nation should be cautiously optimistic but looking forward to getting some more solid proof in the next few weeks.
  • #12 Brooklyn Nets Neutral Site Win:54.1% (Up 5,#14 in offense, #26 in defense): The Nets (when healthy) are a decent-not-great basketball team. With the amount of money that they've spent, it would be embarrassing if they weren't. Evans continues to rise (0.150 WP48). Here's the thing though -- I don't expect them to stay healthy. They'll continue to flirt with five hundred and earn an eventual matchup with the Pacers or Heat (be it in round one or two, and yes, I'm assuming they get every call if healthy in a series against a team that is not Indiana or Miami). It will earn good ratings, but go poorly for them. 
  • #13 Denver Nuggets Neutral Site Win:53.9% (Down 8,#15 in offense, #13 in defense): Boy, the Nuggets performance sure does seem to be tied to how many minutes Faried plays (0.239 WP48). Could it be because he's their best player? I actually think this team is at a crossroads. Ultimately they're a 40 to 45 win team if they try or a 35 win team if they don't. I think the playoffs are somewhat of a longshot at this point so they might just decide to pack it in and tank. I think the Gallo season ending injury might be the final straw there.
  • #14 Dallas Mavericks Neutral Site Win:53.8% (Up 2,#6 in offense, #23 in defense): The Mavs are terrible defensively. Dramatically terrible for a fringe contender. Let me expand on this. They're a well-coached team (all the things I typically look at reflect that) but they're still terrible defensively, which is a reflection of the personnel. The simple truth is they lack size. They're 24th in Orb% and 28th in Dreb%. So they're small, can’t protect the rim and they can't keep people off the boards. They're a good offensive team but they're not good enough to overcome that flaw. They are really the most logical slot for Asik but apparently there is some bad blood there. This is a team that is just a big man away, but that one thing is enough to sink them.
  • #15 Memphis Grizzlies Neutral Site Win:52.8% (Down 5,#19 in offense, #16 in defense): The Grizzlies have done a great job of fighting to a 0.500 record after some serious calamity at the beginning of the season. Adding Miller, Lee, and Johnson to their weakest position (0.155, 0.263, 0.193 WP48) has been a masterful job. What’s the problem then? Well, Marc Gasol really hasn't been himself (0.050 WP48) and neither has Zbo (0.053 WP48). With Marc, it's explainable because of injury and you expect improvement. With Zach, it might be time to think about trades. The Grizzlies remain one of my dark horses, but I do think they need to make a move to get there. Zbo and Koufus for Tyson/Metta/Prigioni, who says no?
  • #16 Atlanta Hawks Neutral Site Win:52.5% (Down 5,#9 in offense, #20 in defense): The Hawks just cannot stay healthy. Now they've lost their backup center (Pero Antic 0.004 WP48). The funny bit is that this should be helpful as his replacements, Brand and Ayon, are significantly better than he is (0.138 and 0.164 WP48, respectively). I actually expect this team to get better, not worse. The 4/5 seed in the East apparently continues to be the Hawks' to lose.
  • #17 Chicago Bulls Neutral Site Win:47.8% (Up 1,#29 in offense, #2 in defense): Weirdly, the Bulls are going to miss Deng (0.139 WP48), but not that much. Dunleavy and Butler are more than capable fill-ins. This is also still the second best defense in the NBA. Honestly, I don't see a scenario (outside of the Bulls front office sending Thibbs on a sabbatical) where this team doesn't go five hundred the rest of the way, win their round one playoff series and make the Pacers or Heat win a war of attrition. Another trade may be coming.
  • #18 Sacramento Kings Neutral Site Win:47.7% (Down 4,#13 in offense, #30 in defense): The Rudy Gay in Sacramento bubble continues apace without collapsing (0.157 WP48). Playing off the ball is apparently helpful to him. I continue to be skeptical of his sudden flirtation with good basketball sense. Isaiah and Boogie continue to be good. The rest of this roster is not.
  • #19 Washington Wizards Neutral Site Win:46.4% (Up 2,#20 in offense, #14 in defense): Let's take a deep look at this wizards roster shall we? Wall, Webster, Ariza, Booker, and Gortat are their best five. Beal, Vesely, and Porter have potential. Nene? I'd be looking to move him. He's too old and too fragile for a team that's on the rise. I'd rather have a few younger pieces and assets that can grow with this roster or some cap room or maybe even a star for a run. Nene, Porter, and Harrington for Pau and Hill: who says no?
  • #20 New Orleans Pelicans Neutral Site Win:45.2% (Down 1,#11 in offense, #27 in defense): The Pelicans are 1-9 over their last ten. It's partly the schedule, but a big chunk of that is due to the crap on their roster. Davis and Aminu are great. Holiday is good. Gordon and Tyreke are trade bait. Rivers, Smith, and Stiesma are crap. Good for the future if they can simply figure how to draft decent, not even great players (then again, they drafted Austin Rivers, so...ugh).
  • #21 Charlotte Bobcats Neutral Site Win:43.8% (Up 1,#25 in offense, #10 in defense): The Bobcats are officially this year's good bad team. They'll be in most games, and on nights when the shots fall, they'll win because their defense is good. Their coach should be on any self-respecting GM's short-list of coaching candidates with how much he's done with so little.
  • #22 Utah Jazz Neutral Site Win:42.9% (Up 1,#23 in offense, #28 in defense): The Jazz are 7-7 over their last 14. They'll continue that improvement. Is it bad that I continue to hope Hayward winds up on the Celtics?
  • #23 Cleveland Cavaliers Neutral Site Win:39.1% (Up 3,#27 in offense, #19 in defense): The Cavs are 3-4 since trading for Luol Deng. Of course, given that they were 12-23 before it, it's an improvement. Let's be realistic here: Thompson, Varejao, and Deng are their three best players (0.118, 0.246, and 0.113 WP48). Kyrie has been a terrible disappointment this year (0.050 WP48). Our numbers have him as the 38th ranked point guard in the NBA. Let's not even get into Jack and Waiters. I'd be very seriously looking at trading any and all of my guards if I were Cleveland (including Kyrie if the offer was good enough). Boogie/Jimmer/Isaiah for Kyrie/Dion/CJ and picks: Who says no?
  • #24 Boston Celtics Neutral Site Win:38.2% (No Change,#26 in offense, #9 in defense): Hey, Rondo is back! Of course, he's still kind of rehabbing (-0.175 WP48) so the Celtics are not seeing any benefit. This team still needs a real center. Kris Humphries should also be playing way more (0.249 WP48). We still don’t know what the goal is here but the point of no return is coming very soon.
  • #25 Detroit Pistons Neutral Site Win:36.9% (No Change,#22 in offense, #15 in defense): Drummond is awesome (0.306 WP48). Everything else must go. Everything else. Of course, knowing Dumars, it's Drummond that's going to get moved. Give him a call Danny!
  • #26 Los Angeles Lakers Neutral Site Win:36.7% (Up 4,#21 in offense, #21 in defense): D'Antoni keeps finding guys off the scrap heap and turning them into players. Kendall Marshall (0.126 WP48) has been a godsend for this team (or a curse, depending on your point of view). This continues to be a terrible roster that's overachieving like mad. All GMs should keep in mind that the Lakers and Philly play at a ridiculous pace. Don't get fooled into overpaying their players next year because of it.
  • #27 Orlando Magic Neutral Site Win:34.7% (No Change,#24 in offense, #22 in defense): 1-12 over their last thirteen. Vucevic still has a "concussion". Still tanking.
  • #28 Milwaukee Bucks Neutral Site Win:31.6% (No Change,#30 in offense, #17 in defense): 2-12 over their last fourteen. Careful there Bucks, the Magic are gaining!
  • #29 New York Knicks Neutral Site Win:30.4% (Down 9,#18 in offense, #29 in defense): Andrea Bargnani is out for the next month. His minutes will be going to the Pablo Prigioni by virtue of the Knicks going small. The Knicks are about to go on a "surprising" run. Why? Because Bargs (-0.052 WP48) is not just a bad player, he's the WOAT (Worst Of All Time). There is no basketball situation in which lack of Bargnani does not make you automatically better. None. His absence will force the Knicks to start their best guard and move Melo to the four, where he actually consistently dominates. But of course the Knicks will cock it up by putting Bargs back into the rotation once he's healthy.
  • #30 Philadelphia 76ers Neutral Site Win:26.9% (Down 1,#28 in offense, #25 in defense): I still feel like this team could be way more competitive if they actually cared to be. They could put together a five hundred roster fairly quickly. They won't and rightly so. They're playing the long game and the future looks bright.

Finally, let's sim out the rest of the season. How does the model think all this plays out over 82 games?

Season Simulation as of 01/23/14
NBA Rank Playoff Seed Team Win Proj Vegas
O/U
Playoff Odds Top 4 Seed
Odds
1 Seed
Odds
Avg Seed
1 E1 Indiana Pacers 63 53.5 100% 100% 98% 1.02
9 E2 Miami Heat 55 60.0 100% 100% 2% 1.99
11 E3 Toronto Raptors 44 36.5 100% 82%   3.72
16 E4 Atlanta Hawks 42 40.0 99% 49%   4.74
12 E5 Brooklyn Nets 41 52.5 98% 38%   5.11
19 E6 Washington Wizards 39 42.0 94% 16%   6.17
17 E7 Chicago Bulls 39 56.5 94% 13%   6.31
21 E8 Charlotte Bobcats 36 27.5 69% 2%   7.87
25 E9 Detroit Pistons 32 41.0 18%     9.86
23 E10 Cleveland Cavaliers 32 40.5 16% 0%   10.04
24 E11 Boston Celtics 31 27.5 11%     10.28
29 E12 New York Knicks 28 49.5 2%     11.88
30 E13 Philadelphia 76ers 26 16.5 0%     12.95
27 E14 Orlando Magic 25 24.5       13.38
28 E15 Milwaukee Bucks 21 28.5       14.68
2 W1 Oklahoma City Thunder 60 50.5 100% 100% 52% 1.62
3 W2 San Antonio Spurs 59 55.5 100% 99% 43% 1.75
6 W3 Portland Trail Blazers 54 38.5 100% 81% 2% 3.62
5 W4 Los Angeles Clippers 54 57.0 100% 74% 3% 3.88
7 W5 Golden State Warriors 51 49.5 97% 27% 0% 5.37
10 W6 Houston Rockets 49 54.5 90% 12%   6.34
8 W7 Phoenix Suns 47 21.5 79% 5%   7.24
4 W8 Minnesota Timberwolves 47 41.0 77% 2%   7.36
14 W9 Dallas Mavericks 45 44.0 46% 1%   8.39
13 W10 Denver Nuggets 40 47.0 6%     10.34
15 W11 Memphis Grizzlies 40 49.0 4%     10.39
20 W12 New Orleans Pelicans 34 40.0       12.70
18 W13 Sacramento Kings 33 31.5 0%     12.96
22 W14 Utah Jazz 31 27.5       13.91
26 W15 Los Angeles Lakers 30 33.5       14.13
                 

 

Have a nice weekend everybody!

Arturo,
I read what you said about Minny. Can you direct me to a site that has history, as well as this season, on teams performance in close games, overtime games, when leading and trailing after 3 qtrs, that sort of thing. I certainly don't want to have to search through the game logs of all 30 teams if I can avoid it.
Ziggy
Can you explain the discrepancy between your model that show Rudbio as a pretty productive player and other models that show Rubio as one of the worst shooters in the league, like Goldsberry: http://grantland.com/the-triangle/courtvision-whats-the-matter-with-ricky-rubio/. Honestly really curious since his value seems to have the most variation depending on the rubric.

Also, Kevin Martin's turnover at the end of the Clippers game Dec 22 seems like a microcosm of their season. Probably a good place to start in trying to figure out their close game woes.
The Clippers starting 5, when healthy, is better than the warriors starting 5 at every position except the 3. I don't get how the warriors have the best starting 5.
DG22, one way to look at Rubio is the points he's involved in creating. Average PG scores 20.3 and assists 17, so 37.3 pts. Rubio scores 13.1 and assists 25, so 38.1. Advantage Rubio.

In addition, he's a great rebounder for position and an absolute thief. He doesn't take many 3pt shots, but this season he converts at 37% and he's good from the line (not that he gets to the line that often). If you're a Wolves fan, you really got a hope he can figure out how to make lay-ups, just for the additional trips to the line.
I disagree that L. Stevenson has been "better" than Paul George. PG has played more minutes and produced more wins. LS is producing more per min, but he's produces somewhat less overall (fewer WP). One might argue that LS should get more minutes. Maybe - or maybe he's playing close to the optimal rate and a heavier workload would exhaust him and diminish his effectiveness per minute. Or maybe not. Either way, it's close. Both players are having terrific years and both deserve to be start in the all-star game (as I posed here several weeks ago). Unless things change, Pacer fans can plan on a very exciting summer.
> Of course, math tells us that a regression to the mean should be > coming and this team should go on a run but at this point, I
> won't trust it until I see it

Gambler's fallacy. An old narrative, and still wrong.
DG, WP likes Rubio because he does a lot of things well, and he keeps the thing he does bad (SHOOOTING) to a minimum.

Now, given that everyone knows he's a terrible shooter anywhere inside the 3pt line, there are probably some negative effects not captured by the WP model.
I posted that Grizzlies/Knicks trade with a link here on the Grizzlies subreddit and got very, very downvoted.

Nate wrote on Sunday
>> Of course, math tells us that a regression to the mean should be >> coming and this team should go on a run but at this point, I
>> won't trust it until I see it

>Gambler's fallacy. An old narrative, and still wrong.

Except that gambling odds in games such as poker can generally be precisely calculated given certain reasonable assumptions. There's more uncertainty in sports, and so the assumptions that one has to make for precise odds calculation are less reasonable. In this case, you have to assume that productivity is evenly distributed across the game clock, but since strategy and personnel tend to change at the end of the game, it's at least plausible that there's some signal in the T'Wolves close game lose rate. It's difficult to demonstrate whether there is or not, but it's not completely unreasonable to withhold judgement on the question of whether or not the T'Wolves anomalous outcomes are completely explained by random variation.
> ....
> Except that gambling odds in games such as poker ...
> whether or not the T'Wolves anomalous outcomes are
> completely explained by random variation.

The passage in question suggests that we should expect a team which has a string of close losses to go on a run, as if there's some sort of conservation of team wins which mystically compensates teams for close losses. On top of that there's a claim that it comes from mathematical reasoning. To round it all out, it confuses the notion of 'reversion to the mean'. (And all this on a site about statistical reasoning an analysis.)

Teams that have more close losses than close wins tend to be better than their record suggests. That doesn't mean that those teams will magically (or even mathematically) start winning games. Without other insight, it's just as likely that the T'Wolves were lucky to be close as it is that they were unlucky to lose.

> ..it's not completely unreasonable to withhold judgement on the > question of whether or not the T'Wolves anomalous outcomes
> are completely explained by random variation.

I'm too lazy to see how frequent close games are in the NBA, but with 32 teams in the league, if each team has 15 close games each year, we'd expect to see a streak like that every 2-3 years on average. (Losing 9 close games is ~ 1 in 512 and 32 teams * 7 9-close-game sequences = 214 chances per year.)
This is further confused by the arbitrary definition of a close win/loss. Keeping with the wolves example: they have an OT win that falls outside of the definition. Regardless of final score, OT wins/losses should always be defined as close due to the tie score at end of regulation.

Sign in to write a comment.