Statistical Analysis. Humor. Knicks.

Monday, July 28, 2014

Changing the Playoff Format

If there’s one thing that stuck out in my mind when making my first round picks for ESPN’s Geek Smackdown, it’s that the East largely consisted of lopsided battles while the West had a couple of really close series. All 8 Western teams finished with a winning percentage of .600 or higher, while only 3 Eastern teams bettered that mark for the regular season. Consequently a couple of Western conference series are of second round quality such as Houston/Utah and San Antonio/Phoenix. And it doesn’t seem fair that a team like Golden State which won 48 games in the more difficult West is watching the playoffs from home while 5 Eastern teams that won fewer games are still playing.

So I wondered what would happen if the NBA discarded the East/West distinction and seeded teams solely based on winning percentage? This year’s playoffs would look like this:

“Russell”
1. Boston vs 8. Philly/Portland
4. Phoenix vs. 5. Orlando

3. San Antonio vs. 6. Golden State
2. New Orleans vs. 7. Cleveland

“Chamberlain”
1. Detroit vs. 8. Toronto
4. Utah vs. 5. Dallas

3. Houston vs. 6. Denver
2. L.A. Lakers vs. 7 Washington

The first thing to notice is the absence of the weaker Eastern teams. Namely Atlanta replaced by Golden State, and Philly needing to beat Portland in a “play-in” game to reconcile a tie for the league’s 16th best record.

In this format, the best first round matchups become the Suns vs. Magic and the Hornets vs. Cavs. Both seem odd when thinking in terms of today’s NBA playoff format, as they are East-West battles. However the series have an intrigue of their own. Orlando’s Dwight Howard, who is currently putting up Russell-esque rebounding numbers against Toronto, would be competing against two huge centers in Shaq and Amare. Meanwhile LeBron wouldn’t have a home court first round matchup as a reward for winning a mediocre 45 games. Instead he would face one of the toughest Western teams in New Orleans.

Looking past the first round yields some better contests than the current format. In this year’s playoffs, the Celtics won’t face a serious challenge until the third round (sorry Cleveland fans). But in my format, they’ll face either Phoenix or Orlando in the second round, and the winner of the Spurs/Hornets in the third round. Meanwhile the “Chamberlain” bracket would be even more competitive, with the Pistons, Jazz, Rockets, and Lakers battling to emerge to the Finals.

Another advantage of this playoff format is that entire conferences wouldn’t be aiming to beat a single team. For a few years, teams in the West sought only to find a way to defeat the Lakers or Spurs. However in a format where teams wouldn’t be assured of going through a certain team to get to the playoffs they can risk finding a different method to win games. Western teams might not have had to waste roster spots to stock their bench with bigmen to counter Duncan or Shaq, and instead become more well rounded to account for any opponent strength.

Additionally teams for the weaker conference wouldn’t be as complacent. This year Eastern conference teams could aim low and still reap the monetary rewards of a couple of playoff home games. Maybe one contributing factor to the East’s recent futility has been forcing 8 teams to go to the playoffs. A sub .500 team will be better off the next year by going to the lottery, instead of the playoffs. And competing against stronger Western teams might force GMs make better moves to improve their teams for a playoff run.

Ultimately this system would correct the flaws of the current system of allowing for bad teams to make the playoffs. Combined with a shuffling of the league’s top teams a non-conference playoff format would make a more balanced playoffs.

55 comments on “Changing the Playoff Format

  1. Matthew

    Hi there – some interesting points and long overdue. The current format has been woefully inept at providing a balanced play-off series that is comparable to how teams performed in the regular season. G. State’s omission this season proves this.

    However, is it not also true to say that Boston’s 66 wins are not as hard-earned at Laker’s 57, given the caliber of teams that Boston often faced? In your format, Boston and also Detroit would then be an undeserved 1-seed I believe.

    One other small gripe could be the extensive travel should a team like Miami / Orlando play Portland / Seattle. Or, if the Knicks ever make it again, playing Lakers or G. State?

    Views anyone?

  2. jon abbey

    yeah, it still wouldn’t be totally fair because of the inequal competition, but it’d be a lot fairer than the current system. I’ve been calling for this for years, I think it actually has a shot at happening this offseason because the West is only going to be better next year, with Oden coming.

    you could get rid of the travel issues by doing the series 2-3-2, but the first round is so spread out between games that it shouldn’t matter much.

    all four West series could easily end in five games or less, that would shatter the conventional wisdom about how competitive those series were all going to be. great win for the Hornets tonight, Julian Wright looked awesome coming off the bench.

  3. W.C.

    I apologize for going off topic in this thread, but I am certain I’ll get the answer I am looking for here.

    When a player gets fouled while shooting, I assume if he misses the shot, the miss still counts against him statistically?

    The reason I ask is that I think I have been evalauting some players improperly. There are some players that drive to to the hoop and draw a lot of fouls. If this happens a few times per game and they miss bcause of being fouled, it would tend to lower their FG% and even potentially make it look like they must take a lot of bad shots (if I know that they are good 3pt shooters for example).

  4. caleb

    The travel would be an issue… though maybe not insurmountable. But I guess you could put me in the curmudgeon bracket — I do not want to see the format changed so dramatically. The system you propose is certainly more fair, but I don’t like it for the same reason I don’t really like interleague play in baseball — it used to be a special occasion, but when league/conference distinctions are erased, it just seems to take some personality out of the game.

    In the spirit of being totally inconsistent, I WOULD change the seeding to eliminate all division distinctions — teams in each conference play exactly the same schedule, so why should a division title mean anything in the seeding?

    Along those same lines, I think a lot of people (like Matt) overestimate the difference in schedules between East and West. I think at least 72 games are identical for every team (each time in your conference 3 times, each team in the other conference 2 twice) — so you’re only talking 10 different games. Even in a year as lopsided as this one, that wouldn’t make more than a 1-game difference on SOS — 2 at the absolute max.

    FWIW, I don’t see why the current imbalance will last – there’s no inherent reason for it.

  5. caleb

    Also off-topic — Dave Berri has organized most of his final player rankings for the year, in one place:
    http://dberri.wordpress.com/

    At this point, he’s only listed top 15 at each position. Without going back to look at past years, it seems the “Wins Produced” model gibes more closely with conventional wisdom this year. For example, his 1st team all-NBA is Chris Paul (his MVP), Kobe Bryant, LeBron James, Tim Duncan (though Garnett rated higher per-minute) and Dwight Howard.

    The only Knick to make a top-15 was, no surprise, David Lee, who comes in at #7 on the PF list.

  6. Rob

    I know alot of people site the “travel” as the biggest hurdle, but if a team can fly cross-country for Finals games, what makes a 1st or 2nd round series any different? During the season, teams are on the road for a week or more at a time. It would need some working out, but its not insurmountable.

    My biggest gripe would be the diminishing of in-conference rivalries. Wiz/Cavs, Sun/Spurs…or going back to Knicks/Heat or Knicks/Pistons. With a revised playoff format, it would be far less likely to happen.

  7. W.C.

    caleb,

    I have a really tough time believing any statistical model that rates Jason Kidd as the 2nd best PG and 3rd best player in the league right now.

    Kidd had some terrific years, but he was never much of an outside shooter. This year however, it has gotten so bad teams just lay off him hoping he shoots from the outside. That has taken a lot away from his ability to create because he can’t get inside as easily and draw other players to him. He’s usually smart enough not to shoot from the outside though because he knows he’s terrible. He looks for a teammate instead.

    IMHO, Kidd has actually been exposed a bit by the trade to Dallas. In New Jersey he had two incredible athletes to work with (Jefferson and Carter). Having those two running up and down the court and leaping above the rim was probably worth about 2-3 assists a night that are tougher to come by with more typical teammates. Before Carter came to town, he had Menyon Martin and Jefferson doing the same thing.

    IMO, right now he’s no better than a competent PG. NJ robbed Dallas. They sent a formerly top notch PG with a wildly enhanced reputation on the downhill slope to Dallas for a younger cheaper player that is already better and on his way up.

    I don’t care what the stats say. Just watch the games. Kidd doesn’t contribute much anymore.

  8. jon abbey

    yeah, that Kidd ranking is hilarious. check out his line from last night, he was one of the worst players on either team in a must-win game for Dallas. maybe Cuban is a Berri fan, but the deal for Kidd was one of the worst deals in recent memory, one that not only hurt Dallas in the long-term, but in the short-term also (they’d be better off with Harris and Diop right now).

  9. caleb

    ” don’t care what the stats say. Just watch the games.”

    I’m not going to spend time arguing the merits of Jason Kidd, who to my mind is obviously not the 2nd best point guard in the league — any more.

    But it’s silly to discount something based on a handful of examples that seem out of whack to you, or to anyone. That’s too ludicrously easy for any any statistic.

    But as Dave Berri himself likes to point out, if you only watch the games you miss 99% percent of what 99% of the playesr are doing at any given time. (unless, I suppose, you Tivo everything and spend every waking hour watching the replays — in which case you might only miss 85 or 90 percent. Oh yeah, except the playoffs, where less of the action overlaps). Statistics might be imperfect or not tell you everything, but they’re certainly a better tool than just sitting down and watching a handful of games. I think it’s true for a lot of reasons, but if nothing else, a larger sample size is always better than a smaller one. I’m not the first person to have explained this to you, seems to have not sunk in, on previous threads…

    Anyway, it’s hard to not at least give serious thought to a method that a) is clearly well-designed; b) generates results that largely match what people “see” intuitively — just look at the 5 best-ranked players this year.

    A more interesting question is: why does this rate Kidd so highly? One not-so-obvious reason: it’s a season-long number; it gives equal weight to early season games, and none at all to playoff games where you’re being schooled by the best player in the league. You’ll recall JK putting up a lot of November-December triple-doubles. Or maybe this: WP seems to look kindly on players who do a lot of things well, rather than a few things very well. Or maybe this: the “team defense” consideration is not enough — WP isn’t responding enough to Kidd’s deterioration on defense. Etc.

    Is there a better stat? I dunno, maybe.

  10. caleb

    p.s. lest this turn into a Jason Kidd debate, You’ll see I called it one of the worst trades ever, the day it happened — an even-up trade of PGs, except one guy being 10 years younger… and then throwing in a top defensive center and two 1st round picks. Wow.

    Chris Paul is awesome, but do you think he’d be having the same series if Harris was checking him, with Diop guarding the basket?

  11. jon abbey

    FWIW, Paul was incredible in the two games in NO, pretty bad in game 3 and adequate last night, they won without him doing too much (Julian Wright was incredible, like a bigger Balkman with a jumper). also, Kidd hasn’t been guarding Paul, Jason Terry mostly has.

  12. villainx

    I’m fine with the playoff the way it is. You have divisions and conferences for a reason. And there is also the draft.

  13. jon abbey

    “You have divisions and conferences for a reason.”

    not a good reason, though. it’s a huge problem in each of the three major sports right now.

  14. MJG

    Doesn’t it seem like changing the playoff format is a permanent solution to a temporary problem?

    Also, the only reason you would make this change is to restore competitive balance, right? But in this model we’re pushing out a young, fairly interesting Atlanta team that took a game from the Celtics. At the same time, the allegedly hyper-competitive West playoff games have been blowouts. (There’s been maybe 2-3 good games out of the bunch, and every series is either 3-0 or 3-1.) I like Golden State as much as the next guy, but, again, in this format they start off against the San Antonio Spurs. Anyone think they have a chance in that series? How about Boston v. Portland? Not sure this is much of a solution.

  15. Thomas B.

    Warning. Sarcasm Alert. Warning. Sarcasm Alert.

    Well it seems that Tim Duncan made the list of top POWER FORWARDS. How could that be? I read so many posts in which posters called Mr. Duncan a center while I argued that he is a PF. How about that?

    Lyndsey Nagle: Do I detect a note of sarcasm?
    Frink: (With sarcasm detector) Are you kidding? This baby is off the charts mm-hai.
    CBG: A sarcasm detector, that’s a real useful invention. (Sarcasm detector explodes)

  16. caleb

    A non-sarcastic note… since Berri used the PF position adjustment, instead of the center adjustment… unless I’m wrong, that gives Duncan a slightly better number than he would have if Berri had considered TD at his actual position of Center. (I don’t know the specifics of his adjustment, but since he considers centers to be the most valuable position, I would think that a center’s numbers would be adjusted “downward” more than those of a power forward…

  17. Sean

    W.C.: I am nearly positive that missed field goal attempts are NOT counted in the stats when the shooter is fouled. I searched the NBA on-line rulebook and couldn’t find it definitively, but that has been my understanding in 26 years of following basketball.

  18. W.C.

    Caleb,

    I don’t have a better model for evaluatiung players using one “know it all ranking formula”.

    Personally, I look at all the same individual stats used in a variety of these models. Then I look at the team they are earned on, the types of players on the team, the pace of their games, and I watch some of the games.

    I think I often come to the same conclusion as the models, but sometimes I don’t and feel very confident I am right.

    I just think the Kidd ranking is silly. That’s why I brought it up. He’s probably getting some extra points for rebounds relative to most PGs, but IMO some rebounds aren’t very significant from that position because if he didn’t get them, one of his bigger teammates often would. So they really don’t have that much value.

  19. W.C.

    >>W.C.: I am nearly positive that missed field goal attempts are NOT counted in the stats when the shooter is fouled. I searched the NBA on-line rulebook and couldn’t find it definitively, but that has been my understanding in 26 years of following basketball.<<

    Thanks.

  20. Mike K. (KnickerBlogger) Post author

    “Also, the only reason you would make this change is to restore competitive balance, right? But in this model we’re pushing out a young, fairly interesting Atlanta team that took a game from the Celtics. At the same time, the allegedly hyper-competitive West playoff games have been blowouts. (There’s been maybe 2-3 good games out of the bunch, and every series is either 3-0 or 3-1.) I like Golden State as much as the next guy, but, again, in this format they start off against the San Antonio Spurs. Anyone think they have a chance in that series? How about Boston v. Portland? Not sure this is much of a solution.”

    No team would matchup well against Boston in the first round, unless the playoffs were reduced to about 4 teams.

    It’s not that the first round series would get much better. In fact this might make them a tad worse. It’s that the rounds after would be superior AND that the right teams make the playoffs.

    I’d bet if you gave a GM the option (with money not having a bearing) if he’d rather make the playoffs as an 8th seed or move up two spots in the draft…

  21. Owen

    Caleb – Duncan does get a small bump from being rated a PF.

    As for Kidd, I don’t think its silly to think that Kidd is a great player. Dallas’ efficiency differential improved by two points after he arrived. However, they still are not a better team than New Orleans. And that’s largely because of Paul, who had the best season of any player in the NBA this year.

    People often say that rebounds that point guards get are usually rebounds another player on the team would have gotten. I think this is totally incorrect. A lot of rebounds they get are long rebounds that would have gone to the other team. And I remember reading a study that concluded that rebounding from the point guard position actually had one of the highest correlations with winning of any stat from any position.

    Jon Abbey – On Berri’s list of top shooting guards you will notice Monta Ellis at the number three spot, something which you predicted and which I did not see coming. I thought he would be a Ben Gordon type, but he morphed into an outstanding shooting guard this year. So a tip of the cap. Going to be a very interesting contract discussion.

  22. jon abbey

    “Doesn’t it seem like changing the playoff format is a permanent solution to a temporary problem?”

    no, it would be a solution that would work whether the conferences were lopsided or they were balanced. I’d also be in favor of a truly balanced schedule, but that’s probably too much to ask. FWIW, I’m pretty sure the NHL did things this way at one point, seeding teams from 1-16 leaguewide.

    “Jon Abbey – On Berri’s list of top shooting guards you will notice Monta Ellis at the number three spot, something which you predicted and which I did not see coming. I thought he would be a Ben Gordon type, but he morphed into an outstanding shooting guard this year. So a tip of the cap. Going to be a very interesting contract discussion.”

    yeah, he had a remarkable season, my favorite player this year. one thing that’s incredible about him is that even though he’s a skinny guy who drives inside constantly, he somehow never gets hit hard, unlike Wade or Iverson or Kevin Johnson, which is good, as that can easily take its toll on one’s career.

  23. W.C.

    >>As for Kidd, I don’t think its silly to think that Kidd is a great player. Dallas’ efficiency differential improved by two points after he arrived.<<

    This further demonstrates that statistics are often very misleading. IMO no one in their right mind believes that Dallas played better ball after the trade. They were desperate to make the playoffs and couldn’t beat a decent team.

    It’s hard to use a statistic to make a point and then use another statistic to support the first one when the very basis of the opposite point of view is that very often statistics do not reflect reality.

    Dallas clearly got worse after the Kidd trade.

    There are also specific things you can point to about his game that explain why. The first and foremost is that he can’t hit an outside shot and no longer even seems to have the confidence to take an important one. He often defers even when wide open on the outside and can’t create easy baskets for himself at all relative to his better days.

    He also can’t create for guys like Dirk and Howard the way he could for athletic freaks like Jefferson and Carter. Those are players that actually create assists for their PG (a stat that doesn’t exist as far as I know) and make them look better than they actually are.

    I’d be willing to bet that Devin Harris’s assists go up for NJ next year by even more than they did in the brief period he was there this year (and playing at less than 100% for part of the way).

  24. W.C.

    I think it’s very difficult to construct a system that ensures that the best 16 teams are in the playoffs when all the teams don’t have the identical schedule. I realize that there’s only a 10 game difference, but if one conference is clearly superior it’s going to matter because the final playoffs spots are usually seperated by only a game or two.

    I think we have to live with the fact that the best 8-10 teams are always going to get in and the chances of one of the 15 or 16 seeds like a Golden State or Portland etc.. winning it all are close to nil.

    It just so happened that Golden State got eliminated and they are a very entertaining team to watch, but it didn’t have to be that way. The 8th team in the east could just as easily have had some young superstar that we were all dying to see sneak in and the west could have a real dud of team that got eliminated.

  25. W.C.

    The clarify further, typically teams on the bubble in the better conference are going to be the teams that will tend to get eliminated if you just use records because they would have had a tougher schedule.

    What happened this year is actually probably a lower probability event because the conference with the better upper end teams also produced a 9th seed with a better record than the weaker conference. That may have happened partially because the very worst 4 teams in the west were worse than the worst of the east.

  26. caleb

    “>>As for Kidd, I don’t think its silly to think that Kidd is a great player. Dallas’ efficiency differential improved by two points after he arrived.<<

    This further demonstrates that statistics are often very misleading.”

    Not at all, the improved offense was outweighed by the crappified defense (partly dropoff from losing Harris, partly dropoff from losing Diop).

    I don’t see the playoff thing being a high-stakes issue – I mean, the best teams make it. I guess re-seeding everything would help eliminate 1 or 2 really bad teams most years (a couple extra this year), but not sure that makes much difference.

    I don’t know that most fans feel this way, but if I were in GS I’d almost be happier having a shot in the lottery, than lining up to get creamed in the 1st round of the playoffs. If I should happen to win a top-3 pick, I’d definitely prefer that to a 1st round wipeout.

  27. J

    First of all, Portland does not have the same record as Atlanta. In this format Atlanta would be out and Portland and Toronto would be in a tie breaker situation for the 15th and 16th seeds.

    Only 58 games are the same, not 72.

    Furthermore, the 4 worst teams in the West were not worse than the worst of the East (Miami and New York with Milwaukee only about 3 games better).

    The West is usually quite a bit better than the East and it will only get worse next year if Oden is the real deal and if Sacramento and the Clippers are actually healthy.

  28. caleb

    “Only 58 games are the same, not 72.”

    You’re right J, thanks for catching it.

    Still, even in this most lopsided year, the East was 582-648 (.473); the West was (obviously) 648-582 (.527).

    A difference of .054 in your opponent’s winning percentage over 24 games would lead to, one would predict, a 1.3 game difference in a team’s overall record (depending on whether they played in the tough West or weak East).

    As for the trend… I just think it tends to even out. Portland will be busting onto the scene, but Denver, Dallas, San Antonio and Houston are at risk for declines…

    Meanwhile, in the East, more teams are likely to improve, than get worse — if for no other reason than that they’re bad to start with.

    Overall, the draft helps smooth things out (over time)… in theory, a conference with more attractive FA destinations would have a long-term advantage, but that’s not really the case… Orlando and Miami sort of balance out Cleveland and Milwaukee, the way Utah, Memphis, New Orleans and Oklahoma City (yikes!) balance out the attractions of LA.

    I do think the East has had more than its share of mega-problems, in New York, Chicago and Atlanta… but even that won’t last forever.

  29. jon abbey

    you can’t quote overall stats regarding the Kidd deal, look at the breakdowns against playoff teams (they’ve been awful) and against everyone else (they’ve been dominant). again, he was one of the worst players on the floor last night, for either team, starters or bench.

    the main reason the playoff seeding needs to be fixed is the regular season. Eastern teams have spent years now coasting for months, just waiting for the playoffs (Cleveland last season is a great example). Western teams, especially this year, have had to play their asses off every single game or risk not even making the playoffs. that’s an extremely unfair advantage in the Finals, and is the main reason that Detroit won a title a few years ago. if you seed 1-16, everyone would have to play roughly equally hard all season long.

  30. Owen

    Honestly, I am not sure Kidd has been the real problem for Dallas. His performance in the series hasn’t been terrible. The contrast with Paul is what makes him look bad, although Paul has cooled off substantially in the last two games.

    I think you would have to lay it at the feat of Josh Howard (15-58) and Avery Johnson.

  31. jon abbey

    Josh Howard has been awful, no doubt. but Kidd was brought in to try to take them to the next level, and they’re 4-13 against playoff teams since he showed up.

  32. cwod

    I agree with the Kidd detractors. That was a dumb trade. Even if it worked, it probably wouldn’t have pushed them over the top, and now they’re kind of screwed moving forward.

    Look at Kidd in this series. He can’t and doesn’t shoot. NO doesn’t have to respect his shot because he refuses to take one. They can wait for Howard to get the ball and miss.

    Kidd’s line for the last game was embarrassing. It’s like he became the PG equivalent of Jared Jeffries or something.

  33. W.C.

    Caleb,

    >Not at all, the improved offense was outweighed by the crappified defense (partly dropoff from losing Harris, partly dropoff from losing Diop).<

    With all due respect, even if you are right about a drop in defense (which I don’t see other than perhaps Diop), the offensive chemistry has been pathetic primarily because Kidd can’t hit the ocean and defenders know that. So they are dropping back off him when he’s outside. That takes away his penetration and ability to create for others. He also doesn’t have any flying Walendas on Dallas like Jefferson and Carter to pass to underneath. Other than not keeping Paul under control in game 1 and 2, it has primarily been the offense that’s the problem….stats or no stats. Just watch the games.

  34. W.C.

    >A difference of .054 in your opponent’s winning percentage over 24 games would lead to, one would predict, a 1.3 game difference in a team’s overall record (depending on whether they played in the tough West or weak East). <

    I don’t think this is a correct assumption. It’s not like the entire east is uniformly worse than the west. There are individual team issues to consider.

    Teams like Boston and Detroit could theoretically be the 2 best teams in the whole NBA and produce a very high win percentage against the west. They contribute to that not so bad .473 over record against the west.

    If so, that means that the rest of conference did much worse than .473 and those are the teams on the bubble. So it was a bigger advantage for them to not play the west as often.

  35. W.C.

    Here’s a Knicks hypothetical (I’ll leave the Kidd debate alone).

    I’m going to start with the assumption that the Knicks have the same team as last year. I think with 2 minor tweaks they can win between 5-10 more games next year.

    Here’s the starting line up.

    PG Marbury
    SG Crawford
    SF Chandler
    PF Randolph
    C Curry

    Strategic change:

    The green light is taken away from Crawford and the offense is run through Curry again the way it was in the prior year. So basically we are lowering Crawford’s shots per game, raising Curry’s shots per game and leaving everything else the same other than substituting a NBA player for a D-league player at the SF.

    I think the combination of a more shots from Curry and less from Crawford (which can be executed because we’ve already done that with Curry and his FG % remained high), a higher FG% (and other things) from Chandler than Qrich, and Marbury at the PG healthy and playing for a new contract is worth almost 10 games.

    Throw in some good coaching.

    Throw in a good draft pick.

    Throw in a positive trade of any kind.

    And I think the Knicks are going to be a lot better next year than we all think.

  36. caleb

    “stats or no stats. Just watch the games.”

    You’re mixing apples and oranges. It’s an objective fact that Dallas’ offense was better after the Kidd trade, despite his inability to shoot.

    On the other hand, it’s obvious that they’ve sucked the big one against the Hornets. No big surprise.

    “Teams like Boston and Detroit could theoretically be the 2 best teams in the whole NBA and produce a very high win percentage against the west. They contribute to that not so bad .473 over record against the west.

    If so, that means that the rest of conference did much worse than .473 and those are the teams on the bubble. So it was a bigger advantage for them to not play the west as often.”

    Oh, I see, the East winning percentage of .473 is “misleading” because two of the teams were really good. How could I not see that?

    We already know that most of the very good teams are in the West. But do Western teams really face tougher schedules than the East? Barely. Sure, they play more really good teams, but they also play 4 truly pathetic teams (Clips, Seattle, Memphis, Minny) instead of just 2 (NYK, Miami). But don’t take my word for it — look at the winning percentage statistics you just quoted.

  37. caleb

    “you can’t quote overall stats regarding the Kidd deal, look at the breakdowns against playoff teams (they’ve been awful) and against everyone else (they’ve been dominant).”

    Before you accuse me of defending Jason Kidd, I thought it was an awful trade in the first place, he’s been awful in the playoffs, yada yada….

    But you’re talking about a pretty small sample size. I’d guess the good team/bad team differential is a fluke. In general, I think teams get equally better or worse across the board – if they improve their percentage .100 against good teams, they do the same against bad teams. (same idea if they get worse).
    Do you think there’s something specific about Kidd’s game, or Dallas’, that makes them uniquely unsuited to play good teams. What do you think that would be?

    Here’s an example, from baseball… Bill James talked about a team (I think it was the Royals) that had a great year, but lost in the playoffs. They were known for being extremely aggressive baserunners. In the playoffs, a ton of their rallies ended when guys were caught stealing or were thrown out trying to take an extra base. James thought the aggressive strategy worked well against bad defenses, but against good teams, it backfired.

    I guess there could be a similar example in basketball.

    But it’s basically the same question as this: what’s a better indicator of team quality: won-loss record, or point differential? Most research comes down on the side of point differential…

  38. caleb

    “I think with 2 minor tweaks they can win between 5-10 more games next year.”

    IMO it’s even easier — if the new coach plays Lee & Balkman starter’s minutes, they’ll win another 5-10 games. This year’s team, with a healthy Marbury, should have been threatening the Hawks – the roster wasn’t nearly as bad as Seattle or post-Gasol Memphis.

    The rest of the team is young, so should improve slightly, naturally. If Marbury makes a decent comeback, that’ll be another couple of wins.

    I don’t think the draft pick will make much difference in the short run. Even if it’s a future star like Derrick Rose, he’ll be an awful NBA player next year.

    37 wins, here we come!!!

  39. jon abbey

    “Do you think there’s something specific about Kidd’s game, or Dallas’, that makes them uniquely unsuited to play good teams. What do you think that would be?”

    well, Kidd’s ideally suited to running the fast break, and he doesn’t have great wingmen to do this with in Dallas, so that’s got to be part of it. more than that, I don’t really know, basketball chemistry is hard to figure out sometimes, and really hard in most cases to add a huge piece midseason and expect to compete at the highest level. I don’t think close to half of a season against playoff teams is an especially small sample size, but to each their own.

  40. W.C.

    >Oh, I see, the East winning percentage of .473 is “misleading” because two of the teams were really good. How could I not see that?

    We already know that most of the very good teams are in the West. But do Western teams really face tougher schedules than the East? Barely.>>

    You are missing the point.

    The premise has been that the difference between the .473 and .527 win percentages is not that significant (maybe worth 1.3 games).

    I am suggesting that that analysis is not correct because the 2 best teams were probably in the east.

    If you subtract out impact of Boston and Detroit from the overall .473 and .527 stats (because they did great against the west) it shows just how inferior the “rest of the east” was to the west.

    Teams in the east only play Boston and Detroit 6 times, but they play the whole west a lot. So the from avoiding games against the west is more than 1.3 games.

    If the west was uniformly better than the east from top to bottom, looking at the overall win percentages of conference vs. conference would be more correct. However, in the case of the two best teams being in the east but the west being better overall, that analysis underates the impact of the different schedules.

  41. W.C.

    >>You’re mixing apples and oranges. It’s an objective fact that Dallas’ offense was better after the Kidd trade, despite his inability to shoot. <

    I have been suggesting one very simple thing from the start.

    Any stat that suggests that Jason Kidd is the second best PG in the league, 3rd best player in the NBA, or that Dallas’s offense improved when he came to the team is flawed because a visual analysis of the team’s performance suggests he is one of the problems and the team’s win loss record against decent teams supports the more subjective analysis.

    If you would like to believe the numbers, you are entitled, but they are wrong.

  42. W.C.

    >well, Kidd’s ideally suited to running the fast break, and he doesn’t have great wingmen to do this with in Dallas, so that’s got to be part of it. <

    Exactly.

    Don’t get me wrong, at his peak Kidd was a top notch PG, but his assist stats and reputation in recent years were somewhat enhanced by having incredible athletes like Jefferson, Carter, and Kenyon Martin scoring from above the rim on ally oop passes and running around like energizer bunnies.

    Without guys like that around, his own limitations (like outside shooting) are having more of an impact and his strengths (open court) are having less of an impact.

    Overall, Dallas’s offense hasn’t looked as good.

  43. caleb

    “…or that Dallas’s offense improved when he came to the team is flawed because a visual analysis of the team’s performance suggests he is one of the problems.”

    “Visual analysis” aside… would you agree that if a team scores more points (and has better offensive efficiency) after a trade, it doesn’t help the argument that the trade hurt its offense?

    You can still argue that it’s luck, a small sample size, etc.

  44. caleb

    Dallas was something like 4-9 against playoff teams in the last part of the season — well below average for the league (compared to other teams’ records against playoff teams). They were 12-4 against non-playoff teams — well ABOVE average.

    Question: Is Dallas an above-average team, or a below-average team?

    Hint: They were 16-13 overall, after the Kidd trade…

  45. caleb

    “>well, Kidd’s ideally suited to running the fast break, and he doesn’t have great wingmen to do this with in Dallas, so that’s got to be part of it. <

    Exactly.”

    Ok, we can agree that Kidd is not a good shooter, but he replaced Devin Harris…

    Harris 3-pt. %: .335
    Kidd 3-pt %: .381

    Harris % jumpers: 66%, eFG .453 (in Dallas)
    Kidd % jumpers; 60%, eFG .425 (in Dallas)

    Not a huge difference.

    Now, Harris gets to the line a lot, which leads to his overall TS% being much higher than Kidd’s… but you’re talking about style of play. Just like Kidd, Harris is much better suited to a fast-breaking team. Considering Kidd’s huge advantage as a passer, it’s not surprising that Dallas actually functioned better on offense, after the trade. (as measured by scoring more points, not by how stylish it looked)

  46. W.C.

    >>would you agree that if a team scores more points (and has better offensive efficiency) after a trade, it doesn’t help the argument that the trade hurt its offense?

    You can still argue that it’s luck, a small sample size, etc.<<

    Yes. But I think the stats can be impacted by schedule, the specific teams played, short term fluctuations that are random etc….

  47. jon abbey

    “Kidd’s huge advantage as a passer” primarily comes in the open floor, and Dallas doesn’t really have the personnel to fully take advantage of that. this may be why he only has 25 assists in 4 games this series, and the total has dropped each game.

  48. W.C.

    >Dallas was something like 4-9 against playoff teams in the last part of the season — well below average for the league (compared to other teams’ records against playoff teams). They were 12-4 against non-playoff teams — well ABOVE average.

    Question: Is Dallas an above-average team, or a below-average team?

    Hint: They were 16-13 overall, after the Kidd trade…<

    Statistical fluctuations like these can typically random when the sample is so small.

    It’s certainly possible that some of the better teams have the resources that can exploit the weaknesses that Kidd brings to the table and the weak teams don’t, but I suspect that’s not the case.

    I think the 16-13 overall record does the best job of evaluating how good they were (but still not perfect because even that sample is too small).

    I also believe 16-13 is worse than their win percentage before the trade and it looks like they actually had an easier schedule after Kidd by playing more games against non playoff teams (which could lead to other stats that misrepresent his value).

    I think the best way to measure things is to look at the stats AND watch the games so you understand where the stats are coming from. That way you know which are meaningful and which are misleading because they are based on other factors.

  49. caleb

    “I think the best way to measure things is to look at the stats AND watch the games so you understand where the stats are coming from. That way you know which are meaningful and which are misleading because they are based on other factors.”

    I completely agree!

    In Dallas’ case, I think it was a terrible trade, but for different reasons… a) defense (Harris was better and they really miss Diop); b) the future (obviously)

    IMO, Dallas has serious problems next year and beyond — they have zero cap flexibility, no chance of improving with the players on the roster now… and minus two first-round draft picks.

    The one bright spot is Brandon Bass — he has been very impressive. But he plays Dirk’s position — what could they get for Brandon in a trade? I think they’re totally screwed.

  50. retropkid

    Josh Howard just disappeared so far in the play-offs….

    How do you feel about the limited sample size of play-off games as a basis for determining if a player steps up or chokes in big games? My view: even with very limited sample sizes, it’s reasonable to use play off game numbers to get a sense of who steps up and who doesn’t….

    As for the original post — if you think the NBA is skewed….look at MLB….given how much the Yankees and Red Sox combined spend on talent, and that they are in the same division, the chance of both of them ever missing the playoffs in the same year is remote…it’s a fix, effectively.

  51. caleb

    Honestly, I think you can blame sample size for Josh Howard’s problems — I mean, last year Howard was great agaisnt GS — doubled his rebound average, upped his scoring from 18 ppg to 21… also had a strong playoffs in the doomed playoff run the year before. But yeah, he’s looked really bad the past few games, especially since he’s being guarded by Peja.

    re: competitive balance, the salary cap has some obvious logic to it, but at the same time it makes it a lot harder to make trades, which makes it harder to turn teams around when they’re in bad shape. No salary cap (or a more flexible cap) would probably mean stronger teams in New York, Chicago, Boston and LA…. hasn’t really hurt baseball, I don’t think. Then again, I’m a Yankee fan and a Knick fan stuck in salary cap hell so I’m comnpletely biased.

Comments are closed.