Brad Miller is the MVP of this series

[Diddy1122]

This series has been full of amazing individual performances. Rose in game 1, Gordon in game 2, Pierce in game 5, and Allen & Salmons in game 6. But the one player who has had the biggest effect on the Bulls this series is Brad "Slo-Mo" Miller.

Now I know the +/- stat is pretty new to Bball, and in terms of starters, it's very skewed because they are playing the majority of the minutes on the court. But I do find it effective when it comes to bench production, since these players play less minutes, making their time on the court even that more valuable. In this series Miller has posted a +9, +19, +21, & a +24. He is leading the team in this category.

Other than the obvious statistics, this is what I'm seeing when Miller is on the court. The ball moves better, the floor is spaced better, the opposing teams offensive rebounds stop. This is what Miller does. It's what he's done his whole career.

And when the chips were down last night and the Bulls seemed all but dead, Brad was on the sidelines during a timeout getting into his teammates and not letting them think for a second that they were out of this game. Thanks to his heady play in the 4th, we are going to game 7 with a legitamate shot to take down the defending champs.

 

[dougthonus]

Brad Miller also has the worst PER of any rotation player on the team. PER actually measures what a player does while +/- just measured what happened while he was on the floor regardless of whether he was directly involved in it.

This isn't to say Miller's been terrible, he was amazing last night, but prior to that I don't think he had a particularly good series.

 

[Diddy1122]

PER is heavily skewed towards offensive numbers, which let's face it, is not the strong suit of Miller's game. It's the little things he's done in this series that are making a big difference, and they don't always show up on the stat sheet.

 

[JakeN7]

In my opinion there isn't one definitive MVP of the bulls in this series. You could make a case for a few guys depending on the night. Derrick Rose got us rolling in game one. Gordon has had some absolutely mind blowing late game heroics, Joakim Noah has had some outstanding plays and has been hitting the boards very well. John Salmons had quite the night last night, and Bradley Miller has brought some toughness and veteran intangibles down the stretch. This series has been unreal.

 

[Manic Devourer]

I don't think so. He has had about 2 games were he made no impact on the game. I think the closest to an MVP is Gordon, though the last 2 games have been average to bad for him, a little like how Pippen was the Finals MVP against the Jazz until the last 2 games or so when he had shockers.

The most consistent player by FAR has been Noah, and it hasn't even been close.

 

[Shakes]

dougthonus wrote:

Brad Miller also has the worst PER of any rotation player on the team. PER actually measures what a player does while +/- just measured what happened while he was on the floor regardless of whether he was directly involved in it.

I don't know, the basketball court is small enough that is there ever anywhere to hide and not be involved? I mean the most extreme case is like when Stephen Curry scored 0 points because he was doubled the whole night and just let the rest of the team win 4 on 3. He would have had a terrible PER that game but was the reason Davidson won.

Miller improving us has been a pretty consistent effect, I think there's something that PER doesn't capture about his play.

 

[Newskoolbulls]

man i cant pick one bull who is MVP of this series because it would be offensive to our other guys since every game we have had a different player step up big.

MVPs: Kirk, Ben, Derrick, Brad, John, Joakim, hell even Tyrus.

 

[dougthonus]

Diddy1122 wrote:

PER is heavily skewed towards offensive numbers, which let's face it, is not the strong suit of Miller's game. It's the little things he's done in this series that are making a big difference, and they don't always show up on the stat sheet.

PER benefits good rebounders a ton as well, so Miller should have a big bonus. The only people PER doesn't really benefit are perimeter defenders. I'd certainly take PER over +/- without blinking, especially yahoo's +/- which is raw +/- rather than net +/-.

 

[dougthonus]

I don't know, the basketball court is small enough that is there ever anywhere to hide and not be involved? I mean the most extreme case is like when Stephen Curry scored 0 points because he was doubled the whole night and just let the rest of the team win 4 on 3. He would have had a terrible PER that game but was the reason Davidson won.

Miller improving us has been a pretty consistent effect, I think there's something that PER doesn't capture about his play.

I wouldn't pretend PER captures everything, but let's say Boston pulls Ray Allen and Kendrick Perkins when Miller comes in (reasonable that they'd give either guy a blow there), then Ben Gordon hits a couple 3s, and Eddie House turns the ball over while Scalabrine does something dumb.

Did Miller do anything to earn the +/- or was he just on the court at a good time? +/- has a lot to do with who you play against and who you play with. Sometimes it can have a lot to do with how good you are, but frequently it doesn't, and the frequency which it's meaningful is small enough that I wouldn't use it as evidence.

Take a look at the year long +/- leaders vs the year long PER leaders and tell me which list is ordered in a more reasonable way for quality players. It's PER, and it's not close.

That's not to say that PER is perfect or great or anything just that it's a lot better than +/-

 

[Diddy1122]

dougthonus wrote:

I don't know, the basketball court is small enough that is there ever anywhere to hide and not be involved? I mean the most extreme case is like when Stephen Curry scored 0 points because he was doubled the whole night and just let the rest of the team win 4 on 3. He would have had a terrible PER that game but was the reason Davidson won.

Miller improving us has been a pretty consistent effect, I think there's something that PER doesn't capture about his play.

I wouldn't pretend PER captures everything, but let's say Boston pulls Ray Allen and Kendrick Perkins when Miller comes in (reasonable that they'd give either guy a blow there), then Ben Gordon hits a couple 3s, and Eddie House turns the ball over while Scalabrine does something dumb.

Did Miller do anything to earn the +/- or was he just on the court at a good time? +/- has a lot to do with who you play against and who you play with. Sometimes it can have a lot to do with how good you are, but frequently it doesn't, and the frequency which it's meaningful is small enough that I wouldn't use it as evidence.

Take a look at the year long +/- leaders vs the year long PER leaders and tell me which list is ordered in a more reasonable way for quality players. It's PER, and it's not close.

That's not to say that PER is perfect or great or anything just that it's a lot better than +/-

I don't even like using +/- stat. Or PER for that matter. I don't think any of them fully work. Especially in basketball. But since I sat thru the lecture I decided to do a little research for the class. I took a look at the year long +/- and PER leaders. Here's what I discovered:

Top 15 PER for 08-09 season:
1. LeBron James
2. Dwight Howard
3. Chauncey Billups
4. Tony Parker
5. Tim Duncan
6. Joe Smith
7. Brandon Roy
8. Linas Kleiza
9. Carmelo Anthony
10. Rajon Rondo
11. Marvin Williams
12. Josh Howard
13. Kobe Bryant
14. Dwayne Wade
15. Lamar Odom

Top +/- for 08-09 season:
1. LeBron James
2. Chris Paul
3. Lamar Odom
4. Dwayne Wade
5. Kobe Bryant
6. Jason Kidd
7. Yao
8. Ray Allen
9. Kevin Garnett
10. Delonte West
11. Rashard Lewis
12. LaMarcus Aldridge
13. Andre Iguodala
14. Brandon Roy
15. Tyson Chandler

It's a tough choice between the two, but I think I'll go with the stat system that doesn't have Joe Smith and Linas Kleiza in the top 10. But thats just me.

 

[sabixx]

linas Kleiza and joe smith being in the top ten is no more silly than Odom being than wade,kobe,yao,kg. West being that high is just as bad as well. so I dont see your point. +/- doesn't work because basketball has uncontested points,free throws.

 

[Diddy1122]

sabixx wrote:

linas Kleiza and joe smith being in the top ten is no more silly than Odom being than wade,kobe,yao,kg. West being that high is just as bad as well. so I dont see your point. +/- doesn't work because basketball has uncontested points,free throws.

It's way more silly. Joe Smith averaged 6.6pts 4.6rebs. Odom, 11.3pts 8.2reb, and is a 6th man. Delonte West is a starter, averaged 11.3pts 3.5ast, and is a good defender. Linas Kleiza, a bench player who hits a bunch of 3's. Smith's offensive effenciency is really good, so is Kleiza's, that's why their PER is so high. Does that mean they are having a bigger effect on the game than say West playing lock down D on Rip Hamilton? Who knows.

All I know is both systems are flawed. I don't like either of them. But just from all the colors I've seen, they're painting a picture of +/-.

 

[Shakes]

dougthonus wrote:

Take a look at the year long +/- leaders vs the year long PER leaders and tell me which list is ordered in a more reasonable way for quality players. It's PER, and it's not close.

The problem I have with this sort of statement(what the sports stats guys typically call the "laugh test") is that it basically says the only good stats are those that tell us what our subjective opinion already agrees with.

To me the point of statistics is to get an understanding beyond what we may be seeing with our own very biased perceptions. I don't think we yet have a full understanding of why some players make teams better than their box score contributions, but the effect seems consistent enough for some players that I think it's pretty hard to deny it does exist.

I'm not saying +/- is the perfect judge of a player, but I also don't believe in coincidences. Brad Miller was good on +/- for us during the regular season, I doubt he's always happened to be in the game when Gordon starts drilling threes and on the bench when he's bricking them.

 

[dougthonus]

The problem I have with this sort of statement(what the sports stats guys typically call the "laugh test") is that it basically says the only good stats are those that tell us what our subjective opinion already agrees with.

To me the point of statistics is to get an understanding beyond what we may be seeing with our own very biased perceptions. I don't think we yet have a full understanding of why some players make teams better than their box score contributions, but the effect seems consistent enough for some players that I think it's pretty hard to deny it does exist.

I'm not saying +/- is the perfect judge of a player, but I also don't believe in coincidences. Brad Miller was good on +/- for us during the regular season, I doubt he's always happened to be in the game when Gordon starts drilling threes and on the bench when he's bricking them.

Okay, well let me add a few points:
Yahoo's box score +/- is completely and utterly useless. If your team averages 20 point losses then you will have a huge -. If you are a role player that may not be much your fault, but if you get traded to a good team, you'll then have a huge + because your team averages 20 point wins. If you play more minutes your +/- will only better reflect the average of what your team does using the yahoo +/-.

Say you are a team with 4 really good players and no depth. The same guy would have a great +/- as the starter and a terrible one as a bench player, because he'd just get the benefit of playing with 4 great players vs playing with more scrubs. It's about line strength. In an individual series (like this one) it depends on the strength of the guys you play against as well.

These are reasons why yahoo/nba/espn +/- is completely and totally worthless. It doesn't analyze well what it's supposed to analyze.

What is quite a bit better is net +/- from 82games.com. This calculates your +/- while on the court and the teams +/- when you are off the court. It then takes those numbers and prorates them to 48 minutes and subtracts them to figure out the net +/- per 48 minutes. Thus if your team is - 20 per game, but only averages -5 per 48 while you are on the floor and -15 per 48 when you aren't, then you'd have a +10 (teh team plays way better when you are out there even if bad).

This gets closer to something useful, but it's still not good.

The reason is because if you look at guys historically, their +/- numbers change radically from year to year. So much that it's hard to draw any correlation between +/- and success consistently.

I've been looking at +/- since 04/05, and when the stat first came out I was very excited about it, but each year that goes and I see the new adjusted +/- stats I see how worthless they are on the whole. Even the ones you think you can draw something from a hard to draw something from because so many are just inconsistent.

 

[engies]

Diddy1122 wrote:

This series has been full of amazing individual performances. Rose in game 1, Gordon in game 2, Pierce in game 5, and Allen & Salmons in game 6. But the one player who has had the biggest effect on the Bulls this series is Brad "Slo-Mo" Miller.

Now I know the +/- stat is pretty new to Bball, and in terms of starters, it's very skewed because they are playing the majority of the minutes on the court. But I do find it effective when it comes to bench production, since these players play less minutes, making their time on the court even that more valuable. In this series Miller has posted a +9, +19, +21, & a +24. He is leading the team in this category.

Other than the obvious statistics, this is what I'm seeing when Miller is on the court. The ball moves better, the floor is spaced better, the opposing teams offensive rebounds stop. This is what Miller does. It's what he's done his whole career.

And when the chips were down last night and the Bulls seemed all but dead, Brad was on the sidelines during a timeout getting into his teammates and not letting them think for a second that they were out of this game. Thanks to his heady play in the 4th, we are going to game 7 with a legitamate shot to take down the defending champs.

As Kush said in his most recent podcast, +/- is quite a flawed stat. Biggest example was when he compared Ray Allen's +/- in Game 6 to Rondo's

As he said, why should someone get - due to something his teammates may have caused or done

Having said that i will say this. I think Miller has been an underrated factor simply coz his inside presence, rebounding & in a few games his shooting has been great for us

its hard to give a definitive mvp since its been such a team effort with BG doing it some games, Rose doing it some games, Salmons doing it for us in terms of defense on Pierce and his great scoring last game, Kirk, Noah. Could make an arguement for 5-6 players on this team

 

[wjb1492]

Wow, interesting thread.

As far as MVP, I agree with the idea that you can't pick just one Bull for the whole series. Derrick unquestionably for Game 1 and Ben for Game 2 had humongous impacts, but have had other games where they didn't play so well. Tyrus was incredible in Game 1 OT. Kirk was huge keeping the Bulls in the early quarters of Game 5, then disappeared after bouncing his head on the floor - so you probably can't call him the MVP of the game even, but without him early the game's likely a total blowout loss. Game 6 you've got Brad MVP of the regular period, then John carrying the team in the first couple of OTs, then the huge plays by Joakim and Derrick to close the game out. The Bulls are hanging far tougher than I think most people thought - almost every game coming down to the wire. Last year the Hawks took them to 7 by barely winning at home and suffering blowouts in Boston. Bulls have been truly competitive throughout. (Yeah, KG is out, but you can only play the team they put on the floor, and Bulls aren't exactly in robust health either). So how do you look at the team effort and single one guy out. Normally I might think it's a cop out, but so far this is honestly one situation where I say let them share the game ball. Now, with Game 7 tonight, there's still the possibility that one guy steps up and is the primary factor in pulling out a win, and maybe that's significant enough that I change my mind, but if it's in line with the other games I say skip the MVP debate and honor the whole darn team.

With the stats argument going on, absolutely every single stat you could pick has its flaws. Sometimes people throw stats around without understanding them or cherry pick the stats that back up their point sure, and sometimes I think we just get lazy and assume everyone else will understand the limitations of whatever stat. I do that far too often probably - stats are a convenient way of saying in shorthand what could be described in words, and when you're already resorting to shorthand like using PER, who wants to include a mini-dissertation on its strengths and weaknesses? But that probably adds greatly to the confusion of what each stat says and when to use them for people who aren't as familiar with stats.

I still love much of the adjusted +/- stuff, as well as the +/- numbers for different lineups and the player-pair stuff over at 82games.com. Sure you still need some context for everything, but what fun data, especially the on-off court stuff and the more in-depth player-pair stuff. I think I enjoy it so much because I tend to root more for the role player guys than the stars - stars get their individual numbers and their importance can rest on those individual numbers to a big extent. But when your favorite player is Kirk Hinrich* and other people are constantly chirping about scoring averages and other individual stats, it's nice to be able to look at the more team oriented measures. That doesn't mean individual stats aren't important and don't contribute hugely to win totals. But for a Kirk fan it's great to be able to look at other numbers and see that everyone else on the team has a better +/- number playing with Kirk than without, or that both the team offense and defense function better overall with him on the court than off, or that the % of assisted baskets are higher with him on the court than off. To me, that means Kirk is doing his job and doing it well. So I like stats that make an effort to measure the impact of play not captured in the typical box score - the guy willing to box out and let his teammate get a rebound, the hockey assist, the strong defenders, etc. But I think people just forget at times that these stats are both very muddy due to the nature of basketball and very contextual to the team and the role on the team. And also that they do not mean a guy who rates highly on these muddy team measures is "better" than a guy who rates highly on individual stats. Heck, the money and the glory come with individual stats, and I'd be highly comfortable betting that most role players would gladly assume the "star" role if they were basketball-capable of doing so.

*Ben Gordon is better.

 

[dougthonus]

But when your favorite player is Kirk Hinrich* and other people are constantly chirping about scoring averages and other individual stats, it's nice to be able to look at the more team oriented measures. That doesn't mean individual stats aren't important and don't contribute hugely to win totals. But for a Kirk fan it's great to be able to look at other numbers and see that everyone else on the team has a better +/- number playing with Kirk than without

Kirk's had a great year IMO. He's really picked up his defense, especially in this play off series as well. I'd note that prior to this season +/- wasn't Kirk's friend if I recall correctly (though it's posssible I don't).

 

[clonetrooper264]

wjb1492 wrote:

*Ben Gordon is better.

Hahaha little parenthetical insert to keep the Ben fans happy I see. Man it's really sad to think that there is still an argument over who is better, but what can you do. Anyway, yes Kirk has played some really solid defense this year and that has been key for us, just as Ben's clutch shooting has been key. Brad Miller has been pretty solid on the boards and has provided some good passing and occasional scoring on the offensive end. I'm not sure that he's the MVP though. I would be more inclined to give it to Gordon, but that's just me. Also, Salmons has been player pretty well the last couple of games, let's hope he can keep it up for tonight.

+/- is a garbage stat and IMO it means nothing. The only thing I see +/- good for is to see who had the best line on a winning team.

 

[wjb1492]

dougthonus wrote:

But when your favorite player is Kirk Hinrich* and other people are constantly chirping about scoring averages and other individual stats, it's nice to be able to look at the more team oriented measures. That doesn't mean individual stats aren't important and don't contribute hugely to win totals. But for a Kirk fan it's great to be able to look at other numbers and see that everyone else on the team has a better +/- number playing with Kirk than without

Kirk's had a great year IMO. He's really picked up his defense, especially in this play off series as well. I'd note that prior to this season +/- wasn't Kirk's friend if I recall correctly (though it's posssible I don't).

But that's all part of the context I was talking about. The team is different. Kirk's role is different. And that's the thing with team stats. They generally suck as a means of evaluating a player's quality and ability as an individual basketball player, but then they aren't designed to measure that in the first place. But they do have a purpose, and it bugs me when people write them off as useless. The team was pretty bad last year, and Kirk personally was awful, so I wouldn't expect to find much that casts him in a positive light. And for the past several years the Bulls have had that issue with the starters not being a bunch better than the bench guys who play, which did give whoever came off the bench a bit of a boost (generally better than bench counterparts) compared to starters (generally even with or slightly worse than starter counterparts). That worked against Kirk in past seasons and benefited him greatly this year.

It's sort of like when people dismiss PER because some low-minute players rank highly. (Anthony Roberson playoff PER of 107.7 anyone?) Yes, PER has a bunch of flaws, and you still need some context to make it useful. But it does have a purpose in using per minute numbers rather than per game numbers, and it does provide a useful summary of all a players' stats. I'm still waiting (and not holding my breath) for the stat that perfectly describes everything without requiring game and team context to interpret.

So ultimately I guess I'm saying there's a good reason for that old saying "lies, damn lies, and statistics," but I still think they're fascinating and a lot of fun. The difficulty lies in all of us trying to make sure we understand what all the different statistical measures intend to do, what their flaws are, and not falling into the trap of using a single stat to determine whether a player is "great" or "sucks," or into the trap of thinking team building is just about trying to put together a bunch of statistically great guys.

 

[dougthonus]

But that's all part of the context I was talking about. The team is different. Kirk's role is different. And that's the thing with team stats. They generally suck as a means of evaluating a player's quality and ability as an individual basketball player, but then they aren't designed to measure that in the first place. But they do have a purpose, and it bugs me when people write them off as useless. The team was pretty bad last year, and Kirk personally was awful, so I wouldn't expect to find much that casts him in a positive light. And for the past several years the Bulls have had that issue with the starters not being a bunch better than the bench guys who play, which did give whoever came off the bench a bit of a boost (generally better than bench counterparts) compared to starters (generally even with or slightly worse than starter counterparts). That worked against Kirk in past seasons and benefited him greatly this year.

Your bottom line explanation is why I think the +/- stat isn't that useful, because it doesn't give you any type of context. Kirk was a good player on teams through 04/05-06/07 in good teams and had pretty poor +/-s for 2 of those 3 years. The +/- as you identified was reflective of the unit he played with and the strength of our bench more so than his own ability.

However, knowing why the number was lower than you would expect, what does the number really support. All it does is make you seek out reasons as to why it doesn't match your visual interpretation of the game.

If I gave you both net and raw +/-s of every player in the league for the last five years, what would you get out of it? Could really tell me you'd discern anything useful at all out of it without seeing them first? Would you tell me you'd want the the five highest guys in raw +/- to be your starters? Would you tell me you'd want the five highest guys in net +/-?

I just don't think I would feel confident (without looking first) in saying anything about +/- leaderboard or lower board or feel confident making any statement whatsoever about that group of players.