PDA

View Full Version : This is ironic.



Hicks
07-25-2007, 10:41 PM
I pretty well dismiss the +/- stat of individual players because I think it's very misleading and basically useless.

However, I find 5-man +/- interesting because it shows how well a specific "unit" of players for the team performed while on the floor together as a team.

Thankfully, 82games.com keeps record of this for us all.

The ironic part? Look at which group is #1 for your Indiana Pacers in 2006-2007. That's right, the same bunch that "led us" to our "downfall/doom/armageddon/onoz/etc.".

Tinsley
Dunleavy
Granger
Murphy
O'Neal

W/L of 16-8, 66% (82 games = 54 wins)

jHjFxJVeCQs

!


Oh, and also funny: The 5-man group with the highest winning percentage? At 5-1, 83%:

McLeod, Dunleavy, Granger, Murphy, O'Neal.

That's easy to write off as it was only 6 games, but the above group played 24 games, 4 more than many on here typically wait going into the start of a season to evaluate the team.

Trader Joe
07-25-2007, 10:56 PM
I'm with the rodent. I can't believe it either.

ChicagoJ
07-25-2007, 11:00 PM
Well, somebody had to be at the top. Doesn't mean they were actually "good."

(However, they were pretty good while Marquis was still available. The downfall wasn't that lineup, it was the dependency on the sixth man to create shots for himself/ teammates.)

Hicks
07-25-2007, 11:19 PM
Well, somebody had to be at the top. Doesn't mean they were actually "good."

(However, they were pretty good while Marquis was still available. The downfall wasn't that lineup, it was the dependency on the sixth man to create shots for himself/ teammates.)

66% winning isn't good? Well, you're also the one who assigns a team something like a "C" for being in the Conference Finals, so maybe I shouldn't be surprised. :cool:

Trader Joe
07-25-2007, 11:27 PM
66% winning isn't good? Well, you're also the one who assigns a team something like a "C" for being in the Conference Finals, so maybe I shouldn't be surprised. :cool:

What it doesn't address is the consistency of the coach to play that particular lineup. Rick may have been very set in his ways during the game, but he had an itchy trigger finger when it came to messing with the starting lineup this year. I think part of that was due to him being used to having every player hurt.

ChicagoJ
07-25-2007, 11:28 PM
66% winning isn't good? Well, you're also the one who assigns a team something like a "C" for being in the Conference Finals, so maybe I shouldn't be surprised. :cool:

Well, the 16-8 can be misleading.

I happen to believe the stretch between the trade and Quis' injury is the only reliable barometer from last season. And that lineup was what, 9-2 during that stretch? (Don't remember, I'm sure somebody will correct me right away though.)

Trader Joe
07-25-2007, 11:33 PM
Well, the 16-8 can be misleading.

I happen to believe the stretch between the trade and Quis' injury is the only reliable barometer from last season. And that lineup was what, 9-2 during that stretch? (Don't remember, I'm sure somebody will correct me right away though.)


We were 29-24 when it all went awry and Quis got hurt. We were 20-20 when Dun, Murph, Diogu, and McLeod arrived. We went 9-4 and then Quis went down and our bench fell apart with no true creator on it.

Anthem
07-25-2007, 11:55 PM
It all went awry when the trade deadline passed and the team realized they weren't going to get any better that season.

Los Angeles
07-25-2007, 11:57 PM
Our schedule for that time period was heavy on home games, heavy on cream-puff opponents and heavy on days off.

I've always felt that losing Marquis was the straw that broke the camel's back, and I was one of the first to say so (IIRC).

But even I have to admit that I have no way of knowing how we would have competed against the March/April schedule even with Quis in the lineup. That would have been the only test.

But even THAT's wrong.

because above all else, I believe that a winning team needs to be established from the first day of training camp. The need to stay together and stay confident and believe in what they are doing all the way to the last game of the finals.

The 06/07 team was wrecked by a MAJOR off-court distraction, twice wrecked by an early disposal of the season game plan and then triple wrecked by a blockbuster trade.

There was no barometer for the season. We were lucky to win 30 games in the first place.

PS - Please, oh Lord, let me root for a team that is capable of going incident free for one season. Please. :pray:

Kegboy
07-26-2007, 12:19 AM
And here I thought this thread was gonna be about us trading Granger right after Hicks bought his jersey. :shrug:

GrangerRanger
07-26-2007, 01:33 AM
Well, even though that may be true, O'Neal was hurt most of the season so we couldn't really see the lineup play that much. I think their was a time right after the All Star break where we won 3 of 4 games with O'Neal in the lineup. He got hurt and bam, we sucked.

Oneal07
07-26-2007, 10:55 AM
WOW. . Interesting!! I Actually Kill people with that Line Up in NBA 2k7 lol

Pacersin2033
07-26-2007, 10:58 AM
Eh, I hate the +/- Stat no matter what it says. Its a flawed statistic because by its very nature it doesn't take into account how the other coaches are playing us.

This stat could just be indicative of other coaches having so little confidence in our bench, they played thier bench more when our starters were in. Letting our starters get a slight advantage. But then thier starters killing us when our bench was in.

NuffSaid
07-26-2007, 01:21 PM
I pretty well dismiss the +/- stat of individual players because I think it's very misleading and basically useless.

However, I find 5-man +/- interesting because it shows how well a specific "unit" of players for the team performed while on the floor together as a team.

Thankfully, 82games.com keeps record of this for us all.

The ironic part? Look at which group is #1 for your Indiana Pacers in 2006-2007. That's right, the same bunch that "led us" to our "downfall/doom/armageddon/onoz/etc.".

Tinsley
Dunleavy
Granger
Murphy
O'Neal

W/L of 16-8, 66% (82 games = 54 wins)

(We can do w/o the rodent here :) )

Oh, and also funny: The 5-man group with the highest winning percentage? At 5-1, 83%:

McLeod, Dunleavy, Granger, Murphy, O'Neal.

That's easy to write off as it was only 6 games, but the above group played 24 games, 4 more than many on here typically wait going into the start of a season to evaluate the team.

I've been hoping that someone would find some stats to support my belief. Finally, prove positive that this team even as currently comprised isn't as bad as most believe. Thanks, Mal! What a great find!

We were 29-24 when it all went awry and Quis got hurt. We were 20-20 when Dun, Murph, Diogu, and McLeod arrived. We went 9-4 and then Quis went down and our bench fell apart with no true creator on it.

Our schedule for that time period was heavy on home games, heavy on cream-puff opponents and heavy on days off.

I've always felt that losing Marquis was the straw that broke the camel's back...
They (Pacers) were pretty good while Marquis was still available. The downfall wasn't that lineup, it was the dependency on the sixth man to create shots for himself/ teammates.)
Three very good points most fans actually do agree with. Losing Quis was a major hit to this team's ability to move forward because it essentially meant the Pacers didn't have a legit 6th Man. Granger tried to fill the role but switching him back and forth between starter and coming off the bench hurt more than it helped. By his own admission, he became lost in trying to grapple with knowing exactly what his role should be and what exactly he was "suppose" to do rather than just going with his instincts.

LA does make two very good points, however:


1) Even I (Los Angeles) have to admit that I have no way of knowing how we would have competed against the March/April schedule even with Quis in the lineup. That would have been the only test.

...which is so true, but I'd like to believe that had Quis been healthy and the Pacers did utilize the starting lineup as outlined above (w/McLeod as the BU/PG) (thanks again for the find, Mal), my boyz would have had a much better record AND made the playoffs.

...and


2) ...above all else...a winning team needs to be established from the first day of training camp. They need to stay together and stay confident and believe in what they are doing all the way to the last game (of the finals).

Which leads us to...


The 06/07 team was wrecked by a MAJOR off-court distraction, twice wrecked by an early disposal of the season game plan and then triple wrecked by a blockbuster trade.

Naptown_Seth
07-26-2007, 02:12 PM
Hicks, if this doesn't prove that I'm about fairness and solid numbers to back up points, nothing will...

I think this is a very interesting stat and perhaps does indicate something really positive. It's tough to resolve the actual winning % when this group was together vs their winning % as a group on-court.

Some may not realize, but the W-L of 5 man +/- isn't the actual wins and losses when they played in terms of the real game, it's in terms of their stints on court as a total game. Basically like "if these 5 played all 48 minutes they would win at this rate".

The stat is pretty legit since they are the top minutes group for the Pacers last year (really???? how is that possible? - oh, because the other main group is split between when Granger started and when Foster started, and both were also winning groups).


Now then, I do have some other questions though. If you total all the W-L rates for the 5 man groups up you get a record of 119-121, which is much better than their actual final record. So all the different 5 man groups were doing well, the 3 biggest minutes groups were actually well over .500, but the team in total was losing like crazy? How is that possible, how can you be losing games if no 5 man group is getting beat very much?

And now we come to a huge issue, one that makes me question the stat keeping itself. If you total the absolute +/- of every group up you come out as +7 as a team.

The problem with that? They were freaking -200 as a team for the year.

edit - clearly these aren't every single group and not the total minutes for the entire team (no Harrison group listed at all for example). So I'm looking into it some more now.


2nd edit - the total points scored and allowed is around 3700 for both. But the season totals are at 7800 and 8000 respectively. So the top 20 minutes units still played less than half the total minutes for the team??? Not very likely, in fact I don't think it's even possible. You take all the various 5 man groups that are possible but not already listed and give them each less than 29 minutes I'm not sure you can reach another 4000 points scored or allowed.

The total minutes shown by the top 20 groups only amounts to 38 games played, meaning "other groups" played more than 50% of the time for the Pacers. If possible and true then I think we've found the problem.

For example, beyond the 11 variations shown for Tinsley here, he played another 800 minutes with another sets of 4 man groups.


The group that Hicks mention only put in 6.7 games total together (in terms of 48 minutes per game), which isn't a lot of time together.

ChicagoJ
07-26-2007, 02:17 PM
Part of this is the selection bias of how 82games reports this. They don't give you the "Eddie Gill" at SF results because that's only a few minutes over the course of the season.

We've all seen a team make a six- eight- point run on a mismatched lineup before the coach can get a deadball or timeout. Some of the five-man debacles only last for a few minutes but have a big negative - not enough minutes to make the 82games list because there are too many five-man combinations to track.

Naptown_Seth
07-26-2007, 03:16 PM
Part of this is the selection bias of how 82games reports this. They don't give you the "Eddie Gill" at SF results because that's only a few minutes over the course of the season.

We've all seen a team make a six- eight- point run on a mismatched lineup before the coach can get a deadball or timeout. Some of the five-man debacles only last for a few minutes but have a big negative - not enough minutes to make the 82games list because there are too many five-man combinations to track.
Yeah, not sure if you posted this before or after my edits, but I realized that despite the logic that would imply that the top 20 five man groups would have a majority of the playing time, that's far from the case.

I'm working an XL of the 82 games numbers right now to figure out just how much PT for each guy was left out.

Here's a list of how much of a guy's playing time is actually represented in the top 20 5 man groups listed at 82 games.

Tinsley 64%
Dun 63%
Granger 55%
Murphy 65%
O'Neal 59%

Jackson 57%
Harrington 55%
Foster 51%
Daniels 13%
Diogu 28%

McLeod 45%
Armstrong 19%
Baston 7%
Williams 19%
Saras 5%

Rawle 0%
Harrison 0%
Greene 0%
Powell 0%

Team total represented 47%

So I guess the final evaluation is "sure that's nice, but what about the other half of the game". In briefly checking other 5 man groups I quickly saw that really even your top group is only going to represent about 10-20% of your playing time. Not nearly enough to decide a win or a loss.

It also shows that individual +/- isn't the non-factor that I think Hicks implies it is. It's great that when Dun was out there with these various 5 man groups but what about when he was paired with other groups, perhaps asked to carry the load more.

Here are the PER 48 MINUTES +/- for every guy when he WAS NOT on the court with one of the top 20 minutes groups of 5, listed from best to worst. This is the individual +/- overall minus the +/- when they played with a top 20 group, divided by the minutes they didn't play with a top 20 group (and *48 of course).

Saras 1.1
Jackson 0.8

JO -0.9
Diogu -2.1
Al -2.2
Army -2.6
Foster -2.8

Baston -3.6
Tinsley -4.3
Quis -4.5
Danny - 4.8

Rawle -5.9
Dun -6.6

Greene -10.0
McLeod -11.5
Harrison -11.6
Williams - 15.1
Troy -18.1
Powell -21.8

First thing I see, it looks like Troy was really being carried by the other main guys he played with. He spent 65% of his time with a main group and was in the black by a fair amount with those groups, but take away that support and he was awful.

Dun didn't fair that well either without big support.

You see JO still able to carry guys away from the starting lineup far better, Jackson too. Shawne I'm fine with because he was a very young rookie, but all the love for McLeod? He benefitted a ton from being with the main core and couldn't translate that to carrying the bench.


Hate to say it, but Saras does have a case for being a legit bench guy, someone who was able to get something out of nothing it would appear. I couldn't tell you why other than he ran a mean PnR with Harrison, but you can't knock results. BTW, for those that hated him at SG, he did put up a +8 while paired with Army at PG, his only minutes in a top 20 group.