Shavlik: Today is decision time

[Fafnir]

Basketball reference on/off court lineups.

Our team defense was terrible when he was on the court. Defensive rating is a far cruder measure that makes several normally assumptions based on box score statistics. These assumptions are flawed overall (but not crippling) and even more flawed in the case of a limited minutes big man.

I also don't trust synergy defensive stats when it comes to team defense. This is because if he fails to rotate or starts the chain of events to cover for his poor hedge then he won't get "discredit" in synergy numbers which are best at evaluating isolation and man to man defense.

What you fail to mention is that our own offensive rating was 113.9 for a net +1.3.  When he was off the court we were -0.6.  We also out rebounded opponents by 2% with him on the court as opposed to getting out rebounded by 5% when he was off the court.  Your own numbers show that we were a much better team when he was on the court than when he was off the court.

Right but that's sustained primarily by his offensive rebounding and putbacks on those offensive boards. While that's useful that's not a skillset that necessarily will translate, especially when you're matched up against backups to put up those numbers.

If you can't defend backups without fouling a ton and the team defense is compromised what will happen if his minutes are extended. Similarly is offensive rebounding and putbacks sustainable for him against sterner competition?

I wouldn't mind giving Randolph some run, but to pretend he's more than a marginal player is to ignore his massive defensive deficiencies.

[Yogi]

Right but that's sustained primarily by his offensive rebounding and putbacks on those offensive boards. While that's useful that's not a skillset that necessarily will translate, especially when you're matched up against backups to put up those numbers.

If you can't defend backups without fouling a ton and the team defense is compromised what will happen if his minutes are extended. Similarly is offensive rebounding and putbacks sustainable for him against sterner competition?

I wouldn't mind giving Randolph some run, but to pretend he's more than a marginal player is to ignore his massive defensive deficiencies.

What you are saying makes no sense.  You ignore his excellent individual defensive stats, PLUS the fact that the team does better when he is on the floor to claim that somehow he is not very good and he is a defensive liability?  There is absolutely no evidence of that. 

You are cherry-picking stats, and then adding your own creative analysis while ignoring the undeniable fact that not only was Shavlik a very productive player, but the team  outplayed the opponent when he was on the floor and got outplayed when he wasn't.

[Fafnir]

Right but that's sustained primarily by his offensive rebounding and putbacks on those offensive boards. While that's useful that's not a skillset that necessarily will translate, especially when you're matched up against backups to put up those numbers.

If you can't defend backups without fouling a ton and the team defense is compromised what will happen if his minutes are extended. Similarly is offensive rebounding and putbacks sustainable for him against sterner competition?

I wouldn't mind giving Randolph some run, but to pretend he's more than a marginal player is to ignore his massive defensive deficiencies.

What you are saying makes no sense.  You ignore his excellent individual defensive stats, PLUS the fact that the team does better when he is on the floor to claim that somehow he is not very good and he is a defensive liability?  There is absolutely no evidence of that. 

Why does the fact that he had a overall slight plus rating in on/off mean that he's not a defensive liability?

When he was on the court the C's scored at an great rate 113 points per 100 possessions, Meanwhile the converse was true the opposition scored at a only slightly less amazing rate of 112.6 points per 100 possessions.

That combined with my own viewpoint of his play when I watched him and his foul rate is what I base my view that Randolph is a defensive liability. He's also a okay energy guy who will roll to the rim and get putbacks in the paint.

I don't feel that having such a massive plus on offensive rating due to offensive boards and a slightly less massive minus on defensive rating is a good indication he's going to help the team getting more minutes.

I'm fine with him getting a shot, but he's definitely a marginal guy.

To restate my view on synergy defensive numbers. They aren't great for team defense and can be shaky overall due to teammate interactions and coding issues. My go to example of this is Ray Allen rated out better than LeBron/Durant I believe his last year with the C's in most defensive categories. He was definitely a defensive liability that year for the C's despite this.

[hwangjini_1]

this thread is becoming almost comical. with the celtics on course for PERHAPS 30 wins, but probably less, does anyone truly believe that shav is going to make a significant difference in the upcoming season? will the celtics win 27 games instead of 24? maybe he will take them to the promised land of 35 wins?

this is not a young talent who has the potential to grow and become a good player.

now, if people believe that shav should stay because they love to watch him play, then sure. but to argue with vigor that shav will make a difference of importance is wishful thinking at best.

shave does not have a track record in the nba that shouts "success". the most optimistic "shout" is "bench player."  in his career of 111 games he averaged 2.7 points, 2.7 rebounds, and 0.3 blocks per game. this ain't "success" in the nba.

in 36 minutes he sits at 11 pts and 11 rbs per game. he also averaged 5.9 fouls by the way. fine, but this is based upon a super small sample size. check out fab melo's 36 minute average. he gets over 7 points a game with 2 blocks. that looks good until you see the reality behind those numbers.

shav is not terrible and might serve as a nice 8-9th guy off the bench. but is he worth the sort of statements being made about him here? i just dont see it.

EDIT: is it is now safe to assume that the celtics didn't chuck shav and that he is still a celtic?

[Yogi]

Why does the fact that he had a overall slight plus rating in on/off mean that he's not a defensive liability?

When he was on the court the C's scored at an great rate 113 points per 100 possessions, Meanwhile the converse was true the opposition scored at a only slightly less amazing rate of 112.6 points per 100 possessions.

That combined with my own viewpoint of his play when I watched him and his foul rate is what I base my view that Randolph is a defensive liability. He's also a okay energy guy who will roll to the rim and get putbacks in the paint.

I don't feel that having such a massive plus on offensive rating due to offensive boards and a slightly less massive minus on defensive rating is a good indication he's going to help the team getting more minutes.

I'm fine with him getting a shot, but he's definitely a marginal guy.

To restate my view on synergy defensive numbers. They aren't great for team defense and can be shaky overall due to teammate interactions and coding issues. My go to example of this is Ray Allen rated out better than LeBron/Durant I believe his last year with the C's in most defensive categories. He was definitely a defensive liability that year for the C's despite this.

1. You keep saying defensive liability simply because that is your opinion. 
2. The offensive rating of the opponent also includes the 4 other guys on the floor with Randolph, so possibly you are talking about Williams, Crawford, D.J. White etc. 
3. Raw offensive rating of the opponent is an irrelevant number.  It doesn't account for pace.  What matters is the net differential.  Even though we had a negative differential for the season, we had a positive differential when Randolph was on the court.  That is really all that matters.
4. Ray Allen has always been a great defender with the Celtics.  Just because he had a certain reputation before doesn't mean it was backed up by the numbers. 
5.  Randolph fouls a lot, but he is never going to be asked to play for 36 min a game, so he does his role extremely well.  The only significant fact is that he fouls smart and he fouls hard. 
6.  You have a theory about offensive rebounding not translating over more minutes, but there is absolutely no evidence that is the case.

[Fafnir]

 
3. Raw offensive rating of the opponent is an irrelevant number.  It doesn't account for pace.  What matters is the net differential.  Even though we had a negative differential for the season, we had a positive differential when Randolph was on the court.  That is really all that matters.

This isn't true, below is the definition of offensive rating straight from basketball reference.

ORtg
Offensive Rating (available since the 1977-78 season in the NBA); for players it is points produced per 100 posessions, while for teams it is points scored per 100 possessions. This rating was developed by Dean Oliver, author of Basketball on Paper. Please see the article Calculating Individual Offensive and Defensive Ratings for more information.

4. Ray Allen has always been a great defender with the Celtics.  Just because he had a certain reputation before doesn't mean it was backed up by the numbers. 

He gave effort and worked hard, but whenever we put in a better defensive SG (Bradley for example) our defense jumped to a completely new level. This was especially true in his last year. Similarly for Miami he was a massive defensive liability this past year.

We're going in circles at this point so I'll get back to work. Randolph like Steamer before him was a productive per minute guy who compromised our team defense to the point where they weren't able to help the team on the floor. You clearly disagree.

[Fafnir]

EDIT: is it is now safe to assume that the celtics didn't chuck shav and that he is still a celtic?

Celticshub says today is the last day to waive him. So once the league office closes he's likely to be on the roster unless we hear something.

[Yogi]

 
3. Raw offensive rating of the opponent is an irrelevant number.  It doesn't account for pace.  What matters is the net differential.  Even though we had a negative differential for the season, we had a positive differential when Randolph was on the court.  That is really all that matters.

This isn't true, below is the definition of offensive rating straight from basketball reference.

ORtg
Offensive Rating (available since the 1977-78 season in the NBA); for players it is points produced per 100 posessions, while for teams it is points scored per 100 possessions. This rating was developed by Dean Oliver, author of Basketball on Paper. Please see the article Calculating Individual Offensive and Defensive Ratings for more information.

4. Ray Allen has always been a great defender with the Celtics.  Just because he had a certain reputation before doesn't mean it was backed up by the numbers. 

He gave effort and worked hard, but whenever we put in a better defensive SG (Bradley for example) our defense jumped to a completely new level. This was especially true in his last year. Similarly for Miami he was a massive defensive liability this past year.

We're going in circles at this point so I'll get back to work. Randolph like Steamer before him was a productive per minute guy who compromised our team defense to the point where they weren't able to help the team on the floor. You clearly disagree.

I was wrong about ORtg not accounting for pace.  But my overall point remains.  It is cherry picking to point to the opponents Ortg and claim that it will translate while looking at your own and claiming that it will not.  The point is we played better.  The type of game we played was maybe a run-and-gun instead of half court defensive game with Pierce and KG.  That may account for the increases in the ORtgs.  It doesn't change the fact that we played better.

I suspect that if you replace most shooting guards with Bradley the defense will be better.  That says absolutely nothing about Ray Allen's defense except that at 37 he wasn't as good as 21 year old Bradley. 

[kozlodoev]

The main fallacy of using OR in the case of a sparingly used reserve player is that your sample suffers from a downward selection bias -- i.e., Ranolph likely saw playing time against the crappiest of competition.

Now, if you tell me that he held his opponents to 20 OR points below their season average, that's a different story.

[danglertx]

The main fallacy of using OR in the case of a sparingly used reserve player is that your sample suffers from a downward selection bias -- i.e., Ranolph likely saw playing time against the crappiest of competition.

Now, if you tell me that he held his opponents to 20 OR points below their season average, that's a different story.

I haven't been following your OR talk, but I will say I watched every game last year and Shav wasn't playing in the blowouts, his minutes were coming against 1st and 2nd units on normal rotations. 

[clover]

The main fallacy of using OR in the case of a sparingly used reserve player is that your sample suffers from a downward selection bias -- i.e., Ranolph likely saw playing time against the crappiest of competition.

Now, if you tell me that he held his opponents to 20 OR points below their season average, that's a different story.

Actually that wasn't as much the case as usual with Shav, because he was resting KG, who'd go out before the other team's starters swapped out.

[Boris Badenov]

Six pages of debate on a guy who's played 305 minutes of NBA ball in the last six years. Yep, I guess we're in rebuilding mode...

[LooseCannon]

The main fallacy of using OR in the case of a sparingly used reserve player is that your sample suffers from a downward selection bias -- i.e., Ranolph likely saw playing time against the crappiest of competition.

You forgot to mention that he is likely seeing playing time with the crappiest of teammates that were available.

[LooseCannon]

Six pages of debate on a guy who's played 305 minutes of NBA ball in the last six years. Yep, I guess we're in rebuilding mode...

It's not a trade Rondo thread.  Glass half full.

[kozlodoev]

Six pages of debate on a guy who's played 305 minutes of NBA ball in the last six years. Yep, I guess we're in rebuilding mode...

It's not a trade Rondo thread.  Glass half full.

Or a Jeff Green is a top 5 SF thread