Daniel Theis - Done for the season

[csfansince60s]

It really sucks to lose Theis at this stage.  His athleticism and defensive timing was unique compared to the rest of the bigs on our roster.  He definitely would have helped in certain matchups come playoff time.

Pretty much this.  Right now the C's would be slated to play Miami in the 1st round.  Monroe is unplayable against their bigs, they will just draw him out to the perimeter and it will be a layup and wide open 3 drill.  The loss of Theis will be greatly missed come playoff time especially against certain matchups.

Is Whiteside injured?

[The Oracle]

It really sucks to lose Theis at this stage.  His athleticism and defensive timing was unique compared to the rest of the bigs on our roster.  He definitely would have helped in certain matchups come playoff time.

Pretty much this.  Right now the C's would be slated to play Miami in the 1st round.  Monroe is unplayable against their bigs, they will just draw him out to the perimeter and it will be a layup and wide open 3 drill.  The loss of Theis will be greatly missed come playoff time especially against certain matchups.

Is Whiteside injured?

Whiteside missed the last game and will not play tonight at Portland.  It sounds like he isn't out to long right now with a hip problem.  Frankly the heat are a much much better team without him.  Their big man rotation of Adebayo/Johnson/Olynyk and small ball lineups with Winslow at the 4 will be big time trouble for some team in the playoffs. 

[Erik]

1. Baynes was absolutely better than Theis the first half of the season. After that, his stamina limits became clear, and Theis made himself not useless, hence things evened out. They're largely a wash right now, but to me having a rested Baynes is as important as having Theis, if not more.

By what metric? You seem to be using the eyeball test and citing advanced analytics as useless, which seems fairly arrogant given the fact that you aren't even a basketball expert. Even basketball experts use personal biases to give incorrect rankings. Just look at ESPN's NBARank every year.

2. I'd use any metric. Just have to understand its limitations. In the particular case of boxscore metrics, you have to realize that you're comparing points, rebounds, assists, steals, blocks, etc. that have been minced through some sort of fancy formula. Unsurprisingly, in the VORP that you presented is largely similar for two players that have, on average, largely similar measurable stats. And given that the theoretical "replacement" has a constant value of -2, VORP ends up by definition a function only of player stats.

Still haven't cited a metric that does a better job at comparing 2 players' production in a given season than VORP. You're the one that stated that it is a useless metric. Defend that position. Find a metric that shows that Baynes has played better than Theis that doesn't just rely on your personal opinion.

3. I'm not upset. I just think everyone around here would rather have Kevin Durant or Stephen Curry over Andre Drummond. But perhaps I'm wrong.

Have? In what sense? In general on their roster? Sure, so would I, due to historical data for expected future value. However, you can't use a metric that is directly weighted by minutes played to compare two players who played a delta of 14 games if your argument is that the expected value of 1 minute of Steph Curry is greater than the expected value of 1 minute of Drummond. I have no doubt that had Curry played the same number of games, that he'd be ranked higher. But, we don't have to doubt... it's right there. VORP = BPM * % of minutes played. BPM shows that Curry performed better per minute. However, I hope we can agree that there is a chance that Andre Drummond could play 1 single season better than Steph Curry. Advanced analytics would show that whereas a subjective test probably wouldn't.

[vjcsmoke]

Sad to see Theis is done.  He made a lot of good contributions this year and I felt he was becoming more comfortable and better in our system every day.

[rocknrollforyoursoul]

What a sucker punch. This season's going down the toilet all of a sudden. I'm SO SICK of all the injuries to this team. I don't ever wish injuries on anyone, but it just boggles my mind how teams like GS and CLE almost never have to adjust to injuries to their major players, but Boston gets hit year after year after year. Going back to KG's knee in '09, the Celtics just can't escape this crap.

I guess you did say "almost" never.   

Losing Theis sucks, but lets try to keep things in perspective.  We have a SEVEN game lead in the loss column on Cleveland with 15 to play.  We are a lock for the #2 seed, and have a month to get all our key players ready for the playoffs.  Theis's absence may limit how flexible our rotation can be in certain situations but it won't be the difference between winning and losing a series this season.  If Kyrie can't get healthy - that's the sucker punch.

Ha! This was the main reason, in fact, that I said "almost never"—I figured someone would bring up the season in which Love and Irving missed time in the NBA Finals.

That said, LeBron is the key for whatever team he's on, and neither the Cavs nor the Heat ever had to deal with a significant LeBron injury. And the Warriors have had virtually no major injury concerns the last four seasons (aside from Curry's current ankle trouble, which doesn't seem too bad).

[kozlodoev]

2. I'd use any metric. Just have to understand its limitations. In the particular case of boxscore metrics, you have to realize that you're comparing points, rebounds, assists, steals, blocks, etc. that have been minced through some sort of fancy formula. Unsurprisingly, in the VORP that you presented is largely similar for two players that have, on average, largely similar measurable stats. And given that the theoretical "replacement" has a constant value of -2, VORP ends up by definition a function only of player stats.

Still haven't cited a metric that does a better job at comparing 2 players' production in a given season than VORP. You're the one that stated that it is a useless metric. Defend that position. Find a better metric.

Which part of "box score based metrics suck, you might as well just look at box scores" wasn't clear? The "metrics" you're trying to ram here posit that what's recorded in a standard NBA box score is sufficient to evaluate players, and that just isn't true.

As you pointed out, VORP, in particular, is essentially BPM. And the BPM model is essentially worthless. The statistical "estimates" of this model fail the commonly used significance criteria so hard that they'll get laughed at by any scientific referee worth their salt. In layman terms, this means that most of the coefficient the model predicts are indistinguishable from zero and therefore completely worthless.

[mgent]

Theis isn't a difference maker.

Smart is.  We need that guy back.

[nickagneta]

However, I hope we can agree that there is a chance that Andre Drummond could play 1 single season better than Steph Curry.

Only if Curry barely plays that year, otherwise, I don't see there being any chance for Drummond to have a better season than Curry.

[Yoki_IsTheName]

Such a shame to lose Theiss for the rest of the year. He's been performing well.

Does this mean more PT for Yabusele?

[SHAQATTACK]

We re screwed .

[bopna]

Injuries are going to be the C's undoing...

They have been since 2009..not one single season we completed the playoffs healthy.. 11 yrs and counting...and sincr then Lebron has ruled the East..

[SHAQATTACK]

Injuries are going to be the C's undoing...

They have been since 2009..not one single season we completed the playoffs healthy.. 11 yrs and counting...and sincr then Lebron has ruled the East..

and Lebron is never injured.  like he is sprinkled with fairy dust.

[Erik]

2. I'd use any metric. Just have to understand its limitations. In the particular case of boxscore metrics, you have to realize that you're comparing points, rebounds, assists, steals, blocks, etc. that have been minced through some sort of fancy formula. Unsurprisingly, in the VORP that you presented is largely similar for two players that have, on average, largely similar measurable stats. And given that the theoretical "replacement" has a constant value of -2, VORP ends up by definition a function only of player stats.

Still haven't cited a metric that does a better job at comparing 2 players' production in a given season than VORP. You're the one that stated that it is a useless metric. Defend that position. Find a better metric.

Which part of "box score based metrics suck, you might as well just look at box scores" wasn't clear? The "metrics" you're trying to ram here posit that what's recorded in a standard NBA box score is sufficient to evaluate players, and that just isn't true.

As you pointed out, VORP, in particular, is essentially BPM. And the BPM model is essentially worthless. The statistical "estimates" of this model fail the commonly used significance criteria so hard that they'll get laughed at by any scientific referee worth their salt. In layman terms, this means that most of the coefficient the model predicts are indistinguishable from zero and therefore completely worthless.

If you aren't looking to box scores to grade a player's performance, what basis did you have to conclude that Theis didn't play better than Baynes? What metric of "hustle points" did you come up with that took box score from incomplete to complete? I'm on the edge of my seat.

However, I hope we can agree that there is a chance that Andre Drummond could play 1 single season better than Steph Curry.

Only if Curry barely plays that year, otherwise, I don't see there being any chance for Drummond to have a better season than Curry.

You can't imagine Drummond ever playing better than Steph? He's 24 years old and this is his first breakout season, much like Steph broke out at 24. I feel that you could assign a non-zero value to this fairly easily.

[Vox_Populi]

3. I'm not upset. I just think everyone around here would rather have Kevin Durant or Stephen Curry over Andre Drummond. But perhaps I'm wrong.

Not sure what the point of this is. I'd imagine most people know judging any metric by the individual placement of players is a bad idea.

[Ilikesports17]

I love this argument. Three questions:
1) I'd like to know if you think that Baynes played better than Theis this season (and you disagree with me), or if you're just doing your usual thing where you follow me around and try to contradict what I say because we don't get along on off topic stuff.
2) Which metric would you rather use to compare two players' current performance?
3) Since you're upset that Steph Curry is ranked lower in a metric that accounts for time played (and he's played 14 less games than Drummond), what do you make of the fact that Theis is ranked +1.1 vs Baynes on 20% less minutes?

Also, let's not forget that Drummond is having an absolute monster season and should have been on the initial all star team had we used a correct metric other than (see your answer for #2), so it's not like he's a bum. VORP doesn't know name value or any other theoretical value. It's the best pure statistical representation of how much worth a player has provided a team this year. There are always a few outliers (Klay Thompson is usually the most significant), but it's generally pretty good -- the voted allstar list and VORP are fairly similar. We could switch to BPM which doesn't account for time played if you'd like. Theis is still much better than Baynes, but by more. That would be a +4.7 Swing (which is nuts).

1. Baynes was absolutely better than Theis the first half of the season. After that, his stamina limits became clear, and Theis made himself not useless, hence things evened out. They're largely a wash right now, but to me having a rested Baynes is as important as having Theis, if not more.
2. I'd use any metric. Just have to understand its limitations. In the particular case of boxscore metrics, you have to realize that you're comparing points, rebounds, assists, steals, blocks, etc. that have been minced through some sort of fancy formula. Unsurprisingly, in the VORP that you presented is largely similar for two players that have, on average, largely similar measurable stats. And given that the theoretical "replacement" has a constant value of -2, VORP ends up by definition a function only of player stats.
3. I'm not upset. I just think everyone around here would rather have Kevin Durant or Stephen Curry over Andre Drummond. But perhaps I'm wrong.

This is about where I stand.

Baynes is our physical defensive presence. That's more crucial than what Theis brings.