RealGM 2023 Top 100 Project - #51 (Draymond Green)

[OhayoKD]

I haven't checked bisected the series the series like that, but sure, I'll take your word for it. Though I'm pretty sure it's the first 3(where the bucks almost went 3-0) vs the last 3 games for defense. I'd say the main defensive issue was Kawhi being left in single coverage(like butler and kd would be after) too much but on the whole they held the raptors 5-points below and were an over-time(with giannig fouling out) away from going 3 games up on contender-level cast(2020) + kawhi.

I am choosing an inclusive flame which factors in pretty much everything including the regular-seasons from teams that were much better in the playoffs and performances with the best players injured.

You are not using any frame thus allowing you to arbitrarily decide what is relevant or not relevant. Should we retroactively boost the Bucks for winning their last 4 games vs the suns and their last 4 vs boston?

The Bucks generally overperform in the playoffs statistically whether you go flat or rolling. If you are going to gerrymander whatever series(or in this case, the portions of series.....), then I think it makes plenty of sense to point out the larger picture.

There is no shortage of caveats and nuances i can push to excuse Giannis or make his good stuff look more impressive, but I don't start with that because ultimately there's no real way to get a nuetral assessment of a player if you're going to start by disregarding the forest so you can paint a narrative on convenient-looking trees.

Looks like the first part we're reaching a natural conclusion.

On the second, well key thing here:

Zooming out to include all information doesn't mean you're going to reach a better conclusion.

If I want to evaluate how the Bucks will fair against contenders, then why would I care about how they did in a series against a team whose big minute players were Luke Kennard & Wayne Ellington? There's nothing arbitrary about saying that the Kennard/Ellington team wasn't a contender. They just obviously weren't, so why would we choose a lens that treats that matchup as if it's just as relevant as later rounds?

[/quote]
SRS is useful because beating up average to decent to merely good teams is relevant to performance vs better teams. Otherwise we are basically working off a sample of 5 where the Bucks did better than they should have in 2.

As is, you can toss out the luke kennard series and the Bucks would still grade out as risers the Bucks also blew to shreads the defending conference finalists +kyrie irving fresh off sweeping a decent team. And the Bucks also got outshot on open looks there too.

FWIW, espn had that one as a toss-up.

I don't think "zooming in" actually leads where you think it does if you zoom-in evenly. The Bucks score +7 vs the raptors with a rating that ignores kawhi improving, the gasol trade and also ignores the sixers trading for one jimmy butler and turning into a top 5 defense.

Re: arbitrary last 4 games. When you lose the last 4 games in a series where the first team to win 4 games wins the series, this is called a "backdoor sweep", and in general means that you got figured out and were the worse team by a clear margin. There are exceptions to this - injuries in all sports, luck in luck-dominated sports like baseball or hockey - but short of that, the idea of looking at all 4-2 series as equally close is just wrong. In a 7 game series, a team's job is not to make sure they don't lose a game, but to make sure they find the best approach so they can close out the other team. And that's precisely what happened in the Bucks-Raptors series.

But luck is also a factor in basketball. The Raptors were a double-overtime away from going 3-0 down after losing 2 games where they outshot the Bucks from open 3's and the last two games were single digit losses. The series differential was 1-point a game and per Ben, van-vleet not getting hot was the difference between the Bucks winning in 6 and losing in 6. If you want to say it wasn't --as close-- because it was a reverse sweep is fine. But it was a close series. The Raptors flirted very seriously with losing despite variance arguably favoring them from the getgo.

This wasn't a "in general" reverse sweep.

Re: gerrymander/narrative-paint/etc. Look man, if you think that my starting point on all this was being biased against Giannis, you're starting with the wrong information, as is almost always the case when you presume that starting from someone's "bias" is the right way to understand their thought process.

Do you really think I "hated" Giannis as he rose to prominence?
Presuming you sanely answer "no", what is it exactly you think led me to be biased against Giannis?
Don't you think it's maybe a bit more likely that when I tell you that his team getting upset quite early 3 times in 5 years concerns me, that it actually concerns me?

As always: I can be wrong and pointing out a different lens of duration might help me see how I'm making a molehill into a mountain, but no, I'm not arguing that beating the Pistons doesn't matter because I hate Giannis. I'm arguing it because beating Kennard/Ellington obviously does not tell us anything about how adept you are at beating contenders.

[/quote]
I never said you hate Giannis, not is it neccesarily to hate Giannis to a skewed conclusion. But when you only apply caveats one way and then filter out results one way and then embelish what works in your direction, your conclusion is probably skewed.

Case-in-point: Repeatedly talking about Luke kennard when you could throw out the first round and the Bucks would still see their srs go up on the back of their defensive performance.

[MyUniBroDavis]

In the context of the bucks defense vs the raptors, don’t remember it all too well but FWIW the 76ers held the raptors below their average offense in 6/7 games while the bucks did in thrice, all in the first 3 games of the series

[Doctor MJ]

Looks like the first part we're reaching a natural conclusion.

On the second, well key thing here:

Zooming out to include all information doesn't mean you're going to reach a better conclusion.

If I want to evaluate how the Bucks will fair against contenders, then why would I care about how they did in a series against a team whose big minute players were Luke Kennard & Wayne Ellington? There's nothing arbitrary about saying that the Kennard/Ellington team wasn't a contender. They just obviously weren't, so why would we choose a lens that treats that matchup as if it's just as relevant as later rounds?

SRS is useful because beating up average to decent to merely good teams is relevant to performance vs better teams. Otherwise we are basically working off a sample of 5 where the Bucks did better than they should have in 2.

As is, you can toss out the luke kennard series and the Bucks would still grade out as risers the Bucks also blew to shreads the defending conference finalists +kyrie irving fresh off sweeping a decent team. And the Bucks also got outshot on open looks there too.

FWIW, espn had that one as a toss-up.

I don't think "zooming in" actually leads where you think it does if you zoom-in evenly. The Bucks score +7 vs the raptors with a rating that ignores kawhi improving, the gasol trade and also ignores the sixers trading for one jimmy butler and turning into a top 5 defense.

- Obviously there's going to be correlation there, but it's absolutely not a causal rule.

- The Celtics with Kyrie were thought to be a dangerous team because of their star power. In reality they were a more dangerous playoff team both prior and later despite having less talent because Kyrie was a nightmare for the Celtics rather than their opponents.

Reading this I feel like you're going to think I'm just dismissing them because it's another thing to dismiss, but to be clear I'm not saying beating the Celtics meant nothing. I'm only objecting to you talking about them like they were a team that lived up to their expectations. This isn't just a situation where a team made some moves after a loss, this was a situation where a franchise was watching in horror as their putative franchise player light the team on fire. That's relevant to the context here.

Re: arbitrary last 4 games. When you lose the last 4 games in a series where the first team to win 4 games wins the series, this is called a "backdoor sweep", and in general means that you got figured out and were the worse team by a clear margin. There are exceptions to this - injuries in all sports, luck in luck-dominated sports like baseball or hockey - but short of that, the idea of looking at all 4-2 series as equally close is just wrong. In a 7 game series, a team's job is not to make sure they don't lose a game, but to make sure they find the best approach so they can close out the other team. And that's precisely what happened in the Bucks-Raptors series.

But luck is also a factor in basketball. The Raptors were a double-overtime away from going 3-0 down after losing 2 games where they outshot the Bucks from open 3's and the last two games were single digit losses. The series differential was 1-point a game and per Ben, van-vleet not getting hot was the difference between the Bucks winning in 6 and losing in 6. If you want to say it wasn't --as close-- because it was a reverse sweep is fine. But it was a close series. The Raptors flirted very seriously with losing despite variance arguably favoring them from the getgo.

This wasn't a "in general" reverse sweep.

There's luck in everything, but basketball is far more deterministic than baseball or hockey.

Re: not "an "in general" backdoor sweep. I think you can say that meaningfully with the fact that a team went 2-0 instead of 1-0 before losing the rest of the games. Beyond that, what I'd say is that losing 4 games in a row doesn't mean you couldn't possibly win a game, but it does mean the odds are vanishingly small that you're actually playing like the better team during that stage of the series.

Re: gerrymander/narrative-paint/etc. Look man, if you think that my starting point on all this was being biased against Giannis, you're starting with the wrong information, as is almost always the case when you presume that starting from someone's "bias" is the right way to understand their thought process.

Do you really think I "hated" Giannis as he rose to prominence?
Presuming you sanely answer "no", what is it exactly you think led me to be biased against Giannis?
Don't you think it's maybe a bit more likely that when I tell you that his team getting upset quite early 3 times in 5 years concerns me, that it actually concerns me?

As always: I can be wrong and pointing out a different lens of duration might help me see how I'm making a molehill into a mountain, but no, I'm not arguing that beating the Pistons doesn't matter because I hate Giannis. I'm arguing it because beating Kennard/Ellington obviously does not tell us anything about how adept you are at beating contenders.

I never said you hate Giannis, not is it neccesarily to hate Giannis to a skewed conclusion. But when you only apply caveats one way and then filter out results one way and then embelish what works in your direction, your conclusion is probably skewed.

Case-in-point: Repeatedly talking about Luke kennard when you could throw out the first round and the Bucks would still see their srs go up on the back of their defensive performance.

Ah, but how do you know it's "in my direction"? What does "my direction" even mean?

Seems to me you're assuming that I'm zooming in what I'm zooming in on in order to justify the player evaluation I prefer, rather than expecting that I'm zooming in on what seems most meaningful to me as a basketball analysis. So, why would you assume I'm doing this as a matter of implicit knowledge in response to a post from me where I'm explicitly saying that this is not what I'm doing?

Perhaps you're just thinking that I can't see my own bias, but of course you could have just said that.
Instead seems like you took us in circles.

Re: Repeatedly talking about Kennard... I'm talking about Kennard because your preferred approach included a team led by Kennard, which shows a categorical issue with it. If you simply wanted to talk about the effective SRS of beating a particular team - like the Celtics - I might push back because of specific context, but the Kennard concern is something will exist to some degree in every season's data and we began this with you applying it across every year in Giannis' career to get a tally implying his teams got better in the playoffs.

I don't mean to say you can't use this as part of a first-pass analysis, but it's not a statement that amounts to some kind of end-proof. No, I don't agree Giannis' teams have been better in the playoffs as a matter of course. I think they've shown to be relatively brittle. It's possible that that brittleness can be explain entirely with correlations to health, but I think this is unlikely.