Greatest Peaks series (Thinking Basketball/Ben Taylor)

[PistolPeteJR]

Conclusion from this video - Curry is consistent hand checker

On a more serious note - it's really good video, one of the most nuanced one with tons of details. I like the off-ball game analysis, but I also like how Ben described his passing and handles - they are excellent, but not historically great.

I also think that his defensive breakdown paints very mixed picture - it shows that Curry has great anticipation and that he usually gives strong effort (which I view similary in my obseravtions) but he's also overly physical and it often leads him to overcommit or make dumb fouls. I know that refs allow small guys like Curry to be more physical, but let's be honest - by modern officiating most of defensive plays Ben showed should be called a foul.

Unrelated to this particular video - I wish he spent as much time talking about Magic's defense for example, because looking at these videos you make come to conclusion that Magic was weak and Curry was clearly positive, which is not something I'd agree with. Magic's effort was sometimes iffy in RS, but in playoffs he was usually clear positive defender and with his size and smarts, he was clearly more valuable than Curry.

This was the second thing to stand out to me. The old refrain of "small players get away with a lot more than bigger players do". It gives them a significant advantage on both ends of the court. The trouble is that there is no real way to quantify that advantage for better or for worse when evaluating players. Sure, refs are imperfect, and always have been, but this is a real problem for sure. I wish it would be addressed already. With all of the technological advances of today, officiating should be getting better and more consistent, not worse. While the game has amped up, we can surely do better than this.

[PistolPeteJR]

I think that Ben didn't show many of his biases for most of this project, but I see the clear difference in Bird/Curry episodes in comparsion to Magic. You can simply feel that Ben prefers their styles of play.

This series was really great. I really enjoyed it.

However, the biggest downside to it to me was how much weight Ben puts into scalability/portability. While it is a legitimate metric for sure in comparisons, I find he puts far too much weight into it, to the point where he pretty clearly seems to prefer those types of players and it clouds his objectivity to an extent. The other side of the coin, heliocentrism, is less preferred by him, and therefore those types of players are penalized due his glorification of those higher on his portability scale. The downside with that logic is that it assumes that those that score higher in scalability are scoring equally as high in heliocentrism as their competition.

[frica]

Conclusion from this video - Curry is consistent hand checker

On a more serious note - it's really good video, one of the most nuanced one with tons of details. I like the off-ball game analysis, but I also like how Ben described his passing and handles - they are excellent, but not historically great.

I also think that his defensive breakdown paints very mixed picture - it shows that Curry has great anticipation and that he usually gives strong effort (which I view similary in my obseravtions) but he's also overly physical and it often leads him to overcommit or make dumb fouls. I know that refs allow small guys like Curry to be more physical, but let's be honest - by modern officiating most of defensive plays Ben showed should be called a foul.

Unrelated to this particular video - I wish he spent as much time talking about Magic's defense for example, because looking at these videos you make come to conclusion that Magic was weak and Curry was clearly positive, which is not something I'd agree with. Magic's effort was sometimes iffy in RS, but in playoffs he was usually clear positive defender and with his size and smarts, he was clearly more valuable than Curry.

This was the second thing to stand out to me. The old refrain of "small players get away with a lot more than bigger players do". It gives them a significant advantage on both ends of the court. The trouble is that there is no real way to quantify that advantage for better or for worse when evaluating players. Sure, refs are imperfect, and always have been, but this is a real problem for sure. I wish it would be addressed already. With all of the technological advances of today, officiating should be getting better and more consistent, not worse. While the game as amped up, we can surely do better than this.

This is a thing in soccer too.

Reminds me of this unit where the small man got a soft call.

Watch on YouTube

[Odinn21]

Well, the series is basically over, only the ranking episode left.
And I have to say that I find Taylor excusing the issues of rather revolutionary players like Curry and Garnett a bit annoying.
I talked to him about Duncan's injuries having an impact on his "against strong defenses numbers" and Timmy having a higher gear in playoffs not showing up in those numbers, and he wasn't doing justice. All he said was "you're not getting what I'm doing", then did the exact same thing for Curry in the last episode. Literally.
TBH, he lost a bit of credit about his objectivity for me when he was doing a Q&A on Discord because he said "after watching so many KG and Duncan videos in quick successions, I was like -hey KG's not doing like that-". I can literally do the same for Timmy but I know that having a mindset like that will have a foregone and probably tainted conclusion.
This is not about me being a Duncan fan and Taylor seeing things differently. He has been either too apologetic or too excusing of his favourite and/or rather revolutionary players ever since Robinson episode.

He spent more time on interpolating things outside of David Robinson's actual peak seasons instead of actually talking about them.
Then Garnett episode was "he's better on a better team" and scoring is rather overrated. Not if you're on a team like 2000s Minnesota Timberwolves.
Then Durant's episode was "he's tall and he can function next to a better scorer than himself".
Then Curry's episode was another excusing episode. The episode ends with "Curry's scoring per 75 in the playoffs without KD&Klay", mate that accounts for one seventh of Curry's entire play time. It is not significant enough to make such a claim.

I've generally liked the videos but I can't really argue with what you posted lol. You can definitely tell who Taylor likes and dislikes when you watch the video. His entire Garnett video was great and all, but he seemed to hand-wave Garnett's lack of scoring as 'not his fault' even though that's probably a reason Garnett's Wolves weren't as successful as they could have been. Meanwhile, he didn't get an episode, but the potshots at Wilt Chamberlain were kind of forced.

I liked the videos too. However I really think that him criticizing some players and not saying "but" in the next sentence, then doing it for the players he clearly favours is a bit too obvious. Either say but for all and make their case, or do not say but at all and leave that to us.

Most of his bias is coming from this skewed sense of portability. According to him, portability is scaling along as the team gets better. Better player on a better team. That's why he's been excusing the players he like better.
But that doesn't make much sense because many of these legends spent a good chunk of their primes on average or worse teams.
For example, Kareem is like the Garnett of the '70s but his situation doesn't get that much attention because Kareem's scoring was higher and that helped his teams to have bigger successes. Should we go "but Kareem was more portable than that" instead and overlooking actual things?..
Non-winning situations, playing on a non-contending team are common even for these great and still fortunate players. On average, non-winning situations are far more common.
So, looking at what it could've been on a better team is not a reason to overlook what happened on a bad team in reality.

---

Let's start to summarize the project:

1. What video changed your perception the most about given player?
2. Which video is the best overall (however you define it)>
3. Which video is your least favorite?
4. Did this project change your perception about 10 greatest peaks post-merger?
5. Who had the best peak ever after watching the project?
6. Pick one player from any era you'd like to see with similar video.

1. I guess none. The only thing I wasn't aware of was Bird's shooting rate dropping against good defenses that hard.
2. Probably one of Olajuwon, Garnett or O'Neal, due to meaningful nuances. I'll probably say Shaq episode.
3. David Robinson episode by far. He spent more time on interpolating stuff than talking about what actually happened. Karl Malone owned Admiral twice in the playoffs from '94 to '96 and I believe that didn't make the news in the episode. It wasn't just Olajuwon that Robinson struggled against.
4. No.
5. I'm probably going to say Jordan. I have '00 Shaq as the goat single season but him not caring enough in 2000-01 reg. season is too big of a deal to overlook. James feels like he falls a bit short of Jordan. It may sound too nitpicky but the competition is so stiff and we're deciding within such close quarters. Kareem having a bad ft shooting series uncharacteristically in '78 is also a weak point against Jordan.
6. Probably Oscar Robertson. Or Willis Reed.

[Djoker]

Let's start to summarize the project:

1. What video changed your perception the most about given player?
2. Which video is the best overall (however you define it)>
3. Which video is your least favorite?
4. Did this project change your perception about 10 greatest peaks post-merger?
5. Who had the best peak ever after watching the project?
6. Pick one player from any era you'd like to see with similar video.

All in all it was truly a fantastic series.

1. Hmm this is tough. Maybe the Shaq video. I never really paid that much attention to his footwork and movement to establish position. It definitely goes under the radar the work that he does before getting the ball. Ben also did a good job explaining that fouling Shaq intentionally in the halfcourt is actually a bad defensive play. 50% free throw shooting is still 1.00 PPP which is very good in the halfcourt at least in that era.

2.The Shaq video was the best and my favorite.

3. Probably Tim Duncan. He focused a lot on his weaknesses and even oversold some like his supposedly poor passing. I always saw and still see Duncan as a good passer not an all-time great but Ben's tone almost sounded like he was mediocre. Even his defense and his playoff scoring was kind of getting underrated. Maybe the whole KG > Duncan narrative soured me for that one.

4. Not a lot, no. Maybe because I already did a lot of research on these players watching historical footage and have been watching the NBA live for almost 24 years now too. One player that took a slight boost for me is Shaq. I always thought Kareem was better but now I'm more open to taking Shaq. Kareem's extra points in efficiency can be balanced by Shaq's higher scoring volume per 75, better offensive rebounding and the havoc that Shaq caused by getting people in foul trouble. Shaq is probably the only one whose valuation changed after this series.

5. Overall I feel like MJ is the best. His peak I really think hasn't been surpassed or even matched by anyone in league history. He has the fewest/smallest weaknesses of any player ever. And he thoroughly dominates the boxscore metrics including the big 4 offensive categories. At his peak 89-91 In the postseason: 32.6 pts/75, +6.1 rTS, great creation and 10.5 TOV%. And his defense while not on the level of great defensive big men is still the cream of the crop for perimeter players. He's the best ever at his peak.

For example compared to Lebron who I think will be #2 for most people. His postseason peak in 12-13 is 27.3 pts/75, +5.4 rTS, slightly better creation and 11.9 TOV%. It's hard to overlook 5 extra points per 75 on similar efficiency that you get from MJ. I just don't see what Lebron could have done on the floor that can make up for that kind of gap.

Like Ben himself said in his top 40 list "Jordan's peak is not unassailable..." but it's not easy to make an argument IMO.

6. Dwyane Wade. He's the one guy I feel should have made this series more than anyone. Wade's standing has been hurt by his lack of longevity. I think peak Wade is clearly one of the 15 best players in the NBA since the merger. As I explained in previous posts I have him over KD, Curry, Robinson, KG and Walton.



One other thing I do wish is that two separate lists are made for perimeter players (PG, SG, SF) and another list for bigs (PF, C).
It's hard to compare completely dissimilar players especially when going down the list.

[Todeasy]

Let's start to summarize the project:

1. What video changed your perception the most about given player?
2. Which video is the best overall (however you define it)>
3. Which video is your least favorite?
4. Did this project change your perception about 10 greatest peaks post-merger?
5. Who had the best peak ever after watching the project?
6. Pick one player from any era you'd like to see with similar video.

1) Non really, although I am even more impressed with Walton’s passing now. Some of the windows he was threading passes through were MINUSCULE.

2) Walton, Shaq, and Bird maybe. Shaq’s a bit above the rest.

3) Hard to say cause I really did like them all, but if forced to pick I’d say Durant’s.

4) Not really no.

5) Jordan by a hair if looking at complete 2 year stretches, maybe Lebron or others if isolating post-season.

6) Mosos Malone, simply because outside of what I’ve read from here I’m least familiar with his game.

[Statlanta]

Let's start to summarize the project:

1. What video changed your perception the most about given player?
2. Which video is the best overall (however you define it)>
3. Which video is your least favorite?
4. Did this project change your perception about 10 greatest peaks post-merger?
5. Who had the best peak ever after watching the project?
6. Pick one player from any era you'd like to see with similar video.

1. The Kareem and Magic videos. He really poked the rebounding and defense of both respectively. Also the transition within the Spurs organization in he late 90’s.

2. Shaq. Total complete video and very rewatchable. From WOWY to demonstrating how a C can anchor an actually good offense to demonstrating how his strengths and flaws fit in-era.

3. Kevin Garnett’s video. Didn’t finish on first watch.

4. It changed that Walton certainly belongs alongside the best.

5. Shaquille O’Neal

6. I personally felt Greatest Peaks meant most dominant runs so I felt Moses deserved a video.

[Blackmill]

Let's start to summarize the project:

1. What video changed your perception the most about given player?
2. Which video is the best overall (however you define it)>
3. Which video is your least favorite?
4. Did this project change your perception about 10 greatest peaks post-merger?
5. Who had the best peak ever after watching the project?
6. Pick one player from any era you'd like to see with similar video.

1. Not a whole late changed for me. However I do think Hakeem's passing is worse than I imagined. I always knew there was a healthy margin for error in my evaluation of his offense since I needed to watch more of him.

2/3. I don't have a good answer to this. I largely agreed with and enjoyed all the videos.

4. I may move Hakeem down a notch.

5. I think LeBron's presentation was the most impressive. Here's Ben's concluding remarks for LeBron

1. One of the best perimeter defenders ever.
2. Strong argument as a top-10 scorer.
3. Strong argument as a top-10 playmaker.
4. Weak off-ball value.

and for Jordan

1. Strong all-league defender but not an all-time defensive wing.
2. Strong argument for best scorer ever.
3. Very good but not elite playmaker.

Based strictly on the presentation and conclusion, I think LeBron looked a step above Jordan, even though I know Ben last had several Jordan seasons above LeBron's peak.

6. Dr. J especially if it includes his ABA seasons. But I would be more interested in team-level analysis. Although maybe that's not Ben's forte. A collaboration between Coach Pyper (Half Court Hoops) and Ben on some of the notable playoff series in NBA history would be amazing. Unfortunately I know Coach Pyper doesn't have time for this (the amount of film study he does to keep up with the current teams is insane).

[yoyoboy]

Let's start to summarize the project:

1. What video changed your perception the most about given player?
2. Which video is the best overall (however you define it)>
3. Which video is your least favorite?
4. Did this project change your perception about 10 greatest peaks post-merger?
5. Who had the best peak ever after watching the project?
6. Pick one player from any era you'd like to see with similar video.

1. KAJ, a bit higher on his post scoring now then before.
2. Shaq, just thought the whole thing was really well done, most well rounded imo.
3. David Robinson, too much of a career perspective to me, completely skipped the matchup against the Jazz in '94/'96.
4. Probably a bit higher on KAJ and Bird, not really lower on anyone in particular.
5. LeBron, unchanged opinion.
6. Draymond, duh.

A Draymond video would be absolutely sick. I wouldn’t mind Ben doing a series or just a video comparing the defensive tactics/abilities of the greatest defenders ever who didn’t necessarily make his best peaks.

[Celtic Koala]

Durant clearly had a greater peak than Magic, KG, Robinson, Kobe, Walton, Bird and Olajuwon. I don't understand why anyone would think his inclusion is controversial unless you never watched any of the above mentioned or you are just considering the player's peaks relative to their eras rather than in a vacuum (then Wilt would be the locked number 1 player).
Durant was a bad fit in OKC and Westbrook with 2 other MVP candidates never performed in any way better in the PS than with KD (3 first round exits and 1 second round exit after a massive carry job against a rebuilding OKC with Chris Paul).
I believe that this season will shed a light on how good KD (even more so than with GS as no one gives him credit for how much he developed his playmaking ability) is once you add some good perimeter players that can benefit from his gravity and he is not forced to take as many bad shots as he did in OKC.

[GSP]

Conclusion from this video - Curry is consistent hand checker

On a more serious note - it's really good video, one of the most nuanced one with tons of details. I like the off-ball game analysis, but I also like how Ben described his passing and handles - they are excellent, but not historically great.

I also think that his defensive breakdown paints very mixed picture - it shows that Curry has great anticipation and that he usually gives strong effort (which I view similary in my obseravtions) but he's also overly physical and it often leads him to overcommit or make dumb fouls. I know that refs allow small guys like Curry to be more physical, but let's be honest - by modern officiating most of defensive plays Ben showed should be called a foul.

Unrelated to this particular video - I wish he spent as much time talking about Magic's defense for example, because looking at these videos you make come to conclusion that Magic was weak and Curry was clearly positive, which is not something I'd agree with. Magic's effort was sometimes iffy in RS, but in playoffs he was usually clear positive defender and with his size and smarts, he was clearly more valuable than Curry.

This was the second thing to stand out to me. The old refrain of "small players get away with a lot more than bigger players do". It gives them a significant advantage on both ends of the court. The trouble is that there is no real way to quantify that advantage for better or for worse when evaluating players. Sure, refs are imperfect, and always have been, but this is a real problem for sure. I wish it would be addressed already. With all of the technological advances of today, officiating should be getting better and more consistent, not worse. While the game has amped up, we can surely do better than this.

Yeah Steph's handchecking and fouling was always blatant as was how much Draymond and Iggy were able to get away with on defense. Not to mention all their illegal screens. Always felt like Klay was the most fundamentally sound and fairly reffed defender out of their team.

[Todeasy]

One thing I want to bring up that is tangentially related to the Curry video is the concept of floor raising and how much one player can truly do. I'm not sure if it's one that's shared in this forum, but a common sentiment in the NBA circle is that one truly transcendent offensive player will have much more of an impact than a truly transcendent defensive player will. I frequently see the team's Ortg and Drtg cited in support of this, and in general the offensives around strong offensive engines seem to be better and more consistent.

Regardless of if you see Curry in that upper echelon of offensive players or the tier a step down, there's no denying that Curry is one of the best offensive players of this generation. Despite all that, according to Hollinger Team Statistics (not sure if this filters out garbage time) the Warriors sit at a measly 22nd in offensive rating. Now in my not so unbiased Warriors fan opinion, there isn't much more that Steph himself could do. He's having his regularly scheduled great scoring season, his off-ball activity is still historic, and although his turnovers have been up his passing seems normal for him. All the impact statistics (RAPTOR, LEBRON, RPM) bear out the same sentiment as well, as he ranks highly in them and the offensive is lifted 22.2 pp/100 with him on the court. Despite ALL THIS, the facts remain. The Warriors are a below average offense.

Seeing the results of this season it makes me wonder how much we have been underrating all time defensive anchors that maybe didn't have as strong of a supporting cast defensively. How much of pointing to middling Drtg surrounding these players as proof of their lesser impact compared to offensive anchors actually grounded in fact?

[parsnips33]

Let's start to summarize the project:

1. What video changed your perception the most about given player?
2. Which video is the best overall (however you define it)>
3. Which video is your least favorite?
4. Did this project change your perception about 10 greatest peaks post-merger?
5. Who had the best peak ever after watching the project?
6. Pick one player from any era you'd like to see with similar video.

1. Maybe not changed my perception per se, but I felt the KAJ video definitely gave me a much better impression of his game. Always knew he was a legend, but it was nice to see an in-depth breakdown of how he did his damage

2. Steph (obvious reasons) or Kobe were my favorites to watch

3. Honestly didn't have one, at least not one that sticks out from the bunch. All very well made and informative IMO

4. Would have to think about it more, didn't exactly have a top ten list in my head before the series

5. Hard for me not to say LeBron, but the Shaq video made a pretty strong case I think

6. Would love a Barkley video, he's one of those 1 of a kind players

[Jaivl]

One thing I want to bring up that is tangentially related to the Curry video is the concept of floor raising and how much one player can truly do. I'm not sure if it's one that's shared in this forum, but a common sentiment in the NBA circle is that one truly transcendent offensive player will have much more of an impact than a truly transcendent defensive player will. I frequently see the team's Ortg and Drtg cited in support of this, and in general the offensives around strong offensive engines seem to be better and more consistent.

Regardless of if you see Curry in that upper echelon of offensive players or the tier a step down, there's no denying that Curry is one of the best offensive players of this generation. Despite all that, according to Hollinger Team Statistics (not sure if this filters out garbage time) the Warriors sit at a measly 22nd in offensive rating. Now in my not so unbiased Warriors fan opinion, there isn't much more that Steph himself could do. He's having his regularly scheduled great scoring season, his off-ball activity is still historic, and although his turnovers have been up his passing seems normal for him. All the impact statistics (RAPTOR, LEBRON, RPM) bear out the same sentiment as well, as he ranks highly in them and the offensive is lifted 22.2 pp/100 with him on the court. Despite ALL THIS, the facts remain. The Warriors are a below average offense.

Seeing the results of this season it makes me wonder how much we have been underrating all time defensive anchors that maybe didn't have as strong of a supporting cast defensively. How much of pointing to middling Drtg surrounding these players as proof of their lesser impact compared to offensive anchors actually grounded in fact?

I mean, this is late Minnesotta Garnett, late-oughts Wade... all over again. Curry at least has a notable player on defense (Green), but they're void on the other front. Even if you're a trascendent player you can't do anything with 70% replacement-level players.

[Todeasy]

One thing I want to bring up that is tangentially related to the Curry video is the concept of floor raising and how much one player can truly do. I'm not sure if it's one that's shared in this forum, but a common sentiment in the NBA circle is that one truly transcendent offensive player will have much more of an impact than a truly transcendent defensive player will. I frequently see the team's Ortg and Drtg cited in support of this, and in general the offensives around strong offensive engines seem to be better and more consistent.

Regardless of if you see Curry in that upper echelon of offensive players or the tier a step down, there's no denying that Curry is one of the best offensive players of this generation. Despite all that, according to Hollinger Team Statistics (not sure if this filters out garbage time) the Warriors sit at a measly 22nd in offensive rating. Now in my not so unbiased Warriors fan opinion, there isn't much more that Steph himself could do. He's having his regularly scheduled great scoring season, his off-ball activity is still historic, and although his turnovers have been up his passing seems normal for him. All the impact statistics (RAPTOR, LEBRON, RPM) bear out the same sentiment as well, as he ranks highly in them and the offensive is lifted 22.2 pp/100 with him on the court. Despite ALL THIS, the facts remain. The Warriors are a below average offense.

Seeing the results of this season it makes me wonder how much we have been underrating all time defensive anchors that maybe didn't have as strong of a supporting cast defensively. How much of pointing to middling Drtg surrounding these players as proof of their lesser impact compared to offensive anchors actually grounded in fact?

I mean, this is late Minnesotta Garnett, late-oughts Wade... all over again. Curry at least has a notable player on defense (Green), but they're void on the other front. Even if you're a transcendent player you can't do anything with 70% replacement-level players.

I think we're saying the same thing really, Curry has good defensive/bad offensive help and the team rating reflects that. This isn't an indictment on Curry at all. However I've observed people saying in regards to great defenders like KG that because his team defense was bad in certain years, he couldn't have been as impactful as someone whose team showed far more consistency ala Duncan.

Not even to say that he necessarily was on that level, I'm simply pushing back on the principles of using team ratings to bolster someone's individual ability without context (a practice I see far more often in defensive anchors than offensive ones), and using Curry's current situation as an example.

[yoyoboy]

One thing I want to bring up that is tangentially related to the Curry video is the concept of floor raising and how much one player can truly do. I'm not sure if it's one that's shared in this forum, but a common sentiment in the NBA circle is that one truly transcendent offensive player will have much more of an impact than a truly transcendent defensive player will. I frequently see the team's Ortg and Drtg cited in support of this, and in general the offensives around strong offensive engines seem to be better and more consistent.

Regardless of if you see Curry in that upper echelon of offensive players or the tier a step down, there's no denying that Curry is one of the best offensive players of this generation. Despite all that, according to Hollinger Team Statistics (not sure if this filters out garbage time) the Warriors sit at a measly 22nd in offensive rating. Now in my not so unbiased Warriors fan opinion, there isn't much more that Steph himself could do. He's having his regularly scheduled great scoring season, his off-ball activity is still historic, and although his turnovers have been up his passing seems normal for him. All the impact statistics (RAPTOR, LEBRON, RPM) bear out the same sentiment as well, as he ranks highly in them and the offensive is lifted 22.2 pp/100 with him on the court. Despite ALL THIS, the facts remain. The Warriors are a below average offense.

Seeing the results of this season it makes me wonder how much we have been underrating all time defensive anchors that maybe didn't have as strong of a supporting cast defensively. How much of pointing to middling Drtg surrounding these players as proof of their lesser impact compared to offensive anchors actually grounded in fact?

Yep. Good post. The difference between the Warriors in 2016 posting a +8.1 relative offense and the Warriors in 2021 posting a -2.5 relative offense isn't Steph playing substantially better on offense in 2016. Steph is playing at probably 95% of the level he did in 2016 on offense once you consider the difference in supporting cast which made his life a lot easier.

People really need to stop using team offensive and defensive ratings as such an important measure of anything to be honest, unless you're applying a hell of a lot of context alongside it. Sometimes it can also be useful to show that a player's game is scalable, but people really overuse the whole "X player is the best offensive/defensive player on the #Y offense/defense in the league, therefore..."

KG is the obvious example of an absolutely elite defender stuck with a putrid defensive supporting cast. But then also, with LeBron, do people seriously think LeBron at 36 years old is having the best defensive season of his career anchoring the #1 ranked -6.3 relative defense despite Davis missing over a third of the season? While he's had other earlier seasons being behind defenses that ranked in the bottom half and even bottom third of the league. Sure there are seasons like 2014 where he obviously expended less effort, but for the most part, I think LeBron is playing just as good on defense as a lot of those seasons on poor defenses.

[Gregoire]

1. What video changed your perception the most about given player? Duncan
2. Which video is the best overall (however you define it)> Curry
3. Which video is your least favorite? Wilt and Russell
4. Did this project change your perception about 10 greatest peaks post-merger? Yes
5. Who had the best peak ever after watching the project? MJ
6. Pick one player from any era you'd like to see with similar video. Barkley

[KTM_2813]

1. Honestly, I learned something new about pretty much every player, from Magic's excellent ballhandling to Shaq's off-ball game. Walton was the player I learned the most about. I had always heard about how great a passer he was, but only when I saw his video did I fully realize what a savant he was.

2. Haven't seen the Curry one yet, but thus far my favorite was Kobe's. It's just so hard to find genuinely good analysis of his game.

3. I don't have one least favorite video, but every time a video mentioned portability, I felt like I only became more and more confused by the concept.

4. I was surprised to see Durant listed above guys like Dirk and Wade. His video made me appreciate him more, that's for sure. I don't think it moved the needle much on the top-ten peaks though. Would be surprised if he made the cut.

5. I felt that LeBron's entry came across as the most impressive, but gun to my head, I would likely still choose Jordan.

6. I would love to see a McGrady and/or Nash video. Those were my two favorite players growing up.

[drza]

Using just raw on/off is not a great way to evaluate impact, as unlike PIPM and RAPM, it does not try to adjust for the teammates you are playing for. And furthermore, if we just follow raw on/off, a past peak David Robinson is better than Shaq, Lebron, etc. It is notable but not the only thing.

A few thoughts about raw on/off for the playoffs. I agree with you that you should never look at a raw on/off (or really any stat) and say "player X is better than players Y, Z and Q". Stats always have to be tied to what they're measuring, and this stat is (to whatever degree noise allows) is trying to correlate a player's presence with the team's scoring margins. That's not the same thing as 'this player is better'.

With that said, I do think people way over-react (and under-appreciate) the information that can be found in playoffs on/off +/-. Yes, it's noisy. And yes, you have to pay attention to sample sizes and be more discerning than if you had a perfect dataset to work with. But as I've pointed out before, there is usable signal within that noise, and I think that signal is important BECAUSE there's such a lack of great postseason +/- information to work with. Even in the larger regular season samples, it is ABUNDANTLY clear that the box scores aren't equipped to accurately estimate overall impact. That's why there's a need for the impact stat approach in the first place. That need doesn't just go away for the playoffs, so relying purely on the box scores in the playoffs is just as lacking as it would be in the regular season...more-so, actually, because of the very small samples. Even box score stats can be gamed in smaller samples, so if we only have the box scores to go off, way too much of the real story is lost.

Anyway. You made a good point about raw on/off not adjusting sufficiently for teammates. It does offer a first-blush, basic correction for teammates that purely raw +/- doesn't, but not enough. So, one quick hack would be to compare an individual's playoffs on/off +/- with the on/off +/- of his teammates. There's several ways to do this, but for the below example I compared the highest playoffs on/off with the second-best on the team (among big-minute rotation players), and note the difference.

So, for this exercise, I use a few general rules to increase the likelihood of useful signal in the midst of the small sample/potential noise.

Rules:
*Teams had to make at least the Conference Finals
*Noting only the highest on-off +/- scores on the team as the ones likely driving the positive margin
*Using difference b/w highest on/off score and second-highest as a means to separate strong individual impact from strong units
*I looked at every conference finalist from 1997 - 2020 for this data
*Setting thresholds (to sort the 96 represented seasons)
I took the average on/off +/- of the top player on each team (+16.4)
I then took the average difference b/w top and 2nd player on each team (+7.7)

Results summary of the "16/8" club
31 player playoffs where team leader on/off > 16, difference b/w P1 & P2 > 8
Only 4 players (LeBron, Shaq, KG, Duncan) made list more than once (accounted for 12 of the 31 16/8 seasons)
Notable 1-offs include 1997 Jordan & 2006 D. Wade

Typically, the non-megastars fit into one of several categories not well captured by boxscores. Few non-intuitive 1-offs.
Defense: '04 Ben Wallace, 02 Doug Christie, '15 Draymond Green, '11 Luol Deng, '13 Marc Gasol, '99 Robinson
Floor stretchers: '00 Reggie Miller, '01 Ray Allen, '10 Rashard Lewis
Floor generals: '02 Jason Kidd, '16 Kyle Lowry, '06 Steve Nash
Mega 6th men: '05 Manu GInobili, '12 James Harden
Past prime stars: '97 Clyde Drexler, '99 Arvydys Sabonis, (Jordan, Robinson & Reggie would fit here too)
('17 Kawhi Leonard makes list, w/ small asterisk since he only played 12 of the team's 16 games that postseason)

Anyway, just a simple exercise like this shows a lot of, IMO, useful info. For example, you mention that by raw playoffs on/off +/- David Robinson looks better than Shaq and LeBron. Well, here we see that '99 Robinson was the only version of late-prime DRob to make this list (by 2001, even though he had a great on/off, he had been surpassed by Duncan), while both Shaq and LeBron were on here multiple times (7 total, between them).

I should note, NOT making this list doesn't necessarily mean a player didn't have a great postseason impact in a given season. They could've had a great on/off in a great unit (e.g. 2011 Dirk met the on/off threshold, but not gap over 2nd because unit was strong). But, on that note...

Focus: you were talking about Steph, in a post that compares him with KD and also mentions Magic, MJ and LeBron's postseason peaks. While it may not be rigorous, it's at least worth noting, to me, that out of all the many times that Steph and KD have played on contending teams (and their entire careers were in the databall era), neither of them had even one postseason run make the above list. Of the others listed, LeBron has the most such seasons on record (4) and even though we only have the last 2 seasons of MJ's Bulls career, he reached the thresholds in '97 and really was barely short of the thresholds to have a second such postseason run in '98 (led Bulls w/ +14.6 postseason on/off, +6.3 higher than 2nd on team). When you factor in that most of the other players in Ben's peak project have seasons like that as well...like I said, I at least take note.

Following up, of the 2, Curry seems to have the more impressive postseason +/- record. As I noted in a previous post, Curry's raw on/off +/- in his prime-ish seasons was +12.6, while Durant's was +5.1. Curry's record looks more like what I described for 2011 Dirk, where he had high numbers but may have been parts of units such that we couldn't ascribe the impact just to him (e.g. the Draymond effect), whereas Durant really doesn't have a postseason record where he stands out even to that extent in the +/- data.

Anyway. All of this is obviously food for thought. But to me it is useful information that adds depth to postseason analysis that you just can't find if you only rely on team results and/or boxscore data.

[LukaTheGOAT]

Using just raw on/off is not a great way to evaluate impact, as unlike PIPM and RAPM, it does not try to adjust for the teammates you are playing for. And furthermore, if we just follow raw on/off, a past peak David Robinson is better than Shaq, Lebron, etc. It is notable but not the only thing.

A few thoughts about raw on/off for the playoffs. I agree with you that you should never look at a raw on/off (or really any stat) and say "player X is better than players Y, Z and Q". Stats always have to be tied to what they're measuring, and this stat is (to whatever degree noise allows) is trying to correlate a player's presence with the team's scoring margins. That's not the same thing as 'this player is better'.

With that said, I do think people way over-react (and under-appreciate) the information that can be found in playoffs on/off +/-. Yes, it's noisy. And yes, you have to pay attention to sample sizes and be more discerning than if you had a perfect dataset to work with. But as I've pointed out before, there is usable signal within that noise, and I think that signal is important BECAUSE there's such a lack of great postseason +/- information to work with. Even in the larger regular season samples, it is ABUNDANTLY clear that the box scores aren't equipped to accurately estimate overall impact. That's why there's a need for the impact stat approach in the first place. That need doesn't just go away for the playoffs, so relying purely on the box scores in the playoffs is just as lacking as it would be in the regular season...more-so, actually, because of the very small samples. Even box score stats can be gamed in smaller samples, so if we only have the box scores to go off, way too much of the real story is lost.

Anyway. You made a good point about raw on/off not adjusting sufficiently for teammates. It does offer a first-blush, basic correction for teammates that purely raw +/- doesn't, but not enough. So, one quick hack would be to compare an individual's playoffs on/off +/- with the on/off +/- of his teammates. There's several ways to do this, but for the below example I compared the highest playoffs on/off with the second-best on the team (among big-minute rotation players), and note the difference.

So, for this exercise, I use a few general rules to increase the likelihood of useful signal in the midst of the small sample/potential noise.

Rules:
*Teams had to make at least the Conference Finals
*Noting only the highest on-off +/- scores on the team as the ones likely driving the positive margin
*Using difference b/w highest on/off score and second-highest as a means to separate strong individual impact from strong units
*I looked at every conference finalist from 1997 - 2020 for this data
*Setting thresholds (to sort the 96 represented seasons)
I took the average on/off +/- of the top player on each team (+16.4)
I then took the average difference b/w top and 2nd player on each team (+7.7)

Results summary of the "16/8" club
31 player playoffs where team leader on/off > 16, difference b/w P1 & P2 > 8
Only 4 players (LeBron, Shaq, KG, Duncan) made list more than once (accounted for 12 of the 31 16/8 seasons)
Notable 1-offs include 1997 Jordan & 2006 D. Wade

Typically, the non-megastars fit into one of several categories not well captured by boxscores. Few non-intuitive 1-offs.
Defense: '04 Ben Wallace, 02 Doug Christie, '15 Draymond Green, '11 Luol Deng, '13 Marc Gasol, '99 Robinson
Floor stretchers: '00 Reggie Miller, '01 Ray Allen, '10 Rashard Lewis
Floor generals: '02 Jason Kidd, '16 Kyle Lowry, '06 Steve Nash
Mega 6th men: '05 Manu GInobili, '12 James Harden
Past prime stars: '97 Clyde Drexler, '99 Arvydys Sabonis, (Jordan, Robinson & Reggie would fit here too)
('17 Kawhi Leonard makes list, w/ small asterisk since he only played 12 of the team's 16 games that postseason)

Anyway, just a simple exercise like this shows a lot of, IMO, useful info. For example, you mention that by raw playoffs on/off +/- David Robinson looks better than Shaq and LeBron. Well, here we see that '99 Robinson was the only version of late-prime DRob to make this list (by 2001, even though he had a great on/off, he had been surpassed by Duncan), while both Shaq and LeBron were on here multiple times (7 total, between them).

I should note, NOT making this list doesn't necessarily mean a player didn't have a great postseason impact in a given season. They could've had a great on/off in a great unit (e.g. 2011 Dirk met the on/off threshold, but not gap over 2nd because unit was strong). But, on that note...

Focus: you were talking about Steph, in a post that compares him with KD and also mentions Magic, MJ and LeBron's postseason peaks. While it may not be rigorous, it's at least worth noting, to me, that out of all the many times that Steph and KD have played on contending teams (and their entire careers were in the databall era), neither of them had even one postseason run make the above list. Of the others listed, LeBron has the most such seasons on record (4) and even though we only have the last 2 seasons of MJ's Bulls career, he reached the thresholds in '97 and really was barely short of the thresholds to have a second such postseason run in '98 (led Bulls w/ +14.6 postseason on/off, +6.3 higher than 2nd on team). When you factor in that most of the other players in Ben's peak project have seasons like that as well...like I said, I at least take note.

Following up, of the 2, Curry seems to have the more impressive postseason +/- record. As I noted in a previous post, Curry's raw on/off +/- in his prime-ish seasons was +12.6, while Durant's was +5.1. Curry's record looks more like what I described for 2011 Dirk, where he had high numbers but may have been parts of units such that we couldn't ascribe the impact just to him (e.g. the Draymond effect), whereas Durant really doesn't have a postseason record where he stands out even to that extent in the +/- data.

Anyway. All of this is obviously food for thought. But to me it is useful information that adds depth to postseason analysis that you just can't find if you only rely on team results and/or boxscore data.

I see why ESPN hired you. Awesome work! This is a really creative, and simple solution that I haven't really seen put into action before.

I also feel the need to make clear, that I don't want people to think I'm a Curry hater. I think he is deserving of a spot in this series, I just suppose that I still can't get behind the idea that he is contending for offensive GOAT. Maybe I am letting total and overall imapact influence but I prefer MJ and Lebron, and even at the PG spot, Magic and perhaps Nash for peak offensive play.

I suppose when I hear the word GOAT, I think of a sacred group, and Curry just doesn't give me that vibe at the moment.