Sansterre's Top Playoff 3-Year Stretches

sansterre · Post #1 » by **sansterre** » Thu Dec 24, 2020 8:23 pm

HCL asked me about this and I thought it would be a fun project for a day off, so here it is!

1) I'm taking *distinct* three year stretches. The same year can't be used twice.
2) I'm using my PSRS calculations that update through the playoffs (with all the ups and downs of that formula)
3) I'm using the PSRS + STD formula that I use for my individual team/season rankings
4) I required a minimum +5 PSRS year to be included, and the team must have been in at least two series. If you're +10, +2 and +10 I'm not accepting it. I require a little consistency for this list. (in retrospect I should have made it a single standard deviation, but I only thought of that 2/3 of the way through)

Only thirty optimal and distinct three-year runs had +5 PSRS.

With that in mind, here we go!

#30. 1968-70 Los Angeles Lakers
#29. 2010-12 Boston Celtics
#28. 1998-00 Indiana Pacers
#27. 1994-96 New York Knicks
#26. 2017-19 Houston Rockets
#25. 1993-95 Houston Rockets
#24. 1978-80 Philadelphia 76ers
#23. 2007-09 Cleveland Cavaliers
#22. 2008-10 Orlando Magic
#21. 2004-06 Detroit Pistons
#20. 2002-04 San Antonio Spurs
#19. 2005-07 Phoenix Suns
#18. 2015-17 San Antonio Spurs
#17. 2005-07 San Antonio Spurs
#16. 1964-66 Boston Celtics
#15. 1987-89 Los Angeles Lakers
#14. 1996-98 Utah Jazz
#13. 1987-89 Detroit Pistons
#12. 2011-13 Miami Heat
#11. 2008-10 Los Angeles Lakers
#10. 1984-86 Boston Celtics
#9. 1980-82 Boston Celtics
#8. 1959-61 Boston Celtics
#7. 1984-86 Los Angeles Lakers
#6. 2000-02 Los Angeles Lakers
#5. 1991-93 Chicago Bulls
#4. 2015-17 Cleveland Cavaliers
#3. 2012-14 San Antonio Spurs
#2. 1996-98 Chicago Bulls
#1. 2016-18 Golden State Warriors

Merry Christmas!

Cavsfansince84 · Post #2 » by **Cavsfansince84** » Thu Dec 24, 2020 8:27 pm

Somewhat surprised to see the 08-10 Magic on there and no early 70's Knicks teams. I imagine if it were changed to 4 year samples that would change.

homecourtloss · Post #3 » by **homecourtloss** » Thu Dec 24, 2020 8:41 pm

sansterre wrote:HCL asked me about this and I thought it would be a fun project for a day off, so here it is!

1) I'm taking *distinct* three year stretches. The same year can't be used twice.
2) I'm using my PSRS calculations that update through the playoffs (with all the ups and downs of that formula)
3) I'm using the PSRS + STD formula that I use for my individual team/season rankings
4) I required a minimum +5 PSRS year to be included. If you're +10, +2 and +10 I'm not accepting it. I require a little consistency for this list. (in retrospect I should have made it a single standard deviation, but I only thought of that 2/3 of the way through)

Only thirty optimal and distinct three-year runs had +5 PSRS.

With that in mind, here we go!

#30. 1968-70 Los Angeles Lakers
#29. 2010-12 Boston Celtics
#28. 1998-00 Indiana Pacers
#27. 1994-96 New York Knicks
#26. 2017-19 Houston Rockets
#25. 1993-95 Houston Rockets
#24. 1978-80 Philadelphia 76ers
#23. 2007-09 Cleveland Cavaliers
#22. 2008-10 Orlando Magic
#21. 2004-06 Detroit Pistons
#20. 2002-04 San Antonio Spurs
#19. 2005-07 Phoenix Suns
#18. 2015-17 San Antonio Spurs
#17. 2005-07 San Antonio Spurs
#16. 1964-66 Boston Celtics
#15. 1987-89 Los Angeles Lakers
#14. 1996-98 Utah Jazz
#13. 1987-89 Detroit Pistons
#12. 2011-13 Miami Heat
#11. 2008-10 Los Angeles Lakers
#10. 1984-86 Boston Celtics
#9. 1980-82 Boston Celtics
#8. 1959-61 Boston Celtics
#7. 1984-86 Los Angeles Lakers
#6. 2000-02 Los Angeles Lakers
#5. 1991-93 Chicago Bulls
#4. 2015-17 Cleveland Cavaliers
#3. 2012-14 San Antonio Spurs
#2. 1996-98 Chicago Bulls
#1. 2016-18 Golden State Warriors

Merry Christmas!

Thank you, Sansterre. As I thought in your Top 100 teams of all time, that Cavs stretch was right near the top.

Lebron played 11 series against teams in this list, including 5 series against at the #1 and #3 stretches.

Do you have the raw numbers?

carlquincy · Post #4 » by **carlquincy** » Fri Dec 25, 2020 1:56 am

#29??

I dont recall Celtics being that during that period, other than '10.

euroleague · Post #5 » by **euroleague** » Fri Dec 25, 2020 2:14 am

That is incredibly high rating for the Cavs/Spurs. How do you justify 15-17 cavs over 91-93 bulls?

or the 00-02 Lakers?

why 84-86 Lakers over 84-86 Celtics? Celtics had the best record every year, and won 2 championships?

homecourtloss · Post #6 » by **homecourtloss** » Fri Dec 25, 2020 4:00 am

euroleague wrote:That is incredibly high rating for the Cavs/Spurs. How do you justify 15-17 cavs over 91-93 bulls?

or the 00-02 Lakers?

why 84-86 Lakers over 84-86 Celtics? Celtics had the best record every year, and won 2 championships?

Did you even read the OP? It’s a pure formula driven “ranking.”

DQuinn1575 · Post #7 » by **DQuinn1575** » Fri Dec 25, 2020 4:06 am

homecourtloss wrote:
euroleague wrote:That is incredibly high rating for the Cavs/Spurs. How do you justify 15-17 cavs over 91-93 bulls?

or the 00-02 Lakers?

why 84-86 Lakers over 84-86 Celtics? Celtics had the best record every year, and won 2 championships?

Did you even read the OP? It’s a pure formula driven “ranking.”

I think some are questioning why a team with all championships get ranked behind teams with only 1.

Jordan Syndrome · Post #8 » by **Jordan Syndrome** » Fri Dec 25, 2020 4:33 am

homecourtloss wrote:
sansterre wrote:HCL asked me about this and I thought it would be a fun project for a day off, so here it is!

1) I'm taking *distinct* three year stretches. The same year can't be used twice.
2) I'm using my PSRS calculations that update through the playoffs (with all the ups and downs of that formula)
3) I'm using the PSRS + STD formula that I use for my individual team/season rankings
4) I required a minimum +5 PSRS year to be included. If you're +10, +2 and +10 I'm not accepting it. I require a little consistency for this list. (in retrospect I should have made it a single standard deviation, but I only thought of that 2/3 of the way through)

Only thirty optimal and distinct three-year runs had +5 PSRS.

With that in mind, here we go!

#30. 1968-70 Los Angeles Lakers
#29. 2010-12 Boston Celtics
#28. 1998-00 Indiana Pacers
#27. 1994-96 New York Knicks
#26. 2017-19 Houston Rockets
#25. 1993-95 Houston Rockets
#24. 1978-80 Philadelphia 76ers
#23. 2007-09 Cleveland Cavaliers
#22. 2008-10 Orlando Magic
#21. 2004-06 Detroit Pistons
#20. 2002-04 San Antonio Spurs
#19. 2005-07 Phoenix Suns
#18. 2015-17 San Antonio Spurs
#17. 2005-07 San Antonio Spurs
#16. 1964-66 Boston Celtics
#15. 1987-89 Los Angeles Lakers
#14. 1996-98 Utah Jazz
#13. 1987-89 Detroit Pistons
#12. 2011-13 Miami Heat
#11. 2008-10 Los Angeles Lakers
#10. 1984-86 Boston Celtics
#9. 1980-82 Boston Celtics
#8. 1959-61 Boston Celtics
#7. 1984-86 Los Angeles Lakers
#6. 2000-02 Los Angeles Lakers
#5. 1991-93 Chicago Bulls
#4. 2015-17 Cleveland Cavaliers
#3. 2012-14 San Antonio Spurs
#2. 1996-98 Chicago Bulls
#1. 2016-18 Golden State Warriors

Merry Christmas!

Thank you, Sansterre. As I thought in your Top 100 teams of all time, that Cavs stretch was right near the top.

Lebron played 11 series against teams in this list, including 5 series against at the #1 and #3 stretches.

Do you have the raw numbers?

The Cavs from 2015-2017 win a title in at least 80% of 3-year stretches in NBA history. They were extremely unlucky in 2015 losing two all-stars against a 10+ SRS team and then 2017 they ran into the GOAT team.

Jordan Syndrome · Post #9 » by **Jordan Syndrome** » Fri Dec 25, 2020 4:34 am

DQuinn1575 wrote:
homecourtloss wrote:
euroleague wrote:That is incredibly high rating for the Cavs/Spurs. How do you justify 15-17 cavs over 91-93 bulls?

or the 00-02 Lakers?

why 84-86 Lakers over 84-86 Celtics? Celtics had the best record every year, and won 2 championships?

Did you even read the OP? It’s a pure formula driven “ranking.”

I think some are questioning why a team with all championships get ranked behind teams with only 1.

Those people should think. Did Pippen and Grant get injured in 1991? Did they play a GOAT team in 1993?

euroleague · Post #10 » by **euroleague** » Fri Dec 25, 2020 5:37 am

Jordan Syndrome wrote:
DQuinn1575 wrote:
homecourtloss wrote:
Did you even read the OP? It’s a pure formula driven “ranking.”

I think some are questioning why a team with all championships get ranked behind teams with only 1.

Those people should think. Did Pippen and Grant get injured in 1991? Did they play a GOAT team in 1993?

They do think. Do injuries not affect a team's top playoff performances? What if Bird was never injured - should the Celtics be considered the greatest team in 86-88, because if McHale and Bird weren't both injured, they probably would've 3-peated?

That logic doesn't make sense...the Cavs played a historically weak eastern conference, and got absolutely destroyed in the playoffs twice. In 2016, Iggy/Bogut/Curry were all injured, and Draymond was suspended.

Furthermore, in1993 they DID PLAY A GOAT TEAM.... Barkley was a superstar MVP transferring into an elite playoff team in the Suns. That team was absolutely one of the greatest to never win, and far more talented than the Cavs were. KJ/Barkley/Majerle were an elite big 3, but more importantly, their supporting cast was excellent.

I didn't realize this list is just an average of playoff SRS, which is extremely flawed as both

a. sample size is too small
b. running up the score gets hugely rewarded

Jordan Syndrome · Post #11 » by **Jordan Syndrome** » Fri Dec 25, 2020 5:54 am

euroleague wrote:Furthermore, in1993 they DID PLAY A GOAT TEAM.... Barkley was a superstar MVP transferring into an elite playoff team in the Suns.

The 1993 Suns were not a GOAT tier team. They were very good. They were heralded by a top 20-25 peak in Charles Barkley while the Warriors had two in Curry and Durant who were arguably better than Barkley (certainly the same tier). Then the warriors had Green who was better than Majerle and Johnson and then a cast of great role players as well.

Let's not pretend like the 1993 Suns were close to the 2017 Warriors--they weren't. The 1993 Suns were a very good team, borderline great but certainly not an all-time great team like the 2017 Warriors.

The list has its flaws just as any list does--whether created on the back of statistical analysis or personal judgements.

euroleague · Post #12 » by **euroleague** » Fri Dec 25, 2020 6:24 am

Jordan Syndrome wrote:
euroleague wrote:Furthermore, in1993 they DID PLAY A GOAT TEAM.... Barkley was a superstar MVP transferring into an elite playoff team in the Suns.

The 1993 Suns were not a GOAT tier team. They were very good. They were heralded by a top 20-25 peak in Charles Barkley while the Warriors had two in Curry and Durant who were arguably better than Barkley (certainly the same tier). Then the warriors had Green who was better than Majerle and Johnson and then a cast of great role players as well.

Let's not pretend like the 1993 Suns were close to the 2017 Warriors--they weren't. The 1993 Suns were a very good team, borderline great but certainly not an all-time great team like the 2017 Warriors.

The list has its flaws just as any list does--whether created on the back of statistical analysis or personal judgements.

Let's not pretend the Cavs in any year were close to the 2017 Warriors... They won 1 game in 2 years.

Draymond was better than KJ - wtf? Draymond got all-nba once, KJ was all-NBA every year as the best player on his team.

I'm not arguing the Suns were better than the Warriors in 17 - however, the Suns in 2015 would've crushed the Cavs. And probably in 2016/2017 as well (Edit: that matchup is close, but I have KJ/Majerle as a far better matchup with Majerle on LBJ than Love on Barkley...)

In 1993, the Suns had 3 all-stars (KJ missed time in the reguluar season, so he didn't get his normal position), they had Tom Chambers, who was all-NBA as recently as 1990 and all-star 1991, coming off the bench... they had Ceballos and Danny Ainge... that team was definitely borderline GOAT level. Led by the experience and hunger of a peak Charles Barkley looking for a ring.

euroleague · Post #13 » by **euroleague** » Fri Dec 25, 2020 6:48 am

To be clear, I’m not criticizing OP - I am not familiar with his formula, so can’t comment on it. Very hard to judge using stats when the conferences are imbalanced, and his list seems informative even if it’s not matching my ranking.

for my top 10, I would have (only judging post 1980, and ignoring regular season)
1. Bulls 91-93
2. Bulls 96-98
3. Lakers 00-02
4. Celtics 84-86
5. Pistons 88-90
6. Lakers 85-87
7. Lakers 08-10
8. Warriors 16-18 (they were really weak in 16 while Curry was injured, and post-injury. 17-18 were dominant)
9. Heat 12-14
10. Rockets 93-95

I omit the Spurs 05-07, who also won 2 in 3 years, with no honorable mention because I think the Horry foul on Nash, and accompanying suspensions, weren’t honorable.

Jordan Syndrome · Post #14 » by **Jordan Syndrome** » Fri Dec 25, 2020 6:52 am

euroleague wrote:Let's not pretend the Cavs in any year were close to the 2017 Warriors... They won 1 game in 2 years.

They weren't but what is your point here? The argument isn't 2017 cavs vs 2017 Warriors...

Draymond was better than KJ - wtf? Draymond got all-nba once, KJ was all-NBA every year as the best player on his team.

Draymond in 2017 was better than Kevin Johnson in 1993--that is the comparison here, right? It seems like you are, based on the first and second segment of your post, attempting to argue something against me where I am not attempting to argue such a thing.

I have 2017 Draymond Green over 1993 Kevin Johnson. I have the 2017 Warriors ahead of the 1993 Suns. I hope this clears up your apparent confusion.

I'm not arguing the Suns were better than the Warriors in 17 - however, the Suns in 2015 would've crushed the Cavs. And probably in 2016/2017 as well (Edit: that matchup is close, but I have KJ/Majerle as a far better matchup with Majerle on LBJ than Love on Barkley...)

That isn't how matchups work. LeBron is so much better in 2016 and 2017 than anything the Suns as a team can accomplish it doesn't matter if Barkley is better than Love.

In 1993, the Suns had 3 all-stars (KJ missed time in the reguluar season, so he didn't get his normal position), they had Tom Chambers, who was all-NBA as recently as 1990 and all-star 1991, coming off the bench... they had Ceballos and Danny Ainge... that team was definitely borderline GOAT level. Led by the experience and hunger of a peak Charles Barkley looking for a ring.

They weren't GOAT level or close to the 2017 Warriors though and since the original comparison was somehow missed by you in this comparison then I am sorry for your misunderstanding.

Have a great night and a happy Holidays. Cheers mate!

euroleague · Post #15 » by **euroleague** » Fri Dec 25, 2020 7:02 am

Jordan Syndrome wrote:
euroleague wrote:Let's not pretend the Cavs in any year were close to the 2017 Warriors... They won 1 game in 2 years.

They weren't but what is your point here? The argument isn't 2017 cavs vs 2017 Warriors...

Draymond was better than KJ - wtf? Draymond got all-nba once, KJ was all-NBA every year as the best player on his team.

Draymond in 2017 was better than Kevin Johnson in 1993--that is the comparison here, right? It seems like you are, based on the first and second segment of your post, attempting to argue something against me where I am not attempting to argue such a thing.

I have 2017 Draymond Green over 1993 Kevin Johnson. I have the 2017 Warriors ahead of the 1993 Suns. I hope this clears up your apparent confusion.

I'm not arguing the Suns were better than the Warriors in 17 - however, the Suns in 2015 would've crushed the Cavs. And probably in 2016/2017 as well (Edit: that matchup is close, but I have KJ/Majerle as a far better matchup with Majerle on LBJ than Love on Barkley...)

That isn't how matchups work. LeBron is so much better in 2016 and 2017 than anything the Suns as a team can accomplish it doesn't matter if Barkley is better than Love.

In 1993, the Suns had 3 all-stars (KJ missed time in the reguluar season, so he didn't get his normal position), they had Tom Chambers, who was all-NBA as recently as 1990 and all-star 1991, coming off the bench... they had Ceballos and Danny Ainge... that team was definitely borderline GOAT level. Led by the experience and hunger of a peak Charles Barkley looking for a ring.

They weren't GOAT level or close to the 2017 Warriors though and since the original comparison was somehow missed by you in this comparison then I am sorry for your misunderstanding.

Have a great night and a happy Holidays. Cheers mate!

Let me clarify the points you are quoting me on, which you seemed to miss:

In 1993, KJ was definitely better when healthy than Draymond in 2017. Draymond in 2017 couldn't shoot, couldn't post-up, couldn't rebound...he was great in 2016, because of his shooting. Other years, not good or even all-star level.

LBJ isn't much better in 2016 or 2017 than Barkley was in 93. Barkley was an MVP, and LBJ wasn't. Barkley was a better scorer than LBJ, and not far off in passing. LBJ was a much better defender, but in 2017 he slacked off quite a lot. But, he was a far better rebounder.

LBJ was that far above any MVP in 2016 or 2017. He wasn't even an MVP. You discredit Barkley quite severely here.

The Suns were a borderline GOAT level team. The 2017 Warriors were the GOAT team. This means that they aren't far off, but aren't equal. You seem to miss the point of my comparison here.

Jordan Syndrome · Post #16 » by **Jordan Syndrome** » Fri Dec 25, 2020 7:09 am

euroleague wrote:Let me clarify the points you are quoting me on, which you seemed to miss:

In 1993, KJ was definitely better when healthy than Draymond in 2017. Draymond in 2017 couldn't shoot, couldn't post-up, couldn't rebound...he was great in 2016, because of his shooting. Other years, not good or even all-star level.

I disagree. I had 2017 Draymond as a clear top 20 player and likely borderline top 10 player. I tend to value playmaking and defense highly.

LBJ isn't much better in 2016 or 2017 than Barkley was in 93. Barkley was an MVP, and LBJ wasn't. Barkley was a better scorer than LBJ, and not far off in passing. LBJ was a much better defender, but in 2017 he slacked off quite a lot. But, he was a far better rebounder.

I have 2016 and 2017 James as inarguably the greatest player ever. Barkley is a fringe top-20 peak to me. I value LeBrons passing to routinely generate open corner 3's and lay-ups as a massive positive where Barkley simply couldn't. It appears where we differ is I view LeBron in this time frame as the GOAT-peak and you do not.

LBJ was that far above any MVP in 2016 or 2017. He wasn't even an MVP. You discredit Barkley quite severely here.

I assume you have 2011 Rose over Dirk and Wade in 2011 and Giannis in 2019 and 2020 over Kawhi and LeBron as well?

sansterre · Post #17 » by **sansterre** » Fri Dec 25, 2020 11:32 am

Cavsfansince84 wrote:Somewhat surprised to see the 08-10 Magic on there and no early 70's Knicks teams. I imagine if it were changed to 4 year samples that would change.

It's the minimum performance requirement that screws the Knicks, and a lot of teams.

The Knicks were really good in the '69 and '70 playoffs, but in '71 they were beaten fairly badly by the +2.5 SRS Bullets, which dropped their PSRS below the threshold for the three-year run. The same thing happens for the 70s Bucks; really good in '71 and '72 but whipped by the Knicks in '70 and knocked out in the first round of '73.

sansterre · Post #18 » by **sansterre** » Fri Dec 25, 2020 11:33 am

homecourtloss wrote:
sansterre wrote:HCL asked me about this and I thought it would be a fun project for a day off, so here it is!

1) I'm taking *distinct* three year stretches. The same year can't be used twice.
2) I'm using my PSRS calculations that update through the playoffs (with all the ups and downs of that formula)
3) I'm using the PSRS + STD formula that I use for my individual team/season rankings
4) I required a minimum +5 PSRS year to be included. If you're +10, +2 and +10 I'm not accepting it. I require a little consistency for this list. (in retrospect I should have made it a single standard deviation, but I only thought of that 2/3 of the way through)

Only thirty optimal and distinct three-year runs had +5 PSRS.

With that in mind, here we go!

#30. 1968-70 Los Angeles Lakers
#29. 2010-12 Boston Celtics
#28. 1998-00 Indiana Pacers
#27. 1994-96 New York Knicks
#26. 2017-19 Houston Rockets
#25. 1993-95 Houston Rockets
#24. 1978-80 Philadelphia 76ers
#23. 2007-09 Cleveland Cavaliers
#22. 2008-10 Orlando Magic
#21. 2004-06 Detroit Pistons
#20. 2002-04 San Antonio Spurs
#19. 2005-07 Phoenix Suns
#18. 2015-17 San Antonio Spurs
#17. 2005-07 San Antonio Spurs
#16. 1964-66 Boston Celtics
#15. 1987-89 Los Angeles Lakers
#14. 1996-98 Utah Jazz
#13. 1987-89 Detroit Pistons
#12. 2011-13 Miami Heat
#11. 2008-10 Los Angeles Lakers
#10. 1984-86 Boston Celtics
#9. 1980-82 Boston Celtics
#8. 1959-61 Boston Celtics
#7. 1984-86 Los Angeles Lakers
#6. 2000-02 Los Angeles Lakers
#5. 1991-93 Chicago Bulls
#4. 2015-17 Cleveland Cavaliers
#3. 2012-14 San Antonio Spurs
#2. 1996-98 Chicago Bulls
#1. 2016-18 Golden State Warriors

Merry Christmas!

Thank you, Sansterre. As I thought in your Top 100 teams of all time, that Cavs stretch was right near the top.

Lebron played 11 series against teams in this list, including 5 series against at the #1 and #3 stretches.

Do you have the raw numbers?

Of course! What specifically are you looking for?

sansterre · Post #19 » by **sansterre** » Fri Dec 25, 2020 11:34 am

carlquincy wrote:#29??

I dont recall Celtics being that during that period, other than '10.

They weren't particularly good, but they were consistently good enough to break the +5 PSRS requirement in those three years. Obviously '08-10 would have been better, except that '09 was a lost year for them.

sansterre · Post #20 » by **sansterre** » Fri Dec 25, 2020 11:40 am

Yeah, sorry about the lack of background. I thought it was interesting enough to make its own article, but I didn't want to do the legwork of posting all my intro content again. Which was lazy of me. As HCL points out, this is purely formula driven from my updating OSRS model. It totally suffers from small sample size (as does all playoff analysis, but it's capable of a lot of nuance, as compared to a "who won the championship" analysis, which is binary. It's what it is, ups and down, I just thought it would be of interest. I am definitely not arguing that this is the be-all and end-all or anything close.