Minnesota Versus Golden State Preview
Western Conference Second Round
#6 Minnesota (49-33) vs. #7 Golden State (48-34)
Season series: Golden State, 3-1
Golden State can win if…Stephen Curry does not wear down, if Jimmy Butler dominates by scoring in the paint/drawing fouls, and if the Warriors' defense contains Anthony Edwards without conceding too many open shots to other Minnesota players. Minnesota reached the Western Conference Finals last year and has homecourt advantage, so Golden State faces an uphill battle in this series.
The Warriors floundered to a 25-26 start this season before acquiring Jimmy Butler from the Miami Heat. After finishing the season with a 48-34 record, the Warriors defeated the Memphis Grizzlies 121-116 in the NBA Play-In Tournament to earn the seventh seed in the Western Conference. Golden State took a 3-1 lead in the first round versus the second seeded Houston Rockets before losing games five and six to set up a game seven showdown in Houston; it appeared as if the younger and more physical Rockets had worn down the Warriors, holding Curry to 13-35 (.371) field goal shooting in games five and six while forcing him to commit eight turnovers--but Golden State won game seven 103-89 as Buddy Hield scored a game-high 33 points while shooting 12-15 from the field, including 9-11 from three point range. Curry bounced back to produce 22 points on 8-16 field goal shooting while grabbing a team-high 10 rebounds (just his second double figure rebounding game in the playoffs since 2022) and passing for a game-high seven assists. Butler added 20 points, eight rebounds, and seven assists.
Curry is perhaps the most unusual of the players who can legitimately be ranked among the 15-20 greatest all-time; he and Jerry West are the shortest players in that group, but West was stronger, could jump higher, and played much better defense. During his playoff career, Curry has been targeted on defense, he has been worn down as series progress, and he has won the NBA Finals MVP just once during four championship campaigns--but he has also produced in the clutch on many occasions, including yesterday's game seven win in Houston and his 50 point masterpiece in a game seven win at Sacramento in 2023.
The addition of Butler has helped Golden State because Butler is not only an efficient scorer but he draws fouls, he is an excellent passer, he rebounds well, he is a very good defender, and he rises to the occasion in big moments. The Warriors would not even be in the playoffs this season without him. Against Houston, Butler ranked second on the team in scoring (18.3 ppg) and assists (4.8 apg), and first in rebounds (6.0 rpg). Butler played for Minnesota in the 2017-18 season before his contentious departure during the 2018-19 season, and this is the first time that he has faced the Timberwolves in the playoffs.
Draymond Green defends well, sets good screens, and is a deft passer. He is also an emotional time bomb that has exploded on many occasions, both to his detriment and to the detriment of his team. His media buddies will say that Golden State cannot win without him, but Green's typical triple single statistics versus Houston speak loudly: 8.0 ppg (sixth on the team), 5.6 rpg (third), and 3.3 apg (third) with .393/.258/.571 shooting splits. Even if one buys the premise that Green does a lot of good things that don't show up in the boxscore, what shows up in the boxscore for Green is often less than impressive, and we have already seen that Green has minimal impact on team success when he is not surrounded by multiple future Hall of Famers, with his triple singles adding up to a 15-50 record in the 2019-20 season.
Hield's game seven performance was an aberration not only in the series--he scored five points or less in four of the seven games--but those 33 points are more than a third of his career playoff scoring total (98 points). The likelihood that he will ever score 33 points again in a playoff game is exceedingly low.
Minnesota will win because...the Timberwolves will use their size to frustrate Golden State at both ends of the court. The Rockets did not consistently exploit their size advantage versus the Warriors and the Rockets lack a defined, elite number one scoring option; the Timberwolves bludgeoned the Lakers in the paint in the first round, and Anthony Edwards is an elite scorer whose young career already includes playoff series wins versus Luka Doncic, Kevin Durant, LeBron James, and Nikola Jokic. Edwards averaged 26.8 ppg, 8.4 rpg, and 6.2 apg versus the Lakers.
Former Laker Julius Randle, who the Timberwolves acquired from the New York Knicks in exchange for Karl-Anthony Towns prior to this series, averaged 22.6 ppg, 5.2 rpg, and 4.4 apg versus the Lakers. Randle can drive to the hoop with power, and he can nail three pointers as well, shooting 11-28 (.393) from long distance versus the Lakers to rank second on the team in three pointers made and second in three point field goal percentage.
Jaden McDaniels played excellent defense versus Doncic while also scoring 17.4 ppg (third on the team behind Edwards and Randle).
Draymond Green loves to mock and belittle Rudy Gobert, but now Green will have to deal with Gobert on the court. Gobert's first playoff matchup versus the Warriors ended in a 4-0 Golden State sweep of the Utah Jazz in 2017, but those Warriors featured prime Kevin Durant, prime Stephen Curry, and prime Klay Thompson, plus three big men (David West, Zaza Pachulia, JaVale McGhee) who each averaged between 10-15 mpg during that series. Gobert averaged 15.5 ppg on .658 field goal shooting in that series, and he led both teams with 13.0 rpg, so the notion that he does not match up well with the Warriors is demonstrably false--and if Gobert can be effective versus the Lakers' trio of Luka Doncic, LeBron James, and Austin Reaves, then he can be effective versus Stephen Curry, Jimmy Butler, and the undersized Warriors. Gobert did not score much during the first four games versus the Lakers, but in game five he had 27 points on 12-15 field goal shooting while grabbing 24 rebounds; the Lakers defiantly stuck with a small lineup, and Gobert destroyed the Lakers in the paint.
Other things to consider: I did not pick either of these teams to reach the second round, so I have to candidly assess what specifically I either underrated about these teams or that I overrated about their first round opponents.
I expected the L.A. Lakers to exploit the scoring/playmaking of Luka Doncic, LeBron James, and Austin Reaves to overcome Minnesota's size advantage--but the Timberwolves pounded the Lakers into submission at both ends of the court while sufficiently containing Doncic, James, and Reaves. The Timberwolves did not fall into the trap of going small to match up with the Lakers but instead punished the Lakers for being too small.
I expected the Houston Rockets to use their size advantage and physicality to wear down the smaller, older Golden State Warriors; the Rockets followed that blueprint successfully for substantial portions of this series--most notably during games five and six--but they could not overcome their anemic offense, scoring 93 points or less in three of the seven games.
The Timberwolves are a bigger, better version of the Rockets team that wore down the Warriors but could not knock out the Warriors; the Timberwolves will finish what the Rockets began.
Minnesota will defeat Golden State in six games.
Labels: Anthony Edwards, Draymond Green, Golden State Warriors, Jimmy Butler, Julius Randle, Minnesota Timberwolves, Rudy Gobert, Stephen Curry
posted by David Friedman @ 4:14 PM
50 Comments:
if the lakers not named luka or lebron played as they had during the regular season, that series would have been much closer. The warriors won't be as rested as the wolves and that will have an impact on game 1 if the refs will swallow their whistles as they did in the first series and allow a higher level of physical play. Unless Ant saves the wolves in crunch time I wouldn't rely on Naz, McDaniels (certainly not Goooobert) to have ice in their veins when it counts.
I looked into the idea that Curry wears down over the course of playoff series, which I've heard often not just from reading you, but surprisingly the statistics don't really back it up. I think perhaps because 2016 is such an enduring memory and stain on his legacy it makes us "feel" like that always happens to him, but in general he's been better in Games 6 and especially 7 than the rest of a series. It's more typically Game 5 where he flounders, oddly:
G1: 27 ppg 5.9 rpg 5.7 apg on 49/45/88 shooting (20-7 record, +7.5 per game)
G2: 25.7ppg 5.1 rpg 6.4 apg on 46/38/91 shooting (19-8 record, +10.2 per game)
G3: 27.3 ppg 4.9 rpg 5.7 apg on 44/40/90 shooting (16-11 record, +4.9 per game)
G4: 28.4 ppg 5.6 rpg 6.4 apg on 46/40/89 shooting (20-9 record, +7.3 per game)
G5: 23.5 ppg 4.9 rpg 6.6 apg on 43/34/87 shooting (14-8 record, +1 per game)
G6: 27.9 ppg 5.4 rpg 6.1 apg on 43/39/86 shooting (10-6 record. +2.9 per game)
G7: 30.8 ppg 7 rpg 7 agp on 47/42/93 shooting (4-2 record, +11.8 per game)
Overall Game 7 generates his highest points, rebounds, assists, and FT%, and his second best FG% and 3pt% only behind G1. Game 6 is his third best points, fourth best rebounds, fifth best assists, but is his second weakest efficiency game behind Game 5.
I considered if the Sacramento game might be inflating the numbers as an outlier, but even with that removed he still averages 27 ppg, 6.8 rpg, 7.2 apg, and shoots 45/43/100, which would drop the scoring rank to 5th, but the rebounds and assists would still be top ranked, and it would still be his second-most efficient game after Game 1. His +- would still be tops at +9.2, as well.
I actually imagine those statistics probably still understate his late series rising a little, given that by definition any Game 6 or 7 is against a team that's good enough to beat you at least twice, and has had by then over a week to "learn" you and adjust, while Games 1-4 are often not, so the degree of difficulty is logically higher.
The Game 5 cratering is interesting, and I theorized it could be because of those Game 5s in which the Warriors had a 3-1 lead which they seem to have a bad habit of punting on, but the numbers don't change much even through that filtering. It seems like contrary to my assumption his Game 5 struggle is not much tied to whether or not the Warriors lead the series.
Lastly I was curious if his minutes changes over the course of a series and if that might explain anything:
G1: 37.5 mpg G2: 35 mpg G3: 36.8 mpg G4: 37.3 mpg G5: 37.8 G6: 39.8 mpg G7: 42.2 mpg
I don't think there's much to learn there. His raw totals and win rate understandably go up as his minutes do, with again an exception in G5, but there does not seem to be a meaningful correlation between playing time and efficiency nor any clue that might explain his comparatively poor Game 5s.
Stats aside I agree the Timberwolves have a significant rest, size, and age advantage. If Gobert can have a similar impact to what Steven Adams did for Houston it will be tough sledding for Golden State, though on the topic of rest it is probably best for the Wolves to try and close things out quickly if at all possible. The series plays every other night until Game 5, when there is a four day break before Game 6 which I'm sure the older Warriors would welcome.
Anonymous:
Luka and LeBron did not play to the level that they needed to play, and that sets the tone for the whole team, but it must also be emphasized that Redick's clueless coaching was a major problem.
Gobert's job is not to be a primary scoring option in clutch situations; his job is to defend, rebound, and convert a high percentage of his scoring opportunities in the paint.
StatDork:
Those numbers are interesting.
It is worth noting that two of the three best point differentials happened in games one and two and the two worst point differentials happened in games five and six. I would also suggest that there are a number of contextual factors to consider; I would not expect Curry to wear down in series versus vastly inferior teams, nor would I expect Curry to wear down in series versus teams that are not physical.
My contention is that Curry is more prone to be worn down by physical play than players who are taller, bigger, and stronger than he is. I acknowledge that when series reach a seventh game Curry often finds a second wind and delivers clutch performances.
In the first few minutes of game one the Rockets suffocated the Warriors with textbook great defense. The Warriors could barely get a shot off and they missed nine of their first eleven shots. Most of those shots were desperation heaves from several feet beyond the arc and I thought that the Rockets had set the tone for the series. They say that defense wins championships, which it does, but so does a reliable offense. A team can have the greatest defense ever but if they have numerous prolonged scoring droughts, and if their greatest defense ever starts to slip, they will not win a championship.
On paper, Minnesota should win this series but the Rockets were also the “on paper” favorites to beat the Warriors and rightfully so based on several matchup advantages. Hopefully, the Timberwolves won’t let the Warriors off the hook by not exploiting their matchup advantages.
To try and verify this theory, I compared fourteen other players. First I checked members of your Pantheon who have +- data available (MJ, Shaq, Duncan, Kobe, Lebron), then I added most of the recent MVPs and a few other Finals level stars who felt like relevant comparisons, all of whom are also larger than Curry (Harden, Westbrook, Durant, Kawhi, Giannis, Jokic, Luka, Wade, Tatum). Despite winning an MVP, I left out Joel Embiid because his playoff record is so injury-marred I was concerned it might mislead the data in Curry's favor. I also left out presumptive MVP SGA because his playoff sample size is still quite small.
So far I have only been able to look at +-, which is a very contextual and noisy stat, but for that stat my basic findings are:
- Curry's +- drop from Game 4 to Game 5 (-86%) is not proportionally larger than most of the other players' largest drops. Only Shaq, Wade, Lebron, and KD do not have a game-to-game drop of at least that percentage.
- However, the worst drop coming in Game 5 is fairly uncommon. For most players sampled the steepest drops are usually in Game 4 or Game 6, and in several cases both (Jordan, Kobe, Kawhi, Jokic). Only Westbrook had a steeper Game 5 drop than Curry (-165%), though Lebron (-55%) and KD (-63%) also show notable Game 5 drops that are smaller than Curry's.
- Curry never drops in consecutive games, but a little over half the players sampled do:
Shaq has a small 26% drop G1->G3, and a larger 76% drop G4->G6
Duncan has a gradual but large decline of 110% G1->G4
Lebron likewise has a gradual 70% decline G2->G5
Durant drops 71% G3->G5
Kawhi has a catastrophic 177% drop in longer series, G5->G7
Harden is even worse over the same span at -226%
Westbrook has a mid-series slump of -144% from G3->G5
Wade has a mild G2->G4 drop of 10%
- If we instead average Games 1-4 vs. Games 5-6, it supports the theory quite a bit better. Curry has the sharpest percentage drop in +- under that sorting, though if we change the rubric to Games 1-3 vs. Games 4-6, he fairs a little better, leapfrogging Harden and Jordan. He, Kobe, Jordan, Giannis, and Duncan are all within 10% of each other on that list, so while it looks like guards generally (Wade and especially Westbrook being exceptions) see a 40%-50% reduction in the second three games vs. the first, Curry's does not seem an outlier.
-Now is a good time to clarify that Jordan's PBP dataset is incomplete, and these numbers only include the last two title runs. It is possible that the younger Jordan had different results.
- Plus minus is very noisy used in this context and I am not sure how much it is actually telling us. For example, Jokic has a +6000% improvement in Games 4-6 vs. 1-3, which is obviously not an accurate reflection of whatever difference may exist in his play. Curry and Jordan have by far the highest +- in the first three games, and also then have the sharpest non-Harden drops in Games 4-6, so it also does seem like this way of looking at things punishes loaded teams that play a lot of series with home court advantage against lower seeds which allows them to run up the early game margins. While much lower than Curry's or Jordan's, Duncan and Giannis have the next best Games 1-3 margins and also the next worst drops vs. Games 4-6, so that appears to be at least somewhat consistent with the obvious caveat of sample size being rather small.
- In a perfect world, I would like to do some secondary research and see how the fluctuations in scoring, rebounding, assists, and shooting percentage for those other 14 players compare to Curry's, which I think would probably be a more instructive comparison. But that is a much larger undertaking and I'm not sure if I will have the time. I'll be back if I do, though!
Reaves, Hachimura and Vincent have been in enough playoff series to know the importance of bringing it every playoff game. Hachimura had one great game (for his standards) but his reaction times are generally slower than what the playoffs require. That's not going to change. Reaves was gassed from playing hard on both ends (although his defense is hardly world caliber) and did not figure out ways to adapt. Luka valiantly played one game while barfing up a storm. Vincent has been inconsistent all year. DFS did what he could but his shooting was off too. Jackson Hayes is too busy complaining (and fouling) to be useful on the court. Reddick is an easy target for his game 5 decision but this is the same guy who coached this team to 3rd place on a flimsy point differential that nearly every pundit used as evidence that the Lakers would fall to a lower seed. Wolves deserve credit but they haven't been tested yet - although the game 6 win on a abysmal 3 point shooting is worth noting (and that's where Gooobert finally had an impact - a guy who was so circumspect his own Olympic coach benched him).
Another point re Curry being unusual: in Jerry West's day the average player's height was much less. So Curry's size disadvantage, relative to West, is really even more
Curry did wear down in game 1 and Butler had a slow start before putting up a nice stat line and GS still dominated. A bad sign for Minnesota.
I'm not sure Curry is unusual for being considered a top 25 player all-time other than he's short for NBA standards at 6-2. He's already played 96 more regular season games than West. The playoffs are longer now than West's era, so it's fair to say Curry has been more durable in his career than West was.
Even with Gobert's career game in game 5 vs the Lakers, he only averaged 8 and 10 in that series. He was overall a non factor in that series. Minnesota still wins the series without Gobert's career game. And in game 1, he goes for 9 and 11 vs an even smaller GS team than the Lakers. He'll likely need to greatly improve from that for Minnesota to win this series.
Green has never been a top 3 option offensively on his teams except for injuries, but overall contributes in several ways offensively. He is, however, a #1 option defensively. He's finished in the top 10 for DPOY 9x out of 13 seasons, including winning 1x. GS is much better with Green than without him. He's clearly a future HOFer. GS was a complete mess in 2020 and it was the Covid season where many strange things happened that wouldn't have. Green only played 43 games that season, too. Nobody would've helped GS much more replacing him for those 43 games. This isn't a good argument to prove Green isn't a #1 or #2 offensive option. We already know that. He's been a very important glue guy similar to how Rodman was, for example.
If Curry is out long-term, Minnesota should win, but I wouldn't be surprised if GS wins especially after they stole game 1. The Rockets were better than Minnesota, so this should be an easier matchup for GS. I don't think GS is easier than the Lakers for Minnesota, but the Lakers didn't play very well, so GS will probably be a tougher out for Minnesota. GS is overall a small team, but they can play 13 guys, so maybe that benefit will help them enough to take this series.
Shams says Grade 1 strain for Curry.
Best the Warriors could hope for, really. Likely out at least the next three games and if they're smart they'll wait until Game 6 since there's a bunch of extra rest after Game 5.
Contingent on them finding a way to win a second game to even get to a Game 6, but in theory that's why you trade for Jimmy Butler, right?
Could even be a blessing in disguise if Butler is somehow able to conjure Playoff Jimmy devil magic to win the series without him and the time off recovering from the hamstring also fixes Curry's thumb and resets his fatigue.
Anonymous:
I am not sure how you would define "much less," but the average height of an NBA player in 1970 was just shy of 6-6, and the average height of an NBA player in 2022 was just a tick above 6-6. The average weight increased from around 205 to around 215 in that same time period.
Anonymous:
Yes, I would say that this season for sure Curry seems to be wearing down.
In general, NBA careers last longer now than in previous eras due to higher salaries and better conditioning. If one controls for those factors, I am not sure that Curry is inherently more durable than West.
Anonymous:
Gobert was a huge factor in game five versus the Lakers, and I disagree that he was a non-factor in the other games. His presence in the paint discourages players from driving and minimizes offensive rebounding opportunities.
I do not dispute Green's value as a role player on a team stacked with Hall of Famers; his skill set meshes well with such players, particularly because he does not demand to shoot the ball a lot. However, my contention is that he would not be great on any team or in any situation, and that distinguishes him from truly great players who would elevate any team that they join. He is a very good defensive player, but not quite as good as some suggest. Steve Kerr recently ranked him above Jordan, Pippen and Rodman as defensive players, which I would say is absurd.
The lethargic and unintelligent way that Minnesota played in game one is why I did not pick Minnesota to beat the Lakers. The Timberwolves are puzzling, and it is difficult to know what to expect from them next. They certainly have enough to beat the Warriors even if Curry is healthy, but it will be interesting to see how Minnesota plays in game two and then on the road in games three and four.
Anonymous:
I think that we disagree about how much credit a coach should receive for finishing third with LeBron and Luka followed by losing 4-1 in the first round.
I am not sure why you keep mocking Gobert's name and his game, nor am I sure what the playing rotations of France in the Olympics have to do with the NBA playoffs. Should we rank Jayson Tatum as an NBA player based on Steve Kerr's Team USA rotations?
StatDork:
I am not yet sure what that data proves or disproves, but it is interesting and I appreciate your effort to pull all of it together.
Just a note on Jerry West and his height. Even though West was listed as 6'2, he always stated he was 6'4.5 feet tall:
https://youtu.be/Oj9jbyUlsPU?si=0jJY5Pz9WYeIUdQm
Whether this was a measurement in shoes or not, I am not sure. Players in the 60s and for most of the 70s were measured without shoes.
Quite frankly I don't really think it proves much of anything, which is why I'd like to dig deeper if time allows. Plus/minus, to me, is a noisy enough stat that it's much more useful as a starting point than a destination. The trends presented suggest possibilities that are worth investigating, but they certainly don't confirm them.
Though at a very broad level, there is probably something to most drops consistently over 100%.
Thought not quite over a100% drop, Curry being demonstrably worse in Game 5 is seemingly backed up by both the raw box score data as well as the eye test. Game 6 is murkier both statistically and anecdotally, and 7 is obviously his high point, but 5 does seem provably weak by his otherwise lofty standards.
Likewise, I trust the late-series Kawhi and Harden trends. Harden is famous for his playoff choking. Kawhi's reputation is fairly saved by his magical 2019 run, but he has otherwise generally been a poor late-series player, with costly mistakes or underperformances leading to Game 6 and/or 7 eliminations in 2013, 2015, 2016, and 2020. He was solid but not exemplary in his Games 6 and 7 this year.
I technically trust the gradual Duncan decline G1->G4, but more in the sense that the data feels easily explainable than as a fundamental truth about Duncan himself. His Spurs were generally a higher seed and as such usually had G1 & G2 at home, and they were good enough that G4 was frequently must-win for a desperate opponent, either to stave off elimination or to avoid a 3-1 hole that a team as potent as San Antonio was not likely to let them out of. Popovich was also a cautious coach with regards to Duncan's minutes, particularly later in his career, and would not ride him as hard in games where the Spurs were already in command of the series.
Kobe has an over 100% drop from G3->G4 as well but I think the Duncan observations above mostly absolves him too. Phil was less likely to cut his minutes than Pop was Duncan's, but I don't think that minor distinction persuades me away from the larger conclusion.
I do not trust any of Jokic's numbers. They are almost all dramatic outliers, and in both directions. To me that speaks to a relatively slim sample size and the erraticness of his two best teammates' availability and performances more than anything else.
The Westbrook mid-series slump feels plausible, but could also be noise. That is something I will keep an eye on if I do dig deeper. It is somewhat plausible that his stylistic predictability makes him easier to counter-scheme as a series progresses, and at least equally plausible that by games 6 & 7 his superior athleticism likely outweighs that disadvantage against tired opponents, but I'd want more corroborating data to confirm both hunches. His '17-'19 run also feels like the sort of extreme usage and team-dependency environment that could poison a larger data sample, so I'd also want to look at how much if at all the trends change with that peak heliocentrism era removed.
The second half of that hypothesis could also explain why Curry rallies so well in Games 6 and especially 7, if his constantly in motion style and best-in-class cardio is simply exhausting defenders into failure by that point in a series, rather than him having any specific clutch gene or late series sorcery. Certainly Dellavedova's hospital trip and Amen Thompson's Game 7 cramping would seem anecdotal supporting evidence for that theory, but as with most of the above I'd want more corroborating data before stating it with any confidence.
Too many anons here and yet me still too lazy to make up a name.
I just wanted to chime in about the height changes. While the average has not changed a lot that is partly because the tallest players have gotten shorter and the shortest players have gotten taller. The average guard now is 6'4 and a half. That is only a half inch shorter than Oscar Robertson, who was considered a very large guard in West's era.
Meanwhile, some of the "centers" in the league this year include Wendell Carter (6'10), Vuc (6'10), Duren (6'10), Sabonis (6'10), Horford (6'9), Okongwu (6'8), Yabusele (6'8), and Draymond (6'6), among others, most of whom would have been power forwards in older eras.
There are still a few skyscrapers like Wemby and Gobert but they have less of a pull on the average in a 30 team league than old timey giants like Wilt and Kareem did when there were only twelve.
Whether you want to blame Lebron, The Warriors, or secret third thing for the move towards "positionless basketball" the increased emphasis on switchability and muti-position nonsense has made guards bigger and bigs smaller.
So West did probably have a more pronounced height advantage in his era against his positional foes, but he also probably played against bigger 4s and 5s patrolling the paint, which was probably a lot more of a pain in the ass to deal with in a world without a three point line.
That makes me more impressed by West's scoring, and a little less impressed
by the gaudy rebounding totals sometimes put up by Westbrook, Harden, and to a lesser extent, Steph.
If Jayson Tatum had been an Olympian as long as Gooobert, then yes. And I stand corrected - he played 12 minutes in the gold medal game. 4 & 5 minutes in the semi and quarterfinals matches.
I will give Draymond his flowers though, I don't think it's nuts to argue he has a case as greatest defender of the modern era, and isn't Kerr kinda uniquely the most qualified dude to judge it? He played with Rodman, Jordan, Pippen, Duncan, Robinson, and Bowen, and he coached Iguodala. He's shared a locker room with most of the serious contenders, give or take Olajuwon and Mutombo I guess.
You talk a lot about how Curry's not as great as X dude because he gets hunted on defense, which is legit, but the Warriors are usually a pretty killer defense anyway and Draymond is the main reason why. Did any of those other guys have to cover for their teammates that much? I guess maybe Duncan had to pick up after Parker some but IDK, feels like pretty much everyone else on those Spurs teams were usually great defenders too while the Warriors tend to have another cone somewhere and have to figure out how to hide a Jordan Poole or a Quinton Post or a Boogie Cousins or a James Wiseman or a post-Achilles Klay or whatever the hell a "Podziemski" is.
He and Rodman are the only two dudes I've ever seen truly "guard one through five" as much as that term gets tossed around, and obvi Rodman's a way better rebounder, but I don't remember him being nearly as active as a helper or off-ball or rotational dude. In fact, I remember Spurs fans getting kinda pissed at him for not helping out while Olajuwon was killing their boy in that infamous 95 series and just camping out for rebounds.
Keith:
Thank you for sharing that clip. It is interesting that even something as simple as a player's height can be in dispute, and even listed incorrectly. I never met West, but I met enough players and retired players to know that listed heights are often wrong by up to two inches in either direction, although typically the taller players preferred to "shrink" to avoid the perceived stigma of dominating just because they were over 7 feet tall while the shorter players preferred to add an inch or two. So, West being 6-4 and change but listed at 6-3 is unusual.
Anonymous:
Even if Curry can come back by game five or game six, his conditioning will be off, and he will be vulnerable to reinjury. Minnesota has a great opportunity to win this series; on the other hand, if Golden State beats Minnesota in this series despite only having Curry for a few minutes then what does that say about Curry's vaunted "gravity"?
Less than what it would say about the Minnesota Timberwolves! I don't think we're gonna need to answer that question, though. I'm pretty sure either Minny will win outright or GSW will drag things out just long enough for Curry to come back and either gut out a win or immediately reinjure himself and kill the vibe.
Warriors winning without Curry would make the Lakers look even worse, though, so maybe we root for it just for that?
But if it did happen, wouldn't the answer depend? Like it looks one way if it's a Miami-style Playoff Jimmy hard carry where they're gutting out 86 to 82 nailbiters and nobody but Butler shoots over 43%, it looks another way if they just slot Buddy Hield into all Curry's sets and get the exact same results they usually get with Curry, right?
StatDork:
I appreciate that you recognize that data has limited value without context and without a proper sample size.
DiffAnon:
You are correct that there has been a size range compression in the NBA: the "tall" players are not quite so tall as they used to be, and the "short" players are a little taller. Your interpretations of what that could mean regarding the relative ease of scoring and rebounding in the respective eras are interesting, but I am not sure if they are correct or provable.
Anonymous:
I still do not understand--and you have not explained--why you find it necessary or relevant to mock Gobert's name and his game, nor have you explained why minutes played in the Olympics is relevant to minutes played in the NBA playoffs.
DiffAnon:
Kerr is qualified to rank defensive players who he has played with/against and who he has coached but are his rankings unbiased? Green is a hotheaded player. Even if Kerr thinks that Green is not as good of a defender as Jordan, Pippen, or Rodman why would he say that in the middle of a playoff run? Kerr is trying to motivate his players, not provide an objective ranking of the best defenders of all-time.
David:
No problem. I tend to think older player's heights are underestimated by at least an inch or so since they were measured without shoes until the late 70s or early 80s, and were wearing low top Chuck Taylors even when they were suited up. I know you've talked about how Bill Walton was a legit 7 footer but preferred to be listed as 6'11. I also suspect that Bill Russell was actually taller than his commonly listed 6'9 height, as he stands pretty even with 6'10 Dwight Howard and only slightly shorter than supposedly 6'11 Kevin Garnett in photos.
I think, amusingly enough, Kevin Durant has finally been listed as 6'11 since he joined the Phoenix Suns in 2022, as opposed to the 6'9 he was listed as for nearly fourteen years that we all knew he wasn't. Late career growth spurt.
"Even if Kerr thinks that Green is not as good of a defender as Jordan, Pippen, or Rodman why would he say that in the middle of a playoff run?"
It's not like he was asked directly to compare them, though. He was just asked a general question about Draymond's impact and volunteered that praise. If he just said "Dray's awesome, best defender in the league," do you think that triggers Green's "hothead" button?
But on the question of unbiased... as great as your predictive record normally is, you're 1-6 on Warriors predictions since KD left, and 0-5 on predicting them to lose (though smart money is you'll be right this time with Steph out and Jimmy moving about as explosively as Mozgov). Do you think there could maybe be a chance you've got a little blindspot against this team yourself as a backlash to all the Curry over Durant narratives or your (legit and justified) distaste for Draymond's antics?
DiffAnon:
I am not sure what triggers Draymond because, quite frankly, even he does not seem to know, nor does he know how to control his reactions even when those reactions demonstrably harm himself and his team. I can understand why Kerr would tread lightly around Draymond.
I did not check the specifics, but I am aware that on more than one occasion I have picked against the post-Durant Warriors and been wrong. I am not sure if those wrong predictions stem from bias or are just situations where I was "right" (in terms of making a logical, well-founded prediction) but turned out to be wrong. For example, I did not expect Curry to score 50 in a road game seven win versus the Kings--and after that game, I candidly stated that this game and some other games have caused me to at least consider the possibility that I may have underrated Curry.
LeBron and Curry are two of the most confounding players who I have analyzed, but for different reasons. LeBron has so many physical and intellectual gifts, yet he has come up short and even seemed to quit more often than should happen with a player as great as he is. After Curry was drafted, I projected a better NBA future for him than most "experts" did, but I did not foresee him becoming the player he became. In general, I don't expect players under 6-6 to be the best player on a championship team. Curry is an anomaly, and anomalies are difficult to categorize/understand.
I would just say that I am way too competitive to pick against a team because I don't like the team or some of the team's players; my goal is to go 15-0 on my playoff series predictions. There may be something about the Warriors that I don't understand, but I am not picking against them because I want them to lose. I am genuinely surprised by some of the playoff series wins that they have had post-Durant. I am still surprised that Tatum and Brown did not lead the Celtics past the Warriors in the NBA Finals, but it was interesting to me to hear Brian Scalabrine--who worked for the Warriors and is now a Celtics' analyst--explain before and after the series why he thought that Golden State was the better team. I still think that the Celtics "should" have won if they had exploited their advantages in the paint.
Legit answer. I think the Warriors are tough for all of us to fully dissect, because what Curry does is so off-brand from how we think of a traditional top flight star. We've had off-ball stars before, but not off-ball MVPs. You made fun of Curry's gravity replying to somebody else upthread, but even if his fans make too big of a deal of it, I think that sh*t does matter some. Even with Butler out there the Warriors offense suddenly looks bush league without him, and even with Durant they looked worse than they should have whenever he missed time. Whether its the "gravity" or something else, it sure feels like there's something about Curry that the box score can't quite capture.
I think we're seeing right now why Tatum and Brown couldn't get it done against him. They just don't really seem to have that extra grit/heart we associate with the all-timers, and they've got a bad habit of crumbling when the momentum turns. It ain't 100%, Tatum did have that awesome Milwaukee series and there've been a couple other big moments. But it's just not as consistent from them as it is from guys like Giannis, Jokic, or Curry who're usually awesome even in defeat. Give or take 2016, I guess.
Boston won last year in spite of that, but they had a lot of help from the injury gods and ran into a pretty flawed Dallas team in the Finals that could not push them defensively in the same way as the Warriors, Knicks, or especially the Butler-era Heat seem to be able to do to make them uncomfortable. Orlando on the other hand had the juice defensively, but just couldn't score enough to make it matter.
As much as the media (and, let's be so for real, Tatum himself) wants Tatum to be the next Kobe, he really just looks like he's actually the next Paul George, only rolling with a lot better teammates that can usually help hide his inconsistent moments. He's even got a badly timed playoff commercial now to uphold the legacy, though at least "Superman" will at least probably not catch on as a sarcastic nickname the same way Playoff P did.
If Paul George feels too harsh, maybe Karl Malone is fairer. Great stats, good enough to get pretty deep into the playoffs. but bad habit of forgetting how to ball when it's time to ball the hardest.
Still a great player, but seemingly not packing the killer instinct that separates the legends from the rest of the pack.
DiffAnon:
What is called "gravity" regarding Stephen Curry used to be called "tilting the floor" or drawing a double team, and superstars have been doing this for decades. In the 1980 and 1982 NBA Finals, the Lakers played a thinly disguised (and illegal) zone defense to cut off Julius Erving's driving lanes while daring his teammates to make open jump shots. I don't dispute that Stephen Curry has "gravity," "tilts the floor," or draws double teams; I dispute the implied notion that he is the first player to do this and that he does so more than any player ever. Wilt Chamberlain regularly had two or three players draped all over him, and he beat those defenses both by prolific scoring and by leading the league in assists. Michael Jordan distorted defenses tremendously (as Steve Kerr knows very well).
Curry is a great player who, through no fault of his own, is the subject of narratives that exaggerate his greatness (a similar thing happened with Steve Nash).
Tatum has had way too much playoff success and sustained All-NBA First Team level play to be lumped in with Paul George, but I agree that Tatum has not proven that he is at the level of Giannis, Jokic, or Curry. Tatum just turned 27 and already has one championship, two Finals appearances and five ECF appearances. He has already accomplished more from a team standpoint than Malone and George did in their entire careers.
I would like to see Tatum attack the hoop more, and I would like to see him convert at a higher rate at the rim. It is surprising that Tatum and Russell Westbrook--two players who dunk with authority--miss so many non-dunks at point blank range. Curry, who rarely dunks, seems to finish better in the paint than those guys.
For sure, stars been drawing doubles since the dawn of time. But I think when people are talking about it with Curry that's not all they mean. Most stars when they draw a double do it in the paint or on their way there. Curry the first guy who really does it thirty feet out, and that creates even more space than the vanilla version. And he can also do the vanilla version.
He's also one of the only guys who gets doubles off-ball. Partly that's because the illegal defense rules used to make that a no-go or I'm sure folks woulda tried it on MJ and Shaq at least and probably a few others. But it's also because he just scares teams more than the oldschool off-ball types did. His schtick isn't that different from Ray Allen or Rip Hamilton, but it's damn sure louder. Teams weren't really worried that Rip or Ray were gonna hang forty on 'em.
It doesn't make Curry better than the guys you don't want him to be better than, but it is pretty unique to him. Maybe it could be the secret sauce of why the Warriors are usually a little better than you think they oughtta be?
As for Tatum, obvi he's had more team success than George or Malone, but that's kinda exactly what I'm saying anyway isn't it? He's been on greater teams. If you take him off the 2024 Celtics and replace him with 2017 Paul George or 1997 Karl Malone, who do you think's beating them? Heat with no Jimmy? Pacers with no Hali? Cavs with no Spida? How much worse than Tatum's 39% can George or Karl really shoot in the Finals to blow three extra games against Luka?
Flip it the other way. Does swapping out Karl for Tatum beat the Bulls? Heck does that version of the Jazz even get to the Bulls anymore? Does dropping him on George's Pacers take down the Heatles? I guess we can't definitely beyond a shadow of a doubt know for sure, but I know where my money'd be.
Obvi he wins on resume, but if you're looking at skillset, what the's difference between him and George really? Tatum's a little better as a rebounder and shot blocker, George has a lot better handle and is a better man defender. Assists are close but Tatum's passing to way better dudes mostly. Probably a slight edge for George there, at worst a tie. Both dudes should be able to get to the rim whenever they want but both dudes would rather shoot a stepback three they forget how to make when they really gotta.
I would grant Tatum's got more total good years, and that margin's only gonna grow, but I'm not really talking about tenure here so much as just their overall bag.
Team success only goes so far. Robert Parish had more team success than Patrick Ewing, but Ewing was still a better player. That's a kinda bad example on my end because I'm not saying George is better, I'm just saying they're comparable, but I still think it makes the point ok.
Course, the big difference is Tatum's still young. He might still grow into being the guy his commercials and PR flaks and Bill Simmons are trying to tell us he already is.
DiffAnon:
Curry had Durant taking the primary offensive load for two of his four titles, and he had Thompson for all four; "gravity" would not matter if Curry did not play alongside other major threats.
As you noted, part of the way that teams guard Curry is related to rules changes. I would add that it is more difficult to guard perimeter players in general because physical defense outside of the paint has been all but eliminated. It would be very interesting to see, among others, Pete Maravich, Julius Erving, Michael Jordan, and Kobe Bryant play in an era with (1) the court spread out due to heavy usage of the three point shot, (2) minimal to no contact allowed on the perimeter, and (3) few legit rim protectors. Prime Maravich averaged over 30 ppg on a weak team with no three point shot. Put him in this era and he could average 35-40 ppg easily, and his assists would also skyrocket due not only to the above factors favoring the offense but also the loosened scorekeeping standards for assists.
I am not impressed by George's handles. I would take Tatum in that regard. Tatum also is a better scorer who already has five 26-plus ppg seasons (George has only one, and is highly unlikely to have any more). Tatum is a much better rebounder (7.3 rpg to 6.3 rpg career, four 8-plus rpg seasons compared to one for George). Tatum's assists are steadily increasing, and this season he averaged more assists (6.0 apg) than George ever has.
Tatum has a 15-7 playoff series with one championship, two Finals appearances, and five ECF appearances. George has an 8-11 playoff series record with no championships, two ECF appearances, and one WCF appearance. In his playoff career, Tatum has two 50 point games and two more 40 point games. He failed to score double figures just six times. George has no 50 point games and one 40 point game. He failed to score double figures 15 times (and he has played in fewer playoff games than Tatum).
Players can't control who their teammates are or who their teams face in the playoffs. Perhaps it could be argued that Tatum "should" have two titles instead of one, but he also will likely have many opportunities to add to his total. It is difficult to make a logical argument that George is a better playoff performer than Tatum.
Based on both career and peak value, I would take Tatum over George without hesitation.
"Curry had Durant taking the primary offensive load for two of his four titles, and he had Thompson for all four; "gravity" would not matter if Curry did not play alongside other major threats"
"Players can't control who their teammates are or who their teams face in the playoffs."
Which is it, man, do teammates matter when we're judging a dude or nah? Curry's gravity only matters because he had a 1st Team level player for two titles, a 3rd Team level player for one more, and then a weaker post-injury version of that former 3rd teamer for one more after that, but meanwhile Tatum's playoff record is more impressive the George's so he must be way better but it's just irrelevant he's getting those wins on All-Star Teams on the best-run organization in the league and George has been on mostly small-market teams with like one good teammate per year or snakebit Clippers squads where the best guy is always injured?
I'm not trying to play gotcha or anything, but the contrast there is too stark not to mention.
I can buy with either take but not from the same store, you know?
DiffAnon:
The "chicken/egg" question with Curry's "gravity" is specifically the extent to which he is making players better versus defenses are stretched thin in no small part due to the abilities of his teammates; to answer that question, the abilities of Curry's teammates matter. We can evaluate Curry's scoring, rebounding, assists, and defense as skill set traits outside the context of his teammates to a large extent, but the focus on his "gravity" seems to intentionally ignore or diminish the difference between creating space for Kevin Durant and creating space for someone who is not an all-time great player.
In contrast, your Tatum-George comparison leans heavily on comparing their respective teammates as opposed to evaluating each player's skill sets and performance levels. I agree that Tatum has had better teammates than George, but I would argue that Tatum has maximized or at least come close to maximizing his opportunities while George has often shrunk in the biggest moments and been the reason that his teams fell short; maybe George did not have enough around him to win, but that argument on his behalf would carry more weight if he did not have a Harden-like propensity to disappear when it matters most.
Food for thought, here's George's best recent playoff run vs. the year Tatum won the title.
https://stathead.com/basketball/versus-finder.cgi?request=1&seasons_type=perchoice&player_id1=georgpa01&p1yrfrom=2021&p1yrto=2021&player_id2=tatumja01&p2yrfrom=2024&p2yrto=2024
Tatum wins the regular season clean but George is hurt half the year, so I'm not gonna kill him for that. Tatum generally way more durable is probably the biggest point in his favor, TBR.
Both played 19 playoff games those years, so sample size is fair. Playoff numbers look a lot alike. Rebounding's virtually tied, George scores more and more efficiently, Tatum's got an extra assist and a better turnover rate.
George's best teammates are 11 games of Kawhi, followed by Reggie Jackson and Marcus Morris. Tatum's got Brown, Holiday, and Horford for the whole push, plus 7 games of Porzingis they didn't really even need because Celtics fans are spoiled.
If level of competition matters, Tatum played the Heat with no Jimmy, the Cavs losing Mitchell in Game 3, the Pacers losing Hali in Game 2, and a healthy Dallas team. George played a healthy pre-Kyrie Dallas team, a Jazz team missing one game of Joe Ingles, and then the Suns missing two games of Chris Paul. He lost Kawhi in the middle of the Jazz series.
PG's scoring goes way up after Kawhi goes down, which isn't weird, but so do all his other numbers and even his efficiency ticks up a little too. I'm not gonna call Wayoff P a playoff riser or anything, but fair shouts to him, at least that one time he at least tried to meet the moment.
Kinda looks like George put up comparable or even better numbers against tougher comp with worse help, IMO, but you might not agree and that's cool.
I'm with you that he's got more total good years and I'd take his career over George's too, but that wasn't the point I was trying to make at all. I'm just saying that Tatum at his best, at least in the playoffs, look way more like Paul George than he does to everyone else people try to put Tatum next to, or any typical championship Alpha this side of like Chauncey Billups, and I don't think one's better enough than the other that NBA history would change much if they swapped places.
IDK, it's just tough to put Curry's gravity all on Durant when he made 3 or 4 Finals without him depending how you count. Had record-breaking offenses without him and so on. Obvi Durant made the Warriors unbeatable. But it's not like no other great player ever played with somebody as good as Klay Thompson before, right? But most of them except maybe the Showtime dudes never played on offenses as good as the 2015 or 2016 Warriors I don't think. And then the 2022 team wasn't as good offensively on account of not really having anybody that was a real star besides Curry but they were still good enough to hang enough points to win on a pretty elite Boston defense.
If StatDork is still reading this thread, maybe he(?) can dig something up? Not sure how you'd even really check that, though. Are there gravity stats yet, and if there are is there any chance they aren't just Hollinger formula slop?
Anyways, as for Tatum and PG I just sent a whole big followup thing before I saw your reply, sorry. I'm not with you that Tatum has maximized his opportunities at all I guess is where we part. Dude was born on third base and mostly hits doubles, and he's been booty in both his Finals. Plus I think he sucked in all three Miami series even though they won one of them. The only time I really remember him rising to the occasion was the one Milwaukee series when Middleton was hurt, which is legit, but George had one awesome elimination game against Lebron's Heat too, I think.
Anyway, it's not important enough to keep going at it, sorry. Just my two cents. Tatum is healthy Paul George in a better situation, and good on him for getting
a ring while Jaylen did the hard part, but it folks could pump the breaks on the Top 5 Player or Face of the League or MVP Candidate convos, that'd be chill.
DiffAnon:
Why should the comparison be limited to just one season from each player's career? George is near the end of his career and has a larger sample size of seasons from which to choose, and we can be confident that he has likely already had his best regular season and best playoff run. Tatum is just entering his prime and he is still showing signs of improvement.
Tatum is bigger and more durable than George, and Tatum has come up with big playoff performances during multiple deep playoff runs. We can't blame George for who his teammates were or who he faced, but we can give Tatum credit for being the best player on one of the most consistent playoff teams of the past several years.
I would take Tatum over George both for peak value and for career, and I don't think that the comparison is as close as you are suggesting; the only way to even attempt to put them on the same level is to cherry pick seasons or compare their teammates and who they faced, because from both a skill set standpoint and an accomplishment standpoint Tatum is ahead by a good margin. There is no way to prove your hypothesis that NBA history would not change much if Tatum and George swapped places, but--for the reasons that I listed in this thread--I disagree with you.
I never put Tatum on the same level as Kobe or other Pantheon players, so it is not up to me to defend that comparison.
Man you're fast, I gotta start refreshing before I reply.
I went with peak seasons because that was the comp I was making. I already said Tatum's more durable. And obvi if Tatum gets better as he ages, the comp won't hold and I'll stop saying it. But so far I don't think he's done anything George couldn't do in the same situation, is all, and he's failed in some very Georgey ways along the way. That's all I meant.
It's cool if you don't wanna go further on it but I am curious now what you think would change? Do you think that George's teams just get a little farther with Tatum, or do you think it actually swings some titles? I won't argue with the take, I'm just curious what it is.
DiffAnon:
I am still reading. I wrote a much longer response but it's over the wordcount limit pretty dramatically and I don't want to main character in David's comments, so I will offer just a few general observations... which ended up still too long, but might fit in two posts instead of the seven or eight the full answer would require.
- There are several stats that attempt to quantify "gravity" but I don't think any of them do it very well. Most are based on player tracking data but struggle to factor in teammate or opponent quality.
- The good all-up offensive team-level impact metrics are much like plus/minus in that I think they are useful as a snapshot or starting point, but are not "proof."
- Those metrics tend to roughly agree that the most impactful offensive players ever are Magic, Nash, and Curry, and sometimes but less consistently Jokic and/or Lebron.
- Metrics that look at smaller sample sizes of 3-5 seasons or individual seasons have more variance. Many players have 3-5 years peaks in those metrics at the Magic/Nash/Curry level but don't sustain them in the same way.
- Most of those metrics like Curry more, not less, when his teammates are worse. 2021 is often his most dramatic "impact" season according to metrics that try to track such things, and that might be his worst supporting cast. So, the metrics don't think he needs Durant or even Klay to be effective.
- It's tempting to assume from that that these metrics reward good players for having bad teammates, but that doesn't really bear out across the wider sample size. Magic's numbers are better when the Lakers are at their best, most of Jordan's strongest impact seasons come alongside Pippen, etc. Curry, Kobe, Hakeem, and Lebron are the players who most tend get "better" with worse teammates, in terms of offensive footprint, in my experience. In Curry's case the variance is usually pretty small, though. Even in the metrics where 2021 is his "best" impact season, his title seasons all also generally still rank as elite, with or without Durant.
- There is a trend across most of those metrics that reward off-ball movement. Reggie Miller consistently ranks much higher across team-level impact stats than he does by any other metric. To a lesser extent the same is true of Ray Allen, Klay Thompson, and Rip Hamilton. The difference between how dominant Curry and Reggie are in these stats vs. Allen, Klay, and Hamilton I believe is deceptively simple: they move more. Ray, Klay, and Rip spent more possessions as stationary spot-up threats, while Curry and Reggie tend to lead the league in miles run per game on offense comfortably in their tracked seasons.
- Interestingly, the running around might be more important than the shooting. Manu Ginobili also does very well by those stats despite not being the same kind of instant offense catch-and-shoot threat. Manu's role as mostly a sixth man makes me suspicious of his data's trustworthiness, but it is still another circumstantial supporting point for the "running around matters" argument. We don't have PBP tracking data for John Havlicek in the same way but he similarly outperforms in these kind of impact metrics relative to what you might expect.
Continued:
- I think a lot of these metrics might slightly underrate Durant, who usually scores well but not at an elite level in them. He's fairly unique in that he's spent virtually his whole career with high-level team creators and as such has never quite been the focal point or thesis of an offensive system in quite the same way as someone like Dirk , whom these metrics love, particularly post-Nash. There's a lot of nuance into the why of how they score so differently I don't have the wordcount for, but in short my hypothesis is that if Durant had a system built around him the same way Dirk did after Nash left, his impact metrics would likely be comparable or higher, but he never really did. Phoenix and GSW plugged him into existing systems, Brooklyn barely had a system at all, and in OKC Westbrook was effectively his version of Nowitzki's Nash.
- I do not think even a purpose-built Durant system would likely reach the Magic/Nash/Curry level of impact, however. He would need to either be a better passer or a more active off-ball mover to replicate those footprints. Even the most elite scorers lag behind those three (and sometimes Lebron or Jokic) because their non-scoring possession don't generate the same leverage.
- Shaq's best 3-4 seasons are by some metrics an exception to what I just said, but it's not consistent, either across his larger career or across all the metrics. But he comes the closest to breaking the "must be an elite passer or off-ball move to crest the mountain" rule, in large part due to his foul-drawing, which has a lot of downstream impacts.
- None of this makes Curry better than Durant. It also does not make Durant better than Curry. It just suggests that Curry has a stronger global effect on his teams' offenses than Durant does. I believe that to be generally true, but the important thing to remember about these kind of metrics is that they are looking at massive sample sizes and don't necessarily mean much in the highest-leverage moments. Curry very likely makes it easier for a guy like Draymond to get layups in the first forty-eight minutes of the night, but in crunchtime of an elimination game you would probably much rather Durant's iso scoring ability than a 25% higher chance of a Draymond Green layup.
- To be fair to Curry, his clutch/crunchtime/iso metrics are pretty comparable to Durant's. To the extent Durant might be better" than Curry it is probably more a function of his defensive or rebounding value than anything that shows up in these metrics, or of something else the metrics can't quite catch.
- The most obvious observation is the one that often goes ignored by either player's fans: they both made each other better, statistically. It was not a one way street in either direction.
- That said, at the risk of feeding the Curry fans, Curry had a much larger statistical impact on Durant than vice versa, and this is consistent no matter how you slice up the numbers and sample sizes.
- I would hesitate to use that to try to prove much beyond that Curry did not "need" Durant to be effective, which you'd think his other title runs would prove anyway. If anything, I think it suggests some offensive dysfunction on the part of Durant's other franchises that they could not as fully set him up for success despite his obvious talent.
DiffAnon:
Yes, I think that Tatum would take George's teams farther than George did, and I think that George would not have led Boston to a title (or five ECF appearances) in Tatum's place. George is not as durable, not as big, not as skilled, and he is often making excuses or looking for ways out (as seen by his team jumping and repeatedly teaming up with other stars, while Tatum never complained or made excuses even when outside voices said that he and Brown would never win a title together).
StatDork:
I appreciate your diligence, but I am skeptical of statistics that purport to measure "gravity," let alone determine how much "gravity" each player has. There are just too many independent variables here to assign credit so specifically.
There is no doubt that Curry has "gravity"/draws double teams/tilts the floor; most great players do this to some extent. I remain skeptical of the notion that Curry does this to a greater extent than any other player. I would need to see a whole lot of numbers paired up with the associated video clips for context to convince me otherwise. I have watched and analyzed a lot of basketball, and I just don't see Curry's "gravity" the way that it is often described. I see Curry being an exceptional shooter, scorer, finisher, ballhandler, and passer. What the Warriors are missing now is not so much Curry's "gravity" as they are missing 25-30 points, five or six assists, and the downstream effects of those missing points/assists: everyone in the rotation has to move up a notch, and TNT's Kenny Smith often talks about the importance of having a balanced rotation in which each player is only being asked to do the things that he does best. Now, the Warriors are missing their best scorer/shooter, their best ballhandler, and their best passer. Of course their offense looks bad. Take away any player who produces 25-30 ppg and 5-6 apg, and his team's offense will crater unless that team has an embarrassment of riches on the roster. We don't need "gravity" to explain that it is difficult to replace 25-30 ppg/5-6 apg, and that Buddy Hield--to cite just one player--is going to not do as well being guarded by a defender who normally would be assigned to Curry. There is an obvious difference between being guarded by the best perimeter defender, and being guarded by a lesser defender who also has some responsibility to hedge toward the 25-30 ppg player.
I am going to stick to the notion that a 6-9 (or 6-10 or 6-11 or whatever height Durant actually is) elite scorer is more difficult to guard (and thus inherently has more "gravity") than a 6-3 elite scorer, and I will say the same about, among others, Julius Erving, Michael Jordan, and Kobe Bryant; if the latter three players played in today's NBA they may not stretch the floor beyond the three point line the way that Curry does, but freed from the shackles of handchecking and attacking a painted area devoid of rim protectors they would each be a nightmare to defend. To cite just one example, Erving averaged 30.3 ppg in the 1977 NBA Finals versus a great team featuring Bill Walton and Maurice Lucas in the paint, and Erving did this with handchecking and without a three point line. What would Erving average in a playoff series versus the Golden State Warriors with no handchecking and with only Green and Looney in the paint? I think that 35-40 ppg is a conservative number. People go nuts over some of Anthony Edwards' dunks, but Erving was bigger, stronger, faster, and dunked harder in traffic over seven footers.
Also, it is not a stretch to imagine Erving developing a good to great three point shot if he played in this era (during his Nets' years, he was the team's go to guy for late game three point shots, but the three point shot was rarely used even in the ABA). Picture Erving in today's game as even just a 35% three point shooter taking 4-6 threes a game to keep defenses honest, and spending the rest of the time attacking the hoop; it would be like watching Luka Doncic with hops, or Luka Doncic at 1.5X speed with hops.
David:
The question you are basing your conclusion off of seems to be "who is harder to guard under the most difficult conditions," and your answer to that makes sense as an answer to that question, but it's not the question these impact stats are interested in. They are asking, roughly, "who provides the most total value over the course of a large sample size?"
Put another way, it doesn’t really matter as much as people like to think who is harder to guard in isolation situations. Both Curry and Durant well over 80% of their points through other means, and both excel in those situations when they do find themselves in them.
Consequently, who’s better in one-on-one or isolation situations is a relatively small part of the puzzle for most of these impact metrics. What they are looking for is not who can get you a bucket when you need it most, but who adds the most total value over the course of a season, playoff run, or career. It is very possible for Player A to be a much better under-pressure scorer and Player B to still have more total offensive value. Magic and Kareem are a good example of this in the early 80s, where Kareem is generally tougher to stop for a single possession, but Magic’s wider skillset and ability to create for his teammates is more important to the offense overall across 48 minutes.
These stats all work differently, so it is noteworthy that they largely agree on Curry. Some look at how a team performs with or without their star player in the lineup, many look at relative scoring or offensive efficiency vs. the rest of the league or other players or lineups, some incorporate +-, others I trust less try to factor in things like “expected shot quality,” and so on.
But off-ball movers like Curry and Miller do very well in these metrics because they’re simply relevant on more possessions than a Durant type is. All players move some off the ball depending on the play-call, but the frequency of it is wildly different.
On an “average” Devin Booker possession, Durant may set one screen or make one cut, or he may just park himself in the corner. He’s still providing value when he does that, as no defender wants to leave Kevin Durant and even just standing idle in the corner will help open the paint, but it does not stress the defense or force them to make a lot of hard choices.
Curry, on the other hand, will often set two off-ball screens and make three or four cuts on a single play. Every time he does, a defensive player has to make a decision or a rotation while also ideally keeping their eye on whatever the ballhandler is doing, and this stresses defenses. Even if they only make a mistake every twenty decisions, if Curry is forcing them to make 3-5 decisions per play, that will add up over the course of a game vs. Durant forcing them to make on average maybe one or two choices per off-ball play.
The most common defensive mistake we see on his off-ball actions is the off-ball double, either when Curry screens for someone else and they both stay with him, or when Curry receives an off ball screen and both defenders chase him leaving the screener wide open. GPII is a frequent beneficiary of this action on the current Warriors, and if you ever see him suddenly all alone at the rim ready for a bunny layup, this is often why. Ten years ago, Iguodala got a lot of those same looks.
The outcome of this kind of constant additional offensive pressure from a non-ballhandler is similar to the impact of a great passer. Against a great defense there may be nights when it doesn’t create any points at all, but there will also be nights when it adds twenty. Over the large sample sizes these stats are concerned with, those possessions seem to add up quite a bit.
There’s a secondary downstream effect as well that this can also just exhaust defenders. He forces them to use more energy on possessions where traditionally they might not have to do anything at all guarding most players, or might be responsible for just one or two sharp rotations. It’s difficult to quantify how much value that has numerically, but it’s something we’ve seen happen enough, and heard defenders complain about enough, that we know it’s “real.”
I’m not really interested in trying to persuade anyone, but I did want to try and explain why those metrics consistently find that Curry (or Miller) has an outsized statistical impact on his team offenses that is greater than what you get from equivalent or superior scorers. I don’t even necessarily think that having a larger “overall” offensive impact necessarily makes someone a better offensive player; James Harden is a great cautionary tale for the extreme version of someone who clearly provides a ton of large-sample size offensive value but also probably guarantees you will never win a title; the playoffs are, after all, a fairly small sample size.
So, two things can be true. It’s pretty clear to everyone who looks at these kind of stats for a living (and to me, who looks at them as an obsessive special interest) that Curry does have some kind of disproportionate gravity that we don’t usually see even in other elite scorers and that shows up in him greater impact metrics than players with otherwise similar box score stats. But for your purposes of ranking players in or out of your Pantheon, it might not actually matter very much.
That’s the trap people fall into with these stats. Just because they’re true, doesn’t always mean they matter. And even when the do matter, it also doesn’t mean that something else they’re not tracking can’t matter even more.
To use the most extreme/easy example, just about every stat we have for looking at this sort of thing would indicate that Curry and his gravity almost certainly adds more offensive value over a regular season than even Michael Jordan. So does the passing of Steve Nash or Magic Johnson. It’s a numbers game that anyone who doesn’t have a major secondary non-scoring offensive impact can’t really mathematically win over a sufficiently large sample size, because they’re influencing so many more plays than a dedicated scorer is usually going to. But for a must win playoff game, I’d argue it’s probably much more preferable to have Jordan’s foul-drawing, ability to scale up his scoring to 50+ points almost as needed, and unstoppable one-on-one game is usually going to improve your chances more than Curry’s ability to create a couple extra good looks for his role players.
Another way of looking at it: Curry’s off-ball gravity is real and valuable, but there isn’t really a way for him to scale it up when he needs to, while Jordan or Durant can always just decide to take fifteen more shots in a big game.
Now, obviously, so can Curry and he has good results when does, but I’m not actually interested in trying to figure out who’s ultimately "better". I’m just trying to explain why the metrics say what they say, and why that doesn’t necessarily mean what either Curry or Durant’s fans might want it to.
My blasphemous opinion is that once you get to the top of the mountain, “better” doesn’t really exist in a way that matters much, or at least, it might vary based on system, matchup, fit, etc. If the Nuggets and Bucks swapped Giannis for Jokic next year it would probably make both teams worse, because Jokic is “better” for how the Nuggets are constructed and Giannis is “better” for how the Bucks are built. So who’s “better?” You could probably say the same for the 2016 Warriors and Thunder; if you traded Durant for Curry before that season I think there’s at least a fair chance it means the Spurs are in the Finals instead, but that’s neither here nor there
TL;DR: Curry’s off-ball gravity is pretty obviously real, or at least there's something about him that makes him "matter" more to his teams than comparable scorers statistically, but it’s also not the magic bullet argument-ender his fanbase seems to think it is. It’s simply one thing that matters in a sport where a million things matter.
StatDork:
Thank you for providing so much background and context. It is refreshing that you understand the limitations of what can be measured and quantified.
I found this statement particularly worth emphasizing:
"That’s the trap people fall into with these stats. Just because they’re true, doesn’t always mean they matter. And even when the do matter, it also doesn’t mean that something else they’re not tracking can’t matter even more."
Many of the people who I derisively call "stat gurus" either do not understand those points, or deliberately ignore those two points because acknowledging those points would negatively impact their ability to monetize their proprietary "advanced basketball statistics."
I love the intelligent use of basketball statistics, and I hate the unintelligent use of basketball statistics. People who use the numbers intelligently are very cognizant of the limitations of the numbers, but people are not using the numbers intelligently insist that the numbers can be used to accurately and precisely rank players and teams.
Well NOW I feel bad about the Paul George comp, damn.
Hopefully just a sprain and not the Achilles it looked for Tatum.
DiffAnon:
Tatum's injury was sad to watch, and it is even sadder to realize that he will likely miss all of next season. This has to be on the short list of most devastating injuries suffered by a defending NBA champion. Off the top of my head, I would put Bill Walton in 1978 at the top of that list. Durant's Achilles injury in 2019 comes to mind. Magic's hamstring injury when the Lakers were going for a three-peat in 1989 does not seem as history-altering as Tatum's injury; the Pistons could very well have beaten the Lakers even at full strength.
The only good thing is that an Achilles injury is not the career-ender it used to be decades ago. Tatum will lose a full prime season, but he should be able to return to elite status for several years after he comes back.
Post a Comment
<< Home