Top 25 peaks of the 2001-25: #11-#12 Spots

[Owly]

My issue with Manu is there is a lot of space between 'highest +/- ever as a duo with Duncan from 05-07' and him playing 28-30mpg in the rs and not having that great of box numbers outside of strong efficiency on low volume and being a strong defender. Could we work Parker into all this +/- stuff as a point of comparison perhaps?

Fwiw on "not having that great of box numbers outside of strong efficiency on low volume and being a strong defender"

Granting lower minutes so rate stats need some accommodation for this (but imo, this is better than raw stats as though a replacement would produce nothing)
Versus a more hyped positional peer (with typically weaker impact numbers)
PER RS peak
Bryant ('06) 28.0
Manu ('08) 24.3

WS/48 RS peak
Bryant ('06) .224
Manu ('07) .246 (excludes low minutes '12 - and if you want to tie Manu to one year and '08 it's .234)

BPM RS peak
Bryant ('06) 7.6
Manu ('08) 8.3

PER playoff peak
Bryant ('09) 26.8
Manu ('05) 24.8

WS/48 playoff peak
Bryant ('01) .260 (if locking to one year and going '09 .238)
Manu ('05) .260

BPM playoff peak
Bryant ('09) 9.1
Manu ('05) 9.2

There are probably people on the board with nicer numbers depending on where you focus, what you trust etc but ... broadly similar numbers here (mileage may vary on accounting for minutes, accounting for defensive effort/impact - alignment of RS and playoffs, weighting of RS and playoffs etc)

Personally, I would be interested in seeing it extend out to like 2014 since Duncan remained effective out until then and some of Parker's best seasons were after 2011. I appreciate the reply though.

The post links to an extend through 2016. You can easily play with different start points and a 2014 end. I don't think it will change the take-home message though. As noted my first set of numbers cut just prior to a year (2011) that really hurts Parker relative to Manu.

Fwiw here's 11-14 as a complement to the first table with '14 as an end point.
https://www.pbpstats.com/wowy-combos/nba?TeamId=1610612759&Season=2010-11,2011-12,2012-13,2013-14&SeasonType=All&PlayerIds=1495,2225,1938
Smaller samples here. Manu still clearly looking strongest (Manu-only now first overall on a decent sample). Older Duncan ... mostly looking weaker though only Duncan in a small sample rises so ... whilst generally looking a bit better ... Parker's "only Parker" number is again out on an island.

[edited to correct two of Bryant's playoff years given for box stats peaks - numbers were correct but years not]

[Doctor MJ]

My question with Manu if you are sort of considering putting him on your ballot is how good do you think his rs actually was in 2005? Like should he have been all nba, should he have been top 10 in mvp voting or what because I think you gotta address that part of it if you are arguing him over guys who won mvps or finished like top 2-3 in given seasons. Plus the obvious thing of whether you can really argue he was the best player on his own team or even close to it.

A good question worth thinking through.

I'll point out briefly that since the MVP ballot only goes 5 deep, placement beyond the Top 5 really isn't anything to take seriously - but as I say this, it's not like I think voters thought of him as a top 10 player.

Re: should Manu gotten far more accolades? Absolutely, the voters in general underrated him by a massive margin... and so did the rest of us.

Re: best player on his team? Over the course of the complete '04-05 season I'd say he was. Generally though he wasn't, because Duncan was. This was a team winning generally with defense, and Duncan was the defensive anchor of the team.

Re: arguing him over MVPs or top 2-3 MVP finishers. This is too general of a statement to get fully addressed, but if you're asking whether I rank him over peak Westbrook or Rose, yes, absolutely. Ginobili's lesser minutes would always hinder his actual MVP candidacy, but there's no question in my mind who I'd rather have if I were trying to win a championship.

[Cavsfansince84]

A good question worth thinking through.

I'll point out briefly that since the MVP ballot only goes 5 deep, placement beyond the Top 5 really isn't anything to take seriously - but as I say this, it's not like I think voters thought of him as a top 10 player.

Re: should Manu gotten far more accolades? Absolutely, the voters in general underrated him by a massive margin... and so did the rest of us.

Re: best player on his team? Over the course of the complete '04-05 season I'd say he was. Generally though he wasn't, because Duncan was. This was a team winning generally with defense, and Duncan was the defensive anchor of the team.

Re: arguing him over MVPs or top 2-3 MVP finishers. This is too general of a statement to get fully addressed, but if you're asking whether I rank him over peak Westbrook or Rose, yes, absolutely. Ginobili's lesser minutes would always hinder his actual MVP candidacy, but there's no question in my mind who I'd rather have if I were trying to win a championship.

Ya, I mean I'd probably take him over WB or Rose at their peaks as well. Also though, keep in mind we're not really that close to WB&Rose getting serious talk yet. We still have Harden, CP3, KD, Nash& Dwight who all won mvps or came close to it. Draymond may also have an argument over Manu as well tbh. On top of Kobe&Dirk.

[Cavsfansince84]

Fwiw here's 11-14 as a complement to the first table with '14 as an end point.
https://www.pbpstats.com/wowy-combos/nba?TeamId=1610612759&Season=2010-11,2011-12,2012-13,2013-14&SeasonType=All&PlayerIds=1495,2225,1938
Smaller samples here. Manu still clearly looking strongest (Manu-only now first overall on a decent sample). Older Duncan ... mostly looking weaker though only Duncan in a small sample rises so ... whilst generally looking a bit better ... Parker's "only Parker" number is again out on an island.

[edited to correct two of Bryant's playoff years given for box stats peaks - numbers were correct but years not]

Ok I looked at the 11-14 data as well and one of the things that I see is that Manu played without the other big 2 as much as the other two did combined. Which I think is explained in part by him becoming more of Pop's secret weapon off the bench but part of this is SA being known for having very deep benches and probably playing a lot of those minutes against other team's benches. Which might impact the numbers we are seeing by him playing quite often against lesser benches rather than starters. Would you agree that this is part of what is reflected in the data? Though at this point maybe its better to just focus more on 2005 and not so much on these other years.

[falcolombardi]

A good question worth thinking through.

I'll point out briefly that since the MVP ballot only goes 5 deep, placement beyond the Top 5 really isn't anything to take seriously - but as I say this, it's not like I think voters thought of him as a top 10 player.

Re: should Manu gotten far more accolades? Absolutely, the voters in general underrated him by a massive margin... and so did the rest of us.

Re: best player on his team? Over the course of the complete '04-05 season I'd say he was. Generally though he wasn't, because Duncan was. This was a team winning generally with defense, and Duncan was the defensive anchor of the team.

Re: arguing him over MVPs or top 2-3 MVP finishers. This is too general of a statement to get fully addressed, but if you're asking whether I rank him over peak Westbrook or Rose, yes, absolutely. Ginobili's lesser minutes would always hinder his actual MVP candidacy, but there's no question in my mind who I'd rather have if I were trying to win a championship.

Ya, I mean I'd probably take him over WB or Rose at their peaks as well. Also though, keep in mind we're not really that close to WB&Rose getting serious talk yet. We still have Harden, CP3, KD, Nash& Dwight who all won mvps or came close to it. Draymond may also have an argument over Manu as well tbh. On top of Kobe&Dirk.

Draymond is like center ginobili in his impact profile but in more minutes, so he prolly should get ranked above ginobili based on that

I disagree on equalizing rose and westbrook or putting manu a tier ahead tho

peak 16 westbrook was either half of elite okc 2-man offense in a very limited spacing/offense roster
And 17 westbrook had a monstruous impact metrics and floor raising season as a lone star in also a limited spacing team

I think his bad ageing curve and frustrstion hair pulling flaws has made people forget westbrook peak a bit, he was a legitimate fringe top 5 player in the league

Rose was a notch below that before the injuries

[Cavsfansince84]

Draymond is like center ginobili in his impact profile but in more minutes, so he prolly should get ranked above ginobili based on that

I disagree on equalizing rose and westbrook or putting manu a tier ahead tho

peak 16 westbrook was either half of elite okc 2-man offense in a very limited spacing/offense roster
And 17 westbrook had a monstruous impact metrics and floor raising season as a lone star in also a limited spacing team

I think his bad ageing curve and frustrstion hair pulling flaws has made people forget westbrook peak a bit, he was a legitimate fringe top 5 player in the league

Rose was a notch below that before the injuries

Well, part of what we're saying is who you'd want to win a series with. With WB maybe his rs in 2017 was good enough to put him over Manu. It's hard to say. One thing I noticed about WB as I was reviewing some of those 11-16 Okc playoff runs is how he'd have some games where he'd get to the line a ton and make almost all of them or some streaky 3pt shooting. It's kind of weird looking at his playoff games. It's like his style of play is so up and down that it's hard to figure out who he really is as a player. There are definitely a lot of series where it seems like he was as good as KD at face value though.

[lessthanjake]

My issue with Manu is there is a lot of space between 'highest +/- ever as a duo with Duncan from 05-07' and him playing 28-30mpg in the rs and not having that great of box numbers outside of strong efficiency on low volume and being a strong defender. Could we work Parker into all this +/- stuff as a point of comparison perhaps?

So I think Owly basically covered what I was going to respond to this with. I’ll just give some similar information as well:

Ginobili’s box numbers are actually really good in rate stats. His regular season BPM in 2005-2007 was 7.2. In those same years, it was 6.8 in the playoffs. There are players with higher numbers than that, but those are pretty great! He was 5th, 6th, and 4th in the NBA those years in the regular season, and he was 3rd in the NBA in playoff BPM in 2005 (behind only two guys who only played one series, so Manu’s was really the most impressive). He was also 7th in the NBA in playoff BPM in 2007 (with two people above him being guys who only played one series). As another box data point, RAPTOR from that era is just a box stat, and from 2005-2007 Ginobili ranked 1st, 2nd, and 1st in the NBA. Similarly, his WS/48 from 2005-2007 ranked 5th, 5th, and 2nd. So yeah, I think Manu’s box stats were actually great. The raw stats just don’t quite look like that because of a combination of playing in a low-pace team in a low-scoring era and playing relatively low minutes. The low minutes is definitely an issue we should account for, but I think we should recognize that his box stats are good on a per-possession or per-minute basis, so the issue of relatively mediocre box stats is really just a product of the minutes rather than being a separate issue.

As another data point on his box stats, I would note that I don’t even really think his volume was all that low on a per-possession basis. In 2005, Manu scored 29.2 points per 100 possessions in the regular season and 34.1 points per 100 possessions in the playoffs. Over the course of 2005-2007, those numbers were 30.3 in the regular season and 31.8 in the playoffs. If we compare to unquestionably great scoring guys on two-star teams, that isn’t actually particularly low volume. For instance, from 2011-2014, Wade scored 33.8 points per 100 possessions in the regular season and 30.0 points per 100 possessions in the playoffs. Wade’s best playoffs in that era was definitely 2011, and he scored 34.6 points per 100 possessions in that playoff run, which is very similar to Ginobili’s 2005 playoff scoring. Similarly, from 2000-2002, Kobe scored 33.8 points per 100 possessions in the regular season and 31.8 points per 100 possessions in the playoffs. Even in Kobe’s best playoff run in that timeframe, his volume (35.4 points per 100 possessions) wasn’t much higher than Manu’s 2005 playoff volume. I won’t bother listing out the numbers, but I’ll also note that the assists per 100 possessions for these three guys in these timeframes were also super similar. So yeah, not the volume of super high-volume guys, but similar to what other incredible players that we’ve already voted in or are about to vote in have output on multi-star teams, particularly in the playoffs.

As for adding Parker to the on-off mix, here’s some data for a couple timeframes. The general gist is basically that Parker isn’t always adding much (and does badly without either of the other two) and Ginobili and Duncan both look great, with Ginobili probably even looking a bit better, particularly in the playoff data.

2005-2007 Spurs net ratings

Regular Season + Playoffs

Ginobili + Duncan + Parker: +14.71

Ginobili + Duncan only: +15.20
Ginobili + Parker only: +9.93
Duncan + Parker only: +6.54

Ginobili only: +6.81
Duncan only: +7.09
Parker only: -8.58

Regular Season Only

Ginobili + Duncan + Parker: +17.02

Ginobili + Duncan only: +16.36
Ginobili + Parker only: +10.60
Duncan + Parker only: +8.22

Ginobili only: +6.63
Duncan only: +10.46
Parker only: -6.75

Playoffs Only

Ginobili + Duncan + Parker: +7.64

Ginobili + Duncan only: +11.55
Ginobili + Parker only: +8.00
Duncan + Parker only: -1.14

Ginobili only: +8.36
Duncan only: -18.85
Parker only: -32.66

2003-2016 Spurs net ratings

Regular Season + Playoffs

Ginobili + Duncan + Parker: +10.86

Ginobili + Duncan only: +12.94
Ginobili + Parker only: +7.07
Duncan + Parker only: +7.01

Ginobili only: +7.98
Duncan only: +4.36
Parker only: -0.16

Regular Season Only

Ginobili + Duncan + Parker: +12.01

Ginobili + Duncan only: +12.57
Ginobili + Parker only: +7.10
Duncan + Parker only: +8.04

Ginobili only: +8.40
Duncan only: +5.57
Parker only: +1.54

Playoffs Only

Ginobili + Duncan + Parker: +7.03

Ginobili + Duncan only: +14.53
Ginobili + Parker only: +7.03
Duncan + Parker only: +1.78

Ginobili only: +4.42
Duncan only: -10.20
Parker only: -20.69

[lessthanjake]

Fwiw here's 11-14 as a complement to the first table with '14 as an end point.
https://www.pbpstats.com/wowy-combos/nba?TeamId=1610612759&Season=2010-11,2011-12,2012-13,2013-14&SeasonType=All&PlayerIds=1495,2225,1938
Smaller samples here. Manu still clearly looking strongest (Manu-only now first overall on a decent sample). Older Duncan ... mostly looking weaker though only Duncan in a small sample rises so ... whilst generally looking a bit better ... Parker's "only Parker" number is again out on an island.

[edited to correct two of Bryant's playoff years given for box stats peaks - numbers were correct but years not]

Ok I looked at the 11-14 data as well and one of the things that I see is that Manu played without the other big 2 as much as the other two did combined. Which I think is explained in part by him becoming more of Pop's secret weapon off the bench but part of this is SA being known for having very deep benches and probably playing a lot of those minutes against other team's benches. Which might impact the numbers we are seeing by him playing quite often against lesser benches rather than starters. Would you agree that this is part of what is reflected in the data? Though at this point maybe its better to just focus more on 2005 and not so much on these other years.

So this is an issue I’ve addressed in an earlier post in this thread (https://forums.realgm.com/boards/viewtopic.php?p=119700820#p119700820). The upshot is that Ginobili’s numbers look incredible even if we filter down to just minutes without the Spurs’ other stars where he is against 4 or 5 opposing starters, as well as in minutes without the Spurs’ other stars where the Spurs have fewer starters on the floor than the other team (even treating Ginobili as a starter for these purposes, regardless of if he did actually start). Like, not only do his numbers on those fronts look better than Duncan’s numbers, but they look better than many of the other guys we’ve already voted in when they played without their co-stars. So yeah, his great numbers aren’t really a function of farming opposing bench units. He just looks incredibly good no matter what the starter state of the game is.

[Owly]

Fwiw here's 11-14 as a complement to the first table with '14 as an end point.
https://www.pbpstats.com/wowy-combos/nba?TeamId=1610612759&Season=2010-11,2011-12,2012-13,2013-14&SeasonType=All&PlayerIds=1495,2225,1938
Smaller samples here. Manu still clearly looking strongest (Manu-only now first overall on a decent sample). Older Duncan ... mostly looking weaker though only Duncan in a small sample rises so ... whilst generally looking a bit better ... Parker's "only Parker" number is again out on an island.

[edited to correct two of Bryant's playoff years given for box stats peaks - numbers were correct but years not]

Ok I looked at the 11-14 data as well and one of the things that I see is that Manu played without the other big 2 as much as the other two did combined. Which I think is explained in part by him becoming more of Pop's secret weapon off the bench but part of this is SA being known for having very deep benches and probably playing a lot of those minutes against other team's bench. Which might impact the numbers we are seeing by him playing quite often against lesser benches. Would you agree that this is part of what is reflected in the data? Though at this point maybe its better to just focus more on 2005 and not so much on these other years.

I think it depends on the year.

I think RAPM should be accounting for this.

In this particular window ... maybe at points?

2011 he's a starter, not many minutes fewer than Parker (2426 versus 2528), he's in the clear-cut highest minute lineup and 4 of the top 5 minutes played lineups. And by box and by impact he's the best player on the team, even though no longer at his absolute best.

I think others discussing this recently have looked at whether he's at some great advantage in terms of opponent starter versus Spurs starter situation and said generally no - if I've understood correctly.

Latterly he's at less than half the game, not starting so some time, probably substantial time may well be in bench units.

I don't think people are or particularly should be looking at that 12-14 window to evaluate his peak too much.

I think his "solo" number in the 11-14 window is good enough that even chipping away at it at the margins in terms of taking it at face value ... it's still really good (and still so much better than maybe-apex Parker's).

I think in the long term it's clear which of the Spurs "big 3" drove significant impact and which ... seemed not so much to do so.

[edited to complete incomplete thought - insert "should be"]

[lessthanjake]

A good question worth thinking through.

I'll point out briefly that since the MVP ballot only goes 5 deep, placement beyond the Top 5 really isn't anything to take seriously - but as I say this, it's not like I think voters thought of him as a top 10 player.

Re: should Manu gotten far more accolades? Absolutely, the voters in general underrated him by a massive margin... and so did the rest of us.

Re: best player on his team? Over the course of the complete '04-05 season I'd say he was. Generally though he wasn't, because Duncan was. This was a team winning generally with defense, and Duncan was the defensive anchor of the team.

Re: arguing him over MVPs or top 2-3 MVP finishers. This is too general of a statement to get fully addressed, but if you're asking whether I rank him over peak Westbrook or Rose, yes, absolutely. Ginobili's lesser minutes would always hinder his actual MVP candidacy, but there's no question in my mind who I'd rather have if I were trying to win a championship.

Ya, I mean I'd probably take him over WB or Rose at their peaks as well. Also though, keep in mind we're not really that close to WB&Rose getting serious talk yet. We still have Harden, CP3, KD, Nash& Dwight who all won mvps or came close to it. Draymond may also have an argument over Manu as well tbh. On top of Kobe&Dirk.

Draymond is like center ginobili in his impact profile but in more minutes, so he prolly should get ranked above ginobili based on that

I disagree on equalizing rose and westbrook or putting manu a tier ahead tho

peak 16 westbrook was either half of elite okc 2-man offense in a very limited spacing/offense roster
And 17 westbrook had a monstruous impact metrics and floor raising season as a lone star in also a limited spacing team

I think his bad ageing curve and frustrstion hair pulling flaws has made people forget westbrook peak a bit, he was a legitimate fringe top 5 player in the league

Rose was a notch below that before the injuries

I wouldn’t say Draymond quite has the impact profile that Ginobili has. For instance, in NBArapm’s five-year RAPM, Ginobili has 5 different spans above 7.0, while Draymond peaks at 7.0. Ginobili has 4 spans in the top 3 in the NBA in five-year RAPM, while Draymond was never in the top 3. If we go to three-year RAPM instead to get at Draymond’s very best years, he still falls a bit lower than Ginobili’s best three-year span, and Ginobili was #1 in the league while Draymond was #3. If we go to impact-box hybrids instead, we see Draymond peaking out at a +5.6 regular season EPM and +6.4 playoff EPM (with those happening in different years). Ginobili peaked out at +6.4 regular season EPM and a +7.3 playoff EPM (both in the same year). Ginobili certainly isn’t light years ahead of Draymond in terms of impact, but I do think he’s ahead. Impact-correlated box data also has Ginobili ahead, and that’s true even when accounting for the minutes difference. For instance, in RS+Playoffs, 2005 Ginobili has a higher VORP (i.e. the non-rate-adjusted version of BPM) than 2016 Draymond, despite playing fewer minutes.

A couple key things for me here also include: (1) in Ginobili’s peak year, his team won the title, while that’s not true for Draymond; and (2) peak Ginobili genuinely has an argument for being the best player on a championship team, while Draymond does not have such an argument (even if we look at the years where his team did actually win the title). I definitely put significant weight on the greatness of winning a title and particularly so if a guy was his team’s best player in doing so. And, I definitely don’t think I’m alone in that. So this is a big factor weighing in Ginobili’s favor over Draymond.

[iggymcfrack]

My question with Manu if you are sort of considering putting him on your ballot is how good do you think his rs actually was in 2005? Like should he have been all nba, should he have been top 10 in mvp voting or what because I think you gotta address that part of it if you are arguing him over guys who won mvps or finished like top 2-3 in given seasons. Plus the obvious thing of whether you can really argue he was the best player on his own team or even close to it.

He was 11th in the league in VORP which I think is his absolute floor since anything with an impact component will obviously rate him significantly higher. I’d say he should have been 1st team all-NBA and finished somewhere between 4th and 7th in MVP voting.

[f4p]

If we look at Duncan and Ginobili instead, here’s what we see from 2005-2007: If Duncan was on the court, the net rating went up +8.15 if you added Ginobili. If Duncan was not on the court, the net rating went up +16.82 if you added Ginobili. If we looked at the full 2003-2016 sample together instead, if Duncan was on the court, the net rating went up +4.64 if you added Ginobili. And it went up +8.88 if Duncan was not on and you added Ginobili. In 2005 specifically, these numbers were +9.85 and +21.18 respectively. In a sense, one could argue that these numbers are conceptually similar, in that Ginobili had a noticeably bigger effect if Duncan wasn’t on the court, but it was a crazy amount of impact going down to a still-massive amount, rather than a lot of impact going down to essentially zero. Seems pretty different to me. And it’s something that I think reflects well on both Duncan and Ginobili.

I think it reflects that the spurs put 2 well fitting teammates together, nothing more. Different franchise situations are different. I don't think we should give people fortunate enough to be in good franchise situations even more bonus points for being in those situations.

[lessthanjake]

If we look at Duncan and Ginobili instead, here’s what we see from 2005-2007: If Duncan was on the court, the net rating went up +8.15 if you added Ginobili. If Duncan was not on the court, the net rating went up +16.82 if you added Ginobili. If we looked at the full 2003-2016 sample together instead, if Duncan was on the court, the net rating went up +4.64 if you added Ginobili. And it went up +8.88 if Duncan was not on and you added Ginobili. In 2005 specifically, these numbers were +9.85 and +21.18 respectively. In a sense, one could argue that these numbers are conceptually similar, in that Ginobili had a noticeably bigger effect if Duncan wasn’t on the court, but it was a crazy amount of impact going down to a still-massive amount, rather than a lot of impact going down to essentially zero. Seems pretty different to me. And it’s something that I think reflects well on both Duncan and Ginobili.

I think it reflects that the spurs put 2 well fitting teammates together, nothing more. Different franchise situations are different. I don't think we should give people fortunate enough to be in good franchise situations even more bonus points for being in those situations.

How do you parse between someone being in a better “franchise situation” and someone just being a better player? It seems like one could always just argue for an inferior player by saying the superior guy just was in a better franchise situation, and then handwave away the evidence of the superior player’s impact by saying it’s a product of the better situation. And that seems especially dubious here given all the data I’ve provided about how well the Spurs did with Ginobili on and the other Spurs stars off the court, even against opposing starters or at a starter disadvantage. That certainly seems like something more than just “the spurs put[ting] 2 well fitting teammates together.” I also do think that Ginobili’s success in the Olympics also provides a reason to be very skeptical of Ginobili just looking great due to the Spurs being a great situation.

[jalengreen]

In two-year RAPM, Draymond peaks at +8.3 in 2015-2016, 2nd behind LeBron James (+8.6). Manu Ginobili peaks at +7.1 in 2005-2006, 1st.

For three-year RAPM, it's Draymond at +7.7 in 2015-2017, 3rd behind LeBron James (+9.9) and Stephen Curry (+8.1). Manu at +8.0 in 2005-2007 which ranks 1st.

Manu's xRAPM | EPM (EW):

2005: +6.4 | +6.4 (+14.2)
2006: +5.6 | +5.7 (+10.7)
2007: +5.8 | +6.1 (+12.9)

Draymond's xRAPM | EPM (EW):

2015: +5.5 | +4.3 (+12.4)
2016: +7.2 | +5.6 (+16.6)
2017: +6.8 | +4.5 (+12.8)

Regular season rate impact looks better for Draymond by xRAPM while it's better for Manu by EPM, though the minute difference puts Draymond ahead in cumulative impact by EW.

Minute difference also means that despite Manu's EPM in the 2005 playoffs (+7.3) being far ahead of Draymond's in 2016 (+5.1), Draymond is only slightly behind in EW (+5.1 vs +4.9). Both players led their teams in both metrics fwiw.

I think the peak impact profile is similar, I'd probably lean Draymond's way, though. For a project like this, I think it's quite reasonable to prefer Manu because of factors like (a) one won the championship while one didn't (not due to Draymond's lack of trying in Game 7, but still - thanks Steph), (b) Draymond lowered his team's championship odds with the suspension, and (c) 2004 olympics are a great signal for Manu. Main thing in Draymond's favor is the minutes, which is certainly substantial. Ballots get tough here no doubt

[f4p]

A couple key things for me here also include: (1) in Ginobili’s peak year, his team won the title, while that’s not true for Draymond; and (2) peak Ginobili genuinely has an argument for being the best player on a championship team, while Draymond does not have such an argument (even if we look at the years where his team did actually win the title). I definitely put significant weight on the greatness of winning a title and particularly so if a guy was his team’s best player in doing so. And, I definitely don’t think I’m alone in that. So this is a big factor weighing in Ginobili’s favor over Draymond.

Are we supposed to blame Draymond for Steph playing like crap in the finals and the other team having peak LeBron instead of having Tim Duncan against peak Chauncey billups?

[lessthanjake]

So the following is admittedly a bit divorced from a discussion of peaks, but since I’ve been discussing the Spurs net ratings, I’ll note an incredible data point I’ve noticed regarding the Beautiful Game 2014 Spurs:

The entire Ginobili/Duncan/Kawhi/Parker group on the court together had a +20.77 net rating in 228 minutes in the 2014 playoffs. Obviously not a big sample and the number was not nearly as impressive in the regular season, but it actually compares favorably to what the 2017 Warriors did in the playoffs with Steph/Durant/Draymond/Klay all on the court (+18.01 in 298 minutes). And we could expand out the sample to the 494 minutes in which Ginobili was on the court with at least one of Duncan/Kawhi/Parker. In those minutes, the Spurs had a +17.57 net rating. That’s only slightly below the +18.92 net rating that the 2017 Warriors had in the playoffs with Steph on and at least one of Durant/Draymond/Klay on. And the above is not really about farming opposing bench units. The net rating with Ginobili on and at least one of Duncan/Kawhi/Parker on against 4 or 5 opposing starters was +15.08. This is a bit below the +19.82 for Steph and the 2017 Warriors with at least one of Durant/Draymond/Klay on. Just historically dominant stuff from the Spurs with Ginobili in the 2014 playoffs. In contrast, in 286 minutes with Duncan/Kawhi/Parker all on the court without Ginobili in the 2014 playoffs, the Spurs only had a +2.25 net rating. This wasn’t peak Ginobili, but he seems to have really been the straw that stirred one of the best playoff teams ever, even when it was outside his peak years.

I wouldn’t say this has much direct value in this thread, because no one will be voting for 2014 as Ginobili’s peak, but I do find myself wondering exactly how outrageous the 2014 Spurs might’ve been in the playoffs if they’d had peak Ginobili instead of 36-year-old Ginobili. Perhaps the above is just a product of noise in small samples and they wouldn’t have been any better, but I guess I kind of feel like the 2014 Spurs were finally playing in a way that was most conducive to Ginobili’s playstyle, rather than centering around Duncan, and the results were incredible but they only did it after Ginobili was old.

[f4p]

Comparing ginobili to people like cp3 and harden and dirk feels like an insult to those guys. If Manu wants to be compared to those guys, he should have thrown a franchise on his back for a decade, played the big minutes, answered all the tough questions, had his game dissected on talk shows, and led a team deep in the playoffs. Or averaged more than 21/6/4. But muh rate statistics isn't going to cut it. 2005 Manu is the supporting actor who stole a movie one time. And maybe even made the movie when we watch it again on TV. And could steal entire scenes in any movie. But there's a reason his name isn't above the title when the studio needs the big opening weekend. Because everyone knows people aren't going to see the movie if it is. And he doesn't get clowned if the movie flops. Manu above those guys is really saying "Manu +Duncan" above those guys and that's not fair to those guys.

Being the main guy means everyone gets to see all your warts because there's no where to hide. Being Manu means being able to take a back seat any time you need if you know Parker or Duncan have it going. It's an easier role that doesn't come with the pressure of the top role.

[f4p]

If we look at Duncan and Ginobili instead, here’s what we see from 2005-2007: If Duncan was on the court, the net rating went up +8.15 if you added Ginobili. If Duncan was not on the court, the net rating went up +16.82 if you added Ginobili. If we looked at the full 2003-2016 sample together instead, if Duncan was on the court, the net rating went up +4.64 if you added Ginobili. And it went up +8.88 if Duncan was not on and you added Ginobili. In 2005 specifically, these numbers were +9.85 and +21.18 respectively. In a sense, one could argue that these numbers are conceptually similar, in that Ginobili had a noticeably bigger effect if Duncan wasn’t on the court, but it was a crazy amount of impact going down to a still-massive amount, rather than a lot of impact going down to essentially zero. Seems pretty different to me. And it’s something that I think reflects well on both Duncan and Ginobili.

I think it reflects that the spurs put 2 well fitting teammates together, nothing more. Different franchise situations are different. I don't think we should give people fortunate enough to be in good franchise situations even more bonus points for being in those situations.

How do you parse between someone being in a better “franchise situation” and someone just being a better player? It seems like one could always just argue for an inferior player by saying the superior guy just was in a better franchise situation, and then handwave away the evidence of the superior player’s impact by saying it’s a product of the better situation. .

I mean do we need to do a lot of digging to say the spurs were a good franchise situation compared to basically any other franchise? They already had a title, a GOAT PF and maybe GOAT coach before Manu ever showed up. And then added great role players, great contracts, got kawhi with a non-lotto pick, etc. Am I supposed to pretend like I can't know if that was a good situation?

[lessthanjake]

In two-year RAPM, Draymond peaks at +8.3 in 2015-2016, 2nd behind LeBron James (+8.6). Manu Ginobili peaks at +7.1 in 2005-2006, 1st.

For three-year RAPM, it's Draymond at +7.7 in 2015-2017, 3rd behind LeBron James (+9.9) and Stephen Curry (+8.1). Manu at +8.0 in 2005-2007 which ranks 1st.

Manu's xRAPM | EPM (EW):

2005: +6.4 | +6.4 (+14.2)
2006: +5.6 | +5.7 (+10.7)
2007: +5.8 | +6.1 (+12.9)

Draymond's xRAPM | EPM (EW):

2015: +5.5 | +4.3 (+12.4)
2016: +7.2 | +5.6 (+16.6)
2017: +6.8 | +4.5 (+12.8)

Regular season rate impact looks better for Draymond by xRAPM while it's better for Manu by EPM, though the minute difference puts Draymond ahead in cumulative impact by EW.

Minute difference also means that despite Manu's EPM in the 2005 playoffs (+7.3) being far ahead of Draymond's in 2016 (+5.1), Draymond is only slightly behind in EW (+5.1 vs +4.9). Both players led their teams in both metrics fwiw.

I think the peak impact profile is similar, I'd probably lean Draymond's way, though. For a project like this, I think it's quite reasonable to prefer Manu because of factors like (a) one won the championship while one didn't (not due to Draymond's lack of trying in Game 7, but still - thanks Steph), (b) Draymond lowered his team's championship odds with the suspension, and (c) 2004 olympics are a great signal for Manu. Main thing in Draymond's favor is the minutes, which is certainly substantial. Ballots get tough here no doubt

All this seems pretty reasonable to me. I will say that the factors you list at the bottom of your post are certainly a much bigger factor here for me than parsing through impact data and deciding I think Ginobili looks a bit better. I do look at the panoply of impact data and impact-correlated box data and conclude that I think peak Ginobili’s data looks better on balance than peak Draymond’s, but as I noted in my prior post I don’t think Ginobili is light years ahead. One could even potentially look at the difference in minutes and decide that they’re similar (or even that Draymond looks better at that point). My view that 2005 Ginobili was solidly greater than 2016 Draymond hinges much more on the fact that peak Ginobili won a title and I think he was his team’s best player that year, while peak Draymond did not win a title and was never his team’s best player. The fact that Manu has that 2004 Olympics signal is also a minor factor, and the point you mentioned of Draymond missing a Finals game also has to be a factor. I dinged 2021 Giannis a fair bit for missing two conference finals games, and I think 2016 Draymond missing a Finals game is a similarly serious issue (particularly when the game Draymond missed was much higher leverage—a later round, against a tougher opponent, in a closer series).

One other thing I will say is that I just really don’t like xRAPM. I’ve cited it occasionally when listing data, but Engelmann using a prior that is previous years’ RAPM basically makes for some very weird results that seem to have lagged effects while he’s purporting to report things out for specific years. Like, if a measure is telling me that LeBron was better in 2011 than 2009 (which xRAPM does), then I think there’s a problem. The problem here often comes when there’s a significant change in a player’s quality year to year. The measure just does a really bad job at picking up on that. And I definitely think 2005 Ginobili is a good example of that. Draymond improved quickly too, but peak Draymond (i.e. 2016) at least had a very good runway of 2015 to improve his numbers (i.e. since 2015 Draymond was better than 2004 Ginobili, this has a big effect on the data that Engelmann is reporting out for 2016 Draymond vs. 2005 Ginobili). I talk about this issue a lot with Engelmann’s stuff, but it’s a pretty persistent issue that I think creates problems quite a lot with that measure. There’s not a lot riding on that particular issue here, since peak Draymond is higher than Ginobili but it’s pretty close regardless, so it doesn’t move the needle a whole lot. But I just feel the need to raise this issue again, as I very frequently do when people cite to xRAPM data that seems to implicate this major issue with the measure.

A couple key things for me here also include: (1) in Ginobili’s peak year, his team won the title, while that’s not true for Draymond; and (2) peak Ginobili genuinely has an argument for being the best player on a championship team, while Draymond does not have such an argument (even if we look at the years where his team did actually win the title). I definitely put significant weight on the greatness of winning a title and particularly so if a guy was his team’s best player in doing so. And, I definitely don’t think I’m alone in that. So this is a big factor weighing in Ginobili’s favor over Draymond.

Are we supposed to blame Draymond for Steph playing like crap in the finals and the other team having peak LeBron instead of having Tim Duncan against peak Chauncey billups?

This seems like an odd way to portray this, given that the 2005 Spurs barely won the Finals against the defending champs, in a series where Ginobili played incredibly well. The degree of difficulty for the Spurs was actually really high, and they very likely don’t win the series if Ginobili wasn’t amazing in the series. Meanwhile, Draymond was great in the 2016 Finals, but he also didn’t even play every game in the series. It’s really not so clear to me that the degree of difficulty was actually higher for Draymond to get the Warriors the win in that series.

But also, yeah, winning the title matters in an assessment of the greatness of a player’s year, even though different players obviously don’t have the same teammates or face the same opponents. The actual achievements a player and his team had in a year matters for purposes of how great their year was.

Comparing ginobili to people like cp3 and harden and dirk feels like an insult to those guys. If Manu wants to be compared to those guys, he should have thrown a franchise on his back for a decade, played the big minutes, answered all the tough questions, had his game dissected on talk shows, and led a team deep in the playoffs. Or averaged more than 21/6/4. But muh rate statistics isn't going to cut it. 2005 Manu is the supporting actor who stole a movie one time. And maybe even made the movie when we watch it again on TV. And could steal entire scenes in any movie. But there's a reason his name isn't above the title when the studio needs the big opening weekend. Because everyone knows people aren't going to see the movie if it is. And he doesn't get clowned if the movie flops. Manu above those guys is really saying "Manu +Duncan" above those guys and that's not fair to those guys.

Being the main guy means everyone gets to see all your warts because there's no where to hide. Being Manu means being able to take a back seat any time you need if you know Parker or Duncan have it going. It's an easier role that doesn't come with the pressure of the top role.

So I don’t actually entirely disagree with this (though I think a lot of the precise wording goes too far). It’s basically why I intend to vote for someone like 2009 Kobe and 2011 Dirk over Ginobili, even though I’m not entirely certain I think those guys were better players than peak Ginobili. The issue is that the guys beyond that all fall into one of a few buckets: (1) great players who unquestionably led their team but didn’t actually win a title with their team (and, in most cases, didn’t even make the Finals); (2) great players who won a title but, like Ginobili, were not unquestionably their team’s best player; or (3) great players that won a title and were their team’s best player but who really didn’t perform well at all in the playoffs (I’m basically thinking of 2024 Tatum for this category). It’s really not clear to me that players in one bucket must be above players in another bucket. You seem to think category #1 is simply above category #2, and I just don’t agree. Ginobili falls into category #2 and actually is probably the one who has the best case for being his team’s best player on a title-winning team. So he seems like a premier option here to me. The other guys I’m most considering in this thread are basically the guys I think were the best out of the category #1 players (Nash, Chris Paul, Harden).

[lessthanjake]

I think it reflects that the spurs put 2 well fitting teammates together, nothing more. Different franchise situations are different. I don't think we should give people fortunate enough to be in good franchise situations even more bonus points for being in those situations.

How do you parse between someone being in a better “franchise situation” and someone just being a better player? It seems like one could always just argue for an inferior player by saying the superior guy just was in a better franchise situation, and then handwave away the evidence of the superior player’s impact by saying it’s a product of the better situation. .

I mean do we need to do a lot of digging to say the spurs were a good franchise situation compared to basically any other franchise? They already had a title, a GOAT PF and maybe GOAT coach before Manu ever showed up. And then added great role players, great contracts, got kawhi with a non-lotto pick, etc. Am I supposed to pretend like I can't know if that was a good situation?

I don’t think anyone would deny that the Spurs were a good situation. But you said that “the spurs put 2 well fitting teammates together, nothing more.” There’s definitely a lot more to what is going on there than that, given that the Spurs also did great with Ginobili on and the Spurs’ other best players off and that that result holds true even in the most difficult starter states we can look at. And, in any event, the Spurs being a good situation is really not mutually exclusive with Ginobili being a better player than someone else, and you’re acting like it is. Even if you think Ginobili was in a great situation, it is very possible for the better player to also be in a better situation. Your position seems to basically just be that if you think a player was in a better situation, then all evidence that he was also a better player must be ignored because it’s just a product of situation. That seems like it’s clearly not a good approach. And, again, it feels like a particularly bad approach when talking about a guy who did really well against opposing starters without his team’s stars with him (at that point, it’s basically impossible to really keep saying his situation was good), and who also led his country to an Olympic gold medal despite not being American.