lessthanjake wrote:So I’m strongly considering voting for 2005 Ginobili in this thread. The more I look into Ginobili’s on-off data, the more impressive it is. Below is a string of data regarding Ginobili, which I think also serves to refute some potential things one might think of to downplay his great numbers:
Ginobili & Duncan had all-time net ratings together in 20051. In the 2005 season, Ginobili and Duncan had a +20.38 net rating in 1380 minutes on the court together. This is a crazy number. For reference, it is higher than the +19.89 net rating that Curry/Durant/Draymond had on the court together in the 2017 season, on the GOAT team. It is also higher than the +19.27 that Curry and Draymond had on the court together in the record-breaking 2016 season. I’m not aware of any star duo that actually had a higher net rating together in a season.
Ginobili & Duncan’s net ratings were incredible over larger samples too2. Okay, but is that +20.38 net rating in 2005 just a product of a small sample size? Well, maybe to a small degree. But the Spurs had a still-massive +16.87 net rating with Duncan and Ginobili together in the three years from 2005-2007. That’s lower, but it’s still a massive number. And there’s reason to believe that 2005 being higher than the overall three-year span isn’t just noise, since Duncan fell off a bit as a player after 2005, while Ginobili had improved in 2005, so 2005 is really the one intersection of them both in the heart of their prime. Meanwhile, in their entire time together from 2003-2016, Ginobili and Duncan had a +12.18 net rating together. For reference, that’s really similar to the +12.56 net rating that Steph and Draymond have together (which will probably end up lower than the Ginobili/Duncan number when all is said and done—and obviously they also played with Durant for years).
Ginobili’s net ratings were fantastic even without Duncan3. Okay, but that’s Ginobili *and* Duncan, and Duncan is a great player, so maybe that’s why the numbers are so good? Well, that’s part of it, of course. But the Spurs were really good in Ginobili’s minutes without Duncan. From 2005-2007, the Spurs were +8.52 when Ginobili was on and Duncan was off. And it was +8.10 in the playoffs in those years. In the 2005 regular season specifically, it was +9.88 with Ginobili on and Duncan off, and +10.27 in those minutes in the playoffs. Over the course of 2003-2016, it was +7.99 in the regular season and +5.69 in the playoffs. So yeah, obviously Duncan made them better, but Ginobili had the Spurs doing great even with Duncan off the court. It was particularly true in 2005, but was also just true across giant samples.
Ginobili’s non-Duncan net ratings were fantastic even against opposing starters4. Okay, but Ginobili often came off the bench, so maybe his minutes without Duncan were just juiced by being against opposing bench players? Turns out that’s not the case. From 2003-2016, against 4 or 5 opposing starters and with Ginobili on and Duncan off, the Spurs were +8.60 in the regular season and +5.15 in the playoffs. These are fantastic numbers to have against starter-heavy units, and are actually virtually indistinguishable from the overall Ginobili-on/Duncan-off numbers. And from 2005-2007, these numbers were +9.84 in the regular season and +18.87 in the playoffs. In 2005 specifically, it was an absurd +17.24 in the regular season and +22.14 in the playoffs (though at this point we’re looking at a small 290 minute regular season sample and 81 minute playoff sample).
Ginobili’s non-Duncan net ratings were fantastic against opposing starters even without the rest of the Spurs’ best players5. Okay, but the Spurs had other good players besides Duncan, so maybe these good net ratings in the non-Duncan minutes against opposing starters are just a product of playing with very good players like Tony Parker, Kawhi Leonard, and LaMarcus Aldridge? Again, we have strong indication that that’s not the case. If we look at 2003-2016, Ginobili played 587 regular season minutes and 46 playoff minutes against 4 or 5 opposing starters and without any of Duncan, Parker, Kawhi, or Aldridge on the court . The Spurs had a +7.03 net rating in those regular season minutes and a +18.79 net rating in the playoff minutes. Combining regular season and playoffs, that comes out to a +7.89 net rating.
It’s worth noting that this compares really favorably with the net ratings against starter-heavy units that top-tier all-time greats and others we might consider in this thread put up without their star teammates. Across RS+playoffs, LeBron had a +4.16 net rating against starter-heavy units in Miami without Wade or Bosh, and a -3.75 net rating on the second-stint Cleveland team without Kyrie or Love. Jokic from 2019-2025 had a +3.74 net rating against starter-heavy units without Murray. Steph had a +0.13 net rating from 2014-2023 without Draymond, Durant, or Klay. From 2001-2004 and 2008-2011 without Shaq or Gasol in these minutes, Kobe had a +1.70 net rating. For Durant, it was +0.36 in 2010-2016 without Westbrook and +2.55 from 2017-2019 without Steph, Draymond, or Klay. From 2012-2017, without Griffin or DeAndre Jordan, it was +2.72 for Chris Paul. From 2005-2010, without Amare or Marion, it was +5.67 for Nash. From 2015-2021 in Houston without Dwight, CP3, or Westbrook, Harden had a +1.62 net rating. And probably most interestingly, Duncan from 2003-2016 without Ginobili, Parker, Kawhi, or Aldridge had a -3.08 net rating. Granted, the sample sizes for a lot of these aren’t super high, but I do think the fact that Ginobili leads the way here highlights just how good the Spurs were with Ginobili as the main guy against the best the opponents had to offer.
Ginobili’s net ratings were fantastic when the Spurs had fewer starters on the floor than the opposing team6. Okay, but that’s just specific starter states against starter-heavy opposition and the sample sizes aren’t enormous. Maybe Ginobili still farmed a lot of his great net ratings just from the Spurs farming bench units with more of their own starters. Well, it turns out that is definitely not the case, with the Spurs doing incredibly well with Ginobili on the court and the Spurs with fewer starters on the court than the other team. I wrote about this at length in the following thread:
https://forums.realgm.com/boards/viewtopic.php?t=2386462. I won’t rehash things that are in that thread too much. But some highlights from that thread: From 2003-2017, the Spurs had a net rating of +9.40 at a starter disadvantage with Ginobili on the floor. It is +8.27 if we treat Ginobili as a starter for these purposes even if he came off the bench. That number was slightly higher than LeBron 2009-2020 at a starter disadvantage, as well as Giannis 2019-2024 and Jokic 2020-2024. Perhaps most pertinently, it is also above what the Spurs did from 2003-2016 at a starter disadvantage with Duncan on the floor.
And even if we narrowed this down to just Ginobili’s minutes at a starter disadvantage with Duncan, Parker, Robinson, and 2014-onwards Kawhi all off the court, the Spurs had a +6.00 net rating from 2003-2017 (note: I realize writing this post that there’s slightly different players being controlled for than what I did above for other data, but I’m just not bothering to re-run numbers to make the controlled-for players exactly the same—especially since PBPstats crashes a lot when I run these numbers, so it takes a long time to pull). I ran these numbers for Steph and LeBron at a starter disadvantage without their co-stars, and Ginobili’s numbers compare favorably. He also compares favorably with Giannis and Jokic if we go back far enough (which seems fair when comparing to 2003-2017 Ginobili). Most pertinently again, he compares favorably to Duncan himself here, who had a +2.64 net rating from 2003-2016 without Ginobili and all those same guys. Even if we add 2001 and 2002 to the mix, Duncan is still slightly behind. So basically, at a starter disadvantage without co-stars, Ginobili genuinely looks better than absolute high-end guys already voted in in this project.
The above-discussed numbers are better for Ginobili than Duncan and Ginobili had huge success outside the Spurs, so it’s not just a supporting-cast thing7. Okay, but maybe all this is just a product of the Spurs having a great supporting cast? Well, to some degree that is surely the case. The Spurs did generally do a good job getting good role players. But, as detailed above, the Spurs did better against starter-heavy units and better at a starter disadvantage with Ginobili on and Duncan off than with Duncan on and Ginobili off. And Ginobili’s numbers look better even in more general terms. From 2005-2007 across regular season + playoffs, the Spurs had a higher net rating with Ginobili on and Duncan off (+8.44) than with Duncan on and Ginobili off (+6.66). In 2005 specifically, this was even more pronounced, with +9.94 with Ginobili on and Duncan off and +5.46 with Duncan on and Ginobili off. Meanwhile, over the course of 2003-2016, the Spurs did slightly better in the regular season with Ginobili on and Duncan off than vice versa (+7.99 vs. +7.54) and a lot better in the playoffs (+5.69 vs. -3.77).
I’d also just note as a quick point that Ginobili being amazing in that era definitely wasn’t contingent on Duncan or the Spurs supporting cast when he led Argentina to the 2004 Olympic gold medal. It’s not NBA basketball, but definitely feels like relevant context.
____________
ConclusionSo yeah, I think it is really interesting to see that 2005 Ginobili was both a part of possibly the most effective duo of stars ever (i.e. the +20.38 net rating with Duncan), while also functioning historically well on his own, even when against opposing starters or with the team at a starter disadvantage. It suggests an incredible level of both ceiling raising and floor raising.
Of course, one response to all of the above is that it’s granular data that all is just part of what goes into RAPM. I do think there’s value in granularity, though. And regardless, it’s not like there’s some negative nuance that RAPM is seeing that this data isn’t. According to NBArapm, Ginobili was #1 in the NBA in RAPM from 2005-2007 (not to mention being repeatedly top 5 in the NBA across a load of different timeframes in his career). He was #1 in TheBasketballDatabase’s three-year RAPM in 2005-2007 as well. The only guys who are in consideration here that have RAPM (whether peak or just overall) as good as Ginobili are Chris Paul and Steve Nash (and with Nash it’s only really a peak thing—Ginobili looks better at a longer-term level). He was also #1 in the NBA in 2005 in EPM—which is probably the best impact-box hybrid measure we have from that era.
Finally, talking purely of impact can obscure the importance of performance in specific high-leverage games—particularly the playoffs. And that is an area where 2005 Ginobili really shines as well. I think there’s a good argument that Ginobili was the best player in each of the Spurs playoff series that year. On paper, the one that’s probably most arguable for that not being the case is the Phoenix series, but I’ll just say that as a huge fan of the Suns in that era I came out of that series having a massive hatred for Ginobili, basically because I thought he’d been their best player in the series. And in the finals, against an incredible Pistons defense, Ginobili was just amazing, putting up a +13.2 rTS% in an incredibly defensive series. To put some additional data on this, the only people who have had a higher playoff EPM than Ginobili had in 2005 are players that have already been voted in. Given the larger-sample impact numbers and the playoff performances, I think there’s actually a pretty good argument that Ginobili was just straight up the best player on the NBA champion 2005 Spurs. Which feels like it would definitely deserve to make the ballot at this point.
