I've just stumbled across this classic in The Guardian from 2018:
Thirty years ago, James Hansen* testified to Congress about the dangers of human-caused climate change. In his testimony, Hansen showed the results of his 1988 study using a climate model to project future global warming under three possible scenarios, ranging from ‘business as usual’ heavy pollution in his Scenario A to ‘draconian emissions cuts’ in Scenario C, with a moderate Scenario B in between.
Changes in the human effects that influence Earth’s global energy imbalance (a.k.a. ‘anthropogenic radiative forcings’) have in reality been closest to Hansen’s Scenario B, but about 20–30% weaker thanks to the success of the Montreal Protocol in phasing out chlorofluorocarbons (CFCs)**.
Hansen’s climate model projected that under Scenario B, global surface air temperatures would warm about 0.84°C between 1988 and 2017. But with a global energy imbalance 20–30% lower, it would have predicted a global surface warming closer to 0.6–0.7°C by this year. The actual 1988–2017 temperature increase was about 0.6°C. Hansen’s 1988 global climate model was almost spot-on.
Classic lack of self-awareness there. The outcome was somewhere vaguely near his mid-estimate, and presumably miles off his lower- and upper-estimates. That's hardly "spot-on", is it?
** The Montreal Protocol was indeed successful in reducing CFC production, hooray. But the pre-existing CFCs are still working their way through the system and are still causing most or all of the current slight warming, reducing the impact of CO2 down to zero.
* James Hansen is the one who came up with the trick of calculating the effective temperature of Earth assuming two-thirds cloud cover, and then contrasting that (low) temperature with with the actual temperature of land/ocean surface (warmer) and ignoring the actual temperature of clouds (colder). See top of third column on page 1 here. He dresses it up in scientific language to cover his traces.
If you include the actual temperature of clouds when calculating the overall actual temperature, it would bring it down to pretty much the same as the effective temperature, i.e. if he had done an honest comparison, like-with-like, he would have had no evidence for a CO2-caused Greenhouse Effect whatsover.