In the second plot you see that the estimated URR has increased a lot in recent years, and it will continue to do so for some time. Only when we see that the points stabilize at an estimated URR value we can have some confidence. Like in the US case.

The parabola method is not good at all at predicting URR before peak year. And we are not past peak year.

The strength of the method IMO is at predicting URR after peak year, and many areas of the world are in this situation. If you go to Graphoilogy you will see that there are 20 countries where the estimated URR has stabilized. There are 21 bad cases like the world case, and there are 9 where it is too early to say.

I don't really see what we learn from this afterall then???
Well, for each region I have made two plots. The first one gives you an estimated URR and the second one gives a rough idea of how reliable it is. For the world case the second plot tells you that URR=1663GB is very unreliable. So that's what we learn. It is important to know when you shouldn't trust your forecasts and when you should.

Of the four plots I have posted, the estimated US URR has stabilized for years, maybe increasing slightly. This tells you that the estimate is quite robust. Mexico has just stabilized at peak year. South Arabia is a very bad case, the estimate is highly unreliable. UK has stabilized recently.

My next project will be to compare HL with this method. I think that we rely to much in the HL, and it is important to have an alternative to compare.