How many elo points are lost starting chess late in life

 I read a study that examined the average age that chess masters started playing chess and it included some other statistics such as "54 percent of those who started playing chess AT the age of 14 reached a rating of 2000 and up. 75 percent of those who started playing chess at the age of 9-12 reached a rating of 2000 and up." I couldn't find the exact data from the study but why didn't they give an exact age for the second statistic as well as the percentiles. If they had said 50 percent of those who started at 9 reached a rating of 2100 and 50 percent of those who started at 14 reached a rating of 2000, then we would have a pretty good idea of the average elo loss, but they stated their findings in a way that makes it complicated. I was wondering if there was anyone out there who could solve this mystery, as I have yet to see anyone talk about it and I think there should be an answer with all the data we have available from the USCF and FIDE databases. 

Those numbers can't be real. Only a small percentage of people who start playing chess ever make it to 2000 FIDE, like 1%, climbing higher for people who take it more seriously. Maybe they mean people who started playing actively including tournament games at those ages, that makes a lot more sense. Plenty of people learned the rules of chess at 9 and not many end up with 2100.

Starting young is, of course, better. Look at GMs and World Champions.

learning rate differs between ages, so i wouldn't be surprised if younger people face more success starting in chess than older people

Yes the participants in the study were all people who took the game seriously, playing everyday in clubs and doing tournaments from the age they learned the rules. So if you were to play for 3 hours a day and do tournaments from the ages of 9-12 you'd have a 75 percent chance of reaching 2000. I also want to talk about Mihai Suba who is the latest starting grandmaster ever. He learned the rules of chess at age 19 and reached a peak rating of 2580 in 1986 at age 40 and was 2500 on average during most of his career. This was all before rating inflation came into effect. In 1985, the 100th ranked player in the world was rated 2500. In 2022, the 100th ranked player in the world was 2650. This is from a chart that showed the average rating of the 1st, 2nd, 5th,10th,20th,50th, and 100th spots on the rating list. For all spots, there is a increase of 150 points in rating over the years starting in 1985. This would mean that if Suba played today, his average rating would be 2650! Assuming that he has the same genetics as possibly the greatest player in the world Magnus Carlsen, and that he started playing chess about 10 years after the age Carlsen learned, that would mean he lost bout 200 points in elo. So we could say there may be a 200 loss in elo for every 10 years missed starting after age 9.

Without the exact data, it's challenging to determine the specific percentiles and ages involved.

but we do lmao