Most of the original (2013) debate took place on these pages viewtopic.php?f=4&t=3994. The formula was reviewed in 2017 and 2019. Not only was the relationship unstable over time, but also dependant on the sample selection. For the purposes of deciding the initial conversion we did want ECF to be equivalent to FIDE levels. The decision was based on the January 2020 lists using adult ratings. The decision was that the old formula was OK. It's not clear whether that decision or the subsequent changes over the following 6 months gave us the situation that Wadih outlines. The original conversion had the added complication of a slope and constant, 2 parameter, fit which now should be just one- the difference (but it isn't and that involves a long story, too long to bring up here).NickFaulks wrote: ↑Mon Apr 17, 2023 11:48 amI wasn't sure of that, but if it really was the case that the old formula was out of date, would that not have been the ideal moment to update it?Roger de Coverly wrote: ↑Mon Apr 17, 2023 11:08 amThey were converted using the formula 7.5* ECF + 700. That was correct by least squares fitting to adults when first devised but had already become out of date by 2019.
I am expecting to review the system annually and the 2013 review is herehttps://www.ecfrating.org.uk/doc/.
