GeorgeFloydOverdosed said: I uploaded a genie scout ratings file for FM26 on page 3 of the thread, which won't be as precise as I've just adjusted it roughly according to HarvestGreen22's FM26 data. But in general, while there are some differences, they're relatively minor. Expand
Not double-checking everything I'm about to say right now, but I would surmise it as two main differences:
1. Orion's data is based on correlations with in-game ratings, whereas I'm focused on actual performance (win rate and goal difference). There appears to be a discrepancy between these two things because ratings overvalue technicals/mentals and devalue physicals, even where the physical players win the games - I haven't derived this from my own testing, it's from a youtuber who did an experiment showing this. I already knew there was something of a mismatch, this would explain why.
2. I weight all attributes, not just a few of them, and try to be as precise as possible taking into account all considerations. This wouldn't matter a huge deal, except that actually certain attributes such as the personality attribute 'pressure' do in fact matter a lot - this comes from Orion's own findings, that the difference between pressure 1 and 10 is something like a 40% performance difference, which is huge.
Yes I'll be updating the main post at some stage. I'm quite busy with life right now Expand
I've been following your test for a month now. I even registered an account just to follow your test and look forward to the final selection criteria. I also want to confirm if FM26 is as effective? Finally, I wish you a Happy New Year! I'm from Macau.
Thank you for your reply and test !!
Not double-checking everything I'm about to say right now, but I would surmise it as two main differences:
1. Orion's data is based on correlations with in-game ratings, whereas I'm focused on actual performance (win rate and goal difference). There appears to be a discrepancy between these two things because ratings overvalue technicals/mentals and devalue physicals, even where the physical players win the games - I haven't derived this from my own testing, it's from a youtuber who did an experiment showing this. I already knew there was something of a mismatch, this would explain why.
2. I weight all attributes, not just a few of them, and try to be as precise as possible taking into account all considerations. This wouldn't matter a huge deal, except that actually certain attributes such as the personality attribute 'pressure' do in fact matter a lot - this comes from Orion's own findings, that the difference between pressure 1 and 10 is something like a 40% performance difference, which is huge.
Yes I'll be updating the main post at some stage. I'm quite busy with life right now
I've been following your test for a month now. I even registered an account just to follow your test and look forward to the final selection criteria. I also want to confirm if FM26 is as effective? Finally, I wish you a Happy New Year! I'm from Macau.