Yes, you are absolutely right.Ok got it - thanks - that does not mean the very last people getting interviewed next year would be much higher than this year, right? The curves I showed in the link were the progression of the current numbers each year. Nothing to do with open or hidden numbers, don't you think?
However, those max numbers do not make a lot of sense. For instance, DV-12 numbers for EU included cutoff 40000 at some point while the max number was only 32000. So, making final cutoff 32000 or 33000 or 35000 or 40000 before current does not allow you to distinguish anything.
In DV-13 we have EU numbers up to 30532, but we saw cutoff 31000 and then 33000 which is even higher, without CEAC data moving up. We need the data for true max number, not for cutoff movement. However, we do not have that data. Only this year I have that data (DV-13), in DV-12 and DV-11 I had this data only for Europe from one of Russian forums that has a lot of EU winners.
Now, I have full underlying data only for years 2007 through 2013. The graph mentions 2006 as an off year. I do not know what happened that year, but I know that years since 2001 until 2007 were years when facial recognition technology was improving gradually. That could trigger 2006 to be an off year. Another thing is regional quota. I do not have reliable sources for information about regional quotas for all years, only for some of them. However, some unreliable sources I have mention EU quota dropped in 2006 compared to 2005 (I have an unreliable source saying in 2005 it was 17248 and in 2006 it was 14875).
Facial recognition technology is stable now. Yes, regional quota is a variable we do not know anything about. Changes in regional quotas could change the picture. However, in dv-14 ALL regions increase max numbers. And the sum of regional quotas did not change - world quota is the same.