This cannot be ruled out.
One of my calculation methods gives the date May 28th (not too far from yours). But I am not yet ready to declare this as a prediction. Let’s see.
I can only more or less definitely say that we will pass halfway on April 15 in the evening. And it’s a day later than I thought before.
691 741 cards to go 
Approximately 36% of the way has already been completed.
I’m sticking with my last prediction for now:
67 mio – 23 May, 2022 at 00:00 (UTC) (midnight from 22nd to 23rd of May)
And we’ll cross half way on April 15, 2022 at 00:00 (midnight from 14 to 15 of April) I guess.
Assuming that the fact exactly matches my forecast (which is not at all guaranteed), then the preliminary rating of the diviners will look like this:
@anon141542 90,09
@Stevyy 85,50
@greenskull 35,29
@Anibaby 30,68
@reiselustig 25,08
@Aguaroble 18,47
@Vozmojnosti_Iriski 18,18
@thibcabe 16,29
@Florallle 13,33
@nnniiina 11,26
@bom_dia 7,17
@Jarana 5,97
@shootingstar7 4,18
@ChocCrossie 3,27
I’ll make a reservation right away that I can be deeply mistaken in my forecasts, therefore this assessment is hypothetical.
I always make the list 
Thank you @greenskull for the details.
As Stevyy and I have such higher numbers next to our names (85, 90, respectively) compared to the majority of you, is that a good thing or not?
It is very good 
But nothing can be 100% sure.
My first prediction (May 19, 2022 at 00:00 UTC) was somewhat impulsive and not well thought out.
The second prediction (May 23, 2022 at 00:00 UTC) is based on a more thorough and deeper analysis. It is based on a hypothesis that is increasingly confirmed by the current trend.
I like Stefan’s predictions, as he has been participating in these countdowns for a long time. And his predictions often come true well. It has happened many times that the fact has fallen somewhere between Stefan’s prediction and mine.
And in this case, my second prediction (May 23, 2022 at 00:00 UTC) and Stefan’s first (and so far only) prediction (May 22 at 05:22 UTC) roughly coincide. It can be assumed that the fact will not be too far from the average point near both of us, although of course this is not guaranteed.
I have detailed the rating calculation method here.
And I made this preliminary calculation of ratings on the assumption that the fact will coincide with my second prediction.
My rating is greatly inferior to yours due to my first frivolous prediction, since all attempts are taken into account in the calculation.
But given the fact that game activity is affected by many factors, including covid, geopolitical events, local holidays, changes in mail forwarding prices in a certain country, etc., it is quite difficult to accurately predict when the next level will be completed.
At least it’s hard to predict from a long distance (let’s say 2.5 months), but a little easier as you get closer to the level passing.
The rating calculation method takes into account both the prediction error and the distance from which the prediction was made. And also all attempts for each countdown participant are taken into account.
PS I forgot to say that I usually send a prize postcard with a Magic Mandala that grants one wish to the diviner whose rating is the highest soon after level reaching
A question, the answer to which is not at all obvious.
Will Postcrossing overcome the level of 70 million registered postcards before the end of this 2022?
I see serious intrigue in this.
Try to guess pls
Thank you @greenskull for your reply, it made for some interesting reading. I will admit that I didn’t fully understand everything but I did get the basis of it. My mind just doesn’t work that way.
With regards to the passing of 70,000,000 this year, my initial thought is NO and we will pass it somewhere around 28th Jan to 2nd Feb 2023. That answer is based on absolutely nothing but my gut feeling.
I think that we will pass halfway (66,500,000) on April 15 at about noon.
As soon as we pass halfway, I will correct my prediction for the level of 67 million. I guess I will have to add 1-2 days to my previous forecast
Half-way there. Registrations are down a bit. I guess folks in usually active countries have different things in mind - like surviving a brutal invasion, or ignoring the wrong direction their ‘president’ is leading them.
Sometimes there’s truth in old cliches. There can be no real peace without justice. And without resistance there will be no justice.
Arundhati Roy
489,620 to go
25.05.2022
05:25 UTC
a card from PA to IN
Registrations down a bit yes. But I think it’s also because of Ramadan, Pesach and Easter. They are all important holidays for the religions. Here in Germany yesterday (Friday) were a holiday and on Monday will be a holiday. So no post will be delivered or sorted. And here are two weeks if school holiday. So Post can take longer than usual
Deviation from the expected trend began to appear from April 5…9, gradually gaining strength.
At the moment it is -15,000 postcards or -2.27%. This amount and percentage of the amount of registered postcards differs from my estimation.
Pink – the estimated trend.
Blue – the fact.
Purple – a hypothetical boundary towards which the estimated trend tends.
April 5…9 is (x=6122…6126).
I believe that the reason for this deviation is still a seasonal factor associated with the Easter holidays.
I assume that this gap may intensify for a few more days. This is connected with the upcoming Easter holidays in a number of Orthodox countries and also with the long May holidays. But I assume that soon after May 10 and up to May 20, Postcrossing will compensate this ~3% gap.
And I think that by the second half of May, when the level of 67 million is expected to be crossed, this gap will be compensated.
So for now I will leave my previous forecast unchanged – 23 May, 2022 at 00:00 (UTC).
But I will watch the trend distortion closely. When it starts to shrink, I can more or less accurately understand whether it will close completely by May 23.
The music is just for fun. This track is playing on my headphones right now.
Assuming we pass 67 million exactly as Stefan predicts, then the Diviners Ranking would be as follows:
@reiselustig 124,59
@anon141542 50,47
@Stevyy 45,70
@Aguaroble 45,65
@greenskull 16,77
@Anibaby 15,64
@Vozmojnosti_Iriski 11,50
@thibcabe 10,74
@Florallle 9,67
@nnniiina 8,56
@Jarana 8,10
@bom_dia 6,03
@shootingstar7 4,98
@ChocCrossie 3,76
But of course, this is just a guess
I don’t think I usually change my prediction but this time I’m going to, probably too far the other way from my original one, but this time to 10:25 UTC on 25th May.
Looks like I’m not a good fortune-teller 
BUT - what if, against all other predictions, finally it will be June, 13th?! 
… so I’ll stay with it 
Time will tell
Some beautiful graphs with explanations. Hope it will be interesting.
This is a trend deviation graph.
I want to please everyone.
It seems that on April 20 (x=6137), the rebound and a compensatory trend reversal began.
Blue is a trend fact.
Light blue is a tube of normal trend volatility.
We see that the intensive trend distortion began around April 05 (x=6122). It is one and half week before Easter (!). And the fall even went beyond Easter for a couple of days.
Green is the rebound and reversal area.
Let’s see how many days it will take to catch up and return to the tube of the trend.
And some inspirational music from my playlist, if you don’t mind.
Darkest night before dawn. I believe in the best.
We see that in a week the Easter gap has already closed by a third.
If it goes on like this, we will be fully back to the trend by the beginning of the second week of May (x=6153 is May 06).
Therefore, I will leave my previous prediction in force.
Well, maybe Stefan is right about 25.05.2022
I can definitely say that we will pass the 67 million level somewhere between Stefan’s prediction and mine.
For now, I will observe a little more, and then I will correct my prediction.
