enchevetrement
Registered Users (C)
Hi raevsky,
Can you please provide more information about where the number 61K came from? Is there any official source that mentions it? how did you come up with it?
Tnanks!
Immigrant NumberVisa Issuances in the Diversity Immigrant
- Thread starter Princedriss
- Start date
Hi raevsky,
Can you please provide more information about where the number 61K came from? Is there any official source that mentions it? how did you come up with it?
Tnanks!
As in the other thread, thank you for that
So to me it sounds like an error. They forgot to tell that beforehands, and realized that later. That is why they informed about that in DV-2013 instructions and will inform in instructions for future lotteries. And they probably would do second batch anyway, but because of the court there was a lot of attention around DV-2012. So they chose to loose the program but to refrain from second batches in DV-2012.
No.
However, what we know from DV-2012 is the following:
1. People with numbers up to EU32K got notifications (I am talking about DV-2012, not 2013)
2. VB listed cut offs in April as 32,000 for EU, but in May for 40000. In June they became current. That means there were numbers above EU40,000
3. Ankara consular schedule website listed a misterious number EU45,xxx.
4. There are no other numbers from EU above 32,xxx that I know of except this one.
So, official sources clearly mention there were cases above EU32K, but they never mentioned the maximum numbers per continent.
Sure. Easily. I'll use DV-2012 data for that.
Everybody knows that probability to win does not depend on the country, it depends only on the region.
Let's consider two countries, Iran and Bangladesh, both from Asia.
There were 289,586 entries from Iran that produced 4,453 wins. Frequency of wins is 1.538%
There were 7,667,030 entries from Bangladesh that produced 2,373 wins. Frequency of wins is 0.031%. The second number is 50 times smaller than the first one.
Do you have an explanation why that happens even though the probability to win is the same?
I do.
Because junk entries cannot finally win and have to be excluded.
Bangladesh has more junk than Iran. Because we know all numbers, it is easy to calculate the following:
- Real probability to win cannot be less for Iran than 1.538%, but it could be more because of junk. Probability to win is the number of wins divided by the number of non-junk entries.
- If Iran has no junk at all (for example), then Bangladesh should have 7,667,030 * 1.538% = 117,919 wins. But it has only 2,373. When would Bangladesh' have 2,373 wins? When the number of non-junk entries is 2,373/1.538% = 154,291. That means that the number of junk entries from Bangladesh is 7,667,030-154,291 = 7,512,739.
- Those junk entries, of course, affect numbering, and they leave holes when junk is removed afterwards. So, out of 8,515,565 entries from Asia, 7,512,739 are junk from Bangladesh. That is 88.22%. That means that resulting numbering would have at least 88.22% holes. Of course, it would be even more, because of two reasons:
- There are countries with frequency of wins larger than Iran, like Lebanon (274 wins, 12,304 entries, frequency of wins is 2.227%). That means Iran has junk entries as well
- Other countries have junk entries as well.
You can sum all that up and figure out that 89.5% of Asian entries are junk. But I showed well enough that at least 88.22% came from Bangladesh and I showed how to calculate the rest.
What would that mean?
That in order to have 15,002 wins in Asia we need to have AS numbers up to about 143,000. Then the amount of junk entries among winners would be equal to 143,000-15,002 = 127,998, what is 127,998/143,000=89.5%, equal to the amount of junk in whole Asia (see above). And finally those 127,998 winning entries are excluded from 143,xxx numbers with 15,002 winning entries left. So, this gives a way to calculate the maximum number form Asia in DV-2012 as 143,xxx
Easy enough?
So, that is exactly what I did for DV-2013 for Europe, and I got 61K.
Last edited by a moderator:
Also, some completely unofficial data from forum (from fox25, a year and a half ago):
What he says is that out of 2250 entries, submitted on October 6th, 1301 won. At the same time, Kirit Amin's declaration, if true, meant that all of them won. That means that only 1301 winners were not hidden winners, the rest were hidden winners. Kirit Amin knew they all were winners, but Fox25 did not. 1301/2250=57.8%
That gives us proportion of total winners to hidden winners. So, the maximum winning number in DV-2012 from EU should be 32K/57.8% = 55,363, so that 32K/55,363 = 57.8%
My calculations for DV-2012 for EU gave 56,700. What do we have? Amazing coincidence (one calculation showed 56,700, another one 55,363)! That is an additional (unofficial) proof that both my technique is correct and Fox25's data is correct.
What he says is that out of 2250 entries, submitted on October 6th, 1301 won. At the same time, Kirit Amin's declaration, if true, meant that all of them won. That means that only 1301 winners were not hidden winners, the rest were hidden winners. Kirit Amin knew they all were winners, but Fox25 did not. 1301/2250=57.8%
That gives us proportion of total winners to hidden winners. So, the maximum winning number in DV-2012 from EU should be 32K/57.8% = 55,363, so that 32K/55,363 = 57.8%
My calculations for DV-2012 for EU gave 56,700. What do we have? Amazing coincidence (one calculation showed 56,700, another one 55,363)! That is an additional (unofficial) proof that both my technique is correct and Fox25's data is correct.
Last edited by a moderator:
enchevetrement
Registered Users (C)
Thanks, this is an amazing information!
Yet, one can think of a valid situation where the % of notified winners is less than or higher than 60%.
The figure in this link illustrates a situation where the number of notified winners is, say 30%. In fact, one can draw a diagram that shows practically any desired ratio. So this data does not refute nor prove the theory.
Yet, whatever the ratio is, the Ankara data may give us some hint about how many actual interviews are there in a given range of case numbers. Of course this is based on many assumptions, any of which can be false but the bottom line is something in the range of 52%. Meaning, for example, if we take the range of AS1000 to AS2000, it represents ~520 interviews (AOS excluded)
From the information and documents you brought in this thread I conclude that KCC do not disqualify many entries, mainly because they are not authorized and do not have the resources and means to do so. So most of the cases should end up with an interview, unless they do not send back their forms.
Vast majority of AS cases in Ankara are Iranians and Iran is what it is and maybe many of them are bogus entries. Yet they are 39% of all Asia entries so they affect us all or at least they affect me...
- I have a high AS number so I'm searching for clues anywhere I can find them....
Vichel
Registered Users (C)
This thread has been fascinating, especially the court records on that egorka site. I laughed at their ineptitude. They had a randomizer program that didn't randomize .... LOL! It isn't the most complex code, so how could they have such rubbish in their IT department? And where was the testing? Goodness me, what a farce. I have a very hard time believing this was just a bug.
The thing is that fox25 mentioned that daily submission was rather quick, it took about an hour. So that would mean that most likely the whole amount of 2250 most likely got before the deadline and all of them are winners. But some of them are hiden winners. 1301 are open winners and the rest are hidden winners. Potentially, the deadline could cut off some of those 2250, but that is unlikely, because the submission was very quick.
So, most likely 1301/2250 is in fact the ratio of open winners to total winners.
Kirit Amin mentioned everybody (maybe except 2%) won among those who made a submission befor certine point in time. Most likely, all 2250 we submitted before that moment of time, and all 2250 are winners. 1301 open winners and the rest - hidden winners.
According to Kirit Amin, he was to blame for everything. He ordered a new development. The new development was completed, and the production people had no idea the new development was done. So, at certain point the code was switched to the newly developed code, Kirit Amin did not order the code tested, and nobody knew the switch was done. So, production people thought nothing has changed and ran newly developed code. They were unaware it was changed.
Kirit Amin was the only one who knew about the new development, but he forgot to order the final touch - switch on the randomizer.
enchevetrement
Registered Users (C)
The inflated case numbers in Asia in the last several DVs can be contributed to Bangladesh, no doubt. Yet the low number of winners can be related to the fact that the total number if immigrants from there came close to 50K in the last 5 years (see Immigrant Visas Issued by Foreign State Chargeability (All Categories): Fiscal Years 2003-2012). This is why they do not participate in DV13
In the last several DVs, Bangladesh was dominant in number of participants and had relatively low number of winners (in DV09 - DV13 is 6,023, 6,001 and 5,999 - very 'round' numbers) I agree that this caused the case numbers to inflate but I would not point the finger at 'junk' entries as the only cause.
According to the logic you present, % of junk among Bangladesh in 2011 is very much lower than in 2012 because the probability of winning then was 0.11% relative to 0.03 in DV12. Does this make sense? such a big change in just one year?
Yet the rule of thumb holds:
the more countries there are with relatively high number of participants and relatively low number of winners, the more gaps there are between valid case numbers.
However the exact reason is not clear. It can be because of 'junk' entries and it can be because of high number of immigrants from those countries in the last 5 years that poses a limitation on how many can win (e.g. Ethiopia, Nigeria)
In DV-2012 the rate of junkness for Bangladesh was 98.12%, and in DV-2011 93.41%. THAt is a big change, it became much closed to 100%
That is exactly the thing - if a country has a large number of psrticipants, the only way it could have low number of winners is if it has a very high junkness.
There are no limitations of the number of wins per country. For instance in DV-1 (DV-1995) the number of wins from Poland was above 50,000. That is all becauise of junkness.
Last edited by a moderator:
That is because of the lawsuit against DOS for erroneous initial results of DV-2012.
First, a number of winners did not know there was a substitute lottery. So, they did not proceed with the visas.
Secondly, because of the court proceedings DOS did not want to release second batch of hidden winners, what usually happens in case the amount of visa applications is low.
First, a number of winners did not know there was a substitute lottery. So, they did not proceed with the visas.
Secondly, because of the court proceedings DOS did not want to release second batch of hidden winners, what usually happens in case the amount of visa applications is low.
Vichel
Registered Users (C)
Wow, such incompetence. I've never heard of an IT department running that way, with people not following proper Systems Development and Project Management principles. And to not test new code, is unforgiveable. Must be an awful place to work.
BTW, how do you know? What are your arguments for that point of view?
Incorrect. It does mean the numbers are consecutive.
https://groups.google.com/forum/?hl=en&fromgroups#!topic/alt.visa.us/Vz0M7tc_ce4