Why we think they cheated in the Iranian elections?

Ahmadinejad : AN, Mousavi : M, Karroubi : K, Rezaei : R

– The ministry of internal affairs (in charge of carrying out elections in Iran) started announcing initial results very shortly after the polls had closed. In Iran the ballots of each polling station is counted right there when the time for voting is up, so usually stations with a smaller turnout will finish sooner. Usually they wait until a district or even province has all it’s votes counted before they announce anything, and then they will announce the outcomes along with the details of where it’s coming from. This time they just announced whatever was counted, as soon as it came, well at least that is what they said. The outcome was something AN: 69%, and M: 28% out of more than 5 million votes, with a ridiculously small support for the other two candidates, particularly less that 1% for K, the other (other than M) reformist candidate. It was the first blow for us, but calculating that these were pretty small polling stations with an average of 500 votes in each was slightly relieving, since generally AN has more support in rural areas due to his populist policies. Anyway we were worried, but tried to be positive and think it’s all just a last effort for the AN people before they accept their defeat, and that everything will change as soon as they count the ballots from the big cities, particularly Tehran.

– Boy were we wrong! They continued announcing results every hour or so, the outcome changed very little, almost keeping the same ratio. They first said that the final outcome would be announced at 10 o’clock, at the same time there were rumors of the other candidates meeting with the supreme leader (Khamenei), who is officially the head of the power hierarchy in Iran, and has the power to intervene. Anyway, we went through some miserable hours that were a mixture of hope and despair, again to no avail. The final results were announced, a few hours later, the ration had changed slightly in every subsequent announcement, but not sunstantially, final alleged outcome: AN: 63%, 24 Million voles, M: 33%. A whopping record-breaking turnout of over 80%, 39 million people voted!

So why are we sure they cheated:

– The ratio of all the announcements were almost exactly the same: AN has almost exactly twice as many votes as M, with an error margin that is so low that it’s ridiculous. In fact, if you plot the graph, it is so close to a straight line that it’s ridiculous (you can find a link to a plot here and here (the text is in Persian, not complete)).
– Not only did they make up the statistics with almost no noise, they also forgot to add invalid and white votes! So usually there is a small percentage of invalid ballots in all elections, because of people showing their dislike for choosing between “bad and worse” by not writing anything on them, expressing their sense of humor by writing stupid and funny stuff, or just having an awful handwriting. These ballots count in the overall percentage of course, but this time there were zero of them!

– My overall analysis of this is that they didn’t even bother adding some ballots in favor of AN, what is probably the easiest and most probable thing to do, they have just made up the numbers, that is more probable than asking all their people to add exactly some amount of AN ballots such that he has twice as much as M. And they forgot to add the invalid votes to their balance. They remembered somewhere towards the end (probably heard it from us complaining), where the guy said they have simply not announced them, but of course they have the statistics, but that doesn’t make sense because they should come for each polling station statistics and be counted in the overall percentage.

– Another fishy thing about the constant ratio is that it is highly improbable to naturally have something like that considering the demographics of the provinces. The outcome of presidential elections in Iran usually differs highly in different regions, it is pretty common to have complete different orderings in different provinces. The most important reason for this is ethnicity, and the best example is the first round of the previous presidential elections where the two candidates that were last in the national race each ranked first in the provinces that they were originally from, or had the same ethnicity as the people living there. So it would have been expeted that M who is Azari, the largest ethnic group in Iran, would have a huge margin in the Azari provinces, which hold somewhere around a fourth of the whole population. No where in the process of adding up do we see some sort of jump that has resulted from the addition of major Azari cities. It just doesn’t add up. It goes so far that the other candidates do not even lead in their hometowns, for instance R has almost no votes in the small village where he was born and is a hero, this is really weird considering the normal attitude that people in such places have. I still am yet to see detailed statistics for all cities and provinces, just a few things like the hometowns here and there.

– Something else that is funny is that the least popular candidates. K and R, have a ridiculously small number of votes in the turnout. If you just add up the votes from K’s campaign members and a small number of people directly influenced by them like, say, a couple of people in their immediate family, you would have more than the two hundred and something thousand votes announced for him.

– The votes from abroad were announced as M: 85,000, AN: 70,000, which is again very weird, since a much higher majority people outside Iran would be anti-AN.

– These are not all the reasons, but probably the most obvious ones, maybe a very important reason would be the wave of Mousavi supporters in the last weeks, and the large crowds of them in the streets before the elections in which people from all social classes and walks of life could be seen. We didn’t see our votes coming out in the turnouts.

– Current status: all hopes on the supreme leader intervening has been lost due to him endorsing the outcome. People are protesting in the streets and there is some violence going on already. This sort of scenario is very much like a coup. M and K have strongly opposed the outcome and are trying to get their voices through and their supporters organized, particularly M who has announced he is the rightful winner of the elections.

– This was written very quickly, I thought now that I’m spending so much time on this, and actually cannot concentrate on anything else, I might as well write my ideas down. I will also post it right now, so I might improve it gradually later.

– What I hope to cover in the next post(s):

– Why is this what happened in the elections are so important, and why it has been made today the worst day of my life and of many other Iranians.

– The ongoing situation in Iran.



  1. […] As I pointed out in the previous post the whole “election” was a […]

  2. Mohammad said

    As an Iranian student living in Tehran, I’m still not convinced that ‘election fraud’ is so evident. Let me elaborate:
    1. At least to me it seems that Ahmadinejad is quite popular in rural areas and small cities. Consider that tehran constitutes 12% of the voters, large cities 21%, small cities 34% and rural areas 33%. So Tehran and large cities constitute less than one-third of the eligible voters.
    2. It’s often said that Iranian polls are not reliable. I passed by a poll done by an American institute named “Terror Free Tomorrow” which is affiliated to “New America Foundation” and was done before the elections. While many of the respondents were undecided, the decideds support Ahmadinejad by a large margin.
    The other surprising result of this poll is that most Azeris would vote for Ahmadinejad, despite Mousavi being an Azeri himself. Again, this is consistent with the official results.
    2. I’m not sure that the constant relation argument holds. The ratios were 69% to 28% at first, and they ended at 63% to 33%. Considering that the numbers were being announced in aggregate (i.e. added to previous counts), the course of the numbers seem reasonable to me.
    Also I have seen arguments against this ‘statistical evidence of fraud’, e.g. here. It shows that this was the case in the US election too!
    3. The invalid ballots were not included in the numbers at the beginning, but they were announced later. These kinds of errors happen everyday in Iran. I understand the sensitivity of the issue, but I don’t see any ‘hard evidence’ of fraud here; it could be that at first they didn’t announce it since they didn’t consider it as important as candidates’ votes. If you live in Iran, you don’t expect that everything goes perfect. But that doesn’t necessarily mean fraud. It could be, but it’s not a ‘convincing’ evidence.
    4. I don’t know exactly about the election and the voting process. But I think the speedy count is not an issue in itself: in average, each voting box hosts less than 1000 votes. Ofcourse there are many boxes with more than that, but there are many more with less than that. They can be counted in less than two hours easily.
    5. It’s true that some candidates had the most votes in their native provinces in 2005 elections. But (other than the poll I mentioned above) I think you should compare this election with 2001 or 1993 elections, when the incumbent president had a ‘natural’ advantage over the others. The poor, uneducated average Iranian has a tendency to vote for the candidate whom he/she knows more, and the incumbent president has been in the TV and mass media for more than 4 years, which a mere 20 days of campaigning won’t change that much. So you should compare this with Tavakkoli’s (and the other losing candidates) votes in 2001 in his hometown: Was he more popular in his hometown than Khatami?
    5. It would be revealing to compare the results with the results of previous elections, and weigh the arguments for and against the fraud. Didn’t such ‘irregularities’ (e.g. the hometown argument, the constant ratio of votes argument, the complaints of closing voting staions or absense of candidates’ representatives in the voting process) happen in the previous elections?
    6. I think the massive rallies of Mousavi’s supporters contributed to the illusion that they are the majority. 50000 people gathering in the streets could make Tehran look like a Mousavi supporting stronghold while that is just 1% of Tehrani eligible voters! They attracted media attention so much and successfully made the atmosphere pro-Mousavi. Ahmadinejad’s campaign gatherings (such as the 600000 gathered at Mosalla in Monday) were often overlooked.
    Let me elaborate on my personal experience. I neither supported Ahmadinejad nor Mousavi, and in my close friends (who were all students) there were 3 Ahmadinejad supporters and 2 Mousavi supporters. But the atmosphere in the university was so pro-Mousavi that the Ahmadinejad supporters were ridiculed and laughed at as soon as they tried to defend him. As a result, they were more silent and shy than the loud Mousavi supporters and didn’t express themselves much, while they weren’t necessarily the minority, even in an engineering university in Tehran.
    7. Mousavi and Karroubi have protested against the results and demanded the cancellation of the elections results. Since they had representatives at the polling stations and the election committee, they might be right. But as the Persian news site Alef, puts it, it could be because of the atmosphere the candidates and their campaigners were in. It states that the same thing happened to Tavakkoli’s campaigners in the 2001 election. It has been called ‘illusion of victory before vote syndrome’.
    8. Before the elections, some analysts (here and here) speculated that we’re witnessing a class conflict in Iran, between the lower-middle-class and the upper-middle-class who have completely different desires and needs. What we’re seeing could be a sign of that conflict.
    9. Don’t forget Ahmadinejad’s PR tactics, especially his fierce verbal attacks against Hashemi Rafsanjani, which made him quite popular in the lower-middle-class and even some upper-class people who dislike Hashemi heavily. He did well in the TV debates and successfully painted himself a ‘lonely candidate who only relied on people and God and was facing unfair attacks orchestrated by Hashemi and his wealthy cronies’. If you’re familiar with Iranian people, you know that how this appeals to them.

    Again, I don’t know if the elections were fair or not. I’m just saying that there’s no convincing evidence that suggests such a mass manipulation of votes.
    The Islamic Republic is terrified by the protests and the prospect of a so-called ‘velvet revolution’, and hastily tried to crack down on them. I think this was a non-constructive move, but it could happen even if the elections were fair. So this is not an evidence of fraud either.
    Sorry for my lengthy comment, as I don’t have a weblog to post all these!

    • Sara said

      I am spending most of my time following and spreading the news from Iran right now. ‘m positive that there has an enormous amount of fraud in the elections. So, quite frankly, my main priority is to find a way out of this situation, not to discuss whether there has been fraud or not. But there are a few points that I would like to address. I will reply to each of your numbers accordingly, and will not capitalize properly from now on, so this could get done faster:
      1. Actually AN had the highest outcome in cities last time. I would suppose he has more in rural areas this time, but none of it is facts of course. Your numbers are right, it really don’t prove anything.
      2. I took a look at the survey you have here, interesting, since I didn’t know it existed, and you know that all attempts by non-governmental parties in Iran has been unsuccessful (the most famous example being the arrest of abbas abdi ghazian for attempting to do just that some years ago). I think it is actually very consistent with my argument. It says:

      “34 percent of Iranians surveyed said they will vote for incumbent President Ahmadinejad. Mr. Ahmadinejad’s closest rival, Mir Hussein Moussavi, was the choice of 14 percent, with 27 percent stating that they still do not know who they will vote for.”

      It was conducted in mid-May, a time when M had basically just started campaigning. M was not in the political scene for around 20 years, and it took a while for people to get to know him again. Many probably didn’t even now that he is a candidate yet. Anybody who has followed the campaigns would agree that his support base was mainly built in the last two weeks, and particularly in the last days before the elections.
      Also another widely known fact in Iranian politics is that the hardliners in Iran have a certain amount of support, a certain base loyal to the Islamic Republic that always systematically votes for them in all elections, and the percentage of these are something around 25% of the whole population (roughly of course). Others vote based on their gut feeling and the particular situation in each election, and most. The outcome depends on those, and even more crucially on people who usually don’t vote. Those who were sure they will vote for AN at that time are not much more than this loyal base (which seems to be shrinking btw, because they may be religious conservatives, but they don’t necessarily support this craziness).
      Most definitely, a large majority of the people that are undecided in this poll did not vote for AN. They basically had 4 years to make up their mind about him, and if they hadn’t yet, they were most likely to vote for one of the other candidates, particularly since it would have been really hard to favor AN after the debates strong initial bias for him, he even lost some of his less fervent supporters after them. There is around 22-23% of people missing in the results of this poll, I would suppose they didn’t answer or didn’t intend to vote, didn’t read all of it, so I’m not sure, but it is not the same as the number of those who didn’t intend to vote(which is 11%, even less than it actually turned out to be and that is particularly weird since this was a record-breaking election in terms of turnout). In any case, it is almost definite that almost all people that usually did not vote and decided to vote this time voted just because they wanted to get rid of AN. So actually these polls can be consistent with M winning the elections in the foirst round wiht a high margin.
      Since what actually happened was that there was a very high turnout, this is highly probable (as I said, I am sure it is the case). This is something I forgot to mention in the original post, might add it to it …
      Also, even this poll predicts that nobody will have a majority in the first round and AN will be elected in the second (the latter argument supposing that voters vote consistently in both rounds which is usually far from true … but lets forget that now), which further supports my point.
      I didn’t find the bit about the Azeri’s results, but something that was very surprising was the extremely high (94%) turnouts reported for them, the azeri’s usually have lower than average turnouts, for instance in the second round of the previous elections they had a lower than 40% turnout in the second round after the Azeri candidate landed somewhere towards the end of the list in the first round.
      Sorry, I just don’t buy any other argument for the Azeri’s.

      second 2!- I think it is not of course theoretically impossible, although highly improbable for Iran. I’ve already seen the link you provide, but I think you can’t really compare that with the 2-party system in the States. I have also heard some things about it being statistically much different when there are more than 2 candidates and another thing about the variances proving some inconsistency, but I don’t know since I’m not a statics expert, but I’m looking into that. I still haven’t looked at the statistics for the individual provinces which have been announced a few hours ago.

      3 – yeah, that is not a proof, none of these may be in itself, but it’s an irregularity, and the people working at the interior ministry organize an election almost every year and are pretty good at it and organized in doing ir, unlike lots of “these kinds of errors” that may be common.

      4- the counting is not, but the accumulation always has taken longer.

      5- i think this is partly answered in the first 2. Just one more point, you just can’t compare Khatami’s popularity with AN. The real base that AN had in the first was the 5-6M he got in the first round last time. In the second round he really got the 17M because of the sentiment against Hashemi, and managed to gain some support up to the 34% in all his 4 years.
      I think that it Tavvakolli was the only famous guy from a small place (don’t know where he is from), he would even have had the lead there, it’s just natural that they would be proud of him.
      second 5! (you seem to be a smart person, but how come you don’t know how to count?)-yeah, would be interesting.
      6- I agree that there is some amount of conformation bias, but I think it affects AN supporters (and is against the argument you are making here) more than M supporters.
      I have seen pretty aggressive AN supporters, in fact most of those who I’ve seen are more aggressive and outward, and they are getting more so by the minute.
      It is definitely not nice to make fun of other people, but maybe that was because the students were so much degraded and oppressed by the AN government that they didn’t find any other channel to express themselves. I myself saw in the protests today in the hauge that there nobody a single person who was tying to insult An was quietened by the crowd, which is a very good sign. We don’t need to make fun of him to prove he’s wrong, that is just so obvious anyhow.
      7- Alef is a conservative site, I don’t consider it a reliable source. If the reformists were to turn the table like this when they lost, they would have done it without putting so much at stake four years ago when they had the interior ministry in their hands, why would they do it now if they had really lost?
      8-there have been many ppl of all classes in the streets for mousavi, just look at the pictures, I think he actually reduced the gap that had been so evident in the runoffs four years ago.
      9-the Rafsanjani bashing has been his only tactic, whih he tried to use again by putting mousavi in the same basket with him, obviously since he had nothing against mousavi himself.
      but it just proved to be awfully unsuccessful for those who were not already on his side.

      I think that this is exactly a coup as I said in the post after this one. period.

      wow! that was a lot, good thing writing this is not my priority.

      • Mohammad said

        Thank you very much for taking the time for a detailed response to my comment.

        Sorry for the numbers issue, I inserted some points in the middle and forgot to update the numbers. And thanks for calling me a seemingly smart person! 🙂 Lotf darid (didn’t know the English idiom)!

        Things have changed since my last comment, and there’s growing hope that the officials are paying attention to the protests and putting more transparency in place. The mass gathering today in the Azadi avenue was great. Alas, the crackdown on the students and protesters was disgusting.
        As you probably know, the Ministry of Internal Affairs has released detailed numbers of each district (“shahrestan”), which is interesting and answers many questions. For example, Mousavi and Rezaei have actually won most of their hometown’s votes (Shabestar and Lali) but Karroubi hasn’t won his hometown (Aligudarz).
        Again, I don’t know what actually happened, but not convinced of a massive manipulation. I’ve seen numerous reports of small-scale vote rigging which seem to happen in every election in Iran, and aren’t necessarily coordinated. For example, there’s a very recommended, healthy and unbalanced discussion going on in Hamed Ghoddusi’s blog where People report their [sometimes conflicting] experiences and analysis.

        On the points I mentioned and you answered, I think the discussion doesn’t need to be continued on every point, since we need more information on most of them and it’s blind to insist on them. The only interesting and new info I found was a poll conducted by “World Public Opinion” in 2008, which finds the ratio of Iranians with different political points of view. It groups surveyed people in 4 groups based on their ‘level of support for the regime’:
        Conservatives: 45%
        Moderates: 24%
        Reformists: 18%
        No Opinion: 13%
        So one should be careful when assessing the votes of the people based on their political views. Specifically, I don’t think that the people who had not voted before, necessarily voted for Mousavi. The feeling against Hashemi is so great (as well as the shock of the debates) that they may have voted just to oppose him (supposedly) and not necessarily to support Ahmadinejad.
        Also, many conservatives who don’t vote in normal circumstances (because of laziness, daily work, etc.) may have voted this time because of the intensity of the competition.

  3. Mohammad said

    I submitted a long comment, but I see no change here. Do comments have to be confirmed by you? Or are long comments automatically rejected? Should I submit it again in pieces?

    • Sara said

      It shouldn’t need moderation any more. Your comment was not too long, it ended up in the spam folder, probably because it had too many links.

      • Mohammad said


  4. […] I made a comment containing what I have explained here on how they really couldn’t use the polls that they mentioned the way they did, And how those […]

RSS feed for comments on this post · TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: