Culture Lifestyle Weirdness

How many people actually went for BERSIH 4? We try to count!

*Tekan sini untuk baca dalam bahasa Melayu

Now that BERSIH 4.0 over, confirm a lot of people ask this question.

“How many people actually went for BERSIH 4.0?”

Well, BERSIH 2.0 says about 500,000 people in total. On the other hand, our PM believed 20,000 people attended on the first day. The police? They say 25, 000 at 3pm on Saturday and 18,000 at 10pm on Sunday.

But once again, if got so many answers (like 1MDB), it can only mean one thing: No one knows who is right or wrong.

Soooooo what we tried to do was find our own numbers. Yeap, CILISOS tried their best to CONSERVATIVELY count how many people went to BERSIH 4.0.


At first, we try manual way…

So initially we tried counting. Yes, the one where you go 1-2-3-4-5. But not count one-by-one all the way from Pasar Seni to Dataran Merdeka la. We tried this instead.

count peopo bersih4

Original image from

See that red box? So we tried counting the number of people in that box, times length and width, and calculate the number of people for the whole of Jalan Tun Perak.

After that we asked the whole CILISOS office to individually count the number of people in that red box…

finding bersih cilisos

Just look at our confused faces. That’s me in background 2cm away from screen.

….and all send at the same time to our FB chat.

estimate count cilisos


Okla, Jo-Lyn probably also counted some people from her home planet, but still…

Susah gila ayam kira sampai buta woi! Let Science kowtim la!

So one method fail di. But actually, everywhere in the world also fail one.

“If you count the number of jellybeans in a jar three times, you’ll probably have three different numbers, because people simply cannot count very large numbers without some error. Now, imagine trying to count a shifting mass of heads, some stooping to tie shoes, some sharing the same umbrella, some arriving late or leaving early. Plus, this is one field in which good intentions are rare.” – Rob Goodier,

Nvm, we found something online called ‘The Jacobs’ Method’ by this guy Herbert Jacobs who’s been credited for modernising crowd-control techniques. So we decided to try that instead. His method has been used for everything from political rallies, to concerts, and is generally accepted as the way to count crowds worldwide. How?

jacobs method popular mechanics


According to this Jacobs fler, this method has one basic principle: area x density. The density is calculated based on how much space a person takes up in a crowd. And for this there are 3 different types of crowd density (basic density rule).

  1. Light crowd – 0.93 square meter/person
  2. Dense but manageable crowd – 0.42 square meter/person
  3. Mosh-pit density crowd – 0.23 square meter/person
What kind of attendance will we expect at Bersih 5? Cilisos investigates...

Because we had to publish this within 24 hours (and had limited reference footage), we decided to use the 2nd one, where one person takes up 0.42 square meter of space (Cos BERSIH 4 was dense but also well managed la, ended peacefully all right?).

ALSO, here are some other things we took into consideration:

  1. We only considered areas that looked PACKED to the MAX.
  2. We used videos taken at specific times (so people can’t say at this time, this area was crowded…etc). So we found great drone shots of these 2 particular time slots (also widely considered the peak time of BERSIH).
    1. 4-6pm Saturday
    2. Midnight on Sunday before the finale

Things we did not take into consideration:

  1. Sparse areas.
  2. Areas not recorded on camera.
  3. Off-peak people coming and going (HOW TO COUNT THAT ONE WEI SIAO AR!?)

Examples below.

bersih not considered

Unedited image from

Why? Because we want to give ugaiz a figure conservative enough that kenot be argued. We’d add about 20% to these figures to account for people going in and out not during peak hours, sparse areas, AND off-camera.

With our methodology done, we started to trace out the dense areas (on this really cool website The traced out maps and jalan kerja for both time slots are as follows.


(view the reference drone videos here & here & here)

map 4pm bersih

Total area: 28377.074 m²

Crowd density: 0.42 m² per person

Jalan kerja: 28377.074/0.42

= 67,564 peeps!


(view the reference drone video here)

dataran merdeka 12am bersih

Total area: 15473.237 m²

Crowd density: 0.42 m² per person

Jalan kerja: 15473.237/0.42=36841.04

= 36,841 peeps!


So why everyone’s figure so off?


Aiya… is like talking about the size of a guy’s bird la (as in, er.. his pet bird. In a cage). The guy who owns it sure say big, other guys sure say small.

As for our figure? We really should mention once again that OUR number does not include:

  1. Sparse areas.
  2. Areas not recorded on camera.

And our numbers also do not, we repeat, DO NOT tell us how many people in total went for BERSIH 4 over the weekend. It only counts the people in CERTAIN AREAS at SPECIFIC PEAK TIMES.

So we reckon Polis figures DEFINITELY off, and BERSIH’s own figures too probably (unless all off-camera). If we had to estimate, we’d say ADD 20% to our conservative figures, mainly to cover sparsely populated areas, off-peak times and off-camera areas.  Closest estimate (and in our opinion, probably the most realistic) to that was actually MalaysiaKini’s estimate of 100,000 on the first day  because they may have been able to include all the things that we couldn’t (the drone footage is mostly theirs after all).

Do ugaiz agree with that number? Let us know.

[interaction id=”55e5b9861cec712a2e4e6083″]

Not happy? You go count yourself lor.


  1. Joe

    03/09/2015 at 2:02 pm

    your conservative of 20% mark-up cannot be used in this case, because,look at the building(a lotttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttttt people is inside buildings around that area,4-6pm could be consider early dinner time), the alley, the not in camera parts, etc etc… so i would say the mark-up shud be somewhere around 50%-60% due to, those above should cover actual on-road people times halve. so yeah.. that is my logic.. just sharing.. because if you are at the rally, you know, any figure less than 150k is a joke…

    • Chak Onn Lau

      03/09/2015 at 2:42 pm

      actually, we would’ve agreed with you before we made our calculations

  2. Nur Damia Danial

    03/09/2015 at 12:22 pm

    Try this method…

    In short, take bunch of numbers from different people and the average will normally be closer to the real answer… Tested it out on those counting competitions before

  3. Ray Tek

    03/09/2015 at 11:38 am

    Just freaking check the LRT ticket sales. We can forgo those who came by buses or other modes and the number will still be huge.


    03/09/2015 at 9:29 am

    I think the peak time at midnight should use the 0.23 density. So it should be 67274.

  5. Mun Shan Koh

    02/09/2015 at 12:18 pm

    Could have been a much bigger area, taking into account the side walks/ pavements and side alleys which were packed with people as well. Also, at Dataran, it was jammed packed, with other smaller stages being manageable.

    • Johannan Sim

      02/09/2015 at 4:11 pm

      Hi Mun Shan!

      We only took into account areas that we could see clearly in pictures or on video at 2 particular time slots. So to get pictures of side alleys at like exactly 4pm a bit tough la. But thanks anyway for your response!

  6. Hobbesianism

    02/09/2015 at 10:23 am

    Good attempt, though I would argue that the use of 0.42 m2 per person for the entire crowded area is not entirely justified. I would argue that at least 60% of the colored or highlighted area were jammed packed with people who at times struggled to move, and the rest of the area being “manageable”. Based on that, from 4-6pm on 29 Aug, you’ll have some 101,000 in just the highlighted area alone and on midnight, 55,000. And that is just two to three hours out of the entire 36 hours with a lot of people attending once and only for a couple of hours either day. So if you ask me, half a million is not really that outlandish.

    • Johannan Sim

      02/09/2015 at 4:16 pm

      Hi Hobbesianism,

      Thanks for your response! We read online that the ‘tight crowd’ category is where people are soooo close to each other that they’re possibly already tripping over each other. While BERSIH was packed, we’re unsure if we can say that the crowd jam until can’t breathe la.

    • Hobbesianism

      02/09/2015 at 5:14 pm

      Hi Johannan,

      I’ve actually went to the extent of measuring the space of 0.42m2 and had myself in the box. Its quite spacious and we wouldn’t have touched one another if its just a person per box. Based on my experience for the two days of protest, the last night was especially packed. You could hardly move towards Dataran.

      Why don’t you do a test, say have 4 people stand inside 4 boxes of 0.42 m2 areas and compare the space with say 8 people in the same total area. Gives you a feel as to which ratio to apply.

      Anyway, all this is just in good fun. No one is right or wrong.

    • Chak Onn Lau

      02/09/2015 at 11:59 pm

      Hi Hobbesianism (can call you Hobbes for short? easier),

      I was there near dataran, and i think it’s a realistic estimate. Many people were sitting down, and there were clear paths even in the densest of areas. Everyone was very cooperative, and I never got shoved once.

      However, like we said, add 20-30% for crowds in off-camera and sparse areas, and you’d arrive at pretty much the same figure everyone else got 🙂 just under 100,000

  7. zaharibb

    02/09/2015 at 8:44 am

    I couldn’t make it or else your guesswork will be wrong because I would bring my family along in support.

  8. Edward Lye

    02/09/2015 at 4:19 am

    Please do the same on 10Oct. Supposed to be 1,000,000 supporters.

    • AEmertis Emeritis

      02/09/2015 at 9:17 am

      That’s KOTOR rally, I’m clean sorry!

    • Chak Onn Lau

      02/09/2015 at 10:07 am

      That one we count brain cells might be easier 😉

  9. Eric Leong

    02/09/2015 at 1:39 am

    “Not happy? You go count yourself lor.”

    this phrase made me burst out in laughter. at 1.37am. in my living hall. when all my family members are asleep…… well my hats off the ground, for cilisos members for attempted the first method (count with eyes 2 cm away from monitor. seriously?)

    good job cilisos!!!

    by the way do you guys have vacancy?

    • Chak Onn Lau

      02/09/2015 at 10:08 am

      Hahaha maybe… Send resume to [email protected]. at the least we’re always looking for freelancers

  10. budleee

    02/09/2015 at 12:51 am

    Bonggok, but I like it and I learned something new. Good article. I don’t give a rat’s ass about the demo since its not my weekend shift. If it is though, at least those who don’t care can know where to go and avoid the crowd.

  11. Choen Lee

    01/09/2015 at 11:26 pm

    i reckon the numbers are more like 50,000+ at any one time, but total people who actually came will be significantly higher than that.

Leave a Reply

Your email address will not be published. Required fields are marked *

Here at CILISOS, we believe that the only way to consume information is with a serious dose of flavour. Our aim is to make mundane things like news and current events entertaining, and informative, hopefully in equal measure. Read More

The Serious Legal Stuff


Cilisos Media Sdn. Bhd. Copyright © 2020. All rights reserved.

To Top
Send this to a friend