Lies, Damn Lies, and Statistics

robbeiflex

robbeiflex

Well-known
Local time
5:55 PM
Joined
Nov 21, 2009
Messages
1,047
If you analyse your photo-sharing site statistics, what entertaining trends do you find, and what do you learn from them?

I've noticed that a few recent threads have discussed the effects of the internet on photography. I know I shouldn't generalize, but they mostly revolve around the photos that formerly ended up in shoeboxes in an attic or would not have been not taken at all, and are now all over Facebook, Flickr, etc. with loads of compliments. This prompted me to take a look through my Flickr stats, and see if they contained any insights into how I might improve my photography.

Here is a quick summary of my own stats to get things started:

The "favourites" don't tell me much and the comments are mostly chit-chat with people I either know personally or through RFF.

The "views" are where things get interesting:

- 1 of a hotel dining room table with well over 3,000 views - I like the photo a lot but I have no idea why it is so popular
- 1 of a bicycle in a shop window in Singapore with over 500 views
- Some portraits with between 200 and 400 views - an interesting guy in Toronto with a weird beard who walked up to me and said "take my picture", my friend where I caught him at the exact instant that his mobile phone camera flash fired, and some portraits of my wife
- A photo of a Citroen 2CV taken with a Lomo with around 300 views
- Over 250 views of a photo of a beer (it is a Guinness though)
- A large number of decent tourist photos with around 100 views or less
- A few experiments with infra-red and various film and developer combinations with around 100 views or less
- A large number of photos (including at least one gallery pick of the week on RFF) with hardly any views

So here is what I learned from this:

- Flickr stats are non-sensical and to be taken with a grain of salt.
- I'm sure that a cute photo of my daughter with a kitten would blow all of the above out of the water.
- Then again if I wanted to improve my stats based on the above I would travel more, and meanwhile take more photos of bikes, interesting people, cars, my wife, beers, landscapes, cityscapes, street photos, etc. mostly favouring obscure cameras and different film-developer combinations.

Hmmmm, maybe I'm on to something....

So what do your stats tell you?

Cheers,
Rob
 
Would you say the same about views at the RFF gallery, or is your comment only about flickr? I find that in general, an image with a large number of views has been posted for a long time and/or it is an interesting looking image and/or it is an attractive looking female.
The last possibility is a software error at RFF.
I am not acknowledging the possibility of many RFF members having poor taste in photography! :)
 
Hi Raid,

All of the above applies only to my Flickr. I tend not to post much at all on other sites, and have only posted a handful of photos on the RFF galleries. I keep telling myself I should post more to RFF but only rarely get around to choosing the real gems and then re-sizing and posting them. I post a lot more to threads by linking to my Flickr.

Thanks,
Rob
 
rob, inspired by your thread, i looked at the 11 latest pages of my flickr account. all four top-view photos were x100 specific. one is a p and s digital showing a coffee mug. iiif and x100, and the remaining three were made with my x100. most, by far, photos in those 11 pages were made with film.

most views at 114:
114.jpg


second most views at 61, x100:
61.jpg


55 views, x100:
55.jpg


53 views, x100:
53.jpg


do we see a trend here?
 
Flickr observations

Flickr observations

I think you are fairly correct in your observations. The types of photos that I have on my Flickr account is quite different (no beer :(, no attractive females -- no people, actually :D), but my top photo (2,483 views to date) has also been there the longest (more or less). It helps to have lots of labels/tags and geo-tagging for lots of views.

Favorites and Comments appear to be mostly a function of the groups they get added to, and/or the number of "contacts"...
 
Last edited:
Dear Rob,

my most-viewed shots here are of a pretty young blonde drinking champagne on a dirt road along a canal. This message will up the score on those shots dramatically, I reckon. :p

And for obvious reasons if one looks at the general population around here. No clue as to what the demography of the Flickr crowd is but I suppose it will have a huge effect on why some shots are more viewed than others. As will tagging, used camera, and composition on a small thumbnail size!

In general my shots aren't viewed much, a situation I someday soon hope to change on my conditions, not Flickrs. But it would need a re-invention of my flickr sets etc, since its a digital dumpster now, I slap anything I need to get online on there.
 
My most-viewed photo on flickr is one that I didn't even take:


What does a blonde do after a serious car crash?

A blonde woman brushes her hair by a roadside car wreck
photo received by e-mail and of unknown provenance (certainly not taken by me!), resized and some basic photoshopping done (levels, curves, saturation)
Click to expand...

My 2nd-most viewed photo is of a Konica Hexar RF. Quite different subject matter indeed:

http://www.flickr.com/photos/mfunnell/302450335/

...Mike
 
My stats tell me my images don't stand out enough as thumbnails, or that I don't tag them vigorously enough. Also, that a picture of a Leica attracts far more attention than a picture taken by one ;)
 
So do you suppose Flickr fabricates the stats? If so why would they bother?

Are you saying Flickr stats are mis-calculated which makes them inaccurate?
 
The ones with high views are likely to be reposted either on someone else's blog or in another forum.

My highest viewed shots are the ones that show up in top 3 results in google for some relatively popular keywords. So whenever people search for those keywords, they click on my flickr.
 
willie_901 said:
So do you suppose Flickr fabricates the stats? If so why would they bother?

Are you saying Flickr stats are mis-calculated which makes them inaccurate?
Click to expand...

Nope. I'm not accusing Flickr of any misrepresentation. I just liked the phrase popularised by Mark Twain as a catchy title for the thread. What we tell ourselves about the statistics, now that's where the potential for lies is rife. :)

Thanks for your posts everyone, this is fun!
Rob
 
haempe said:
There is a nice phrase in German:
Don't believe in a statistic, which you didn't faked yourself...
;)
Click to expand...

Ah, that's beautiful! :)

There is the reason why I refrain from posting on Flickr and other similar sites. But even stats on my own blog tell their share of lies. In the beginning (5 months ago) I used to get 10 - 30 visits a day, now 50 - 300 a day. That's far from a sign of increasing appreciacion. More like increasing amount of links to it lying around (facebook, RFF and so on).

Occasional visitors who just click in and out mean nothing to me. But when someone leaves a comment or clicks "follow" to receive updates by email, that makes my day, as it does show that they've been through more than one page and found my stuff interesting.
 
I think comments are far more reliable way to measure the merits of your images than views. View numbers can be distorted in so many ways; it may be a image is linked to some other site or to a popular image pool for example.

People don't usually bother to analyze images very deeply but they may comment it like "great shot" just to point out they think your image is better than average among the image stream they are following. You get comments if you yourself comment other peoples images. That's just fair I think. You tend to be rewarded for activity.
 
raid said:
Would you say the same about views at the RFF gallery, or is your comment only about flickr? I find that in general, an image with a large number of views has been posted for a long time and/or it is an interesting looking image and/or it is an attractive looking female.
The last possibility is a software error at RFF.
I am not acknowledging the possibility of many RFF members having poor taste in photography! :)
Click to expand...

Hi,

You left out fluffy kittens and puppies with big sad eyes...

Regards, David

PS I know how many have looked at my pictures but I'd like to know how many were pleased and how many went "yuk" and wondered why they'd wasted their time on it.
 
David Hughes said:
... I know how many have looked at my pictures but I'd like to know how many were pleased and how many went "yuk" and wondered why they'd wasted their time on it.
Click to expand...
Your wish is plausible...
But I hate these like/dislike buttons, which allows everybody out of the armchair to judge others effort whithin seconds ...
This is also often not really meaningful.
 
Constructive criticism is always useful - rarely given though, perhaps because it takes thought and effort as well as having the experience.
I never look at flickr comments or the number of viewings because I just use the site for getting pictures from one place to another.

jesse
 
You must log in or register to reply here.
Back
Top Bottom