Statistical Analysis of top 10 photos on DPC

Sony SAL-18250 18-250mm f/3.5-6.3 High Magnification Zoom Lens

03/07/2007 05:46:26 PM · #1

What makes a top 10 photo on DPC? Do models help or hurt a photo make the top 10? If you shoot in a studio or use clever props will you have better chances of reaching the top 10 than if you shoot what nature gives you to work with? How often do photos of children make it in the top 10? How often to photos of flowers make it in the top 10?

I sought answers to these questions and conducted a small test to see what the numbers suggest. Using a random number generator I sampled 30 challenges, taking only the top 10 shots of each challenge under consideration, for a total of 300 photos total by selecting 5 challenges from the first 100 challenges, 5 challenges from the next 100 challenges, 5 challenges from the next and so on for the first 600 challenges (there are currently around 635 challenges on DPC). Using my own subjective judgment I listed each photo as either being a �natural setting� or a �studio setup/prop shot.� Next I listed each photo as containing a �model� or �not containing a model.� Lastly, since I believed photos of children and flowers would appear much less frequently, I recorded the number of photos which contained a child or a flower as a significant part of the image. The results are posted here for you to argue over.

For those interested in my criteria, I determined that a �natural setting� was any photo for which I feel the photographer just shot what appeared naturally before him or her without manipulating the elements himself. For example, a person shooting a photo of a plane going over head likely didn�t ask the pilot to fly by that position or tilt the wing as he went by so I marked it as a �natural setting.� Studio or prop shots are those photos I felt that the photographer either shot in a studio or brought in props of some sort to achieve the photo they wanted. For example a perfectly placed baseball helmet, glove, and ball on a bench in a dugout would probably get listed as a �studio/prop shot� since it seemed important to me that the ball, mitt, etc were positioned exactly as they were and were likely manipulated by the photographer to achieve the effect. Shots of sunsets, waterfalls, etc were most often counted as a �natural setting� while a shot of the same waterfall with a rubber ducky going over the falls would likely to be listed as a studio/prop shot. Obviously this is highly subjective so take it with a grain of salt. For the models my criteria was any living creature or limb of a creature including humans, human hands or feet, pets, animals, birds or insects in which the living creature is a significant focus of the photo. For those interested, I�ve broken out the numbers of those involving only human models as well. My criteria for �children� was any subject of a human child which appeared to be under the age of 12. My criteria for �flowers� included any flower or plant for which the plant was the major focus of the picture. This included photos of leaves, weeds, and close up of tree branches.

Here is my analysis of the top 10 photos for 30 �random� challenges:

Natural Settings � 56% of the time
Studio/props � 44% of the time.

Models � 40% of the time (humans made up 62% of the models)
No Models � 60% of the time

Children only appeared in 7.5% of the top 10 photos sampled.
Flowers only appeared in 8.3% of the top 10 photos.

Any comments OTHER than I have too much time on my hands?

yanko

Carl Zeiss 50mm ZE f/1.4 Planar T for Canon

03/07/2007 05:49:43 PM · #2

Maybe I missed it but did you include challenges where a specific subject was the focus of the challenge? For example, a flower challenge would yield 10 entries of flowers in the top ten. It would seem to me you would ignore such a challenge for this analysis since not every subject has had it's very own challenge devoted to it.

Message edited by author 2007-03-07 17:50:19.

Gordon

RIM BlackBerry Curve

Canon EF 85mm f/1.8 USM

03/07/2007 05:51:38 PM · #3

So within the bounds of the error of your analysis, personal subjectivity etc, you could easily say a +-10% error on any results.

Which means it is a wash, for the factors you considered. Model/ no model was 50/50. Natural vs Studio was 50/50

Outcome of that is that something else matters, not these factors.

Like if the picture was good or not ? ;)

levyj413

Nikon D7100

Tamron SP AF 17-50mm f/2.8 Di II LD Aspherical (IF) for Nikon

03/07/2007 05:59:33 PM · #4

Interesting. That tells me that having a model or not, and studio vs. natural, is pretty irrelevant. Kids and flowers, on the other hand, may be tougher to get right.

Oh, and you have too much time on your hands. :P

Muppet

Canon EOS-400D Rebel XTi

Canon EF-S 18-55mm f/3.5-5.6 II

03/07/2007 06:01:02 PM · #5

you have too much...oh forget it, you know where i was going...

noraneko

Canon EOS-7D

Canon EF 28-135mm F/3.5-5.6 IS USM

03/07/2007 06:01:31 PM · #6

Did you do an analysis to see how many of those photos include Icelandic northern lights and/or Canon Beach?

soup

Canon EOS-30D

Canon EF 17-40mm f/4L USM

03/07/2007 06:02:14 PM · #7

cut him some slack guys, he obviously spent some time coming up with this in his quest 4truth ;}

Sony SAL-18250 18-250mm f/3.5-6.3 High Magnification Zoom Lens

03/07/2007 06:02:29 PM · #8

Actually I did not ignore challenges that were biased toward one thing or another (i.e. I wouldn't ingore a self portrait challenge full of models nor would I ignore a flower challenge full of flowers). I felt that in order to be random I had to take what I got. The only adjustment to the "random" I made was the random generated created the same numbered challenge twice in which case I increased the number by one and grabbed the next challenge so I'd still have 300 photos.

I have to agree with another comment made here. It's probably a wash. I'd love for someone to do a different comparison to figure out what it is but something tells me it is even more subjective like (use of lighting and composition).

kirbic

Canon EOS R5

Canon RF 24-70 f2.8 L IS

03/07/2007 06:05:45 PM · #9

What would be very interesting is to take a truly random sample from those same challenges and compare the proportions. For instance, if shots with children make up 7.5% of the top ten, and 8% of the photos overall (estimated by taking 8% of a random sample from the same challenge) then we can't say that having a child in the photo either helps or hurts. If, however, there are children in 20% of the shots from that challenge, well, different story. See what I mean?

nadiaC

Nikon D70s

Nikon AF Zoom-Nikkor 28-105mm f/3.5-4.5D

03/07/2007 06:07:37 PM · #10

Originally posted by noraneko:

Did you do an analysis to see how many of those photos include Icelandic northern lights and/or Canon Beach?

BAHHAHHAHAHAHHAHA!

seriously.

:o)

Gordon

RIM BlackBerry Curve

Canon EF 85mm f/1.8 USM

03/07/2007 06:10:53 PM · #11

Originally posted by kirbic:

See what I mean?

Not really, because it still ignores the fundamental question of if the picture was any good or not.

Subject matter certainly can matter, but I think it would be a far secondary effect, after things like is it in focus/ exposed correctly/ correct size/ no noise and all the other pet things that people assume are automatically 'wrong'

Even then, you are still looking at what makes a 5 not a 3 or a 4.

The shots that score over 6 have something going for them other than the subject matter. It can be good lighting, it can be a clever trick. It on occasion can even be an emotive, interesting photograph. But I doubt a simple listing of subject choice is going to ever show anything useful.

Consider a 'rubber ducky' challenge. Does subject choice matter there ?
All the other challenges are equally similar. Each challenge often has many shots with the same idea, same subject matter etc, and those shots score a wide variety of final votes.

because the subject mostly doesn't matter. What you do with it is what counts. This is right up there with the idea that you have to move to Iceland to take winning pictures, or live on Cannon Beach.

You could take 300 typical entrants, fly them all to the same location and still see the same variance in quality of the end result.

Message edited by author 2007-03-07 18:13:16.

yanko

Carl Zeiss 50mm ZE f/1.4 Planar T for Canon

03/07/2007 06:13:16 PM · #12

Originally posted by levyj413:

Kids and flowers, on the other hand, may be tougher to get right.

Especially if you're not Scalvert or Ursula...

Sony SAL-18250 18-250mm f/3.5-6.3 High Magnification Zoom Lens

03/07/2007 06:15:15 PM · #13

Gordon, here is why I think it matters. I've heard a friend of mine tell me that he can't do well at DPC because we just don't have pretty landscapes around here. This random test would demonstrate that he doesn't have to have a pretty landscape to do well at DPC since it's a 50/50 wash one way or another. I've also had a friend tell me that he can't do well on DPC because he can't afford a studio and lights like the big dogs have. Again I say the numbers suggest it doesn't matter that much if you have a studio or not since half the shots are done without props and lights, etc. It SHOULD give people less excuses for why they can't compete and more realization that it has to do with the photographer more than the location in which you live or the money you can afford to spend.

levyj413

Nikon D7100

Tamron SP AF 17-50mm f/2.8 Di II LD Aspherical (IF) for Nikon

03/07/2007 06:25:10 PM · #14

Originally posted by Gordon:

You could take 300 typical entrants, fly them all to the same location and still see the same variance in quality of the end result.

I agree, but only to a point. Some of my highest-scoring shots could've been taken by anyone with a bit of exposure knowledge at that location at that time.

For example, put 300 people at this spot and point them at the Jefferson Memorial, and I'd bet you'd get 300 very similar scores:

Which pretty much tells you why I try my best not to shoot things like that too often - I want to find something that's unique to me.

On the other hand, there's beauty, emotion, ugliness, chaos, serenity, and a million other variations all around all of us, wherever we live.

Message edited by author 2007-03-07 18:25:58.

ursula

Fujifilm X-T2

Fujifilm XF 55-200mm f/3.5-4.8 R LM OIS

03/07/2007 06:37:35 PM · #15

Flowers less than 10%? Bummer. I'm gonna have to change my strategy :(

neophyte

Nikon D810

Nikon AF-S NIKKOR 24-70mm f/2.8G ED

03/07/2007 06:54:40 PM · #16

Originally posted by kirbic:

Amen...

While the data that was analyzed is impressive its only a start and there's too much missing (like what kirbic mentions) to draw any realistic conclusions. cool stuff though (I'm such a nerd at heart but what I can I say, I majored in Economics....)

Gordon

RIM BlackBerry Curve

Canon EF 85mm f/1.8 USM

03/07/2007 07:01:06 PM · #17

Originally posted by lkn4truth:

Yup, we are both saying the same thing. But photographers always use those excuses, until they start to realise good photography doesn't have much to do with the thing being photographed, other than in the most trivial sense.

It takes different people different amounts of time to come to that realisation, usually its about the time that the constant quest for new equipment fades.

Some people never make it.

Sony SAL-18250 18-250mm f/3.5-6.3 High Magnification Zoom Lens

03/07/2007 07:27:14 PM · #18

I was kind of hoping someone would see the flaws in my statistical analysis and become inspired to do their own analysis which would give us something even more interesting to discuss. It doesn't look like anyone is taking the bait though...probably too busy getting a shot for the next entry. In any case, I was curios if there were any trends worth discussing.

gg3rd

Canon EOS-500D Rebel T1i

Sigma 17-70mm F2.8-4 DC Macro OS HSM

03/07/2007 07:33:31 PM · #19

Originally posted by ursula:

Flowers less than 10%? Bummer. I'm gonna have to change my strategy :(

Your strategy seems very good to me, considering your number of ribbons... ;-)

Message edited by author 2007-03-07 19:34:21.

Elvis_L

Canon EF 17-40mm f/4L USM

03/07/2007 07:43:58 PM · #20

Gordon makes some good points. it matters how good it is . voter friendly matters but good overcomes that allot. Joey beat lots of nice shots with a shot of him sawing his foot off. several have ribboned with guns or American flags that people always say voters hate. People say kids score low. My wife has 8 top tens half of those are of children. Scalvert has more ribbons than anyone and many of his are of kids. (i would give the accurate number but don't have the desire to count that high).

howzit

03/07/2007 07:48:21 PM · #21

I think I sense a possible podcast in the works (hint hint) :)

Sony SAL-18250 18-250mm f/3.5-6.3 High Magnification Zoom Lens

03/07/2007 07:54:09 PM · #22

Of course there are people who do really well shooting flowers or really well shooting children but that of course doesn't change the fact that pictures of children or flowers tend not to place in the top 10 as often (that seems to be a statistical fact). The advice you might give that person then is, if you want to have a successful flower shot you should check out ursula and see what she's doing to get such high success with this category or check out scalvert and see what he's doing to achieve such success with children.

I like Joey Lawrence's stuff but as general rule if someone was considering entering something like a person sawing off their foot I would say be warned that unless you do it exceptionaly well it probably won't score well at all. Just because he got a ribbon with a shot like that doesn't mean you are going to have a snowballs chance in hell if you try to duplicate it.

My point is that simply because there are people who have mastered a technique or subject (like flowers or children) it doesn't mean that as a general rule the masses can repeat it.

bmartuch

Nikon D7200

Nikon AF-S DX Nikkor 35mm f/1.8G

03/07/2007 09:38:24 PM · #23

What about people that CONSTANTLY take shots of themselves. That way, everyone knows who it is. I think many people boost the score a little if the photographer is someone popular. JMO.

yanko