Voting Supidity Trends - DPChallenge Forums

Voigtlander Ultron 35mm f/1.7 Aspherical MF Wide Angle - Black

08/12/2011 01:26:25 AM · #1

The last 4 Challenges that I have entered the first vote is always a 4 then 10 views later the image may have risen to 4.3 and at the end of the challenge after 100 + votes 5.5 if the image is lucky.

Based on a scale of 20 votes @ 4 @ 5 @ 6 @ 7 @ 8 the average = 6 after 100 votes

I know you are going to say the 4 voters go back and may bump, to late after 100 votes the mathematical rise is minimal .1 .2 0r .3

Based on the example above if 60 voters rate your image 6 or above.

So how do photos finish 7+ at the end of voting.

My stats are
Votes Cast: 37,363
Avg Vote Cast: 5.4974
Votes Received: 98,665
Avg Vote Received: 5.1719

I have seen profile stats Avg cast 4.5
Avg Rec 6.8

This is not really DPC friendly in MHO.

Paul

Leica SL

08/12/2011 03:11:48 AM · #2

Getting 20 votes at 8 will put you in ribbon territory. Why would you expect flat voting patterns rather than a normal distribution?

Of course there are big differences between average vote given and average vote received for some photographers. Their work is better received.

I don't really understand the point you are trying to make....

BrianR

08/12/2011 06:41:03 AM · #3

Originally posted by paulbtlw:

The point I am trying to say is if can vote 6 7 8 on an image why is there such a flat voting by the remainder of the voters.
And these voters that receive high votes I have noticed those that only give 4's or 5's yet receive high votes do they go through only voting on images they know that DMC or will score poorly.

I notice your out going vote cast is 5.1 but you have received 5.8 in return. with almost the same number of votes cast or received.

Message edited by author 2011-08-12 06:46:04.

Mike

Tamron SP 24-70mm f/2.8 Di VC USD Canon Mount

08/12/2011 06:48:32 AM · #4

you should try your hand at weather prediction, you may have more success.

Apple iPhone 5

Nikon AF-S DX Zoom-Nikkor 18-55mm f/3.5-5.6 ED

08/12/2011 06:52:54 AM · #5

This theory might hold true if every voter voted randomly on every image.

Paul

Leica SL

Voigtlander Ultron 35mm f/1.7 Aspherical MF Wide Angle - Black

08/12/2011 07:21:47 AM · #6

Originally posted by BrianR:

Originally posted by paulbtlw:
Getting 20 votes at 8 will put you in ribbon territory. Why would you expect flat voting patterns rather than a normal distribution?

Of course there are big differences between average vote given and average vote received for some photographers. Their work is better received.

I don't really understand the point you are trying to make....

The point I am trying to say is if can vote 6 7 8 on an image why is there such a flat voting by the remainder of the voters.
And these voters that receive high votes I have noticed those that only give 4's or 5's yet receive high votes do they go through only voting on images they know that DMC or will score poorly.

I notice your out going vote cast is 5.1 but you have received 5.8 in return. with almost the same number of votes cast or received.

Using number of ribbons as measure of 'success' - I currently sit at 38th place among the DPC community. Given that, you'd actually expect my average received to be significantly higher than 5.8; but I mix my entries up with some that I know will be less well received.

Defending my 5.1 given is easy - I don't vote selectively, I vote on entire challenges and I give out more 1-5s than 6-10s. There's a multitude of reasons for that.

08/12/2011 08:16:24 AM · #7

there is no reason to defend 5.1 vote cast. Just look at any challenge and open your eyes. Most images are Avg or worse. If you do not vote selectively ( only the good or the bad in your opinion) 5. something is where you should be..

BrianR

08/12/2011 08:58:13 AM · #8

Originally posted by Basta:

So what you are saying those that vote low early in the voting gives a photo little chance of growing. In a current challenge comments are showing votes and reasons for scoring a 9, unfortunately the photo has no chance of a high score due to the continuous early voting below 5, voting is now over 100 votes, does this mean according to you that the high voters see something the low voters don't???

Marc923

08/12/2011 09:01:36 AM · #9

I think it doesn't matter. There are going to be low voters no matter how good the picture is. Look at the top 3 Lego shots. They all have low (1's, 2's, and 3's) votes.

ShutterRev

08/12/2011 09:39:21 AM · #10

I don't get your point at all Brian. People voting averages are what they are for each individual reason.... and each individual looks at photographs differently. Not everyone put's good photographs in the challenges, and they are rewarded with the results of what everyone else thinks. Some people score all photos in a challenge as honestly as possible... after all this is a challenge not a therapy group where we hold hands and tell each other we are genus, right? Then there are others that seemingly vote only on a few in each challenge and vote really high to keep a high votes given avg so that they don't look bad which I think is damaging because your not giving feedback to those in need of improvement.... everyone is different it's the way it works. Instead of trying to blame your vote received frustration on voters, blame it on your photography, that's what I do for my own. Then you must adapt and overcome, learn how to take photographs that will do well and score high, or don't. But realize that there is no trick science to voting other than we are each individuals looking at a set of pictures through our own filter voting as such.

I guess the general rule of thumb is... if after 100+ people vote individually on YOUR image, and the score is not high.... IT'S NOT THEM. IT'S YOU. And you need to be your own worst critic and try to get better with each challenge, it's what I do.. I don't have ribbons or anything but I am learning and understanding that there are things that I can do better to score higher... AND HAVING FUN DOING IT!

Nick

08/12/2011 10:00:44 AM · #11

Originally posted by BrianR:

Originally posted by Basta:
there is no reason to defend 5.1 vote cast. Just look at any challenge and open your eyes. Most images are Avg or worse. If you do not vote selectively ( only the good or the bad in your opinion) 5. something is where you should be..

So what you are saying those that vote low early in the voting gives a photo little chance of growing. In a current challenge comments are showing votes and reasons for scoring a 9, unfortunately the photo has no chance of a high score due to the continuous early voting below 5, voting is now over 100 votes, does this mean according to you that the high voters see something the low voters don't???

was not the point, but if you think about it. If a voter votes consistently low on every entry in the challenge it will not effect where given image ends up overall. Yes, it may have a .0 something lower avg, and that is about only difference.

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

08/12/2011 10:01:49 AM · #12

Originally posted by BrianR:

So what you are saying those that vote low early in the voting gives a photo little chance of growing....

Actually Brian the early votes are a poor indicator of where an image will finish. The average can swing wildly in the early stages of voting:
How long does it take for an average to stabilize?

Kelli

Canon EOS-D60

08/12/2011 10:10:34 AM · #13

Originally posted by DJWoodward:

Originally posted by BrianR:
So what you are saying those that vote low early in the voting gives a photo little chance of growing....

Actually Brian the early votes are a poor indicator of where an image will finish. The average can swing wildly in the early stages of voting:
How long does it take for an average to stabilize?

I know nothing about statistics (other than that they can be twisted). But, recently I've noticed that a)there hasn't been enough votes on most challenges to get a score that doesn't change much no matter what the vote and b)that my own score, at least, seems to be dropping significant amounts on the last day with most challenges after holding steady for most of the week. In pastels I held a 6.1 the entire week until the last day and ended with a 5.8. In Lego's, I was steady at a 5.8, even through yesterday afternoon when I last looked, but woke up to a 5.6. Now I'm sure they ended up where they belong in percentile ranking, but why such large drops at the end lately?

Voigtlander Ultron 35mm f/1.7 Aspherical MF Wide Angle - Black

08/12/2011 10:33:43 AM · #14

Originally posted by Kelli:

Originally posted by DJWoodward:

Originally posted by BrianR:
So what you are saying those that vote low early in the voting gives a photo little chance of growing....

Actually Brian the early votes are a poor indicator of where an image will finish. The average can swing wildly in the early stages of voting:
How long does it take for an average to stabilize?

I know nothing about statistics (other than that they can be twisted). But, recently I've noticed that a)there hasn't been enough votes on most challenges to get a score that doesn't change much no matter what the vote and b)that my own score, at least, seems to be dropping significant amounts on the last day with most challenges after holding steady for most of the week. In pastels I held a 6.1 the entire week until the last day and ended with a 5.8. In Lego's, I was steady at a 5.8, even through yesterday afternoon when I last looked, but woke up to a 5.6. Now I'm sure they ended up where they belong in percentile ranking, but why such large drops at the end lately?

OMG.. I can see a good conspiracy theory developing... Someone must be steeling all the high votes just before rollover..I can see the headlines " DPC Rollover criminal strikes again"...

and about statistics Mr. Mark Twain said: "There are three kinds of lies: lies, damned lies and statistics."

Paul

Leica SL

08/12/2011 10:34:03 AM · #15

Removed by the scrubber? People cherry picking the images they like during voting but then not doing a full 20%. There are probably more such people than those looking for a few images they don't like just to vote low.

posthumous

Canon EOS-40D

Canon EF 50mm f/1.8 II

08/12/2011 10:46:19 AM · #16

Brian and Paul's difference is a classic Keirsey conflict between an "F" (feeler) and a "T" (thinker).

Brian puts emotional value on the scoring numbers. A low vote is a personal insult.

Paul sees the numbers more mathematically, and so do I. For "T"s, the actual numbers don't matter, as long as we vote consistently.

Because of this fundamental difference in how the numbers are perceived, this argument can go back and forth endlessly, as so many DPC arguments do.

This kind of stuff fascinates me. :)

Kelli

Canon EOS-D60

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

08/12/2011 10:54:34 AM · #17

Originally posted by paulbtlw:

Most likely, because I had very few low votes on either entry. But in the case of the pastels entry, I watched it fall during the last day. It seems that's when I received all of my low votes.

FourPointX

Canon EOS-60D

Canon EF 50mm f/1.8 II

08/12/2011 11:08:09 AM · #18

after only 2.5 years on DPC, i can make the following observations re: voting trends:

they change constantly, from challenge to challenge, from day to day, and there will always be complaints. nearly everyone has seen 4.x early voting turn out to be 5.7 finishes, and seen 7.x early starts drip down to 5.9

with the steady "turnaround" of memebers and/or registered users, i think it's nearly impossible to predict where a voting pattern will emerge on each entry. GENERALLY you'll see your score stay within .4-.5 after the first 20 or so votes.

my own voting pattern can change from challenge to challenge. in the past month most challenges i have voted a 7.x average on. however there have been a couple that i've given a 4.x average vote for.

lastly, you hopefully reach a point where a thoughtful comment, or a favorite, is all you need for a challenge and the voting seems distant. i know after a 4.x finish i'll look and see 3 tens given for that image, and i'll be happy knowing 3 people loved it as much as i did.

Sigma 70-300mm f/4-5.6 DL Macro for Nikon

08/12/2011 12:06:41 PM · #19

Originally posted by FourPointX:

With totally NO reference to your images...Maybe 3 ten's did not like that photo, but knowing it did not have chance they give it 10 to boost given vote Avg? Point is , you will never know without the comment by them...

Nadine_Vb

Nikon D2X

08/12/2011 12:28:16 PM · #20

Originally posted by Basta:

Could be, but I honestly don't think that there are voters out there giving those nice "10-because-otherwise-you'll-get-zip" votes.
Why should there be such voters? If even the best photographs are rewarded with multiple 1 votes so honestly I don't think there are good Samaritans among us doing the opposite.

08/12/2011 01:19:26 PM · #21

Originally posted by Kelli:

Most likely, because I had very few low votes on either entry. But in the case of the pastels entry, I watched it fall during the last day. It seems that's when I received all of my low votes.

Perhaps this (see bold above)is true?

I ran an experiment with your scores from the Pastel challenge. I created a macro that would randomly order the 137 votes in the histogram on your image. I wanted to see if I could randomly recreate a drop from 6.100 to 5.883 (your final score) with those votes. Now I donâ��t know how many votes you received in the last day but I decided to look for the drop to start at the last 20, last 10 and last 5 votes for the simulation.

I let the simulation run ~10,000 times and the biggest random drop I could create was from 6.076 at vote 117 to 5.883 at vote 137 (last 20 votes). This is pretty close but not quite what you described. But even in a randomly generated order the drop can be nearly duplicated. (Perhaps if I had tried starting at the last 25 it would have worked?)

So to force the worse possible scenario I simply ordered all of you votes from highest to lowest. Doing this I could create a drop from 6.110 in just the last 10 votes. But for this to happen, your absolute lowest votes would have to have come at the very end of the voting. I guess this is possible but I think it would be unlikely. No twisted statistics here just simple math :-)

If this drop occurred then I suspect that the vote scrubber took away some high votes or a handful of voters downgraded their votes in the last day exaggerating the drop.

crowis

Canon EOS-60D

Canon EF 50mm f/1.4 USM

08/12/2011 01:26:37 PM · #22

I don't want to add fuel to the fire here, but please let me state one thing about the Pastel Challenge. I track my votes, often on a 1 by 1 basis as they fall. On Tuesday night I was sitting at 6.0000 with 136 votes. When the challeng finished I was at 140 votes with a score of 5.9000. That would indicate that the last four votes of the challenge had an average of 2.5, which seems to be beyond the norm for what one could expect on a 6.0 shot. Not impossible, just not what you'd expect to see. My thoughts on the issue was that one of three things happened on those last four votes. People who had voted earlier, changed their votes. . .The last four voters hated my shot (legitimately or not)....or the averaging program was not 100% before that point--or all of the above. I don't want to cast stones at the voters or immediately step into "cheat" land with it, but, it was just odd.

08/12/2011 01:33:38 PM · #23

Originally posted by crowis:

On Tuesday night I was sitting at 6.0000 with 136 votes. When the challeng finished I was at 140 votes with a score of 5.9000.

How long before rollover were you at 136? Do you know how many votes you had just as rollover began? Was it 140 or were there more that and the scrubber removed some?

Just curious

08/12/2011 01:42:28 PM · #24

Originally posted by crowis:

How do you know some of your higher votes did not change there mind in the end?