DPChallenge: A Digital Photography Contest You are not logged in. (log in or register) 
 

DPChallenge Forums >> Challenge Results >> Voting down others to improve your rank
Pages:  
Showing posts 26 - 50 of 80, (reverse)
AuthorThread
 05/07/2008 01:04:15 PM · #26
Originally posted by PapaBob:
Congrats on your photo and your honesty when voting and keeping your integrety intact. Do I personally think some people vote unfairly yes, do I want to win a ribbon that way no! Anyone who wins a ribbon by cheating will always know they are not good enough to win any other way and to me that has to be a hollow victory.


What papabob say's is absolutely right, what is the point in cheating?! I've just won a blue ribbon in the sepia challenge, but I requested to have my image unsubmitted a couple of day's prior to the end of voting, even though I knew I was a dead cert for a ribbon! Apparently I have to sel-disqualify myself which I've just finished doing! I want to win as much as the next person, but on merit and not on sculduggery - what's the bloody point???

As regards the voting, why dont dpc introduce a system whereby the lowest and highest 5-10% of votes are excluded, thereby removing these anomalies! Its seems the only logical way of dealing with this problem!

05/07/2008 01:07:45 PM · #27
People cheat all the time. For every person who asks "Why?" there's one that asks "Why not?". I have noticed lately that there is a bunch of really negative voters out there attacking images like a pack of dogs and dragging good images into the dirt. Case in point is the recent Sepia challenge:

Average Score: 5.18502
Highest Score: 7.142
Median Score: 5.2154
Lowest Score: 2.8659

There was a lot of good images that got slammed and a lot of shots that were scored below a five which I cannot fathom. The site average here is 5.5 and the average for this challenge was 5.18 That might not seem to be a lot but given the large number of entries in that challenge (385) the average should have been a lot closer to normal. It should also be pointed out that the top image received 169 votes chich isn't even half of the number of entries. (Most challenges I think I get about 200 votes typically).

I'm not saying that there is overt cheating going on but something smells awfully fishy to me.

edit: grammar

Message edited by author 2008-05-07 13:09:19.
 05/07/2008 01:08:49 PM · #28
no worries,cheaters never win no matter what bottom line,and although the challenges are fun and exciting its the comments and the loads of experience that we can learn from here that are really important...
 05/07/2008 01:09:57 PM · #29
I know we all assume that is why the low vote happened but keep in mind there are times when there are reasons someone voted it lower than we feel it should have been. I know I personnally gave an entry a 1 once and it ended up winning the challenge, my reasoning was not great but at the time I felt strongly the image did not meet the challenge, the shot was a fantastic bug shot (I hate bug shots) in a rain challenge but the rain looked like mist not rain or else the bug had to be huge. I changed my vote several times from a 5 to a 1 debating which way to go and ended up giving it a 1. In hind sight a 5 would be the vote I would use now based on the criteria I use now. Long story short do not assume that is the only reason for the low vote, sometimes people have reasons other than self interest.
 05/07/2008 01:17:33 PM · #30
Originally posted by PapaBob:
I know we all assume that is why the low vote happened but keep in mind there are times when there are reasons someone voted it lower than we feel it should have been. I know I personnally gave an entry a 1 once and it ended up winning the challenge, my reasoning was not great but at the time I felt strongly the image did not meet the challenge, the shot was a fantastic bug shot (I hate bug shots) in a rain challenge but the rain looked like mist not rain or else the bug had to be huge. I changed my vote several times from a 5 to a 1 debating which way to go and ended up giving it a 1. In hind sight a 5 would be the vote I would use now based on the criteria I use now. Long story short do not assume that is the only reason for the low vote, sometimes people have reasons other than self interest.


I might agree when a single shot gets a few low scores but when a bunch get slammed with an inordinate number of them? Go to page 20 of the Sepia challenge and look at a few images. I personally don't see any that deserved votes below 4. However there is a large number of them with 2 and 3's. And not just few either.

People vote how they want and that's fine but when I start wondering if I am playing ping pong on a hill...(weird analogy I know). Anyways, I should leave this alone before I start wondering what other conspiracy theories I can dream up.

Message edited by author 2008-05-07 13:30:25.
 05/07/2008 01:25:35 PM · #31
its slippys squirrels im tellin ya...
 05/07/2008 01:28:06 PM · #32
I agree with PapaBob, I think people have different criteria and it doesnt necesarily mean they're trying to hurt others intentionally, or trying to get a higher place in the challenge. I personally try to be fair, but I accept that I'm very tough when judging. At the same time, I usually know when my photos arent as good as they should be, and I'm happy with the results.
05/07/2008 01:32:20 PM · #33
Originally posted by Moose408:
Interestingly I had a bunch of low votes right before rollover. All week my score was 7.3xxx until 15 minutes before rollover when it dropped to 7.2657 due to several really low scores. Normally my photos go up in the last 20-30 minutes of a challenge.

As I said no big deal but I find it interesting.


I have noticed this as well, I had a 6.8ish score drop to a 6.725 in the last 15 minutes, I think the trollers do this to avoid having their votes discounted. I know the SC disregards people who only use 1, 2, or 3. But if they do this at the last minute they will get counted. The only solution I can think of is an hour delay from end of voting until announcement. That way someone can delete the trolls votes.
 05/07/2008 01:39:40 PM · #34
Originally posted by AndyMac24:
... I think the trollers do this to avoid having their votes discounted. I know the SC disregards people who only use 1, 2, or 3. But if they do this at the last minute they will get counted.

Are you sure this is true? I thought the vote scrubber worked automatically, after the vote closes at midnight EST.
 05/07/2008 01:42:11 PM · #35
Originally posted by FourTDSean:
I didnt believe it at first but after watching 15-20 consecutive votes that lowered my score, I know it is true. I just feel bad for people that feel the need to win a free contest that badly. Hell let them win if that is what they need to be a better photographer.


How does seeing 15-20 consecutive low votes show that some kind of cheating is happening? How in the world would 20 cheaters know to get on at the same time and vote your image down? Why would it matter if they all voted you down at the same time? Couldn't they all just do it any time during the challenge?

Now... do I think some people deliberately vote others down? Yes. Do I think it makes much difference in the rankings of a challenge? No. As the example above shows, the only place it really makes a difference is when the top 3 or 4 are very close in score. I believe it would be VERY EASY to implement a automatic check to find the most likely suspect for such actions: High scoring members who want to win very badly. Because I'd bet the members at the top of the DPC heap are also the very best at selecting the top 5 or 10 photos in any given challenge. Something you'd have to be able to do to affect the scoring by voting them down. So in reality to materially affect your score in a challenge you'd already have to be pretty damn good!

Do I believe there are high ranked DPC members doing this type of thing? No.

But then I remembered Rikki.
 05/07/2008 01:48:59 PM · #36
I believe there some who vote their top three picks 10s and everyone else 1s.

Also there are some who are overly critical with no merit in photogrpahy at all. The list goes on and on.
 05/07/2008 01:54:48 PM · #37
Originally posted by JaimeVinas:
I believe there some who vote their top three picks 10s and everyone else 1s.



If the vote scrubber doesn't already detect this it should be implemented at once. I suspect it already checks for this.
 05/07/2008 01:57:50 PM · #38
Originally posted by JaimeVinas:
... The list goes on and on.

Exactly. That's the nature of this particular beast. If you want different standards for scoring, you might have to try a different type of site. We all know (or soon learn) the pros and cons of DPC voting methods.
 05/07/2008 02:01:41 PM · #39
ill admit i normally dont score as high as i do when i dont have an entry in the challenge, but scoring drasically lower is another story, thats just wrong
 05/07/2008 02:09:17 PM · #40
Originally posted by Citadel:
I might agree when a single shot gets a few low scores but when a bunch get slammed with an inordinate number of them? Go to page 20 of the Sepia challenge and look at a few images. I personally don't see any that deserved votes below 4. However there is a large number of them with 2 and 3's. And not just few either.

I agree with you! I went back and looked at my scoring on that page, and I voted 4 fours, 3 fives, and 3 sixes. I don't understand it either.
 05/07/2008 02:11:33 PM · #41
Originally posted by alfresco:
Originally posted by AperturePriority:
It's a simple solution. Langdon should implement logic where you can not vote in challenges in which you are entered.


A simple solution is to not let people see their own score while voting is in place.


Hey - you're still out there Good to see you posting JP. Now post some new pics in your portfolio :P

I agree with JP on this. It has always been my feeling that you should not be able to see your score until YOU have completed your voting. Now this does not mean you have to vote on every entry. If we had a "show my score" button, once you click it you could see your score, but no longer be able to vote (and unable to change votes already cast).

Message edited by author 2008-05-07 14:14:53.
 05/07/2008 02:32:28 PM · #42
Originally posted by fir3bird:
Originally posted by FourTDSean:
I didnt believe it at first but after watching 15-20 consecutive votes that lowered my score, I know it is true. I just feel bad for people that feel the need to win a free contest that badly. Hell let them win if that is what they need to be a better photographer.


How does seeing 15-20 consecutive low votes show that some kind of cheating is happening? How in the world would 20 cheaters know to get on at the same time and vote your image down? Why would it matter if they all voted you down at the same time? Couldn't they all just do it any time during the challenge?

Now... do I think some people deliberately vote others down? Yes. Do I think it makes much difference in the rankings of a challenge? No. As the example above shows, the only place it really makes a difference is when the top 3 or 4 are very close in score. I believe it would be VERY EASY to implement a automatic check to find the most likely suspect for such actions: High scoring members who want to win very badly. Because I'd bet the members at the top of the DPC heap are also the very best at selecting the top 5 or 10 photos in any given challenge. Something you'd have to be able to do to affect the scoring by voting them down. So in reality to materially affect your score in a challenge you'd already have to be pretty damn good!

Do I believe there are high ranked DPC members doing this type of thing? No.

But then I remembered Rikki.


Actually the answer to how this is deliberate is in statistics...
20 votes ago I had 120 votes and my score shortly before this had been oscillating by about .05
at 30 votes, I may have a bloated score of 8 and most of my scores from there on will be low. (or vice versa)
By around 120 votes, my score was pretty much static, regardless of new votes, and there had been no trend up or down since around vote 60 or so.
one could assume that at around 120 votes, I reached my compensation score and the amount of positive vs negative votes should be about in balance.
So the weight of the 120 votes should by then support the current score and the votes shouldn't fall to one side or the other.

If anyone is still following this, so the model predicts that upon reaching compensation score, total + vs - votes will continue to be equal unless...... acted upon by a force. OR another very simple way to say it..... the more votes you have, the more stable your vote should be.

This force is in my case 16 votes in a row below my score of "6"
the model predicts that the each vote has a 50% chance of being higher. we should see approximately 8 pos, and 8 neg votes. Instead there are 16 neg votes. which brings me to my final conclusion....

this is the probability for 16 negative votes in a row to occur after reaching compensation score
the prob for first neg vote was: 1/2, for it to happen 2x:1/4. 3X:1/8, 4x:1/16 5x:1/32 6x:1/64 7x:1/128...... 16x:1/65536 so the probability that 16 consecutive votes will be all negative when they should have been 50/50 is 1 out of 65,536

I may be a nerd but I am not that unlucky.

This was developed using probability theorem, and modeled after the power series and productivity compensation depth and i did this in about 5 mins so I apologize for any errors I have.
 05/07/2008 02:38:53 PM · #43
Originally posted by FourTDSean:
Originally posted by fir3bird:
Originally posted by FourTDSean:
I didnt believe it at first but after watching 15-20 consecutive votes that lowered my score, I know it is true. I just feel bad for people that feel the need to win a free contest that badly. Hell let them win if that is what they need to be a better photographer.


How does seeing 15-20 consecutive low votes show that some kind of cheating is happening? How in the world would 20 cheaters know to get on at the same time and vote your image down? Why would it matter if they all voted you down at the same time? Couldn't they all just do it any time during the challenge?

Now... do I think some people deliberately vote others down? Yes. Do I think it makes much difference in the rankings of a challenge? No. As the example above shows, the only place it really makes a difference is when the top 3 or 4 are very close in score. I believe it would be VERY EASY to implement a automatic check to find the most likely suspect for such actions: High scoring members who want to win very badly. Because I'd bet the members at the top of the DPC heap are also the very best at selecting the top 5 or 10 photos in any given challenge. Something you'd have to be able to do to affect the scoring by voting them down. So in reality to materially affect your score in a challenge you'd already have to be pretty damn good!

Do I believe there are high ranked DPC members doing this type of thing? No.

But then I remembered Rikki.


Actually the answer to how this is deliberate is in statistics...
20 votes ago I had 120 votes and my score shortly before this had been oscillating by about .05
at 30 votes, I may have a bloated score of 8 and most of my scores from there on will be low. (or vice versa)
By around 120 votes, my score was pretty much static, regardless of new votes, and there had been no trend up or down since around vote 60 or so.
one could assume that at around 120 votes, I reached my compensation score and the amount of positive vs negative votes should be about in balance.
So the weight of the 120 votes should by then support the current score and the votes shouldn't fall to one side or the other.

If anyone is still following this, so the model predicts that upon reaching compensation score, total + vs - votes will continue to be equal unless...... acted upon by a force. OR another very simple way to say it..... the more votes you have, the more stable your vote should be.

This force is in my case 16 votes in a row below my score of "6"
the model predicts that the each vote has a 50% chance of being higher. we should see approximately 8 pos, and 8 neg votes. Instead there are 16 neg votes. which brings me to my final conclusion....

this is the probability for 16 negative votes in a row to occur after reaching compensation score
the prob for first neg vote was: 1/2, for it to happen 2x:1/4. 3X:1/8, 4x:1/16 5x:1/32 6x:1/64 7x:1/128...... 16x:1/65536 so the probability that 16 consecutive votes will be all negative when they should have been 50/50 is 1 out of 65,536

I may be a nerd but I am not that unlucky.

This was developed using probability theorem, and modeled after the power series and productivity compensation depth and i did this in about 5 mins so I apologize for any errors I have.
dude,you dont need a camera,you need the Hubble Telescope,lol jk.:)meaning your a scientist/mathmatician

Message edited by author 2008-05-07 14:39:45.
 05/07/2008 02:42:23 PM · #44
actually I'm a biologist
 05/07/2008 02:43:54 PM · #45
Originally posted by FourTDSean:
actually I'm a biologist
kool!
 05/07/2008 02:49:34 PM · #46
hmm...my thought as an engineer is that is flawed logic. Since the person voting has no idea what your score is, your rank, or anything about you other than the photo....you have an equal probability of getting any score at any given time.

So no matter if it's after 100 votes or 1000 votes. Your probability of getting any given number is 1/10. Keeping in mind that this isn't like a die where there is no thinking involved. It's much harder to classify any real probability based off thinking or feeling.

The only thing your score does is if you were guessing (from your end) you'd have a higher chance of guessing what a person gave you, but it doesn't work in reverse....ie...if your score were a 5.5, you could make an educated guess that you'd receive a 5 or 6 on the next vote...but it doesn't change the probability of you getting a 2.

okay I'm just really bored right now......
 05/07/2008 02:54:39 PM · #47
I have noticed that there is a tendency for the low votes and high votes on a given entry to come in groups. My assumption is there is something in common between the folks that like or dislike certain types of images.

My initial thought is that it may be based on geographic location. For instance, the members in Europe may have a tendency to prefer certain styles, and those in North America have different tastes. Most of the voting will be going on during evenings and lunch hours, hence you get large groups of similar votes in a short span. Just a theory.
 05/07/2008 02:57:22 PM · #48
Originally posted by citymars:
Originally posted by AndyMac24:
... I think the trollers do this to avoid having their votes discounted. I know the SC disregards people who only use 1, 2, or 3. But if they do this at the last minute they will get counted.

Are you sure this is true? I thought the vote scrubber worked automatically, after the vote closes at midnight EST.


citymars, you are correct. It is an automated process that happens after voting closes and before the results are posted.
 05/07/2008 03:06:57 PM · #49
Originally posted by yospiff:
I have noticed that there is a tendency for the low votes and high votes on a given entry to come in groups. My assumption is there is something in common between the folks that like or dislike certain types of images.

My initial thought is that it may be based on geographic location. For instance, the members in Europe may have a tendency to prefer certain styles, and those in North America have different tastes. Most of the voting will be going on during evenings and lunch hours, hence you get large groups of similar votes in a short span. Just a theory.


That is something I have wondered about, maybe if I can develop an international style.......hmmmmmmm
 05/07/2008 04:00:56 PM · #50
I didn't explain that too well I guess, so I will use bigger numbers. And you will know this to be similar to Alternating series or convergent power series.

IF PEOPLE ARE ACTUALLY VOTING and voting with honesty
All estimates below
With the first vote, a 5 is chosen
the 2nd vote brings your total to 7
the 3rd vote total equals 7.33.......... early on, predictions on score are futile, but as early as the 15th to 20th vote,most already know within one point of where they will be come final
at 20 votes your score is 7.1, and we can estimate roughly your final score will fall between 6.1 and 8.1
around 50 votes your score may or may not have changed,7.4, but you know with more confidence that your final perse +/- .1 pts
By 100 votes, we will say you have a vote of 7.5, and chances are that your final will be within +/- .01
at the hypothetical 200 votes, the score is still 7.5, but now your final should be with in +/-.001
If 1000 votes were honest your score would still be 7.5 but the final will be +/- .0001 or so
Why is this..... its the same reason we predict who is president with less than 1% of the voters polled. Statistically an accurate score is known usually well before it voting is over using only a porting of total pop. This would be due to people on average valuing the photo similarly. theoretically your score will level out.

If the voter was blind, yes we would have to calculate this as all scores having same probability, but being that we all can judge, or most of us at least, some of those numbers are selected for more than others based on quality.
the average should have settled after alternating and at some pt become semi static. to break the alternating series from static convergence would take repeated scores to one side.

In a final scenario to bring my horrible writing together, if you skipped the above READ THIS
after 120 votes in FR, I estimated that my score of 6.5 was static and would not deviate much.
(MAY WANT TO SKIP: how calculated if 6.5 is set a 0 and all of the votes are plotted as + or - from mid point 0, one will find mass on either side of axis is equal. Both sides of line are of equal weight but different ranges on number line so since the range is smaller above 6.5, there would actually have to be more votes over 6.5. But score was no longer changing and for simplicity, It was declared static at 120 votes)
approx 60 of the 120 votes where above 6.5
and 60 of the 120 votes where below 6.5
all this says is that

around some large # votes, the changes in your score should ultimately equal 0.
At that point your score becomes static.
The more votes obtained, the less likely it is to have a string of one sided votes.
while 5 votes in a row is probable
22 votes one way this far into voting seems to be statistically imposible and seems to point to 22 non conventional consecutive votes to purposely lower scores. But Im not even good at math, this is just an idea.
Pages:  
Current Server Time: 06/01/2025 06:45:13 PM

Please log in or register to post to the forums.


Home - Challenges - Community - League - Photos - Cameras - Lenses - Learn - Help - Terms of Use - Privacy - Top ^
DPChallenge, and website content and design, Copyright © 2001-2025 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Current Server Time: 06/01/2025 06:45:13 PM EDT.