| Author | Thread |
|
|
12/10/2003 01:16:04 AM · #1 |
Did anyone else's score get a hammering in the last few hours of the voting ('Money' in particular)?
Mine went from 6.1453 when I woke up this morning (7 hours before voting closed) to 6.0751 just before closing. This is quite a big drop, considering there was over 230 votes. Thankfully, it jumped back up to 6.174 with the throw out of the <20% / voting pattern cleanout.
Anyone else cop a hammering before closing?
|
|
|
|
12/10/2003 01:27:10 AM · #2 |
Heck, mine got hammered from the first day ;) My second lowest score.
Actually, many people like me do a last minute "re-eval" of photos at the end. I did, and in fact, I raised your score 1 point in the last 30 minutes of the challenge. So it's just last minute ratings. (I always look at the field "relatively" at the end, and make final "grades" on a curve.) |
|
|
|
12/10/2003 02:08:18 AM · #3 |
Originally posted by nshapiro:
Actually, many people like me do a last minute "re-eval" of photos at the end. I did, and in fact, I raised your score 1 point in the last 30 minutes of the challenge. So it's just last minute ratings. (I always look at the field "relatively" at the end, and make final "grades" on a curve.) |
I think I saw your increase of 1 in the last 30 minutes (yep, I'm one of those stats junkies who checks every few mins and calculates everything!).
However, this was more that the votes in the last few hours averaged less than the previous votes.
Eg: The drop I mentioned, from 6.1453 to 6.0751 over the last few hours, was over 19 votes, and in those 19 votes, the average was 5.2. This is quite a low average in quite a few votes. If you consider that some people, such as yourself, may have increased their votes in the last hour, the actual 19 votes may have been lower still.
Maybe it's people tending to rush, so they can get all their votes in? Who knows.
|
|
|
|
12/10/2003 02:13:12 AM · #4 |
I don't if my score was effected in the last few hours, but it did jump .6 points in 4 days though.
|
|
|
|
12/10/2003 05:27:46 AM · #5 |
mine never ever go up at the end of voting! 20% throw out? Never happened to me ;(
|
|
|
|
12/10/2003 09:34:16 AM · #6 |
Speaking of scores... I'm a bit surprised at how people are manipulating their scores. I'm guessing they either have multiple accounts or have a circle of friends who all vote on each other's pics. Proof? Well, look at the last place money shots (insert Paula Abdul here).
10s? Really. I can't comprehend who would give those a 10 except for the photographer him/herself or their buddies.
Oh well, I guess it's bound to happen, I just find it really tasteless. |
|
|
|
12/10/2003 10:52:08 AM · #7 |
Originally posted by Ten13:
Speaking of scores... I'm a bit surprised at how people are manipulating their scores. I'm guessing they either have multiple accounts or have a circle of friends who all vote on each other's pics. Proof? Well, look at the last place money shots (insert Paula Abdul here).
10s? Really. I can't comprehend who would give those a 10 except for the photographer him/herself or their buddies.
Oh well, I guess it's bound to happen, I just find it really tasteless. |
That a few people do not share your particular esthetic sense makes them neither idiots nor cheaters/manipulators.
PLEASE refrain from throwing out non-specific and unnamed general accusations -- I find it insulting and demeaning to the whole site.
If you have SPECIFIC KNOWLEDGE of cheating or vote manipulating, please report it to the admins. If you do not, to idly speculate that "people" are cheating is wrong and destructive to the community here.
We do not offer a fixed voting criteria, therefore no one's voting criteria can be "wrong." The site already has measures in place to detect vote manipulation. Your "guesses" imply not only cheating by the members, but incompetence by the site administrators. I suggest, as the saying goes, guess again. |
|
|
|
12/10/2003 01:25:09 PM · #8 |
Originally posted by GeneralE:
We do not offer a fixed voting criteria, therefore no one's voting criteria can be "wrong." |
Actually the site does....
"While voting, users are asked to keep in highest consideration the topic of the challenge and base their rating accordingly."
I think what Ten13 is getting at is that the photos that contained currency didn't meet the challenge and therefore shouldn't have been given a 10 as we are told keep in highest consideration the topic. The challenge rules reasonable imply that a photo that doesn't meet the challenge can't be "perfect". |
|
|
|
12/10/2003 02:36:30 PM · #9 |
Originally posted by GeneralE:
That a few people do not share your particular esthetic sense makes them neither idiots nor cheaters/manipulators. |
Just for the record, "Idiots" and "incompetence by the site administrators" is your verbiage, not mine.
"The site already has measures in place to detect vote manipulation."
"We do not offer a fixed voting criteria, therefore no one's voting criteria can be "wrong."
Aren't these 2 statements contradictory? If I vote on all of the images in the challenge as 1s, would they be thrown out? If nobody's voting criteria can be wrong, then that's not a valid statement.
I apologize if I offended the Admins or anyone else, that was not my intention. I was just surprised that winning images would get 1s and last place images would get 10s (fairly consistently) from anyone voting truthfully on this site.
I guess some voters see images much differently than the majority at times.
Again, sorry to have offended.
Message edited by author 2003-12-10 14:39:47. |
|
|
|
12/10/2003 03:31:46 PM · #10 |
My photo took a jump up at the end of about .2 and surprised me a bit.
How are the votes from voters with less than 20% handled? Are they added in during the run and removed at the end or are a persons votes held back from the averaging until they reach 20%?
If the former, that could explain some of the large jumps at the end of the challenge.
Also, I know that when I am finished going through all pictures, I go back and make a few small adjustments in comarison voting. Usually before the last day though.
|
|
|
|
12/10/2003 04:13:53 PM · #11 |
A voter's criteria for applying any particular vote to any particular photo is considered a private matter between the voter and their esthetic conscience. However, a PATTERN of voting showing no relationship between any criteria and the votes, and seemingly done for the sole purpose of influencing the outcome, can be considered and dealt with. I think it's a subtle but valid distinction which prevents my earlier statements from being mutually contradictory.
And what voting guidelines there are (the scale) only labels 10 as "good" and not as "perfect" -- winning photos with ones and losing photos with tens are not evidence of cheating or vote manipulation.
For the record, no one but the admins (not even SC members) knows just how the vote-examining/purging algorithm works. |
|
|
|
12/10/2003 04:24:01 PM · #12 |
Originally posted by GeneralE:
.....And what voting guidelines there are (the scale) only labels 10 as "good" and not as "perfect"..... |
Actually.....From the Challenge Rules located here:
"Users should rate each and every photograph in the challenge on a scale of 1 to 10 (10 being a perfect photograph)." |
|
|
|
12/10/2003 04:45:29 PM · #13 |
| Well, I just hope I can recognize perfection when I see it then. |
|
|
|
12/10/2003 04:49:14 PM · #14 |
Originally posted by GeneralE:
...snip...
For the record, no one but the admins (not even SC members) knows just how the vote-examining/purging algorithm works. |
So do you know of any instances in which it was invoked?
How was it "validated" as working properly
Is it still effective, given the changes to the way people vote (you can see that by walking through the history of challenges)
In other words, how do you know it is really working (said the software engineer and scientist in me)?
Seems to me there should be statistics posted so everyone can see, not the specifics, but the "effectiveness" of it. Then perhaps people won't complain about it. Those statistics could be automatically generated and part of the overview of the challenge statistics.
While I'm on that, some other things that might help:
1) Standard Deviations along with the individual and population averages
2) I think the breakdown of votes from users with cameras versus no is not very interesting. I think it would be more useful to see the votes from people who have entered a photo, and those that hadn't. That should prove very interesting!
Edited: To clarify, I mean, entered a photo into the challenge being scored...
Regards--Neil
Message edited by author 2003-12-10 16:50:47. |
|
|
|
12/10/2003 05:42:24 PM · #15 |
Originally posted by sleekr:
Eg: The drop I mentioned, from 6.1453 to 6.0751 over the last few hours, was over 19 votes, and in those 19 votes, the average was 5.2. This is quite a low average in quite a few votes. If you consider that some people, such as yourself, may have increased their votes in the last hour, the actual 19 votes may have been lower still.
Maybe it's people tending to rush, so they can get all their votes in? Who knows. |
This isn't much different than (as often happens to me) checking your score after 20 votes and having a 6.5, then waking up the next morning to a 5.0. That there were 19 votes averaging only 1 point lower than your final score seems like a pretty reasonable variation. I wouldn't be surprised if you went back and crunched your numbers (I'm an on-and-off vote logger myself), you might find a spot somewhere in there where you got 20 votes in a row that averaged 7 or above.
For myself on the money challenge, my score settled in pretty early around 4.8 and stayed constant most of the week, so I lost interest in tracking towards the end of the week. So I don't know if I took any dips - though I doubt it since I actually ended up about a tenth of a point up from where I had been for the last couple of days. I have had last day dips like you're describing in past challenges. While they are frustrating, I haven't seen any trends though. |
|
|
|
12/10/2003 06:08:13 PM · #16 |
Originally posted by Ten13:
Speaking of scores... I'm a bit surprised at how people are manipulating their scores. I'm guessing they either have multiple accounts or have a circle of friends who all vote on each other's pics. Proof? Well, look at the last place money shots (insert Paula Abdul here).
10s? Really. I can't comprehend who would give those a 10 except for the photographer him/herself or their buddies.
Oh well, I guess it's bound to happen, I just find it really tasteless. |
A couple of possible explanations: Perhaps some voters had the same problem as most of those last place finishers - they didn't read the description, and therefore didn't know about the limitation on currency. Perhaps there was no colusion, but there was some sympathy from one person who entered currency for other entries that fell into the same trap. Or perhaps someone was just sympathetic to currency shots because of the (mild) beating they took in the forums and gave them all a 10.
Or, closer to your line of thinking, maybe in fact a couple of folks know each other and give each others' photos automatic 10s. Based on those last place money shots, I don't think any of them got more than a couple of 10s. This affects the final score by maybe .05 of a point. Yeah, I suppose that could make a difference of a place one way every now and then. But it's not going to make a mediocre shot a winner, so it doesn't seem worth getting too worked up about, especially since you can never really prove anything anyway. In the end, maybe someone, for some reason, just really loved those shots. Go figure.
Oh yeah, and one other possibility - there are some who are regular contributers who've made it clear that they give no weight for or against a photo based on meeting the challenge. They view the challenge topic as having the sole purpose of spurring creativity. They feel that once the voting starts, the topic becomes irrelevent. I personally don't agree with this, but the fact is there are some who vote that way, and it's their perrogative to do so. Maybe some of those folks were the ones who gave these shots high marks. |
|
|
|
12/10/2003 06:09:03 PM · #17 |
Originally posted by nshapiro:
Originally posted by GeneralE:
...snip...
For the record, no one but the admins (not even SC members) knows just how the vote-examining/purging algorithm works. |
So do you know of any instances in which it was invoked?
How was it "validated" as working properly
Is it still effective, given the changes to the way people vote (you can see that by walking through the history of challenges)
In other words, how do you know it is really working (said the software engineer and scientist in me)? |
I was trying to answer this when we had a momentary power outage ... the basic answer is that it's essentially free and fun and educational, so the site admins consider the current system adequate, and since they own the site they won't change it unless they want to.
If we were paying entry fees and competing for valuable prizes, I think we'd be more deserving of some kind of independent audit, but as it is, I feel fine about trusting whatever D+L are doing. |
|
|
|
12/10/2003 06:12:58 PM · #18 |
| Also, my son Isaac hasn't been voting in quite a while, but when he did he pretty much would tend to exercise a six-year old's tendency to see things as polar opposites -- he'd vote a lot of ones and tens, and had a hard time interpolating and articulating the in-betweens. |
|
|
|
12/10/2003 06:15:06 PM · #19 |
Originally posted by ScottK:
That there were 19 votes averaging only 1 point lower than your final score seems like a pretty reasonable variation. I wouldn't be surprised if you went back and crunched your numbers (I'm an on-and-off vote logger myself), you might find a spot somewhere in there where you got 20 votes in a row that averaged 7 or above.
|
I disagree that a full 1 point variation over 19 votes is reasonable - 19 votes is enough to consider that the average for these votes should be close to the average overall. Sure, if two or three of these votes were 1's, it would push the average down, but that just validates my point - that the late votes seemed to be low.
Oh, and I did crunch the numbers - where I could trace 19 or 20 votes in a row (and thats most of the time), the highest average for those was about 6.5 - less than 1/2 a point above overall average).
|
|
|
|
12/10/2003 07:09:25 PM · #20 |
| Of course this is entirely anecdotal, but I went and found one of my entries (the Insect challenge) which ended with a score of 6.7 with about 150 votes. I was able to find one stretch of 15 votes where I got an average score of 7.9 - a 1.2 point difference from the total. That's a fairly similar sampling (about 10%) to the 19 vote span you had. So its not out of the question that a 1 point difference within a 10% span of the voting can occur - in either direction. |
|
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: 08/08/2025 02:02:14 AM EDT.
