Why, hello there. I am so happy to meet you!
Allow me to introduce myself. I am the new Worth1000. Is this your first time meeting the new me? We should get reacquainted then. So many things are different about me now. Come and learn more :).
I always check the vote histories to see if I was on target. In this contest I had a .06 deviation and got FOUR karma points. I thought the max was 3? I also saw some people that had a greater deviation than me (but still good) that got 6 points. What's up with that?
I also checked that out and I didnt get any karma. I was not humping or think i voted unfairly. Whats up with that?
I had a deviation of 0.12 and Boofhead had a deviation of 0.01 and even he didnt get any karma. V177ain had the same deviation as myslef (0.12)and yet had a karma increase of 5. Am i doing something wrong please help, I am confused
I didn't have any karma change either and I was 0.046 deviation. I can't see that that could be right that my karma didn't change. Unless I'm not voting as well as I thought I was, but I'm pretty sure that's not right either.
I didn't vote any differently then normal so this would suggest my voting patterns haven't been on target :( :( :(
Link to the contest in question and we'll help you figure out how to better align things with the voting scale.
http://fx.worth1000.com/vote-history/24171/critter-superheroes-3
I noticed those with 6 karma changes were off in deviation (as much as 0.6) but used a greater range of the scale. It's not necessarily what let's say I would have thought the entry (being a 3 versus a 4) deserved, but the deviation in the past as least indicated to me that I was along the lines of the general populace in terms of where the average lay
The new system seems to suggest that I intentionally vote an entry or two higher or lower so that my karma reflects a greater use of the scale
If the new karma system is made to give 0 karma to the bumpers then your karma change obviously can't be decided by the overall deviation;)
(If the overall deviation decided the change then people could continue voting only 5's and have a karma raise)
I know the knew Karma system is on trial at the moment - has the FXB "Hot Rod" contest the correct Karma, according to the new system? When comparing votes. I can see a voter with almost exactly the same vote as me, except I have four more '5' votes, they have four more '7' votes. They have a +4 Karma, I have no change. Seems a little odd.
that implies that most contests - which end with middle of the road averages will give no one karma but those that splat their votes all over the place will be rewarded regardless of whether it adheres to what the majority thing. Karma humpers are not the norm really. I really don't think there are many intentional ones.
The fact is, most entries do fall within 4-6 with a number over that, and the occasional one beneath. The average will always fall in that area.
People have opinions. I don't think we should be punished for it. I might look at a chop and see things from a vantage point that another does not. WHat my be a 6 to you cold be in my opinion an 8 for it's execution, or vice versa.
Conformity to single idea of what is good is not in itself a great idea
Let's be clear about something: there is no punishment here. We just want to encourage people to use more of the scale. Voting heavily in the middle of the scale has almost the same statistical impact as not voting at all. We hope the new system will encourage you to spread your votes out a bit more (in a good faith way). There is less punishment, more reward for doing that now. 5 is the enemy. :)
It seems to me to now require a quota for each contest. Certain number of 2's, 3's, etc. I could see the point of it that there may be more needed separation than what we are giving, but if it's sole purpose is to punish humpers, I think it's doing more punishing than what is deserved? Are there safeguards to prevent new karma humping by just spreading the vote throughout the scale?
I absolutely love the fact that low-ballers and karma-humpers won't get that much karma gain, if any, but I think maybe the max karma requirement seems a little harsh. I know that I have max karma in fx but it still says I would've gained 4 karma points by the way I voted. I used the scale from 3 through 9. So does that mean that I would have to vote at least one 1 or 2 and/or a 10 to get 6 karma points? Like I said though, I have the max so it doesn't matter too much for me in fx but some people might think they'd need to vote 1's and 2's so they could use more of the scale to get the max amount of karma given.
Look at the Beginner Hot Rod contest and how many experienced voters got ZERO. I voted on a range of 4-7 which I thought was pretty fair considering the entries had a deviation of .04 and got a zero for my effort.
You're over thinking it. The point is not to figure out how many 2's or 3's you need to vote to gain karma. Just vote fairly according to the scale. Apply the scale more thoughtfully. Instead of just slapping a 5 on it, view full to see if you can award it a little more or maybe that one could have used a little more work.
And don't compare your karma gain to some "normal." The old gain of 3 is still in effect - that is "normal." Gains above that are earned by special circumstances. Not every contest will have the quality range of entries that would earn you 6 points - those extra 3 are for voting broadly and fairly on those contests which do.
And I have to take exception with:
The fact is, most entries do fall within 4-6 with a number over that, and the occasional one beneath.
The average entry may fall in that range but the average does not define MOST entries. If you are voting 90% of time a 4, 5 or 6 - you are karma humping. 90% of entries are not average. Some are better, some are not as good, but in a contest with 40 entries, if 36 of your votes are in the 4 to 6 range - take a good look at your voting. (I mean this as a pattern, not the voting in a specific contest - the RARE contest may have entry distribution like this).
There are not many contest I don't view full on each entry. I think the fine tuned voting being asked of us can only come through experience and expertise. I can see the point of getting people to do that (view full and take more time) but I think the truth is many entries may fall in that 5-6 average score?
Ok= I'm going to go out on a limb here and suggest something- uh- different. Currently the statistical occurrance of a normal bell curve has meant that it's quite easy for some less ethical people to "cheat" the system- and i'd like to suggest that spreading the scores around will be adapted to quite quickly. The problem is not so much the voting system- but the vote cheats who are ruining it for everyone else.
So what's the solution?
I'm going to suggest that each "voter" organise the competition in the order they believe the final result will be. So move a bad entry down the list.- raise a good one up. Sure in comps with 80+ entries- that's kinda insane... but you can't just "hump it"- sure you could just leave it as is- but I'd suggest there'd be alot of variation between a "left" vote and a properly thought about one. Each position on the scale would have a vertain amount of points assigned to it- for example- Top of the list might be worth 20 points in a 20 entry comp. This should give most points to the entry that most people place at the top. So how does karma fit in here? Make it a multiplier of pointsas it is now. So a top entry worth 20 points from a voter with 100 karma is worth 2000 points.
Averages are based on your position in the final scale- so the top of the scale actually recieves a 10- crazy perhaps- but that seems to be where it's headed now anyway- a great diversity.
While there are certainly problems I've not worked out yet- this is as close as i can think of to "unhumpable". Each entry that is "out of place" is given points of variation depending on how far out you are- and depending on how many points you depends on whether or not you get karma.
Anywho- just a thought :)
[edit] Just thinking more about averages- Perhaps it could be worked out as percentage of total points by your entry. Thus to get a 10- you would need 100% of the possible points available- aka- every single voter put you at the top of the list- if however your total point value was only 82% of the possible max points- you've just scored an 8.2. Thus the scale is consistant over any size comp as it's worked out on total vote points available in the comp.
I voted between 3 and 7 with a pretty even distribution and also ended up with zero karma change. Considering 7 of my votes were 7's I'm unclear how it would have been considering humping. If you look at the numbers, it's not even a bell curve of votes.
You mentioned how experienced voters are getting zero. Experienced voters should be the ones the karma system should be based on - assuming they've been around long enough and know enough about chopping. still think 'the masses have spoken' counts for more then an algorithm to see if you fit a formula
You've got former jurors getting 0 change or at least less then six. I think that should indicate a problem in the calculation
Qofcheez said
Considering 7 of my votes were 7's I'm unclear how it would have been considering humping. If you look at the numbers, it's not even a bell curve of votes.
The 7 votes were 5s in the chart I looked at.
This was your break down. Am I reading this incorrectly?
0 0 0 3 7 2 3 1 0 0
In beginner hot rod:
0 0 1 3 6 5 7 0 0 0
I got zero change and that doesn't look like a hump to me at least
the one you quoted was for critter superheros
