Hearthpwn

Lol this is bugging me too mush plz halp DX

Quote from LarryMoments >>

Lol this is bugging me too mush plz halp DX

Quote from Mewdrops >>

Quote from LarryMoments >>

Lol this is bugging me too mush plz halp DX

It's 'Put forth', as in 'put forward', not 'put four of them'.

I think.

Quote from Mewdrops >>

Quote from LarryMoments >>

Lol this is bugging me too mush plz halp DX

It's 'Put forth', as in 'put forward', not 'put four of them'.

I think.

Quote from Viridian >>

Quote from Phoenixfeather >>

Quote from CheeseEtc >>

Quote from Schlauchneid >>

How many votes do you need to get further?

 Let's wait until the comp is advertised on the front page and the subsequent influx of voters which will no doubt completely change the ranking.

By the way, I'm currently doing stats on ALL THE BIG COMPS THAT HAVE EVER BEEN MADE to see to what extent people on the top receive more votes than people on the bottom, and that doesn't look pretty...

 It's an unfortunate circumstance that can't really be changed without completely revolutionizing the voting system as a whole. It's only natural that submissions that near the top get more votes as they're viewed more often. I'm sorry to anyone who ends up being disqualified by this manner.

Fortunately, this appears on the front page soon. That generally brings in a large influx of votes, and should even the odds of the competition.

The top ~26 entries will continue. We'll see about wildcards.

(ab) / c = x is the current formula. a = your votes, b = number of submissions on the page, c = number of total votes on the page

one idea: d = your submission's position on the page (from 1 to 20, always less than or equal to b, excepting page 1, where d is 1 to 19)

((ab) / c )+ d/b = x

that gives you a fixed bonus for being further down the page, the higher up on the page the lower scored you are. You can fiddle with the strength of that...

((ab) / c )+ 0.5(d/b) = x

another idea, or even an additional idea, is to add a variable that tracks which page your submission is on. e = the page your submission is on, f = total number of pages. Starts at 1 and goes up (we don't want 0)

((ab) / c )+ 0.5(d/b) + (e/f) = x

you could also weight this with a fixed number.

I'd say trying all three at once on one competition... figure out which seems most fair (overall you'll have to use some extra research— like what Cheese is doing regarding the upvote pattern down a page and across pages— to figure out how you should modify the weighting of the d/b and e/f ratios), and then implement it for future contests. You'd have to make these 'bonuses'/counterbalances very small however, as otherwise people will try and game the system by posting farther down and along.

Now this I can agree with, and it seems like Sinti's latest idea would do quite well to fix it; submissions deleted and then all simultaneously undeleted. A 'blind submission' period.

Quote from nurgling13 >>

Quote from Viridian >>

Quote from Phoenixfeather >>

Quote from CheeseEtc >>

Quote from Schlauchneid >>

How many votes do you need to get further?

 Let's wait until the comp is advertised on the front page and the subsequent influx of voters which will no doubt completely change the ranking.

By the way, I'm currently doing stats on ALL THE BIG COMPS THAT HAVE EVER BEEN MADE to see to what extent people on the top receive more votes than people on the bottom, and that doesn't look pretty...

 It's an unfortunate circumstance that can't really be changed without completely revolutionizing the voting system as a whole. It's only natural that submissions that near the top get more votes as they're viewed more often. I'm sorry to anyone who ends up being disqualified by this manner.

Fortunately, this appears on the front page soon. That generally brings in a large influx of votes, and should even the odds of the competition.

The top ~26 entries will continue. We'll see about wildcards.

(ab) / c = x is the current formula. a = your votes, b = number of submissions on the page, c = number of total votes on the page

one idea: d = your submission's position on the page (from 1 to 20, always less than or equal to b, excepting page 1, where d is 1 to 19)

((ab) / c )+ d/b = x

that gives you a fixed bonus for being further down the page, the higher up on the page the lower scored you are. You can fiddle with the strength of that...

((ab) / c )+ 0.5(d/b) = x

another idea, or even an additional idea, is to add a variable that tracks which page your submission is on. e = the page your submission is on, f = total number of pages. Starts at 1 and goes up (we don't want 0)

((ab) / c )+ 0.5(d/b) + (e/f) = x

you could also weight this with a fixed number.

I'd say trying all three at once on one competition... figure out which seems most fair (overall you'll have to use some extra research— like what Cheese is doing regarding the upvote pattern down a page and across pages— to figure out how you should modify the weighting of the d/b and e/f ratios), and then implement it for future contests. You'd have to make these 'bonuses'/counterbalances very small however, as otherwise people will try and game the system by posting farther down and along.

You can't use a system like this. You're assuming there is some intrinsic effect to being on the top of the page, but that's not the issue. The issue is how long was there between the time that the person at the top submitted and the person at the bottoms submitted. That varies wildly depending on both the level of participation in the competition and how long the submission page was up. It is exacerbated in the second phase of the large competitions because there are so few participants and the submission page is up for so long.

The best solution is to not have the submission thread up for so long. If you had a week to plan Phase II, but everyone posted in the same 24 period, there would be virtually no difference in upvotes between the tops and bottoms of pages.

Quote from Mewdrops >>

Quote from nurgling13 >>

Quote from Viridian >>

Quote from Phoenixfeather >>

Quote from CheeseEtc >>

Quote from Schlauchneid >>

How many votes do you need to get further?

 Let's wait until the comp is advertised on the front page and the subsequent influx of voters which will no doubt completely change the ranking.

By the way, I'm currently doing stats on ALL THE BIG COMPS THAT HAVE EVER BEEN MADE to see to what extent people on the top receive more votes than people on the bottom, and that doesn't look pretty...

 It's an unfortunate circumstance that can't really be changed without completely revolutionizing the voting system as a whole. It's only natural that submissions that near the top get more votes as they're viewed more often. I'm sorry to anyone who ends up being disqualified by this manner.

Fortunately, this appears on the front page soon. That generally brings in a large influx of votes, and should even the odds of the competition.

The top ~26 entries will continue. We'll see about wildcards.

(ab) / c = x is the current formula. a = your votes, b = number of submissions on the page, c = number of total votes on the page

one idea: d = your submission's position on the page (from 1 to 20, always less than or equal to b, excepting page 1, where d is 1 to 19)

((ab) / c )+ d/b = x

that gives you a fixed bonus for being further down the page, the higher up on the page the lower scored you are. You can fiddle with the strength of that...

((ab) / c )+ 0.5(d/b) = x

another idea, or even an additional idea, is to add a variable that tracks which page your submission is on. e = the page your submission is on, f = total number of pages. Starts at 1 and goes up (we don't want 0)

((ab) / c )+ 0.5(d/b) + (e/f) = x

you could also weight this with a fixed number.

I'd say trying all three at once on one competition... figure out which seems most fair (overall you'll have to use some extra research— like what Cheese is doing regarding the upvote pattern down a page and across pages— to figure out how you should modify the weighting of the d/b and e/f ratios), and then implement it for future contests. You'd have to make these 'bonuses'/counterbalances very small however, as otherwise people will try and game the system by posting farther down and along.

You can't use a system like this. You're assuming there is some intrinsic effect to being on the top of the page, but that's not the issue. The issue is how long was there between the time that the person at the top submitted and the person at the bottoms submitted. That varies wildly depending on both the level of participation in the competition and how long the submission page was up. It is exacerbated in the second phase of the large competitions because there are so few participants and the submission page is up for so long.

The best solution is to not have the submission thread up for so long. If you had a week to plan Phase II, but everyone posted in the same 24 period, there would be virtually no difference in upvotes between the tops and bottoms of pages.

 Actually, I'd argue that there is an intrinsic effect to being on top of the page. You're among the first submissions people see, and therefore the one they pay most attention to. As you scroll through several walls of text to get to, say, the 15th entry, you get kinda bored and stop caring as much, something that may seem 'pretty cool' at the top of the page starts seeming just kinda 'meh'. Especially when each entry is a massive wall of text and images.

Quote from nurgling13 >>

Quote from Mewdrops >>

Quote from nurgling13 >>

Quote from Viridian >>

Quote from Phoenixfeather >>

Quote from CheeseEtc >>

Quote from Schlauchneid >>

How many votes do you need to get further?

 Let's wait until the comp is advertised on the front page and the subsequent influx of voters which will no doubt completely change the ranking.

By the way, I'm currently doing stats on ALL THE BIG COMPS THAT HAVE EVER BEEN MADE to see to what extent people on the top receive more votes than people on the bottom, and that doesn't look pretty...

 It's an unfortunate circumstance that can't really be changed without completely revolutionizing the voting system as a whole. It's only natural that submissions that near the top get more votes as they're viewed more often. I'm sorry to anyone who ends up being disqualified by this manner.

Fortunately, this appears on the front page soon. That generally brings in a large influx of votes, and should even the odds of the competition.

The top ~26 entries will continue. We'll see about wildcards.

(ab) / c = x is the current formula. a = your votes, b = number of submissions on the page, c = number of total votes on the page

one idea: d = your submission's position on the page (from 1 to 20, always less than or equal to b, excepting page 1, where d is 1 to 19)

((ab) / c )+ d/b = x

that gives you a fixed bonus for being further down the page, the higher up on the page the lower scored you are. You can fiddle with the strength of that...

((ab) / c )+ 0.5(d/b) = x

another idea, or even an additional idea, is to add a variable that tracks which page your submission is on. e = the page your submission is on, f = total number of pages. Starts at 1 and goes up (we don't want 0)

((ab) / c )+ 0.5(d/b) + (e/f) = x

you could also weight this with a fixed number.

I'd say trying all three at once on one competition... figure out which seems most fair (overall you'll have to use some extra research— like what Cheese is doing regarding the upvote pattern down a page and across pages— to figure out how you should modify the weighting of the d/b and e/f ratios), and then implement it for future contests. You'd have to make these 'bonuses'/counterbalances very small however, as otherwise people will try and game the system by posting farther down and along.

You can't use a system like this. You're assuming there is some intrinsic effect to being on the top of the page, but that's not the issue. The issue is how long was there between the time that the person at the top submitted and the person at the bottoms submitted. That varies wildly depending on both the level of participation in the competition and how long the submission page was up. It is exacerbated in the second phase of the large competitions because there are so few participants and the submission page is up for so long.

The best solution is to not have the submission thread up for so long. If you had a week to plan Phase II, but everyone posted in the same 24 period, there would be virtually no difference in upvotes between the tops and bottoms of pages.

 Actually, I'd argue that there is an intrinsic effect to being on top of the page. You're among the first submissions people see, and therefore the one they pay most attention to. As you scroll through several walls of text to get to, say, the 15th entry, you get kinda bored and stop caring as much, something that may seem 'pretty cool' at the top of the page starts seeming just kinda 'meh'. Especially when each entry is a massive wall of text and images.

 

I used to think this was true, but I'm pretty sure that effect is actually negligible. Position on the page doesn't seem to have any effect for advancing to Final Polls in the WCDCs. 

Quote from Mewdrops >>

Quote from nurgling13 >>

Quote from Mewdrops >>

Quote from nurgling13 >>

Quote from Viridian >>

Quote from Phoenixfeather >>

Quote from CheeseEtc >>

Quote from Schlauchneid >>

How many votes do you need to get further?

 Let's wait until the comp is advertised on the front page and the subsequent influx of voters which will no doubt completely change the ranking.

By the way, I'm currently doing stats on ALL THE BIG COMPS THAT HAVE EVER BEEN MADE to see to what extent people on the top receive more votes than people on the bottom, and that doesn't look pretty...

 It's an unfortunate circumstance that can't really be changed without completely revolutionizing the voting system as a whole. It's only natural that submissions that near the top get more votes as they're viewed more often. I'm sorry to anyone who ends up being disqualified by this manner.

Fortunately, this appears on the front page soon. That generally brings in a large influx of votes, and should even the odds of the competition.

The top ~26 entries will continue. We'll see about wildcards.

(ab) / c = x is the current formula. a = your votes, b = number of submissions on the page, c = number of total votes on the page

one idea: d = your submission's position on the page (from 1 to 20, always less than or equal to b, excepting page 1, where d is 1 to 19)

((ab) / c )+ d/b = x

that gives you a fixed bonus for being further down the page, the higher up on the page the lower scored you are. You can fiddle with the strength of that...

((ab) / c )+ 0.5(d/b) = x

another idea, or even an additional idea, is to add a variable that tracks which page your submission is on. e = the page your submission is on, f = total number of pages. Starts at 1 and goes up (we don't want 0)

((ab) / c )+ 0.5(d/b) + (e/f) = x

you could also weight this with a fixed number.

I'd say trying all three at once on one competition... figure out which seems most fair (overall you'll have to use some extra research— like what Cheese is doing regarding the upvote pattern down a page and across pages— to figure out how you should modify the weighting of the d/b and e/f ratios), and then implement it for future contests. You'd have to make these 'bonuses'/counterbalances very small however, as otherwise people will try and game the system by posting farther down and along.

You can't use a system like this. You're assuming there is some intrinsic effect to being on the top of the page, but that's not the issue. The issue is how long was there between the time that the person at the top submitted and the person at the bottoms submitted. That varies wildly depending on both the level of participation in the competition and how long the submission page was up. It is exacerbated in the second phase of the large competitions because there are so few participants and the submission page is up for so long.

The best solution is to not have the submission thread up for so long. If you had a week to plan Phase II, but everyone posted in the same 24 period, there would be virtually no difference in upvotes between the tops and bottoms of pages.

 Actually, I'd argue that there is an intrinsic effect to being on top of the page. You're among the first submissions people see, and therefore the one they pay most attention to. As you scroll through several walls of text to get to, say, the 15th entry, you get kinda bored and stop caring as much, something that may seem 'pretty cool' at the top of the page starts seeming just kinda 'meh'. Especially when each entry is a massive wall of text and images.

 

I used to think this was true, but I'm pretty sure that effect is actually negligible. Position on the page doesn't seem to have any effect for advancing to Final Polls in the WCDCs. 

 Yeah, maybe. But I can't help but feel like this effect would be more pronounced when each post is 10 cards + a bunch of other stuff instead of 1 card and a few lines of text.

Finally through the submission topic! Here were my ten favorites for this phase. Just in order of submission, not preference:

Broeck's Illusionist, ercjlee's Loner, NecroaV's Geomancer, Sinti's Gladiator, Turkeybag's Scribe, Tomerick's Jester, PupleMD's General, Cooler's Earthshifter, maxlot's Summoner, and Zanywoop's Berserker

Looking forward to seeing who moves on! =D

Ranking time as of today 1 PM GMT+1.

List without disquals (I believe there's a chance mods will be lenient on the "example cards" detail)

List with disquals taken into account (except the nuts/bolts one because I can't be sure of them)

EDIT: All of this supposing there won't be a score merging of page 2 and 3

Quote from MarioToast >>

Why...why did nobody tell me about this?