Hearthpwn

So many good entries look like they're gonna get axed this round, which is really sad.

I don't know if my entry is counted among the good ones, but it certainly looks like its counted among the axed ones.

Loser's bracket, anyone?

The only completely fair voting system would be to have a panel of judges which vote on / score every entry, and closing the voting process to the public.

That said, I'm kind of considering an unofficial loser's bracket where people do wild card sets instead. That sounds legitimately super fun. If I'm allowed to do such a thing 🤔

Quote from Teknician >>

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.

Quote from Teknician >>

Quote from Sinti >>

Quote from Teknician >>

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.

Im not sure this is the best idea, given the formula "has to" be visible for everyone for the transparency sake. Small bonuses or not, ppl would try to game the system. That just a fact.

Also one thing to note about what Cheese is doing, i dont think he can rly gather any reliable data from the previous competitions. Since i dont think he can possibly account for the quality/likeability of a submission.

So for example when supposedly bad submission is higher on a page and supposedly very good submission is lower on a page, how can you determine, if the votes that each submission have r mostly cause of the position? I know the position does matter, there is no doubt about that, but what im saying is, there is no way to actually determine how much, since you cannot objectively fator in the overall quality of a submission.

 I suppose you're right about how subjective this all is. But saying that revolutionizing the whole voting system is possible basically was just asking for something like this. With this explanation, it really isn't possible is it? lol

On another note, I guess I'm actually going to lose on page 3. Surprised and saddened yet somehow expected. :( Perhaps I should've shown some different cards above the fold.

Quote from CheeseEtc >>

I won't have time to do more analysis, but from what I've found so far a submission at the bottom of a page has on average 60% [more or less depending of what entries and pages I decide to exclude] of the votes of the submission at the top of a page. The worst spot is #15 though (40%), probably chance.

@Sinti: When you take into account lots of pages, the quality of a submission is a given spot will even out so there's no reason to expect that the absolute quality of a submission is correlated to its position on a page. The ONLY way a submission's quality could be correlated to its position is if we expect veteran card makers (who tend to make better submissions) to strategize and wait for a new page to pop to post their entry, but I don't think it's a big factor.

If someone more stat-savvy than me is interested, I can give you the raw data.

Quote from Opal736 >>

Quote from CheeseEtc >>

I won't have time to do more analysis, but from what I've found so far a submission at the bottom of a page has on average 60% [more or less depending of what entries and pages I decide to exclude] of the votes of the submission at the top of a page. The worst spot is #15 though (40%), probably chance.

@Sinti: When you take into account lots of pages, the quality of a submission is a given spot will even out so there's no reason to expect that the absolute quality of a submission is correlated to its position on a page. The ONLY way a submission's quality could be correlated to its position is if we expect veteran card makers (who tend to make better submissions) to strategize and wait for a new page to pop to post their entry, but I don't think it's a big factor.

If someone more stat-savvy than me is interested, I can give you the raw data.

Actually I think making comparisons between the first post and the last ones in a page is not the best way to do it. I'd rather consider the posts in blocks of 5 (4 blocks per page), because it's obvious that the second post can receive the same if not even more upvotes than the first one (which might just be a really bad post, or just an average post (and as Sinti said you don't have enough data to avoid random results)).

Still I think that the position in the page is not the real issue. Maybe the right thing would be considering them day by day, which would give other interesting results (if 30 people were to submit in day 1, then 10 in day 2, it's obvious that the first 10 of the second page will receive more visits).

That said I agree with Sinti that a completely fair method would have to take too many of these factors into account to be fine.

Edit: I was wondering... couldn't you hide everyone's submission until submission time is over and show them all at the same time? Just to know if you have considered the idea and if it's possible for the site.

Quote from Opal736 >>

Quote from Sinti >>

Quote from Opal736 >>

Quote from CheeseEtc >>

I won't have time to do more analysis, but from what I've found so far a submission at the bottom of a page has on average 60% [more or less depending of what entries and pages I decide to exclude] of the votes of the submission at the top of a page. The worst spot is #15 though (40%), probably chance.

@Sinti: When you take into account lots of pages, the quality of a submission is a given spot will even out so there's no reason to expect that the absolute quality of a submission is correlated to its position on a page. The ONLY way a submission's quality could be correlated to its position is if we expect veteran card makers (who tend to make better submissions) to strategize and wait for a new page to pop to post their entry, but I don't think it's a big factor.

If someone more stat-savvy than me is interested, I can give you the raw data.

Actually I think making comparisons between the first post and the last ones in a page is not the best way to do it. I'd rather consider the posts in blocks of 5 (4 blocks per page), because it's obvious that the second post can receive the same if not even more upvotes than the first one (which might just be a really bad post, or just an average post (and as Sinti said you don't have enough data to avoid random results)).

Still I think that the position in the page is not the real issue. Maybe the right thing would be considering them day by day, which would give other interesting results (if 30 people were to submit in day 1, then 10 in day 2, it's obvious that the first 10 of the second page will receive more visits).

That said I agree with Sinti that a completely fair method would have to take too many of these factors into account to be fine.

Edit: I was wondering... couldn't you hide everyone's submission until submission time is over and show them all at the same time? Just to know if you have considered the idea and if it's possible for the site.

I mean we could technically do that. Tho from Phase III the upvotes wont matter, there will be poll topic(s) to determine who advances to next stages. So we can talk options, but it would be most likely implemented only in the next big comp.

One variant could be to have a (pre)submission thread and every time a person would submit their post, we would move it to a "hidden" thread in our staff section where you wouldnt be able to see it and after the submission period was over, we would move it back. Problem is it would be a lot of extra work and more importantly i dont think we can delete upvotes, so if someone "by accident" upvoted an entry before we would be able to move it, that person would get an unfair advantage. Also given how long this stage usually lasts (about a week), having "nothing" in a submission thread might give a wrong idea to potential voters.

Other variant could be to just send your cards to a designated mod(s) and after submission period was over a thread would go up with all the submission posted by a mod all at once. Again, this would be a LOT of extra work. Same thing here about the wrong idea that voters could get from there being no submission thread for a week and then bam! one for 2-3 days to upvote stuff and then poof! and nothing for a week again. Dunno.

With that said, we could lower the amount of work (for the mod(s) in question in the second variant) by having a standardized format of submissions, which is something i was goin to talk about with FC mods to consider for the next CCC anyway, since i feel like whoever can photoshop or format their post better has extra advantage. You know very well that if you see a horribly formated post with walls of text etc. you often dont even bother looking over the cards or reading anything.

I mean presentation skills r definitely nice, but at the same time, if you dont have them, you shouldnt be penalized for that and being at a disadvantage just cause u r bad at using this forum's tools. One could argue that it's everyone's own fault by being "lazy" and not givin their submission post enough attention, but i dont know, feels like this particular part shouldnt be the main reason why your submission fails. Ppl do often help others by creating banners and other stuff for them, but that is still not ideal imho, its not fair to ask them to put even more work into the comp by helping the less fortunate up the quality of their submission post to a "new standard", even if they offer to do it themselves.

So what i was thinking about is that until the finals, there would be a standardized format for submissions where all would "look the same" and in finals where each finalist makes their own thread thats where there would be no restrictions and they could sparkle it up as much as they would want to. You can tell me if you think this is a total BS, but it's one of the things i thought would create a more of a fair play environment.

It could also to some extent help with some other stuff, like ppl not including their 0/10-cost card as one of their example cards etc. I mean i think that in a first place, we should have fun creating our classes, but at the same time, this is a competition after all and if we have rules, we should adhere to them and not forgive everyone who breaks them in the end anyway. My point is, i dont want anyone to be disqualified on a technicality, but it feels wrong to just let everything slide as well. Just my opinion. (not talking about anything specific here, just in general)

No, actually I think that your idea would be completely fair. It certainly would decrease the originality you can put in your submission (the way you shape it and so on...), but well, even those who used to put a lot of care in their posts will just have more time to think of their actual submission as they don't have to bother about the looks. And for those who have old generation devices, who already have troubles when they try to write their post (10 minutes to type a word, or well even opening the site is a pain sometimes) it's one less thing to worry about.

And well, as long as you can customize your entry in the finals I guess it's fine, all in all I hope something like that is implemented. Making it easier to see whether you've satisfied all the challenges or not is also a valid point.

I see. And here I thought it was easier to do what I was talking about. There are some sites in which mods can just choose to make all the posts hidden for a certain thread and I was wondering if that option was available here too. I think we could try finding a solution that doesn't involve mods having to do all that work. Maybe there could be a bot to do all the cut/paste part? But again I don't know how much can be implemented on this site.

Quote from Sinti >>

Quote from Opal736 >>

Quote from Sinti >>

Quote from Opal736 >>

Quote from CheeseEtc >>

I won't have time to do more analysis, but from what I've found so far a submission at the bottom of a page has on average 60% [more or less depending of what entries and pages I decide to exclude] of the votes of the submission at the top of a page. The worst spot is #15 though (40%), probably chance.

@Sinti: When you take into account lots of pages, the quality of a submission is a given spot will even out so there's no reason to expect that the absolute quality of a submission is correlated to its position on a page. The ONLY way a submission's quality could be correlated to its position is if we expect veteran card makers (who tend to make better submissions) to strategize and wait for a new page to pop to post their entry, but I don't think it's a big factor.

If someone more stat-savvy than me is interested, I can give you the raw data.

Actually I think making comparisons between the first post and the last ones in a page is not the best way to do it. I'd rather consider the posts in blocks of 5 (4 blocks per page), because it's obvious that the second post can receive the same if not even more upvotes than the first one (which might just be a really bad post, or just an average post (and as Sinti said you don't have enough data to avoid random results)).

Still I think that the position in the page is not the real issue. Maybe the right thing would be considering them day by day, which would give other interesting results (if 30 people were to submit in day 1, then 10 in day 2, it's obvious that the first 10 of the second page will receive more visits).

That said I agree with Sinti that a completely fair method would have to take too many of these factors into account to be fine.

Edit: I was wondering... couldn't you hide everyone's submission until submission time is over and show them all at the same time? Just to know if you have considered the idea and if it's possible for the site.

I mean we could technically do that. Tho from Phase III the upvotes wont matter, there will be poll topic(s) to determine who advances to next stages. So we can talk options, but it would be most likely implemented only in the next big comp.

One variant could be to have a (pre)submission thread and every time a person would submit their post, we would move it to a "hidden" thread in our staff section where you wouldnt be able to see it and after the submission period was over, we would move it back. Problem is it would be a lot of extra work and more importantly i dont think we can delete upvotes, so if someone "by accident" upvoted an entry before we would be able to move it, that person would get an unfair advantage. Also given how long this stage usually lasts (about a week), having "nothing" in a submission thread might give a wrong idea to potential voters.

Other variant could be to just send your cards to a designated mod(s) and after submission period was over a thread would go up with all the submission posted by a mod all at once. Again, this would be a LOT of extra work. Same thing here about the wrong idea that voters could get from there being no submission thread for a week and then bam! one for 2-3 days to upvote stuff and then poof! and nothing for a week again. Dunno.

With that said, we could lower the amount of work (for the mod(s) in question in the second variant) by having a standardized format of submissions, which is something i was goin to talk about with FC mods to consider for the next CCC anyway, since i feel like whoever can photoshop or format their post better has extra advantage. You know very well that if you see a horribly formated post with walls of text etc. you often dont even bother looking over the cards or reading anything.

I mean presentation skills r definitely nice, but at the same time, if you dont have them, you shouldnt be penalized for that and being at a disadvantage just cause u r bad at using this forum's tools. One could argue that it's everyone's own fault by being "lazy" and not givin their submission post enough attention, but i dont know, feels like this particular part shouldnt be the main reason why your submission fails. Ppl do often help others by creating banners and other stuff for them, but that is still not ideal imho, its not fair to ask them to put even more work into the comp by helping the less fortunate up the quality of their submission post to a "new standard", even if they offer to do it themselves.

So what i was thinking about is that until the finals, there would be a standardized format for submissions where all would "look the same" and in finals where each finalist makes their own thread thats where there would be no restrictions and they could sparkle it up as much as they would want to. You can tell me if you think this is a total BS, but it's one of the things i thought would create a more of a fair play environment.

It could also to some extent help with some other stuff, like ppl not including their 0/10-cost card as one of their example cards etc. I mean i think that in a first place, we should have fun creating our classes, but at the same time, this is a competition after all and if we have rules, we should adhere to them and not forgive everyone who breaks them in the end anyway. My point is, i dont want anyone to be disqualified on a technicality, but it feels wrong to just let everything slide as well. Just my opinion. (not talking about anything specific here, just in general)

No, actually I think that your idea would be completely fair. It certainly would decrease the originality you can put in your submission (the way you shape it and so on...), but well, even those who used to put a lot of care in their posts will just have more time to think of their actual submission as they don't have to bother about the looks. And for those who have old generation devices, who already have troubles when they try to write their post (10 minutes to type a word, or well even opening the site is a pain sometimes) it's one less thing to worry about.

And well, as long as you can customize your entry in the finals I guess it's fine, all in all I hope something like that is implemented. Making it easier to see whether you've satisfied all the challenges or not is also a valid point.

I see. And here I thought it was easier to do what I was talking about. There are some sites in which mods can just choose to make all the posts hidden for a certain thread and I was wondering if that option was available here too. I think we could try finding a solution that doesn't involve mods having to do all that work. Maybe there could be a bot to do all the cut/paste part? But again I don't know how much can be implemented on this site.

Actually, thinking about it now, i suppose we could just delete the entries and then undelete them, lol. That might work.