Thoughts on the public voting and discussion


RPG Superstar™ General Discussion

1 to 50 of 62 << first < prev | 1 | 2 | next > last >>
RPG Superstar 2012 Top 16 , Marathon Voter Season 6, Marathon Voter Season 7, Dedicated Voter Season 8

I'm curious as to how the public voting period is going to play out here on the messageboards. Especially if individual items are going to be discussed. Right now, there don't seem to be any limitations on what we can say about the items we have seen during voting.

I can easily see myself going trough the voting booth again and again, until it stops showing me new items. And likely make my own "Keep" folder in google docs or evernote where I copy/paste my favorite items to see if my choises correspond with the judges.

I figure the debate ranging from broad term talk about popular themes among the items, over "Wow, that Everfull Ladle of Awesomesauce sure was neat!" to people doing full Spicer-style line-by-line reviews of selected items.

There is of course also the risk of contestants getting their friends to hype their submission in the hopes of boosting votes.

What do you all think?

Paizo Employee Chief Technical Officer

We'll be publishing rules and a voting FAQ prior to voting that will address some of those things.

But please feel free to share thoughts on this topic—you might come up with something he haven't thought of yet!


I'm wondering how it will be significant. It seems to me the number of votes would have to be exponentially larger to cover even a middling amount of permutations. Has anyone done the math?

RPG Superstar 2010 Top 32, 2011 Top 4 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Marathon Voter Season 9 aka DankeSean

2 people marked this as a favorite.

As far as voting goes, I plan to try and be hardcore about it and keep refreshing the link until I've gone through every. single. item.

I do plan to copy items to a personal doc for longer-term review than the fairly brief time I'll have when deciding between two items. Partly because I expect I'll have to make some hard choices (two great items that I'll have to decide between, for instance.) along the way and after the fact I'll be curious to see which of my almost-picks made it to the top 32. And partly because I'm obsessive.

I don't know how much I personally want to engage in ongoing commentary on the various items while they're being voted on, though. It just feels a bit weird to publicly debate the merits of something while the anonymous authors are out there cringing inside as every tiny flaw and error gets scrutinized. Plus I'm not sure how much real effect discussion on the items will have; since anyone who can knowledgeably discuss an entry has presumably seen that entry and already voted. It's not like the usual judge chamber talk where the judges actually debate the merits, change each others minds, and actually shuffle items around between the reject bin and the keep folder. We'll have seen the items and already hit 'keep' or 'reject' on them and there ain't no going back.

I suppose one effect of public discussion is that there could be certain 'hot' items that become buzzworthy and have voters eagerly looking forward to seeing them pop up on their screens. Not sure if that's a benefit or drawback, though, since it might artificially inflate the hype of certain items just because they happen to get the luck of the draw early in the voting process with some of the more vocal posters.

Don't get me wrong; I'm interested to see whatever kind of dialogue opens up during voting. Just not sure how much I see myself dialoguing in the process.

RPG Superstar 2009 Top 16, 2012 Top 32 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Marathon Voter Season 9 aka Epic Meepo

1 person marked this as a favorite.
Sean McGowan wrote:
...anyone who can knowledgeably discuss an entry has presumably seen that entry and already voted...

You're assuming people will be reading entries for the first time when they enter the voting booth. I could easily imagine a circumstance where some voters gain access to the full list of entries and decide on favorites long before they visit the voting booth.

We're talking about the internet, here. It's just a matter of time before someone decides it would be fun to collect the full text of every entry in the contest and post it somewhere online. Convincing people the list was genuine would probably require more time and effort than actually compiling the list in the first place.

RPG Superstar Season 9 Top 32, RPG Superstar 2014 Top 16, RPG Superstar 2013 Top 16 , Marathon Voter Season 6, Dedicated Voter Season 7, Dedicated Voter Season 8, Star Voter Season 9 aka RainyDayNinja

Sean McGowan wrote:
I suppose one effect of public discussion is that there could be certain 'hot' items that become buzzworthy and have voters eagerly looking forward to seeing them pop up on their screens. Not sure if that's a benefit or drawback, though, since it might artificially inflate the hype of certain items just because they happen to get the luck of the draw early in the voting process with some of the more vocal posters.

I heard of a study, in which people were asked to use a website to rank how well they liked certain songs. One group had a social media aspect to the website, where they could see how other people had rated the songs, while the control group did their rating in a (figurative) vacuum. When it was over, and they looked at the top 10 from each group, it turned out only about half of them were the same. Food for thought.

Grand Lodge RPG Superstar 2012 Top 32 , Dedicated Voter Season 6, Marathon Voter Season 7, Dedicated Voter Season 8, Star Voter Season 9

I've heard about that study, too, and that's why I've always been a bit cricital of the exit poll threads. "Popularity breeds popularity". Open discussion could potentially lead to a situation where people with friends who have good lobbying skills are more likely to get into the top 32.

On the other hand, I really do want to discuss really cool items, or at least post my "top 32" if it's allowed. If not, I'll just copy-paste my favorites in a file to see how many make it to the top 32. It's a good way to improve your "spot mojo" skill.

RPG Superstar 2012 Top 16 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Dedicated Voter Season 9 aka motteditor

I think after voting is done, there's no reason not to have people post a Top 32 thread, though I imagine very few will see every item. Like Sean, I plan to do my best to get to all of them if only to make sure my item gets at least one vote. : )

I imagine it could also turn into a sort of Critique My Item thread, for those that don't make the top 32, since there'll be no judges' comments for us to ask for for most of the items.

While voting's going on? I feel like that's a little dodgier...

RPG Superstar 2012 Top 16 , Dedicated Voter Season 6, Star Voter Season 7

3 people marked this as a favorite.

Personally, I intend to treat this like rounds 2-5 from previous years. That is, as a competitor it's bad form for me to discuss or comment on other competitors' submissions before voting ends.

As soon as the votes are in, I'm all for as much discussion as possible.

Star Voter Season 6, Dedicated Voter Season 7, Star Voter Season 8

Since all entries are owned by Paizo, if any entries show up in third-party products--either through plagiarism or resubmission--what would happen?

If we see an item we designed show up in a third-party publication, what's the process of bringing that to Paizo's attention, and would it be realistic to expect any action to be taken?

RPG Superstar 2013 Top 32 , Dedicated Voter Season 6, Star Voter Season 7 aka Standback

I wanted to add that I'm very impressed with this twist to the contest. I think it's a brilliant idea that acheives several worthy goals at once - reduces load on the judges, gives voice to community opinion, shakes up the contest, shows everybody the general body of submissions, and occupies the forum until the Top 32 are announced :) Neil also referred to the possibility of offering a new Open Call task next year if this works out, since crowdsourcing the winnowing process might let Paizo choose something that takes a few a larger wordcount. All in all, very cool.

I'm also very pleased at the catch made, correcting the original statement that the Top 32 would be selected from the Top 64 ranked. I think that's an important and necessary safeguard until we've seen how the public voting pans out.

One suggestion I might have for future years would be to consider two-phase voting - the first phase voting on all entries, the second only on the top 10% (as determined by the first phase) (for "10%," substitute any percentage of your choice). I expect that at the moment, the voting process will primarily seperate entries into clear strata - rough layers where all the items rank "pretty much the same". For example, the group of "Really Awesome Items" will almost always get votes vs. any "lesser" item; whereas the ranking within this group will be much less significant.

So there might be value to voting once the field has been winnowed down some, and voters can give more attention to voting between really good items, and not just seperating between the really-good and the not-as-good. (Obviously, under the current setup, the chance of voting on two items which are both in the top 10% will only be about 1%.)

On the other hand, the rough separation might be more useful than a full community ranking of the top items. Also, a two-phase voting scheme as I described it would necessarily tip your hand as to which entries are still in the running... But I think it might be worth considering. (A weighting mechanism, where the "top" entries feature more frequently as voting proceeds, would have similar advantages.)

Dark Archive RPG Superstar 2012 Top 16 , Star Voter Season 6, Dedicated Voter Season 7, Dedicated Voter Season 8 aka FaxCelestis

2 people marked this as a favorite.
Jacob Trier wrote:
Everfull Ladle of Awesomesauce

Yessssss.

*makes notes for next year's entry*

RPG Superstar 2012 Top 16 , Dedicated Voter Season 6, Star Voter Season 7

1 person marked this as a favorite.
Garrett Guillotte wrote:

Since all entries are owned by Paizo, if any entries show up in third-party products--either through plagiarism or resubmission--what would happen?

If we see an item we designed show up in a third-party publication, what's the process of bringing that to Paizo's attention, and would it be realistic to expect any action to be taken?

Garrett, if you have genuine concerns over abuse of Paizo's IP then you could PM or email Vic Wertz. As for taking action, I'm sure they consider a large number of factors, but it has happened before.

Marathon Voter Season 6, Marathon Voter Season 7, Star Voter Season 8

what will happen if by chance we are given two atrocious objects where in we don't think either should advance... are we going to be forced to 'like' one of them anyway (choosing the lesser of two evils so to speak) or are we going to get a none of the above option.

Same question but reversed what if we get two amazing items, both of which ooze superstar? are we going to be forced to pick one over the other, even though both might be way better than any item we have seen in the last dozen or so pairings we were given to vote on?

Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Dedicated Voter Season 9

Adventure Path Charter Subscriber; Pathfinder Rulebook, Starfinder Adventure Path, Starfinder Maps, Starfinder Roleplaying Game Subscriber
cwslyclgh wrote:

what will happen if by chance we are given two atrocious objects where in we don't think either should advance... are we going to be forced to 'like' one of them anyway (choosing the lesser of two evils so to speak) or are we going to get a none of the above option.

I think in this case you just refresh the voting page to get your next two items. You will only get to see each item once during the voting period, assuming your refresh enough times :)

At least, that is my understanding from how the proposed system should work.

Marathon Voter Season 6, Marathon Voter Season 7, Star Voter Season 8

1 person marked this as a favorite.

I have blocked out Thursday for voting (the whole day) so I am hopeful I can get through them all.

RPG Superstar 2013 Top 16 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Dedicated Voter Season 9 aka Darkjoy

cwslyclgh wrote:

what will happen if by chance we are given two atrocious objects where in we don't think either should advance... are we going to be forced to 'like' one of them anyway (choosing the lesser of two evils so to speak) or are we going to get a none of the above option.

Same question but reversed what if we get two amazing items, both of which ooze superstar? are we going to be forced to pick one over the other, even though both might be way better than any item we have seen in the last dozen or so pairings we were given to vote on?

This is my fear as well, from experience (the critique my item threads) I know there will be a slew of sub-par items.

I want to vote for great items, I can even make a choice between two great items, one will always be better. I don't want to vote between two horrible items. I hope the F5 option is available.

Dark Archive RPG Superstar 2012 Top 16 , Star Voter Season 6, Dedicated Voter Season 7, Dedicated Voter Season 8 aka FaxCelestis

My fear is finding two excellent items at the same time, wanting to keep both, and not being able to do so.

Scarab Sages Contributor , Star Voter Season 6

James Raine wrote:
My fear is finding two excellent items at the same time, wanting to keep both, and not being able to do so.

I'm also wondering about this. Does anyone know how the items will be paired? I'm assuming at random, but maybe there will be a chronological or alphabetized aspect to it?

RPG Superstar 2013 Top 4 , Marathon Voter Season 6, Star Voter Season 7

I can be mistaken, but for this kind of voting system to work, it's important to vote in all the items you can, to give the system more data when compairing and ranking the items. Even if they are both terrible.

I mean, with enough votes, all items will be eventually paired against each other (hopefully more than once) and the average preference wil prevail; the more data acquired, the more accurate the outcome.

Because we are more accostumed to ranking things on our own, this system may seem a little different - after all, if someone used the results of his individual votes, he would not end up with any kind of rank at all, only with hundreds of "keep"/"reject" pairs. So, obviously, if he looked at his choices, there would be items that got a "keep" that are worse than others that were "rejected", due to the random pairings. But when all paired votes are analyzed together, this anomaly disappears (because the system can then compare how item X did against every other one, in average).

RPG Superstar 2013 Top 16 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Dedicated Voter Season 9 aka Darkjoy

1 person marked this as a favorite.

I understand the system, but I don't want to vote for crap.

Anyway, the system should / will also record that a pair was shown but recieved no votes. I don't know how Paizo will deal with this scenario, but it is a very real possibility.

RPG Superstar 2009 Top 16, 2012 Top 32 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Marathon Voter Season 9 aka Epic Meepo

4 people marked this as a favorite.
James Raine wrote:
My fear is finding two excellent items at the same time, wanting to keep both, and not being able to do so.

You're not voting to keep one and reject the other. You're voting to determine the order in which the items are presented to the judges.

If you get two items you dislike, just pick one at random. Your choice between two terrible items won't lead to either one getting kept. It'll just determine the order in which those items are placed at the bottom of the barrel, far from the sight of the judges.

And if you get two awesome items, just pick the one that you think is slightly better. The one you think is second best won't get rejected. It'll just be placed slightly lower down in the list of items that go straight to the judges.

Every possible combination of items will be assigned at random to different voters. For every pair of terrible items you see, a hundred other people are seeing each of those same items paired up against awesome items, and is voting the terrible items down. For every pair of awesome items you see, a hundred other people are seeing those same items paired up with terrible items, and is voting the awesome items up.

Scarab Sages Contributor , Star Voter Season 6

Eric, thanks for clearing that up!

RPG Superstar 2013 Top 4 , Marathon Voter Season 6, Star Voter Season 7

Hey, even crappy items deserve some loving. Behind them, there's still someone trying. I promise to read them all. ;)

RPG Superstar 2013 Top 4 , Marathon Voter Season 6, Star Voter Season 7

Eric Morton wrote:
For every pair of terrible items you see, a hundred other people are seeing each of those same items paired up against awesome items, and is voting the terrible items down. For every pair of awesome items you see, a hundred other people are seeing those same items paired up with terrible items, and is voting the awesome items up.

...that was much more didactic.

Paizo Employee PostMonster General

1 person marked this as a favorite.

Items will be selected randomly, biased toward items which have been viewed the least.

We haven't 100% decided on the labeling for the buttons, but you'll be given three options: Prefer this one, prefer that one, or prefer neither.

There will also be the ability to flag an item for breaking the rules. Items that get flagged for breaking the rules will be reviewed by the judges and if they determine that an item does in fact break the rules it will be removed from further consideration.

RPG Superstar 2013 Top 16 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Dedicated Voter Season 9 aka Darkjoy

Gary Teter wrote:

Items will be selected randomly, biased toward items which have been viewed the least.

We haven't 100% decided on the labeling for the buttons, but you'll be given three options: Prefer this one, prefer that one, or prefer neither.

There will also be the ability to flag an item for breaking the rules. Items that get flagged for breaking the rules will be reviewed by the judges and if they determine that an item does in fact break the rules it will be removed from further consideration.

Cool, like the neither option.


I’m a little concerned over the significance of the data set from a public vote. Not that I don’t trust people…I do…it is just that a single pairing system will force voters to be very prolific. I mean if there are just 500 entries…a lowball I assume…that would mean there are approx 124000 (assuming no repetitions) permutations (my math might be a little off—this is going back 20 years). If you limit the data exposure to one view per voter that only allows a single voter 250 votes: meaning you would need 496 people to vote 250 times (or 1240 people to vote 100 times…or more likely …12400 people to vote 10 times) to get a single set of permutations.

Now, I know, you aren’t looking for a full set of permutations because you don’t need it to grade (think of sports tournaments); but, through the randomization, you will seriously skew the caliber of submission through the lack of permutations. If you assume that only 10% of submissions are of short-listing caliber any lack of permutation would increase the possibility of short-listing. Meaning, there is a greater chance of poorer quality items being shortlisted and better quality items being eliminated (goes both ways). You are essentially changing the goal-posts…I can’t do the math without specifics; but, it would be similar to saying instead of any one item having a 10% chance of being shortlisted they now have a 10-15% chance of being shortlisted: not because of the lessening of standards (or ratio…that always stays the same—and probably the reason why this was pitched) but because of the significance of the data. Many models are quite happy to assume a +/- 5% variation…but without at least one set of permutations I’m not sure this model can even get close to 5%.

(Just for another comparison if there are 1300 entries there are 844000 permutations; meaning 1298 people would have to vote 650 times; or 8440 people would have to vote 100 times; 84400 people would have to vote 10 times; for just a single set of permutations)

Thoughts? Am I missing something? My stats class is approx 20 years old…so I’m more than happy to hear where I messed up.

Dark Archive RPG Superstar 2012 Top 16 , Star Voter Season 6, Dedicated Voter Season 7, Dedicated Voter Season 8 aka FaxCelestis

I think you're underestimating the determination of gamer nerds, for starters. I also think you're missing the point: it doesn't matter specifically what items are ranked against each other, just that they are ranked. If an item is preferred over one, it gets a +1 to its score, and the other is given -1. "Prefer neither" would give a +0 to both. Final score would then determine the order of presentation, with items that are largely loved being shown first and those that are largely disliked shown last.

RPG Superstar 2009 Top 16, 2012 Top 32 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Marathon Voter Season 9 aka Epic Meepo

Gary Teter wrote:
We haven't 100% decided on the labeling for the buttons, but you'll be given three options: Prefer this one, prefer that one, or prefer neither.

I stand corrected. I wasn't expecting a "prefer neither" option, but there you have it.

RPG Superstar Season 9 Top 32 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Marathon Voter Season 9 aka theheadkase

The only twist to that, Wild Gazebo, is their algorithm that is going to pair items based on least viewed (from what Gary said). So that set of potential permutations will decrease by 2 until all have been voted on, then it will start over.

This is where I'll get a little fuzzy...if the algorithm takes into account the preference selected and pair them based on that, then the set of permutations decreases exponentially for each of the sets left.


What I'm saying is that there is a far wider margin for items of less-than-normal quality to gain more points because of the significance of the data gathering--and the opposite.

I'm absolutely certain it will giver a us a good list of candidates and that many gamers will spend hours combing over the entries. But, I thought, since this was the topic, I would share my thoughts on the significance of the data. I have a feeling the data won't be significant because of the large percent of error it will employ.

RPG Superstar 2013 Top 16 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Dedicated Voter Season 9 aka Darkjoy

Wild Gazebo wrote:

What I'm saying is that there is a far wider margin for items of less-than-normal quality to gain more points because of the significance of the data gathering--and the opposite.

I'm absolutely certain it will giver a us a good list of candidates and that many gamers will spend hours combing over the entries. But, I thought, since this was the topic, I would share my thoughts on the significance of the data. I have a feeling the data won't be significant because of the large percent of error it will employ.

There is a reason that we have some judges at the end of the voting.


Pendin Fust wrote:


This is where I'll get a little fuzzy...if the algorithm takes into account the preference selected and pair them based on that, then the set of permutations decreases exponentially for each of the sets left.

Yes, I was assuming they weren't doing that because of that. It would be more like a single elimination type of tournament.


Darkjoy wrote:
Wild Gazebo wrote:

What I'm saying is that there is a far wider margin for items of less-than-normal quality to gain more points because of the significance of the data gathering--and the opposite.

I'm absolutely certain it will giver a us a good list of candidates and that many gamers will spend hours combing over the entries. But, I thought, since this was the topic, I would share my thoughts on the significance of the data. I have a feeling the data won't be significant because of the large percent of error it will employ.

There is a reason that we have some judges at the end of the voting.

Hurrah for people!!!!!!

Its funny. This is just a discussion for me. I'm really not worried. Comparing this to the subjectivity of judges is really quite silly--in terms of math.

I'm just interested in what people think...and if it was thought of.

RPG Superstar 2009 Top 16, 2012 Top 32 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Marathon Voter Season 9 aka Epic Meepo

Wild Gazebo wrote:
(Just for another comparison if there are 1300 entries there are 844000 permutations; meaning 1298 people would have to vote 650 times; or 8440 people would have to vote 100 times; 84400 people would have to vote 10 times; for just a single set of permutations)

I suspect there will be more than 1298 people voting, and I don't think it will be hard for any of them to vote several hundred times over the course of a few weeks.

Heck, if the voting booth is smartphone-friendly, everyone interested in voting could go through a handful of item pairs every time they have to wait for the bus (or whatnot). That's an almost ideal system for wracking up a large number of votes over time.

RPG Superstar 2013 Top 16 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Dedicated Voter Season 9 aka Darkjoy

I meant that the judges act as a balance against your perceived lack data significance.

RPG Superstar 2012 Top 16 , Dedicated Voter Season 6, Star Voter Season 7

3 people marked this as a favorite.

Wild Gazebo, we don't need to speculate about how well the voting method will work. Gary's said that they're using a modified Condorcet method for voting, and there is extensive academic research and literature on how that works. To summarise, the method works exceedingly well for exactly this sort of ranking problem, even though most voters only see a tiny subset of the available permutations.

Here is a good starting place if you want to read more, but note you could take graduate level courses in this stuff.


Darkjoy wrote:
I meant that the judges act as a balance against your perceived lack data significance.

Oh, Darkjoy, I wasn't calling you silly! I know what you meant. I was calling myself silly. Meaning, even if there was a 15% margin of error it would still be damn close to the subjectivity of a judicial review.

RPG Superstar 2013 Top 16 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Dedicated Voter Season 9 aka Darkjoy

Wild Gazebo wrote:
Darkjoy wrote:
I meant that the judges act as a balance against your perceived lack data significance.
Oh, Darkjoy, I wasn't calling you silly! I know what you meant. I was calling myself silly. Meaning, even if there was a 15% margin of error it would still be damn close to the subjectivity of a judicial review.

Well, it did work well for 5 years.


Will Cooper wrote:

Wild Gazebo, we don't need to speculate about how well the voting method will work. Gary's said that they're using a modified Condorcet method for voting, and there is extensive academic research and literature on how that works. To summarise, the method works exceedingly well for exactly this sort of ranking problem, even though most voters only see a tiny subset of the available permutations.

Here is a good starting place if you want to read more, but note you could take graduate level courses in this stuff.

I like to speculate. I'm not trying to drum up a revolution. This is just a conversation.

I'm very familiar with that method, because of some political volunteering I did. Unfortunately, it assumes a different method. Now, as I understand it form some brief reading, outside of Wikipedia, there is another method that employs single tally collection; but, that method would employ multiple permutations...in fact, it would need half of the total amount of entries in permutations of the subsets. So, I don't think they will employ that method.

I imagine they will simply rank 1pt or no point and only allow one view per voter and bracket the randomization based on the subset, or number of votes, so that they get an even field of 'times of votes'. Otherwise they would have to apply a ratio of pts/#votes on the weighting of every entry to properly assign a rank.


I am intrigued by this system.

Crowdsourcing the initial ranking and the "does this break the rules" issue takes an amount of busywork out of the judge's hands. I like that it gives the judges more time to do other stuff, like focus more on the items and less on rules. Or on the mythic playtest.

sips tea


Darkjoy wrote:
Wild Gazebo wrote:
Darkjoy wrote:
I meant that the judges act as a balance against your perceived lack data significance.
Oh, Darkjoy, I wasn't calling you silly! I know what you meant. I was calling myself silly. Meaning, even if there was a 15% margin of error it would still be damn close to the subjectivity of a judicial review.
Well, it did work well for 5 years.

It sure did. And they all deserve nothing but respect for it. This contest is great for the gaming community.

RPG Superstar 2012 Top 16 , Dedicated Voter Season 6, Star Voter Season 7

Wild Gazebo wrote:

I like to speculate. I'm not trying to drum up a revolution. This is just a conversation.

I'm very familiar with that method, because of some political volunteering I did. Unfortunately, it assumes a different method. Now, as I understand it form some brief reading, outside of Wikipedia, there is another method that employs single tally collection; but, that method would employ multiple permutations...in fact, it would need half of the total amount of entries in permutations of the subsets. So, I don't think they will employ that method.

I imagine they will simply rank 1pt or no point and only allow one view per voter and bracket the randomization based on the subset, or number of votes, so that they get an even field of 'times of votes'. Otherwise they would have to apply a ratio of pts/#votes on the weighting of every entry to properly assign a rank.

Good on you for volunteering and getting involved. And I'm all for speculating, even more so when it leads me to learn about interesting new things.

It looks to me like they'll end up with a large set (1 per voter) of +1 / 0 / -1 voting matrices. Many will be sparse, with only a few item-pairs evaluated and every other vertex set to 0. Some will be more fully populated (looking at the completer-finisher gamers out there). Then a simple addition of all the matrices, over thousands of voters, gives a really good basis for ranking, even though each individual matrix may be sparse, and even though some items may have been voted on more times than others.

Paizo Employee PostMonster General

This page gives a reasonably plain-english description of how we're planning on ranking the entries.

RPG Superstar 2012 Top 16 , Dedicated Voter Season 6, Star Voter Season 7

Thanks Gary, very interesting, particularly around the beatpath algorithm. Fun times in the coding department.

RPG Superstar Season 9 Top 32 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Marathon Voter Season 9 aka theheadkase

Thanks Gary! Good read!

RPG Superstar 2008 Top 16 , Marathon Voter Season 6 aka Core

As far as discussion of the items, a common sense approach would likely be best. That is, discuss items without being a overly negative or a door-to-door salesmen.

I suspect that the bizarre-yet-memorable novelty items may win out on the discussion threads however. After all, a well-balanced and sensible item may not give rise to discussion like an enlarged barnacle that attaches to one's back, hands the user enchanted arrows, and allows the wearer to occasionally lay an egg.

RPG Superstar 2012 Top 16 , Marathon Voter Season 6, Marathon Voter Season 7, Marathon Voter Season 8, Dedicated Voter Season 9 aka motteditor

2 people marked this as a favorite.

I say we just roll a handful of dX (where X is the number of entries) to pick the top 32 because all the math is making my mind hurt. : )

(I knew there was a reason I ended up going into liberal arts.)

Marathon Voter Season 6

2 people marked this as a favorite.

I've taken way too many levels in Barbarian for this to not make my head hurt.

1 to 50 of 62 << first < prev | 1 | 2 | next > last >>
Community / Forums / Archive / Paizo / RPG Superstar™ / General Discussion / Thoughts on the public voting and discussion All Messageboards

Want to post a reply? Sign in.