Metanet Forums

Me, I'm all for the top-rated coming back, but there are tons of maps now with a perfect 5 rating, and I just wondered which you think would get knocked off the top first and which one would eventually take the top spot.
Discuss.

well it really helps if you wanna check what the most popular maps are, but if sniping becomes a problem, then it pretty much ruins the whole point. im for bringing it back, cause it was cool

At this point, I'm not so sure I would want it back, cause it wouldn't really change anything anymore.
The only thing is most new users will miss out on some great maps, but hey, that's why there's featuring maps now.
I'm probably 50/50 on this, but really could careless.

Top rated was always a bit of a shambles. People felt they were justified dropping (sniping) maps if they didn't agree with their positioning. It really didn't mean as much as it suggested.

If it were re-implemented, almost all (if not every single one) of the perfectly rated maps would soon be dropped down to 4/5.

I don't see the need for it at all, but I wouldn't be entirely against it if the majority were in favour.

atob wrote:Top rated was always a bit of a shambles. People felt they were justified dropping (sniping) maps if they didn't agree with their positioning. It really didn't mean as much as it suggested.

If it were re-implemented, almost all (if not every single one) of the perfectly rated maps would soon be dropped down to 4/5.

I don't see the need for it at all, but I wouldn't be entirely against it if the majority were in favour.

I agree with atob, in that although it would be nice, ultimately it would be a short lived victory. If anything, it would just justify sniping and possible dash the community.

southpaw had an idea about this some time ago...it was like a hierarchy top-rating system, organized according to how many maps the person had. E.g. a top rated page for newbies with less than 20 maps, another for promising newcomers with 21-50, another for regular members with 51 to something like 150 or 250, and then a final one for the veterans with more maps than that.

How would that work?

incluye wrote:southpaw had an idea about this some time ago...it was like a hierarchy top-rating system, organized according to how many maps the person had. E.g. a top rated page for newbies with less than 20 maps, another for promising newcomers with 21-50, another for regular members with 51 to something like 150 or 250, and then a final one for the veterans with more maps than that.

How would that work?

This is an interesting idea, my only real concern with this is that it might encourage people to make more crappy maps and less good maps just to reach the higher hierarchy. Besides, there are some mappers who make brilliant maps and don't have huge map totals.

Also, bringing back the top-rated system would probably just spotlight the same few maps that were always spotlighted before in the old NUMA. Maps like Afternoondragon, crazyminer, etc.


What I think could work is a top-rated maps of the month section. I don't know how sensitive the voting and rating systems for NUMA are, but if there was a way to see which maps are the highest rated for a period of time, say a month or a fortnight, then I could see top-rated maps being a viable component.

incluye wrote:southpaw had an idea about this some time ago...it was like a hierarchy top-rating system, organized according to how many maps the person had. E.g. a top rated page for newbies with less than 20 maps, another for promising newcomers with 21-50, another for regular members with 51 to something like 150 or 250, and then a final one for the veterans with more maps than that.

How would that work?

Yes, I did suggest that. And to epigone, I think the main issue is that one would have to make consistently good maps to be rated the highest in each bracket, so that even if a shitty author made 250 maps they would obviously be nowhere near the top of the list.
I also suggested another portion to that for those who disable ratings on all or some of their maps, such as atob or GreenEggsAndHam.

But with the current Hot Maps system I consider this to be the Top Rated of the Moment. It isn't as expansive as epigone's idea is, but it gets the point across that the maps closer to the top for the longer period of time are usually better than the maps below it. Now, this doesn't always work like one would hope it did, but it discourages massive sniping, as only one or two maps are at the top and will eventually fall by the wayside rather than having a map with 6 votes that sits above a map with 680 votes. Then we see sniping happen and we watch as a single vote causes it to drop 250 places on the Top Rated list.

My only concern with a Top Rated system is that the votes be weighted more. Instead of having every map out of 5 make it out of 10. Or make it like this new idea I just came up with where it is out of 1/5 of the total votes. For something with 100 votes it'd be weighted over 20 votes instead of 5. That way all maps are weighted according to the number of votes they have received and not by their overall average rating.

Manus Australis wrote:

incluye wrote:southpaw had an idea about this some time ago...it was like a hierarchy top-rating system, organized according to how many maps the person had. E.g. a top rated page for newbies with less than 20 maps, another for promising newcomers with 21-50, another for regular members with 51 to something like 150 or 250, and then a final one for the veterans with more maps than that.

How would that work?

But with the current Hot Maps system I consider this to be the Top Rated of the Moment. It isn't as expansive as epigone's idea is, but it gets the point across that the maps closer to the top for the longer period of time are usually better than the maps below it. Now, this doesn't always work like one would hope it did, but it discourages massive sniping, as only one or two maps are at the top and will eventually fall by the wayside rather than having a map with 6 votes that sits above a map with 680 votes. Then we see sniping happen and we watch as a single vote causes it to drop 250 places on the Top Rated list.

My main problem with Hot Maps is that even if a map deserves100 votes of 5, people are soon going to stop rating it, and as more maps are submitted the farther the map will fall down the queue.

I think the best idea of this sort would be a kind of Hall of Fame system, wherein once a map gets a certain number of votes it's admitted, and can no longer be rated. Maybe the system could send a request to the author for it, and the author could deny the request or accept it. This way the Top Rated page wouldn't get cluttered with maps that have already been proven great.

I'm also thinking a dA-like time system (8 hours, 12 hours, 1 day, 1 week, 1 month, All Time, and Way Back) could be applied...that would take a heck of a lot of coding, though.

incluye wrote:

Manus Australis wrote:

incluye wrote:southpaw had an idea about this some time ago...it was like a hierarchy top-rating system, organized according to how many maps the person had. E.g. a top rated page for newbies with less than 20 maps, another for promising newcomers with 21-50, another for regular members with 51 to something like 150 or 250, and then a final one for the veterans with more maps than that.

How would that work?

But with the current Hot Maps system I consider this to be the Top Rated of the Moment. It isn't as expansive as epigone's idea is, but it gets the point across that the maps closer to the top for the longer period of time are usually better than the maps below it. Now, this doesn't always work like one would hope it did, but it discourages massive sniping, as only one or two maps are at the top and will eventually fall by the wayside rather than having a map with 6 votes that sits above a map with 680 votes. Then we see sniping happen and we watch as a single vote causes it to drop 250 places on the Top Rated list.

My main problem with Hot Maps is that even if a map deserves100 votes of 5, people are soon going to stop rating it, and as more maps are submitted the farther the map will fall down the queue.

I think the best idea of this sort would be a kind of Hall of Fame system, wherein once a map gets a certain number of votes it's admitted, and can no longer be rated. Maybe the system could send a request to the author for it, and the author could deny the request or accept it. This way the Top Rated page wouldn't get cluttered with maps that have already been proven great.

I'm also thinking a dA-like time system (8 hours, 12 hours, 1 day, 1 week, 1 month, All Time, and Way Back) could be applied...that would take a heck of a lot of coding, though.

Wait, what?

incluye wrote:wherein once a map gets a certain number of votes it's admitted, and can no longer be rated.

So you're saying that something like Chase. wouldn't be able to gather any more votes once it hit a certain "vote cap" and then be added to a hall of fame? no no no, that's just wrong. I love it when I see a comment on an old map of mine that has a lot of votes, not only because it's great to see people going back and playing some of my maps but because it means people are still interested in rating the old maps and not just the new. Older maps with more votes that you cannot vote on is simply saying that you can't show your support for older maps with anything other than a comment.
People like to rate. I think the 1/5 of votes idea is much better than telling people once they reach 50 or 100 (or even 300) votes that you can't vote on the map anymore.

Bring back the Top-Rated page! We are only able to see one Featured Map at a time, and with the Top-Rated page, we can see more. Bring it back! It was the joy of the community!

PNI wrote:Bring back the Top-Rated page! We are only able to see one Featured Map at a time, and with the Top-Rated page, we can see more. Bring it back! It was the joy of the community!

There is a Featured page which is the same style as other pages. Ten maps at a time.

And who said you speak for the community?!

incluye wrote:My main problem with Hot Maps is that even if a map deserves100 votes of 5, people are soon going to stop rating it, and as more maps are submitted the farther the map will fall down the queue.

Absolutely. This is why there needs to be separate new and top-rated sections. The float/sink system was interesting but really just shortchanges both features (albeit one more than the other).

incluye wrote:I'm also thinking a dA-like time system (8 hours, 12 hours, 1 day, 1 week, 1 month, All Time, and Way Back) could be applied...that would take a heck of a lot of coding, though.

That would definitely be awesome, but I imagine it would require a major overhaul in the coding. :(

southpaw wrote:Or make it like this new idea I just came up with where it is out of 1/5 of the total votes. For something with 100 votes it'd be weighted over 20 votes instead of 5. That way all maps are weighted according to the number of votes they have received and not by their overall average rating.

I don't understand your wording, but I agree with the idea. You're saying that it's dumb that a map with 5 votes of 5/5 should be placed ahead of "true" top-rated maps that have a 4.6 average from 70 votes. I absolutely agree. There should be a formula to determine rank that takes into account # of rates as well as average rating. It shouldn't weigh the former too heavily though. 3/5 by 100 people shouldn't be above 5/5 by 20 people.

Also, there are other alternatives. I suggested (somewhere, sometime...) that there could be a top-rated system that shows maps by rounded rating, and then date. That is, the newest map with 5/5 would be at the top of the list, while the oldest map with 5/5 would be at the end of the 5's. Then the listing would move down to the 4's and so on. I like southpaw's idea better, but this is also a possibility.

Ultimately I still think my trinity idea is the best way to go: top-rated of all time, recently top-rated (or top-rated for the week, whatever), and new maps. It's like Newgrounds, or like a simplified version of DeviantArt. This satisfies everyone who can reasonably be satisfied.

And those of who you are still against a top-rated system need to think back to when you first came to this community (unless you were here from the beginning, in which case just imagine you came later). You can play maps off the newest page endlessly, but ultimately if you want to see the best maps around, a top-rated page is the way to go. Arguments against it range from "it doesn't change enough!" to "it changes too easily, with one vote sending a map from 2nd to 250th!", but either way, it showcases some of the community's most highly-rated maps, and it should absolutely be available. If you've seen all of those maps, fine, don't visit the top-rated page then. That doesn't mean everyone else should have to do without, particularly people who haven't seen those maps yet.

incluye wrote:I'm also thinking a dA-like time system (8 hours, 12 hours, 1 day, 1 week, 1 month, All Time, and Way Back) could be applied...that would take a heck of a lot of coding, though.

...

...dude.

That's the solution I keep hearing.

"Hey, make up a system that shows the best maps; but because you can't possible show the best maps of all time, make it so that it shows you the best maps for a given period!"

...

...that's the system I came up with. arachnid and I talked it over and it had the best of both worlds. It didn't lie to members by saying that the top 10 maps on the page were "the best ever". In fact, it exposed the ratings flaw that confused the way that people rate now with how they rated way back when.

Instead, the new system gives you top-rated maps that you haven't seen before. It is designed to help people who need good maps fast. Who intends to see a map that they've seen before? Only rating-crazies and people who like to stalk pages for attack/defense purposes.

The new NUMA isn't a realtime strategy game, it's a website.

blue_tetris wrote: ...dude.

That's the solution I keep hearing.

...that's the system I came up with. [...] The new system gives you top-rated maps that you haven't seen before. It is designed to help people who need good maps fast.

No, your solution is NOT what incluye suggested. And stop claiming that it is (you've done so before). Why isn't it?
(1) It's like having only the "today" tab from DeviantArt. Even maps that are 5/5 by 20 people eventually get pushed off in <24 hours by the continuous submission of new maps.
(2) Frequently there is no "top-rated" map on the Hot Maps page.
(3) Usually such maps have <10 votes, which is hardly "top-rated". Maybe "new and good" or "top-rated of the day", but not "top-rated".

Hence:

Avarin wrote:The float/sink system was interesting but really just shortchanges both [the top-rated and new maps listings] (albeit one more than the other).
[...]
Ultimately I still think my trinity idea is the best way to go: top-rated of all time, recently top-rated (or top-rated for the week, whatever), and new maps.

blue_tetris wrote:Who intends to see a map that they've seen before? Only rating-crazies and people who like to stalk pages for attack/defense purposes.

Whoa, calm down and think a little. The top-rated is not necessarily for those that have seen it before! That's what a "recent top-rated" page would be for. A top-rated of all time would be for people who haven't seen it before (or, more accurately, who haven't gone through it before). Make sense?

How many times do I have to state perfectly good points before people stop ignoring them? :/

Do you think the ratings system really allowed the Top Rated page to show the absolute best ten maps on the front page?

I think it showed the least-sniped maps, perhaps. I think it's a bold lie to say those maps were better than the other 100,000 maps on NUMA just because people rate differently now.

Tell me, straight-up, what you thought the "Top Rated" maps were reflective of? Was it quality? Was it recent (i.e. not yet sniped) maps with quality? Why were the people checking the Top Rated page those who had visited the site hundreds of times before? Are the best maps really changing that often? Do old maps suck that bad?

the brackets thing would be good. what if we had a top rated system where it weighted the rates more? so if you made rates as heavy as they could be, it would essentially be the old top rated thing, except some might have faded by now, but if you adjusted the sensitivity you could have good maps from the last week, to 6 months ago.
just a thought.
the dA system would work better though, that or the author brackets.

count = 50 still works in the url btw, for those wining about 10 maps per page.
bleh i dont know. id like it back, but modified heavily. if it were brought back as it were it would be a snipe-fest again.

Sorry everyone. Discussing ideas takes a crapload of text. In the interest of improving NUMA, please read this and give me your thoughts.

Section 1
First off, I have a couple new ideas about how a "top-rated of all time" system could be better. For example, instead of dealing with making a formula, you could simply have a top-rated system that ignored all maps with less than 20 or 30 votes. Alternatively, perhaps maps couldn't get on that list until they were 2 weeks old (after they have been obligatorily removed from the "recent top-rated" page). Yet another idea is to use a different exact average that negates all 0's, because really, if a map is rated so highly by so many, it shouldn't be worth a 0 to anyone. I would understand a 1 and definitely a 2, but 0? On any half-decent map, the only real use for a 0 is for sniping or plagiarism. I suppose that is my opinion, though I imagine most would agree with me.

Section 2
Now to address b_t:

blue_tetris wrote:Do you think the ratings system really allowed the Top Rated page to show the absolute best ten maps on the front page? I think it showed the least-sniped maps, perhaps. I think it's a bold lie to say those maps were better than the other 100,000 maps on NUMA.

I assume you're not merely suggesting that this feature was rightfully removed simply because it didn't function properly, but rather that it is impossible for the feature to function properly. In that case, here's my response:

Yes, it showed the best ten maps, but in a technical way that no one really cares about: in the sense that "best" means "has the highest average rating at that given moment".

The thing is, this entire community needs to deal with their ideas of a rating system. I'm going to make another thread to deal with this issue (and a few others that have similar origins), but for now I'll just say that it seems to me that people are thinking along these lines:
"Because everybody likes different things, ratings are rather useless because they don't represent how well a particular user will like a particular map."

I know I certainly used to think like this. Or rather, when thinking about the issue of ratings, I couldn't get around this issue. But now I have. Here's what I say to it:

This isn't Netflix, people. NUMA isn't supposed to take your ratings and run it through complicated systems to produce modified averages that try to predict whether or not you'll like maps. NUMA runs on a normal rating system, which allows the community the quantify their feelings about maps. Because of this function, it has the potential to serve as a rough indicator of how well you'll like a map, but ultimately it depends on how similar your tastes are compared with the wider community. Now keep in mind that this is assuming that ratings are perfect - that is, if everyone in the community weighed in on every map. Of course, that idea is ludicrous. But for top-rated, it starts to come back into play. Generally, a map can be considered as being closer to this magical "community consensus" the more rates it has. This is how top-rated is ideally supposed to be a rough indicator of the maps that have the best community consensus. Or at least, it is if you filtered out maps that are 5/5 from 5 people.

So that's my argument as to why top-rated roughly does serve the function it's intended to serve (or rather, that it can serve that function with some modification). Furthermore, you really really REALLY need to understand that the top-rated page needn't present the absolute ten best maps on the first page. There's genres to consider, and there's a "Next" button. This is why putting up the Dronies lists can't possibly supplant the top-rated page: the top-rated listing has pages upon pages upon pages of highly rated maps. There are just a handful of Dronie-winning maps each year. Clearly, top-rated is more extensive.

Section 3
This leads us straight to:

blue_tetris wrote:Why were the people checking the Top Rated page those who had visited the site hundreds of times before? Are the best maps really changing that often? Do old maps suck that bad?

People are checking the top-rated page because it served as BOTH top-rated of all time and recent top-rated. Separating it into two top-rated sections (recent and not recent) more or less solves that.

Furthermore, if you had really played all the top-rated you cared to, then yes, in my suggested system you would probably only ever check the recent top-rated and new maps listings. HOWEVER, there are still a huge mass of people who have not gone through the top-rated listings to their liking, largely comprised of all the current and future newcomers to this community. In trying to rid yourself of what you considered an eyesore, you screwed over pretty much everyone in that group. And for the love of all that is holy, can someone please acknowledge that concept? I've repeated it like 5 times total spread over 2 or 3 threads and none of you anti-top-rated people has said anything in response to it. :/

Section 4
Now I also have to apologize. Communicating these ideas takes so much friggin text and I bet people are sick of me typing up encyclopedias when I post. I know I just spent forever defending my ideas, but I never really addressed an issue that seemed very important to you, b_t, and that's sniping. Sniping on the top-rated page happens all the time, true. But really, it's more of side issue. If you are saying that you would be fine with a top-rated listing if snipers didn't exist, then holy hell, please say so and we'll work from there.

I see two ways to deal with snipers:
(1) Make the top-rated listing less vulnerable to them. For example, this could be done by using new averages to rank maps - these could automatically not include all 0's, or remove the highest and lowest 10% of votes. Or you could simply show a histogram of all the votes a map has received on its page. Perhaps seeing that a top-rated map has already been sniped 10 times will dissuade sniping (either true snipers, who may realize that they're activities aren't so novel, or regular members who really don't think the map should be rated so highly and are sniping in an attempt to return the average to a more reasonable number).
(2) Make sniping less possible. I've seen many an idea shot down because of potential abuse by snipers. I've got an idea - why don't we actually DO something about sniping in the first place? If >20% of a user's placed rates are 0's (once they have a sufficient # of rates), can't we safely assume they aren't rating properly? I don't know, it just seems like this community has an incredibly laissez-faire attitude towards sniping. I have nothing against that on its own, but it becomes a serious problem when that attitude serves as an excuse to not implement good ideas.

Section 5

blue_tetris wrote:...just because people rate differently now.

Someone else said that recently as well. I don't understand what you guys mean. Are you saying that there is a significant difference in the way people rate now that rating is confined to whole numbers only? And that this difference affects top-rated?

you know its not just about the logistics of the page...
i enjoyed going to it :D

Vastly hugely and seriously off-topic, but I think it would be cool if we could have a random function, which only took into account the highest rated maps. For instance, a 5Ninja section, which just gives you 10 random maps with a 5 rating. Obviously you could make the pool bigger or smaller, as I understand the system measures average ratings to a far greater degree of accuracy.

This would mean that you would be able to look at the "top rated" maps, but they aren't necessarily ranked so that any one map can be seen to be the highest rated map. As such, the system would be as anti-snipe as it could be, as it would throw up a different 10 maps each time anyone viewed the section.

x

Avarin wrote:

southpaw wrote:Or make it like this new idea I just came up with where it is out of 1/5 of the total votes. For something with 100 votes it'd be weighted over 20 votes instead of 5. That way all maps are weighted according to the number of votes they have received and not by their overall average rating.

I don't understand your wording, but I agree with the idea. You're saying that it's dumb that a map with 5 votes of 5/5 should be placed ahead of "true" top-rated maps that have a 4.6 average from 70 votes. I absolutely agree. There should be a formula to determine rank that takes into account # of rates as well as average rating. It shouldn't weigh the former too heavily though. 3/5 by 100 people shouldn't be above 5/5 by 20 people.

Also, there are other alternatives. I suggested (somewhere, sometime...) that there could be a top-rated system that shows maps by rounded rating, and then date. That is, the newest map with 5/5 would be at the top of the list, while the oldest map with 5/5 would be at the end of the 5's. Then the listing would move down to the 4's and so on. I like southpaw's idea better, but this is also a possibility.

I think you twisted the point of my idea.

It was this: That a map that has been rated 100 times will be averaged over 20 votes instead of 5. Personally I think that until a map reaches 25 votes it will be averaged over 5 votes, from there it is averaged by 1/5 of the total votes.

I'll explain more late, as I have class.

mintnut wrote:Vastly hugely and seriously off-topic, but I think it would be cool if we could have a random function, which only took into account the highest rated maps. For instance, a 5Ninja section, which just gives you 10 random maps with a 5 rating. Obviously you could make the pool bigger or smaller, as I understand the system measures average ratings to a far greater degree of accuracy.

This would mean that you would be able to look at the "top rated" maps, but they aren't necessarily ranked so that any one map can be seen to be the highest rated map. As such, the system would be as anti-snipe as it could be, as it would throw up a different 10 maps each time anyone viewed the section.

x

Sounds good with me.

I think the top-rated authors page should be brought back, but seperated into join date by year, so all the inactive greats (lucidium, formica, etc..) wouldn't swamp the page.

I wish you would've been around back when Nmaps.net was getting remade. Everyone had complaints and no one had solutions. I stepped up, offered solutions, and willingly worked hand-in-hand with arachnid to get things done.

arachnid is busy again now, so even if you've got an amazing solution, it's likely not going to get implemented soon.

I kinda wish people were happier from all the work we (mostly he) did. Instead, they complained before and they complain now. People had their time to fix NUMA and most used the time to come up with fresh new complaints to volley.



Anyhow, for your first point: If there was an even array of votes, the top-rated page would need to have 20,000 results. That's 20,000 maps with 5's, which have the possibility of being the best of the best. If there was even a sharp bell curve, you'd still probably need 2,000 of the 100,000 NUMA maps on the top of the Top Rated page with roughly equal ratings.

Ratings have a grand margin of error. If any 5 stands out beyond another 5, it's due to this margin of error.

I also hear the argument "Well, the page would still give you the highest rated maps, even if they're not the best. Isn't that okay in its own right?" No. It ain't. What's the purpose of hoarding ratings if they have no meaning? If we can't put the ratings through a system that utilizes the ratings for an end, then the ratings are pointless. There's no need to classify ratings for their own sake.



About your comment on the "community involvement" aspect. Guess what the ratings histogram for all the top rated maps looks like. It's a bunch of 5's and a bunch of 0's. Only the maps all over the top rated page gained this distinction. Were some maps worth 3's and 4's while others were worth 5's and 0's? Maybe. We didn't think so, though. We figured that the huge spike in ratings garnered by seeing a map on the top rated page were more about politics than analysis. Let me explain.

When a map hit the top rated page, it pretty much stopped getting normal votes ranging from 3's to 4.5's and the like. Suddenly, it became this odd sort of battleground. It was hard to tell the quality of the map beyond the network of the map maker, those lining up behind him with 5's and those pitting against him with 0's. The top rated page didn't end up showing the best 10 out of 100,000 on NUMA. If you believe that those were the best 10 out of 100,000 on NUMA, you have a problem. Instead, it showed the not-so-valiant struggle of man against ratings and the demise of a workable system. It showed users gaming the system for the sake of one page.

You say it was meant to represent what's good now, which is fine in theory. But it held permanent ratings and compared them against maps throughout all of history. So, in order to map "Top Rated" equal "Current Top Rated", people had to either slowly increment the ratings they were giving to all maps (thus ensuring that tomorrow's Top Rated really will make the page) or snipe the maps on the Top Rated to make room for new ones. Again, the Top Rated page without categorical distinctions just gave an opportunity for people to game the system.

You may be okay with that. We weren't.



The solutions you propose for sniping are not feasible. Dumping "non-average" ratings ruins the system, by giving earlier votes more power than later votes, which suggests that early voters somehow know better than later ones. It's a fix that weakens the existing system.

The sniping solutions are all ones we heard before and we came to the humble realization: Everyone is allowed one vote. We don't care what the vote is, but you only get one. That's really the only rule.

Avarin wrote:Someone else said that recently as well. I don't understand what you guys mean. Are you saying that there is a significant difference in the way people rate now that rating is confined to whole numbers only? And that this difference affects top-rated?

Nope. New ratings are higher than old ratings. We examined the average over time and it turned out that recent votes were averaging in the 4.1 range, while the oldest votes averaged in the 3.2 range. Either everyone got better at making maps (and you all need to go tell that to maximo yourselves, because I'm not breaking the news) or the votes just keep getting higher. Either way, the individual ratings for solitary maps collected en masse over the last four-plus years are growing less comparable to one another.

Top Rated does a disservice.



Here's a solution:

If you guys think that maps aren't staying on the Hot Maps page long enough, tell us how long you figure they need to stay. A few days? A week? Those float values can be easily changed and it helps make the Hot Maps page into the new "Top Rated Recent" page.

I think it could work well and it had zero testing. Maybe instead of redesigning the site, we can just switch up one or two numbers?

I think it really boils down to that and I don't think we've got enough manpower to even come eye-to-eye with all the complaining. Maybe it'd be good to meet the hardworking designer somewhere halfway?

Uh, anyway, to sum up. The final vote average doesn't need to be numbercrunched. You said it's "not Netflix". We take the raw data and then we generate an average. That's all that's really needed. We don't monitor votes. We don't hold people's hands.

On the display side of things, we need to find a way to take this data and use it for something meaningful. The float/sink system is meaningful--it just needs to be tweaked to fit precisely what you want from a Hot Maps page. If you want 5'd maps to stick longer, say the word. It's an easy fix.

displaying a graph like the attached would be nice. green is the rate made, red is the resulting average. the graph would only show to the public when the ratings were public.
that and yeah, a random "high rated" would be nice. like if we were allowed to search by rating, random would always be true.
im sure they arent that tricky to get working, though the graph might be harder. it makes identifying a snipe easy though.