Forged Alliance Forever

Re: The Problem With "Multiple Choice Polling"

2018-12-07T23:05:28+02:00

I just want to say the polling software is godawful. Use radio buttons instead of a scoll-list please?

Statistics: Posted by Quant — 07 Dec 2018, 23:05

Re: The Problem With "Multiple Choice Polling"

2018-12-07T03:03:06+02:00

Yeah dude, maybe if instead of placing an ad every 30 minutes on the FAF client's most commonly looked at screen, we instead place it on a part or the UX that is heavily glossed over or even skipped entirely ONCE and we will get a +200% increase in votes?

Sure dude, Gorton plans to add something anyways. Will it end up in more norfair dissenters? maybe if the population sample retains the current opinion set. Will we be expecting another thread from you if that is the case?

Medicraze wrote:
@Biass regarding the overwatch comment:

I am well aware of who seagull is and etc. I do think however that the idea that they ignore pick rates and etc throughout anything but the pro (or even just the T1 and not the tendies or open div scene) is most likely incorrect. I also fail to see where you were leading with it so i apologize.

Statistics: Posted by biass — 07 Dec 2018, 03:03

Re: The Problem With "Multiple Choice Polling"

2018-12-06T23:01:39+02:00

"We can't ask a poll to show up on the 'What's New' page of FaF unless there is a promotions councilor"

- The FaF ladder team

Statistics: Posted by Morax — 06 Dec 2018, 23:01

Re: The Problem With "Multiple Choice Polling"

2018-12-06T20:38:26+02:00

Why does literally every discord snip on this god forsaken server intentionally cut out the context that is mere PIXELS about the snip?

The comment is about the poll, ftx mentions that the other poll was correct, he also mentions that column that had an error was related to people voting.there is no plausible way you somehow correlate that comment with being the actual map score unless you were intentionally misleading people into having lower faith in the ladder teams level of competency.

I have one answer to the poll turn out.
P R O M O T I O N C O U N C I L O R

Statistics: Posted by biass — 06 Dec 2018, 20:38

Re: The Problem With "Multiple Choice Polling"

2018-12-06T19:23:42+02:00

Turinturambar wrote:
that quote is out of context. the actual results are (afaik) used correctly. the thing that gorton was refering to there was not the actual rating which is used to pick the pool.

As far as you know, but it doesn't seem like it would be far from the truth since the poll was never advertised.

How can you explain that you guys are okay to use the poll info when not even that many people are made aware? It seems 1 poll was successful in turnout, then the rest are just "whatever." Can you give some more insight as to what critical number of votes will be needed to allow it to have an effect? E.G. if one poll has 300+ responses verse one with 40? Why should that count as much? It seems like a misuse of information.

Statistics: Posted by Morax — 06 Dec 2018, 19:23

Re: The Problem With "Multiple Choice Polling"

2018-12-06T19:18:26+02:00

that quote is out of context. the actual results are (afaik) used correctly. the thing that gorton was refering to there was not the actual rating which is used to pick the pool.

Statistics: Posted by Turinturambar — 06 Dec 2018, 19:18

Re: The Problem With "Multiple Choice Polling"

2018-12-06T19:13:26+02:00

Medicraze, it's not just the execution, but there are errors in the polling itself. Their "polls and statistics" are questionable and this only gives me further reason to believe they're not very helpful:

Statistics: Posted by Morax — 06 Dec 2018, 19:13

Re: The Problem With "Multiple Choice Polling"

2018-12-06T04:03:25+02:00

@Biass regarding the overwatch comment: This information is based on A_Seagull (former E-sports pro player for Overwatch). Obviously, take this claim with a grain of salt, but he states this information came from blizzard employees. He stated the balance was determined based on corresponding pro player affinity for a character's win percentage. Reportedly, the team would review what made some tactics so successful, and some of the abilities were adjusted based on what they gathered. I'm too lazy to look through his hundreds of videos, but I do remember this being said because I found it an interesting approach.

@morax Yes, there are flaws, but I thing it's mostly in the execution. Bare in mind, there are many barriers needing to be addressed, because we are dealing with a small sample size, and of that size, what motivates a player to respond, will that skew the data compared to those who dont, etc. There's many variables. That's why I would do that "youtube Polling" bullcrap ad that pops up when you launch a video. It's mildly inconvenient, but doesn't really keep you from doing what you want. But, being a game people want to play and improve on, I'm sure this would really help.

Statistics: Posted by Medicraze — 06 Dec 2018, 04:03

Re: The Problem With "Multiple Choice Polling"

2018-12-05T22:59:53+02:00

It's not about whether only two maps are too poorly rated to be considered, but whether they get reviewed after fixes.

Those of us that like to improve require feedback, so if there's a "one chance" policy on map discussion, it's quite frustrating for us folk.

To Medicraze point, I think you guys definitely should improve the polling in FaF. It comes off as quite low-effort, but yields big results.

I'm glad Evan and he agree there are flaws here.

Statistics: Posted by Morax — 05 Dec 2018, 22:59

Re: The Problem With "Multiple Choice Polling"

2018-12-06T04:48:37+02:00

EvanGalea wrote:
I guess this is a good place to ask the question: What happens if there's a huge conflict between ladder team and playerbase opinion? Like say the ladder team think a map is really good but the majority of voters hate it? does it kinda cancel out and the map shows up an average amount of times?

Vote tally doesn't directly influence the score edit: "yet" as far as we are concerned, a conflict will however force a reevaluation of said score amongst the team (who are just as happy to argue about maps as anyone else) we haven't had a major issue yet, but i would assume team members would attempt to find out why by asking around if there was one and we didn't understand the problem.

Also, what happens if a new version of a bad map is uploaded, but before the new version, it was panned so heavily it's set to almost never appears? Is the map given a second chance in another vote or pool?

Never thought about that, and that should be "my area" Yikes.
I will know if a new version is uploaded, however if a mapper has decided there is a major change they should let us know directly through any channel. Map scores can be reevaluated at any time and are constantly adjusted. If I were to adjust a score mere seconds before someone generated a pool, the changes would apply.

To be fair, I only know of two maps rated too poorly to be considered. Let's just say that we try to avoid cluttering the spreadsheet if the time investment was less on the map then it was on said sheet. You get me?

overwatch

Not only is only balancing around the pro scene just not true, why is this even here?

Are you trying to confer a point about considering low rated opinions on maps? We do already attempt that. having FtX on the team makes it easy kappa

Statistics: Posted by biass — 05 Dec 2018, 20:17

Re: The Problem With "Multiple Choice Polling"

2018-12-05T18:39:23+02:00

Okay, as a program planner with experience in this domain, I'd like to address a couple things.

A small list of player barriers related to surveys:
1: complexity: This being the balance between holistic data collection with survey engagement. So while finding a good balance provides data, it may be incomplete. This is something we will just have to deal with, as a complete survey is more complicated, which reduces sample size.
2: Accessibility: No one wants to make an account for a survey, but most have a gmail. Using google forms for data sets may be a good idea, although, I'm sure this has been considered before.
3: Perception of Influence: Players need to know they are being listened to. One of the most effective is videos (people are lazy and like passive information). Integration of a video acknowledging feedback in the client would be helpful in persuasion. It would also provide for increased perception of "our opinion maters".

One work around is intermittent topic based surveys: for example -> Survey player opinion on simple topics such as maps with inclusion of qualitative data. Simple questions such as "rate these 5 maps from least favorite to most" With a picture of the map (no one wants to open the map vault and remind themselves of names, easy = more involvement). Bite sized weekly surveys over time will increase involvement. You can have a pop-up on client launch that says "help us improve the game! Spend 2 min answering these questions. We listen. Tell us how we can help! This week's topic, matchmaking.". Something like that. Consider the 1 vote per user...idk how you'd implement...I'm gonna stop here, this is likely falling on deaf ears.

I do want to note: Although you may reserve your opinion; Overwatch is balanced based on top tier player performance. I think balancing based on this principle is a good idea due to the condition of mastery. Although lower rated players who haven't completely refined their skills may find changes to certain units "OP/nerfed" secondary to their perception/knowledge of viable strategies. In time, they would come to find effective counters, and this perception can be minimized by acknowledgement of the changes, through (you guessed it) and informational video that is short, and to the point. The simple fact is, a unit on paper is detached from the user's skill handling them.... I just realized I'm ranting about balance and not about surveys. I'm still going to post this because now I'm laughing lol

Anyway, if you want an actual breakdown of how to do a proper survey, let me know. I am trained in it. I do have a degree in public health... that's all we did. Bare in mind, my availability is spotty, but I'm here to help.

Statistics: Posted by Medicraze — 05 Dec 2018, 18:39

Re: The Problem With "Multiple Choice Polling"

2018-12-05T18:19:10+02:00

I guess this is a good place to ask the question: What happens if there's a huge conflict between ladder team and playerbase opinion? Like say the ladder team think a map is really good but the majority of voters hate it? does it kinda cancel out and the map shows up an average amount of times?

Also, what happens if a new version of a bad map is uploaded, but before the new version, it was panned so heavily it's set to almost never appears? Is the map given a second chance in another vote or pool?

Statistics: Posted by Evan_ — 05 Dec 2018, 18:19

Re: The Problem With "Multiple Choice Polling"

2018-12-05T14:34:42+02:00

Mozart, it's not opinion; he is legit, plain wrong in many of his assumptions.

You can't deny that fact and opinion exist within this game.

Read up, and yeah, it's okay to exclude people who consistently demonstrate an inability to understand facts.

Statistics: Posted by Morax — 05 Dec 2018, 14:34

Re: The Problem With "Multiple Choice Polling"

2018-12-05T03:29:11+02:00

Morax wrote:
You have to consider the fact that the first poll, which received ~150 votes, could have been anyone: not necessarily a 1v1 player, someone who plays a ton, some random dude who has played the maps and just wants to share his opinion. When you have people like "moses the red," who go on and on about their values and opinions, while 90% of the community disagrees with them, you are getting bad data.

I like this part the most, so the dude is legit part of the community and has his opinion. Maybe he even plays ladder. It just so happens that you dont like him and disagree with him, so his opinion is not valid now as a player now So we should build a better system that cuts off his opinion on what maps he dislikes to play, and instead focus on getting valid opinions that go in line with your vision of how ladder should be? Whew, whiff that!

Statistics: Posted by Mad`Mozart — 05 Dec 2018, 03:29

Re: The Problem With "Multiple Choice Polling"

2018-12-05T00:11:08+02:00

If I may respond to Morax's original post and some of the things being said here.
Video link: https://www.youtube.com/watch?v=5yfd1HJZBzM
New title: Good map, not enough time
New Description: It has come to my attention that this is being interpreted as a negative view on the map. I want to be clear on this. This is a great map, but I had no time for a long game. I wanted a quick one and done.
I featured this video because of the delicious bomber.

Morax, the creator of this map and active member of the community (and general good guy) pointed this out.

I will leave this video up, because I believe it is at the least entertaining.

Again: NOT ripping on the map. THIS IS A GOOD MAP THAT IS PROPERLY MADE. SOMETIMES YOU JUST DON'T HAVE TIME FOR A LARGE MAP!

negative-ad campaigning
Response: I don't like negative ad campaigns. This brings nothing productive to the table. My execution was poor. I meant nothing by this video outside for entertainment and awesome bomber. I asked you if you wanted me to take the video down via PM on slack, to which you stated it was fine. My apology if this has offended you. I've changed the title. PM me if you'd rather I remove it. I bear no hard feelings towards you or anyone. Frankly, people just don't know how much work and soul is required for map making, and I want to say this is one of the better ones.

Gauge player opinion of map
Response: This is a statistical variable as part of a NOIR survey (Nominal-Ordinal-Interval-Ratio) aspect of survey. It may be limited and incomplete, but it is still a viable variable.

I don't have time to play a 20km and okay onto the next game
Response: Game was over quickly, and I wanted to chance a quick 5x5 or something. Don't let the heuristic bias fill the voids.

This video didn't need to exist
Response: But it was entertaining?

SIDE NOTES:
I would spend the time to point out some things regarding the poll and some of the fallacies associated with proper program planning (which includes survey) but my carb alarm just rang, so it's time for my run.

I do want to note, at least something is being assessed, and for that I command those of you who volunteer your time to even be bothered. You are cherished. (that includes you Morax)

Check out the FDA program guidelines, there may be some insightful information there. I'll edit this comment after I go for my run and include links and tidy up... If I remember. lol

Statistics: Posted by Medicraze — 05 Dec 2018, 00:11