Idea: Good Game, Bad Game vote button.
Ironsoul
Join Date: 2011-03-12 Member: 86048Members
Join Date: 2011-03-12 Member: 86048Members
Hello, I have a simple idea that may help give Unknown Worlds a more useful glimpse into the status of their game.
The idea is simple, just like Vote Concede, we have a Vote Good Game and Vote Bad Game options. If 51% vote good game, the game is logged as a good game, if 51% vote bad game, the game is logged as a bad game. If not enough vote, the game is logged as Neutral Game.
The statistics would be sent off somewhere (ns2stats?) to be turned into a pie chart of some sort for later consumption.
Players can vote any time during the round, and they can change their vote any time they want as well. The votes are only tallied up at the end of a round.
I believe this simply feature will greatly change how ns2 is developed, instead of focusing mainly on wins/loss ratios it should be focused on good games/bad game ratios.
This wouldn't rank games, it wouldn't rank players, it would be purely "Did I have fun that last round?" If a game isn't fun/enjoyable, why play it?
Please post whether you agree, disagree or don't care so the thread doesn't die.
The idea is simple, just like Vote Concede, we have a Vote Good Game and Vote Bad Game options. If 51% vote good game, the game is logged as a good game, if 51% vote bad game, the game is logged as a bad game. If not enough vote, the game is logged as Neutral Game.
The statistics would be sent off somewhere (ns2stats?) to be turned into a pie chart of some sort for later consumption.
Players can vote any time during the round, and they can change their vote any time they want as well. The votes are only tallied up at the end of a round.
I believe this simply feature will greatly change how ns2 is developed, instead of focusing mainly on wins/loss ratios it should be focused on good games/bad game ratios.
This wouldn't rank games, it wouldn't rank players, it would be purely "Did I have fun that last round?" If a game isn't fun/enjoyable, why play it?
Please post whether you agree, disagree or don't care so the thread doesn't die.
Comments
UWE can look at the stats go "hmm, looks like people are enjoying the game more this patch, let's find out what we did" or "hmm, looks like only 3% of people actually enjoy the game... let's find out what's wrong". Instead of just going (50/50 win/loss ratio, games balanced, job done).
Bogging down the system with comments is kind of useless, the forums are the place for comments, so is other various places I've heard about. In the game we just need something simple to give a bit of feedback to say whether we're currently enjoying the game or not.
With this in mind, I don't think Valve or UWE would gather any good data with this feature.
It might be possible that if UWE analized the data of good and bad games,they can find the main similarities between all good games and between all bad games and than see the diffrences between good and bad games.
It might help them a little bit in the development procces.
I cant see it doing any harm to the game so why not give it a shot.
Presenting a survey after each game would be tedious. I don't see a place for such things in NS2, since you start the next round from ingame, the ready room.
With the recent post by UWE about their statistics, how they gather game times for example, I'd be interested in seeing game time being related to whether a game was classified as good or not. Are games that go for 18 minutes mostly "good games"?
You have a valid point about the "without reason, this could be useless", but as NS2 is iterative, I don't see the gg, bg vote system ending there, I see it improving... naturally over time.
I'm just being the devil's advocate. Most people probably won't vote. If they do, it's more likely they'll only make the effort when their experience was bad, because people that are happy, are most often less inclined to convey this message (see psychology textbooks). Add to that the possbility of getting 'trolls' that want to influence the system..
I don't think there's a way to make this reliable enough. If you can't "control" your target audience (e.g. - make sure that the average demographic is represented), no conclusions can be drawn from the results.
That being said, I do definitely like the suggestion itself and what you aim to achieve with it.
The real problem is, as Angelusz alluded to, is how do you DEFINE a good game? Ask a dozen people and you'll get 13 answers. Even if you get more specific - Was this a "fun" game? - you still run the risk of different people having different definitions of 'fun'.
The one 'poll type' thing I *would* like to see in NS2 is similar to what they have in L4D2 on the main menu, where the devs can do polls of the gamers. Those polls reach ALL players, as compared to forum polls which only reach a fraction. I think that would be a better way to go since you could properly define the question, which would make the answer meaningful.
I think internet forums about any subject are proof of this. There's always more negativity on them then positive remarks, even for good products. Unless we can get the positive people more motivated to vote as well, I'd be highly skeptical about the reliability of such a system. Not even mentioning how easy it would be to try and influence it one way or the other.
Humans suck and the sooner we realize this, the better off we all are!
Maybe if they click good game or bad game it leads to feedback tree. This would all be optional, of course. If they don't access the feedback system within some arbitrary number of seconds, the feedback system disappears.
Here's a sample of what one branch of the feedback tree might look like:
Obviously a certain set of people post on the forums, and its clear that UWE values this feedback. However, what about the players that play for the first time and decide never to play again? Wouldn't it be nice to know why?
And emotion at time of polling (which is to say, during and immediately after gameplay) is the entire point of the game. The function of the game is to produce pleasing emotions in the playerbase, if it is not doing that, it is not fulfilling its function.
And no, that isn't all it tells you, you can combine it with the huge number of other metrics to find trends. Compare it with game length, whether the player won/lost the last round, server performance, server size, map, whether x lifeform/upgrade was present in the game, whether the player used x lifeform/upgrade.
About the only statistic which isn't tracked at the moment is whether players actually enjoy any given game played, and that's possibly the most important thing to track because it ultimately lets you build a map of how the playerbase is experiencing the game. If long games are universally disliked, work towards making games shorter. If one map is universally preferred, look at that map and try to figure out what makes it so good, and put it in the other maps. Perhaps you may even notice a trend that causes people to say a game was good even if they lose, perhaps if the game gets to a point where both sides are fielding good gear then all players tend to say the game was fun, even if they lose, and from that you can go on to say that getting good gear on the field is a massively important thing for player enjoyment.
A simple binary choice like that adds so much context to the other statistical data that I can't imagine for one minute why it wouldn't be implemented. It is incredibly short sighted to dismiss it as useless because 'people would just vote based on win or lose'. If they do that that's still valuable information, it tells you that the game itself doesn't matter so long as you win, and the only thing of value is ensuring a 50/50 win rate because that is evidently the only thing that matters at all to the playerbase.
I think, however, you would find it a bit more nuanced than that.
Here's a quote:
Which is saying exactly my point in one, concise sentence. We need both, we currently have the objectivity side, but not the subjectivity side. I see no other ideas on how to get this into the game.
I would really love to hear some developer's thoughts on the idea.
Emotions at the end of the game (time of polling) are the most important ones of all. They determine whether the individual continues to play, or is compelled to return to play at a later time.
I think it may be surprising how much information can be garnered from such a simplistic and binary choice as this one. You can compare enjoyment to so many things, like overall game time, resources gained, marine or alien k/d, overall tech level, total number of kills. These things could or could not have a direct impact on how much enjoyment a player experiences, but a system like this would identify any existing patterns, and give you a certain level of evidence.
The case of winning players voting good, and losing players voting bad game could be an issue, but I don't think it would be. This is because if that is the case, and the norm is for losers to vote bad and winners to vote good, then you can isolate situations where this is not the case, and identify patterns in this limited data set that contrasts to the ordinary case.
But I think this system should track individual experience. That will prevent a lot of games from being voted neutral.
It also helps to solve the problem of rage voting of the loosing side.
As for tracking individual experiences in general, while I think it would be interesting to go really in depth and see where the players were that disliked the round, what they did, how much time they spent alive, dead, as a certain class, firing their weapon, reloading all that good stuff.. I think a binary (was the game liked by most people, or was the game disliked by most people) choice, being simple as it is, is the best place to start out, as that alone can be linked to other statistics already being gathered to give huge insight to the quality of the experience of the game.
Tracking individual experiences would be too much for the developers to look at, and possibly too much for their system to handle, or at least it would be unnecessary for the most part.
@|strofix| you get my idea completely, the developers can match up the 3 possibilities (gg, bg, neutral) with other variables and figure out patterns.
I personally don't think that's all too necessary but I wouldn't mind such a system, Youtube used to use a star rating and in all honesty, I preferred that to the pure like/dislike system. Of course, youtube isn't a FPS / RTS with complex player interaction mechanics.
I guess you right about individual vote, that it will be too much data. Although I feel it can give some interesting conclusions)
So with that said, I also think that the neutral vote should be "weighted" significantly lighter than the other two options, I like ratios, so I would have a say that half is a nice weight for the neutral vote.
On an 18 player server for example, 10 neutral votes is equivalent to 5 active votes. So on the same server, if 4 / 18 voted bad game, 4/18 voted good game and the rest said "whatever" and don't vote. The game would be logged as neutral. But if 5 people vote good game. That's a tie, there's a problem. I would say the game would still default to neutral if it's not beaten by over 50%. Meaning, 6 people would have to actively vote for good game in order for the game to be logged as a good game.
If people leave before the votes are counted, their vote gets tossed.
I would go on to discuss the possibility of a heavy neutral vote, or an active one, where it's worth the same as an active gg or bg vote, but I think that might be a bit too complicated for the needs of this idea. This system needs to be elegant to use and to code and to analyse. SIMPLICITY.