120 replies to this topic

[ticklemyiguana]

If, like you suggest, your MM matches people with 60% win rate, in time they'll get normed to 50%. If like worldt of tank, you team 60% with 40%, in time you'll get 50% for everyone. 50% is meaningless. W/L ratio is not like the age distribution amongst a population. It's only two states, win or lose, and one create the other, so it's symetrical. So assuming infinite time, like you said, everyones W/L ratio would tend to 50%. With or without any matchmaker.
 
So what's the purpose of matchmaker? It's to provide interesting matches for everyone. Multiple criteria. Team composition, number of players, repartition of skill/experience, number of parties, even language.
 
I'm not a douche, I'm kind enough to explain that to you. I point out that Hawken's playerbase has been heavily modified at least three time with three events, the worst being steam launch. It's situational awareness in the real world. Whatever you're testing, MMR is a useless stat, so broken that the core playerbase farmed it to the point they're not even able to play the game. You're not a new player, you cant recreate a new user experience. You cant monitor your experience yourself. You're just trying to be a brainy smurf.

In terms of the math you're presenting that's still very wrong, unless we're speaking of truly infinite time where infinity times one win out of every hundred is equal to infinity times 100 wins out of every 100. This is a silly assertion to make, as it literally requires people to play the game forever. Like I said, you are assuming that the statistical average is equivalent to the average of every player.

For any reasonable period of time, the worst player in Hawken is never going to beat the best player. They're not going to beat the second, third, fourth, fiftieth, or three hundredth best either.

They're going to beat someone of equal skill roughly fifty percent of the time, but that pool of people is one, because this is the worst player in Hawken. So since every person playing Hawken is better than them, a lack of matchmaker putting them consistently with people at least near their skill, is going to make the chances of getting those people that this player has a chance of beating in a match together go way way down, and as such, this player is going to almost never win a match.

The same thought process can be applied to the best player, or any in between. In fact, the only people who are likely to have a 50% win loss ratio are people in the 50th percentile. The median gamers - and they, for the large majority of their experience are going to win without any challenge whatsoever, or will get utterly stomped.

Now, the one thing pulling this argument away from the whole truth here is that each team has six players. This will have the effect of averaging things out, but it will only slow the process down. The most likely composition of five other random players is one that will tend toward an average of the average skill of players in the game - meaning from the standpoint of anyone playing the game, they are still going to be the one skewing the balance for or against their teams over time.

This all being said, that's not even close to what the thread is about. Nor is it about recreating new player experience. A chemist doesn't magically become his chemistry set every time he wants to explore, he uses the tools available to him.

My results in the past have overwhelmingly expressed complaints about the grind. That has changed now to some extent, and I would like to see if that has been solved in any measure.

You might be happy to know that I've received far less complaints this time around, meaning capnJosh's pricing modifications seem to have had something of the desired effect.

Currently, the complaints I hear are about not knowing how to play the game, meaning the next issue I will likely stress on the forums will be changing the tutorial.

That is how testing works. One experiments and gathers results, and then shares them with his or her peers.

[talon70]

The reason your smurfs are gaining mmr slower is because you are  playing other smurfs and all the constant smurfers have gotten better. You Constant-make-a smurfers are not testing or helping or being constructive to the new player retention issue at all and you know it. Spinning it otherwise is bs.

[ticklemyiguana]

The reason your smurfs are gaining mmr slower is because you are playing other smurfs and all the constant smurfers have gotten better. You Constant-make-a smurfers are not testing or helping or being constructive to the new player retention issue at all and you know it. Spinning it otherwise is bs.

Repeating the same nonsense as everyone else in the thread who's said the same thing only means you're repeating the same nonsense as everyone else in the thread who's said the same thing. You're in great company.

Though you do bring up a point I'd like to address. I encounter practically zero smurfs until about the 2000 range. This has been the norm on every account I've made. This leads me to believe that the "anti-smurfing" campaign is rather overzealous in their interpretation of the issue.

And that shoe fits. You are literally trying to tell me what's own experience is despite everything in the thread saying it's not.

Edited by ticklemyiguana, 20 June 2015 - 10:02 AM.

[Call_Me_Ishmael]

The reason your smurfs are gaining mmr slower is because you are  playing other smurfs and all the constant smurfers have gotten better. You Constant-make-a smurfers are not testing or helping or being constructive to the new player retention issue at all and you know it. Spinning it otherwise is bs.

 

I still top the match, and still get full value for it.

 

As for testing, I answered the question:  MMR didn't change.   I don't use that test account anymore, after 11 matches with it.

 

Crafty made a 4-page request for me to use my main for a while (Lunatic_Fringe).  I'm dropped into the SAME EFFING MATCHES by the Matchmaker, but with a 2200-ish MMR.  How is that helping anything for newer players?

 

I've explained how MMR works.  Me joining a 1700-level server with a 2200 MMR will really hurt anyone not in the top 50% by more than if I hadn't joined the match. Using an alt with a lower MMR is kinder in this regard.

 

All I can do is offer critique/advice/teach during such matches.  Many of the new players are salty/swagger-y and don't want to hear it, or I get accused of haxx.  I come to the conclusion that most of those who are NOT complaining are alt/smurfs.

 

Eventually, due to the lack of higher-level servers outside of TPG/WW, I will move back over about 2250/2300, and then won't find matches once again.  What do I do then, Talon?  Reduce my MMR deliberately?  How is that any different than using an alt with a lower MMR?

 

One thing about using an Alt.  No one recognizes me, and gives me grief for playing in the only server available to me/matchmaker-matched server.

Edited by Call_Me_Ishmael, 20 June 2015 - 10:00 AM.

[ticklemyiguana]

I still top the match, and still get full value for it.
 
As for testing, I answered the question:  MMR didn't change.   I don't use that test account anymore, after 11 matches with it.
 
Crafty made a 4-page request for me to use my main for a while (Lunatic_Fringe).  I'm dropped into the SAME EFFING MATCHES by the Matchmaker, but with a 2200-ish MMR.  How is that helping anything for newer players?
 
I've explained how MMR works.  Me joining a 1700-level server with a 2200 MMR will really hurt anyone not in the top 50% by more than if I hadn't joined the match. Using an alt with a lower MMR is kinder in this regard.
 
All I can do is offer critique/advice/teach during such matches.  Many of the new players are salty/swagger-y and don't want to hear it, or I get accused of haxx.  I come to the conclusion that most of those who are NOT complaining are alt/smurfs.
 
Eventually, due to the lack of higher-level servers outside of TPG/WW, I will move back over about 2250/2300, and then won't find matches once again.  What do I do then, Talon?  Reduce my MMR deliberately?  How is that any different than using an alt with a lower MMR?
 
One thing about using an Alt.  No one recognizes me, and gives me grief for playing in the only server available to me/matchmaker-matched server.

While I agree with everything here, I'm not sure that playing at 2200 is less kind. The game expects you to trounce on that account and as such won't make drastic adjustments, whereas on a 1700 acct, the game should go "woah something's wrong here" and adjust yours higher and others lower.

[Call_Me_Ishmael]

If I move the server average over a player's MMR, and that player scores in the bottom half, he gets penalized more.  By joining, I move the point where the curve inverts over his head.

[ticklemyiguana]

If I move the server average over a player's MMR, and that player scores in the bottom half, he gets penalized more. By joining, I move the point where the curve inverts over his head.

Hmmmm. Pending an edit.

 

Ok, so to me this doesn't make sense. If the average MMR in the game is higher than a given player's, and he or she performs poorly relative to the average, that's expected, and so it doesn't make sense to lower that player's MMR substantially.

 

However if your account doesn't raise the average MMR by much, and that player falls at the average or above it, yet fails to perform at the expected level relative to that average, then I'd expect their MMR to go down.

Edited by ticklemyiguana, 20 June 2015 - 02:14 PM.

[SS396]

There are better official ways to get feedback of new player experiences and grind without harm. They are called called questionnaires, let Reloaded do their own homework.

[ThirdEyE]

If I move the server average over a player's MMR, and that player scores in the bottom half, he gets penalized more.  By joining, I move the point where the curve inverts over his head.

That's a pretty weak justification.  Say you're a 2200 player joining an 1700 rated server with 11 people in it.  After you join the server rating will be ~1742, and it's highly unlikely that more than 1-2 people fall between 1700 and 1742 given the typical server MMR spread.  The difference in server rating is basically negligable.

 

The autobalancer will do a much better job of creating a fair and fun match when everyone in the server has an MMR representative of their skill.  Now that matchmaker will place you in a server regardless of your MMR it's best to play on the account that best represents your skill.

 

SS396, how would you go about distributing questionnaires to the new or low-rated players who don't visit the forums?

[CraftyDus]

Crafty made a 4-page request for me to use my main for a while (Lunatic_Fringe).

 

 

SMH 

+

all my facepalm

_______________

 

= no I didn't

[Call_Me_Ishmael]

 

SMH 

+

all my facepalm

_______________

 

= no I didn't

yeah, we chatted after I posted, and I forgot to go edit this.

 

Sorry.  Have a beer?

[Call_Me_Ishmael]

That's a pretty weak justification.  Say you're a 2200 player joining an 1700 rated server with 11 people in it.  After you join the server rating will be ~1742, and it's highly unlikely that more than 1-2 people fall between 1700 and 1742 given the typical server MMR spread.  The difference in server rating is basically negligable.

 

The autobalancer will do a much better job of creating a fair and fun match when everyone in the server has an MMR representative of their skill.  Now that matchmaker will place you in a server regardless of your MMR it's best to play on the account that best represents your skill.

 

SS396, how would you go about distributing questionnaires to the new or low-rated players who don't visit the forums?

 

 

Not really justifying it, just noting it.  Maybe 1-2% of players are exposed if I (or another) did this, tops?

 

 

It's worse if Automatch puts me into a 4-person 1500-level server, fer sure.

Edited by Call_Me_Ishmael, 20 June 2015 - 05:41 PM.

[Call_Me_Ishmael]

Hmmmm. Pending an edit.

 

Ok, so to me this doesn't make sense. If the average MMR in the game is higher than a given player's, and he or she performs poorly relative to the average, that's expected, and so it doesn't make sense to lower that player's MMR substantially.

 

However if your account doesn't raise the average MMR by much, and that player falls at the average or above it, yet fails to perform at the expected level relative to that average, then I'd expect their MMR to go down.

 

 

The system WILL penalize you for trying.... like in Chess.

[SS396]

SS396, how would you go about distributing questionnaires to the new or low-rated players who don't visit the forums?

Well, I am currently at a Diamondbacks ball game, so my wall of text algorithm is going to have to wait, but I'll post it later when I get home. But one thing I do is go to streams and it's very easy to get information from new players that way. I've also run into both Tiggs and Josh doing the same thing.

Another idea would be to use spectator mode so that you wouldn't effect the gameplay but can still communicate using in game chat.

But yeah, the amount of data that the game collects on players makes it very easy to sort into a match which would satisfy your question nicely.

It reminds me of one of assignments in my prolog class. God what a horrible language.

[IareDave]

SMH
+
all my facepalm
_______________

= no I didn't

Nah u just got busted bro

[CraftyDus]

yeah, we chatted after I posted, and I forgot to go edit this.

 

Sorry.  Have a beer?

 

deal

[talon70]

To make a smurf, play it , and give it to someone must be against tos and you are not doing anyone any favors imo. also @ 2000mmr there are still smurfs called alts.

 

I doubt mmr calculations have changed yet but I think maybe parameters of what is considered a 3 star server are tightened up again and that is affecting how fast mmr is gained or lost due to closer in game mmrs. It seems you get stuck at  the levels you are forced to play at in the pub servers in a way.

[SS396]

SS396, how would you go about distributing questionnaires to the new or low-rated players who don't visit the forums?

Ok, back from the disappointing ball game.    

First off, after reading your question, I'm not sure if you meant a random player (Tickle) distributing questionnaires or Reloaded themselves. I am just going to say that I don't think its any players responsibility to do that, so I'm going to just assume you meant Reloaded themselves as the random player having access to this account information isn't very likely.

Now, I'm not exactly up to date on what all the fields the game holds on each and every account, but I know for a fact they hold an e-mail address. I would simply use that address to send a questionnaire to every player that matches a specific set of cases.

Those cases are:
1) Have an account created after some random date, like take the date of when Reloaded bought Hawken as an example.
2) Have an IP address that does NOT link the new account (above) to any other account. (not sure the player database holds IP address fields, it should for banning purposes so I think it might, but its also a valid way to help identify smurfs so they should at least start tracking it, yes I know its not perfect but its better than nothing at all)

They probably are not the best cases but you asked me for something, so this is what I thought up during the ball game. In simple terms a new account that isn't a veteran in a smurf.

If case 2 cross referencing by IP address isn't possible, one other way would be to remove accounts on the first list (case 1) by high win percentage, MMR, and K/D ratio, as smurfs normally excel above the rest of the population in those stats during their relatively short playtime.

But one thing I do is go to streams and it's very easy to get information from new players that way. I've also run into both Tiggs and Josh doing the same thing.

Another thing I just want to add, I notice a great amount of the time when I do see Josh or Tiggs, they are not asking questions or bothering the player AT ALL, just observing and taking notes. I myself have been noticed asking loaded questions, which is understandable, I have a tendency to do that. It has also come to my attention they might be observing and not logged into accounts, and/or they (Josh/Tiggs and others++) also have smurf accounts on twitch for lurking "ahem" I mean observational purposes. lol.

I will remind everyone of that APB players comment on Reddit where they said LadyTiggs is always watching. XD

Trust me, they know more about the new players experience than any of these so called experiments do.

[ticklemyiguana]

Trust me, they know more about the new players experience than any of these so called experiments do.

 

I appreciate the write up, but I'm betting, 100% that this is inaccurate. Tiggs has other games to concentrate on. Josh has the game to concentrate on. Neither of them are well versed in the game, its progression, the stereotypes it's addressed, the differing opinions of new players and old ones, the different builds it's had, the regional differences that exist, what questions have been asked 1,000 times before and which are just surfacing, and certainly not what the game feels like at the high end of play and how to preserve the good stuff up there while still addressing new players.

 

And I'm betting that despite that not all being new player experience, it's not the easiest task in the world to sort out what is and isn't.

 

Short and simple, as good a job as they've done, regardless of whatever esteem I have in them, they don't know Hawken as well as most of us.

 

That's part of the reason these forums are around - so they can get a slice of the experience of players. The problem is, is that these forums are biased, they tend toward people who are better than average, have been playing longer, and have invested more time and money into the game.

 

I feel it's worth getting a view of the other side of the coin.

 

This thread has encouraged me to start documenting my experiences, as it has certainly not changed my mind on that what I do has a net positive result, rather quite the opposite. It has informed me that most negative opinions are grounded in uneducated rhetoric that goes directly against my observations- and I'd like to share these things with more than anecdotal evidence.

Edited by ticklemyiguana, 20 June 2015 - 11:31 PM.

[SS396]

I appreciate the write up, but I'm betting, 100% that this is inaccurate. Tiggs has other games to concentrate on. Josh has the game to concentrate on. Neither of them are well versed in the game, its progression, the stereotypes it's addressed, the differing opinions of new players and old ones, the different builds it's had, the regional differences that exist, what questions have been asked 1,000 times before and which are just surfacing, and certainly not what the game feels like at the high end of play and how to preserve the good stuff up there while still addressing new players.
 
And I'm betting that despite that not all being new player experience, it's not the easiest task in the world to sort out what is and isn't.

Ok, I can dig that, however there are others than just Josh and Tiggs. I've had some private messages with Josh where he mentioned and alluded that there were more than just those two observing us and reading our posts and comments in twitch.

[M4st0d0n]

In terms of the math you're presenting that's still very wrong, unless we're speaking of truly infinite time where infinity times one win out of every hundred is equal to infinity times 100 wins out of every 100. This is a silly assertion to make, as it literally requires people to play the game forever. Like I said, you are assuming that the statistical average is equivalent to the average of every player.

For any reasonable period of time, the worst player in Hawken is never going to beat the best player. They're not going to beat the second, third, fourth, fiftieth, or three hundredth best either.

They're going to beat someone of equal skill roughly fifty percent of the time, but that pool of people is one, because this is the worst player in Hawken. So since every person playing Hawken is better than them, a lack of matchmaker putting them consistently with people at least near their skill, is going to make the chances of getting those people that this player has a chance of beating in a match together go way way down, and as such, this player is going to almost never win a match.

The same thought process can be applied to the best player, or any in between. In fact, the only people who are likely to have a 50% win loss ratio are people in the 50th percentile. The median gamers - and they, for the large majority of their experience are going to win without any challenge whatsoever, or will get utterly stomped.

Now, the one thing pulling this argument away from the whole truth here is that each team has six players. This will have the effect of averaging things out, but it will only slow the process down. The most likely composition of five other random players is one that will tend toward an average of the average skill of players in the game - meaning from the standpoint of anyone playing the game, they are still going to be the one skewing the balance for or against their teams over time.

This all being said, that's not even close to what the thread is about. Nor is it about recreating new player experience. A chemist doesn't magically become his chemistry set every time he wants to explore, he uses the tools available to him.

My results in the past have overwhelmingly expressed complaints about the grind. That has changed now to some extent, and I would like to see if that has been solved in any measure.

You might be happy to know that I've received far less complaints this time around, meaning capnJosh's pricing modifications seem to have had something of the desired effect.

Currently, the complaints I hear are about not knowing how to play the game, meaning the next issue I will likely stress on the forums will be changing the tutorial.

That is how testing works. One experiments and gathers results, and then shares them with his or her peers.

 

You're the one bringing infinite time and limits in the conversation. Of course it's sterile and abstract. So in your maths, there's infinite time, and truly infinite time, and they are different? You're talking about duels now, it's totally pointless. Matchmaker is used for TDM and objective based modes. MMR is a stat for the match maker. Just ask for a fecking ranked ladder to get some skill stats to tryhard.

 

You just need to be aware of a few things. Testing the grind with new pricing is an honorable idea. But as a community opinion leader you should just start to think about how smart it is to present smurfing as a testing tool to help the devs. There is data collected. Smurfes are making the data worthless, because player experience cant be monitored anymore. I've yet to see any Excel table you and your "peers" provided about gamedata. You're just thinking to do it now... Talk about uneducated, fool.

 

And just one little thing. You're not a tester. You're not a unique snowflake. You are a consumer already, like everyfuzzy bunnyngone else here.

[Call_Me_Ishmael]

Mast,

 

I'm curious.  The community includes some surprising people. Do you feel that there is no one in the community qualified to analyze?

[CraftyDus]

To make a smurf, play it , and give it to someone must be against tos and you are not doing anyone any favors imo

 

I did Space_ a favor during TPG s1 giving him a smurf PewPewPewTM to play.

His regular account had a glitched out friendslist which prevented him from joining on friends and participating in any organized scrims or matches.

ADH wasn't ever going to fix it.

This guy was a former QuakeWars pro, money LAN player.

It would have been a shame not to have him play.

[Flifang]

One way they could get new player experience could be to post a button in the default garage linking a form to fill out if they so choose and state it would really help them to know what they want

[ticklemyiguana]

You're the one bringing infinite time and limits in the conversation. Of course it's sterile and abstract. So in your maths, there's infinite time, and truly infinite time, and they are different? You're talking about duels now, it's totally pointless. Matchmaker is used for TDM and objective based modes. MMR is a stat for the match maker. Just ask for a fecking ranked ladder to get some skill stats to tryhard.

No. I said that what you were saying only worked with infinite time, where nothing makes sense for our purposes. I said truly infinite time after you responded and treated infinity like it was just a long period of time. I needed to clarify that what you were referencing was not possible. I'm not sure what the rest of that stuff means.

You just need to be aware of a few things. Testing the grind with new pricing is an honorable idea. But as a community opinion leader you should just start to think about how smart it is to present smurfing as a testing tool to help the devs. There is data collected. Smurfes are making the data worthless, because player experience cant be monitored anymore. I've yet to see any Excel table you and your "peers" provided about gamedata. You're just thinking to do it now... Talk about uneducated, fool.

And just one little thing. You're not a tester. You're not a unique snowflake. You are a consumer already, like everyfuzzy bunnyngone else here.

To be fair, this thread, I'll say it again, was brought up due to an observation about the rate at which MMR was being gained, and it conflicting with what appears to need to be done. It was not brought up to debate smurfing either in general or in specific, yet there were so many misconceptions presented that debates were then required.

My testing of the grind has no real need or method of being documented quantitatively, as the pricing update was a surprise, and there are no numbers to compare it to. That being said, there was no need. I've heard a single complaint about the grind itself since the update, and that player wasn't even aware that the pricing structure had changed. Could you please refrain from trying to draw my ego out with insults?

As far as testing is concerned - we're all testers. This is beta.

Ok, I can dig that, however there are others than just Josh and Tiggs. I've had some private messages with Josh where he mentioned and alluded that there were more than just those two observing us and reading our posts and comments in twitch.

I would very much hope this is true.

Edited by ticklemyiguana, 21 June 2015 - 08:35 AM.

[M4st0d0n]

Mast,

 

I'm curious.  The community includes some surprising people. Do you feel that there is no one in the community qualified to analyze?

 

There is plenty of very qualified people in this community who could analyse data. IMO you're just not one of them. This whole smurfing to test nonsense for exemple is as good as using random names to label samples. If you're really serious about contributing to the development of this game, just send them your resume. Dont ever work for free. We've already seen how ADH "valorized" some testers in the past with Hawken Hero and Advanced Battalion. Some of them are still so pissed they'll use every forum they'll find to fuzzy bunny on the old devs.

[Call_Me_Ishmael]

In your opinion, I'm not one of them?

 

What do you know of me, Mastodon?  What about my professional life, education, and experience?

[M4st0d0n]

In your opinion, I'm not one of them?

 

What do you know of me, Mastodon?  What about my professional life, education, and experience?

 

I've formed my opinion from what I read and it was enough. But if it can make you feel good, opinions are like arseholes, everyone got one. I'm not working for Reloaded nor planning to so you still got all your chances. Smurf.

[ticklemyiguana]

Ah fuzzy bunny it. You've exceeded my limit for caring what your opinion is. You're not a person I need to justify actions to or hand my resume.

Edited by ticklemyiguana, 21 June 2015 - 11:21 AM.

[Call_Me_Ishmael]

So, let me get this straight.  You think anything I may state is invalid because I am too good at this game for you to play with me, so you pejoratively use the word 'smurf' to discount anything I might say?

 

In your eyes, anything I might say is invalid?

 

I mean, I think I'm a pretty successful guy.  I have a few degrees, a nice title, and just a few people working for me.  I've somehow managed to do that all for a while, I must kinda know what I'm doing, wouldn't you think?  Between you and me, I think I fake it pretty well, and yet, I feel like I kind of have a handle on statistical analysis.

 

You could ask around... there are a few folks who know what I do here on the boards.

[JeffMagnum]

You could ask around... there are a few folks who know what I do here on the boards.

He's actually a professional game ruiner and fun burglar

[PoopSlinger]

He's actually a professional game ruiner and fun burglar

Shut up smurf

[StubbornPuppet]

One way they could get new player experience could be to post a button in the default garage linking a form to fill out if they so choose and state it would really help them to know what they want

It would be great... except that I think 75% of the forms would come from long term players creating alt accounts so that they can fill out the form... again and again... to make sure that Reloaded knows "what players really want."

 

Anyhow, it may not be with 'MMR' specifically, but something has certainly been different for me over the last 2-3 weeks:

 

For one, I can barely find matches any more.  It doesn't matter much what day or time of day I log in, there are never more than 4-5 matches available - and those are almost always DM and TDM.  If I want to play Siege or MA, I have to actually select an empty server and sit and wait for 15-20 minutes while the same 10 people join and quit and rejoin and requit until they realize that this is the only match.  It could be that our player base has really tanked in the last couple of weeks, but I really don't feel that's the case.  If I look at the closed matches, it seems that there are plenty out there which I just can't get access to.

 

Second, although it's been said that Josh and team rejiggered the matchmaking, even though there are fewer matches, when I do get in I still see mostly SRD's of 1450 - 2450 and 1650 - 2550.  That's the same as it's been for over a year - except now the "star rating" doesn't seem to represent anything (before it was 1 star = I'm top mmr, 2 star = I'm bottom mmr and 3 star = I'm middle mmr).

 

Third, since I'm starting almost all the matches I play now, and I'm an avg of 1970-2000 mmr, I can never figure out how the first person to join my matches is usually either 2400-2500 or 1400-1500.  If 'they' tightened up the matchmaking deviation, I suppose the only way to explain it is a 1400 or 2500 player sitting in the general matchmaking queue for 90 seconds... and I'm the only MA or Siege match in the whole region.

 

Finally, where as my MMR used to actually fluctuate and change from day to day (dramatically on the odd occurrence where I was highest MMR in the lobby or lowest MMR in the lobby), it now sits almost totally stagnant.  I can play exceptionally well in a 1600-2400 lobby for several hours, topping the carousel, getting lots of MVP, etc. and it only moves about 5 points up over the course of 8 matches.

 

The only thing that I can actually see that seems specifically different is that there are a whole lot of level 20 and lower players in every match I play.

[Flifang]

So, what you're telling me is that long term players who have stayed this long and would probably want the game to succeed: would go out of their way to skew data on suggestions on improving the game? The age average for this community is quite high (higher than most games) and I'm pretty sure plenty of people would know this could do possible damage. Who's to say the answers would even be taken seriously to begin with? Really what we need are players that are interested in improving the gameplay for themselves so they might stay, and I think getting people to the forums or even just an in-game survey would help achieve that. Hell, some promotional codes to use for getting boosts and in-game currency would achieve this. People love promo codes and I remember that's how I got involved on the forums Eons ago. Again, this is merely about holding people's interest not actually getting quick concrete fixes to things.

[StubbornPuppet]

So, what you're telling me is that long term players who have stayed this long and would probably want the game to succeed: would go out of their way to skew data on suggestions on improving the game?

Yup, that's what I'm saying.  I'm not pointing any fingers and have no idea which people might be inclined to do so... but that's human nature; "My point of view is right... I don't understand why people can't see it the same way... so many people seem to be so dumb that they can't see the same truth that is obvious to me... if only there were a way to make sure that these people don't ruin it for me..."

 

I do not believe that the people would feel at all, as you say, that they are "going out of their way to skew data on suggestions on improving the game" - but I just believe that human nature would lead them to believe they're doing the right thing, for the greater good.  Heck, I don't even fault them for it - we've all had moments where we overlook objectivity to pursue our own interests because we think we're right.

 

I love your suggestions, Flifang, about the promo codes and I agree getting and retaining players is more important than any of the other stuff.

[SatelliteJack]

Alright, now I'm starting to get upset. I just played three rounds, and won all of them. I checked my MMR before I started, and after each round. I started out 1721. Then I went down to 1715. Then down to 1703. After the last round, it went up to 1715. This leaves me with a three win streak, and an MMR 6 lower than what I started with.

[ticklemyiguana]

Alright, now I'm starting to get upset. I just played three rounds, and won all of them. I checked my MMR before I started, and after each round. I started out 1721. Then I went down to 1715. Then down to 1703. After the last round, it went up to 1715. This leaves me with a three win streak, and an MMR 6 lower than what I started with.

Wins don't necessarily mean a rise in MMR. There's some sort of curve in each game that the algorithm responsible for MMR sort of sets as a guess as to where players will score based on their current MMR. If you violate that curve your MMR is effected.

 

Basically the game is telling you that it's disappointed in you.

[SatelliteJack]

Wins don't necessarily mean a rise in MMR. There's some sort of curve in each game that the algorithm responsible for MMR sort of sets as a guess as to where players will score based on their current MMR. If you violate that curve your MMR is effected.

 

Basically the game is telling you that it's disappointed in you.

 

But I was in the top 3 every game! D:

[ticklemyiguana]

But I was in the top 3 every game! D:

Dude. You have no idea. On my highest accounts, those ranging from 2300 to 2500, if I get into the wrong game I can outscore the next guy down by 2 or 300 points and still lose MMR.

 

The game just expects you to stomp harder.

[SatelliteJack]

Dude. You have no idea. On my highest accounts, those ranging from 2300 to 2500, if I get into the wrong game I can outscore the next guy down by 2 or 300 points and still lose MMR.

 

The game just expects you to stomp harder.

 

This game is a like a demanding Asian father.