Jump to content

The Last Match Maker Thread We Need


248 replies to this topic

#41 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 09 June 2019 - 11:54 AM

View PostLordNothing, on 09 June 2019 - 11:49 AM, said:

make psr mean something and so on. your analysis while thorough has not revealed any new information that can be used to improve the game.


Saying something or other is as valuable as a fart in the wind. What is valuable is metrics and accuracy.

#42 LordNothing

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 17,224 posts

Posted 09 June 2019 - 11:58 AM

View PostNightbird, on 09 June 2019 - 11:54 AM, said:


Saying something or other is as valuable as a fart in the wind. What is valuable is metrics and accuracy.


all the metrics in the world wouldn't convince pgi to make the necessary changes, even if its just a number in an xml file. when the ship is sinking is not the time to do a study on vessel design optimization.

Edited by LordNothing, 09 June 2019 - 12:01 PM.


#43 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 09 June 2019 - 12:06 PM

View PostLordNothing, on 09 June 2019 - 11:58 AM, said:

all the metrics in the world wouldn't convince pgi to make the necessary changes, even if its just a number in an xml file.


This thread is likely the first professional level analysis on the entire forums, with others being degrees of informed (or not so informed) opinions. If anyone with business background or instinct reads this OP post though, they will face palm. That I guarantee.

P.S. I'm not going for a change, I'm going for the FP.

#44 LordNothing

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 17,224 posts

Posted 09 June 2019 - 12:16 PM

View PostNightbird, on 09 June 2019 - 12:06 PM, said:


This thread is likely the first professional level analysis on the entire forums, with others being degrees of informed (or not so informed) opinions. If anyone with business background or instinct reads this OP post though, they will face palm. That I guarantee.

P.S. I'm not going for a change, I'm going for the FP.


no its not. this game community is made up of a lot of 35+ players, most are professionals at this point in their life. some have jobs in the stem fields and are more than capable of doing their own analysis. they pop up from time to time and get immediately ignored by pgi. they likely have their own metrics and graphs and professionals who do that sort of thing yet are still facing a game in decline.

#45 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 09 June 2019 - 12:20 PM

View PostLordNothing, on 09 June 2019 - 12:16 PM, said:

no its not. this game community is made up of a lot of 35+ players, most are professionals at this point in their life. some have jobs in the stem fields and are more than capable of doing their own analysis. they pop up from time to time and get immediately ignored by pgi. they likely have their own metrics and graphs and professionals who do that sort of thing yet are still facing a game in decline.


Being in a STEM field doesn't mean having a stats background. If they don't have a degree in, and produce stats models for a living, they'll be no more accurate. Being a dentist doesn't mean you can suddenly do heart surgery.

#46 LordNothing

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 17,224 posts

Posted 09 June 2019 - 12:24 PM

View PostNightbird, on 09 June 2019 - 12:20 PM, said:


Being in a STEM field doesn't mean having a stats background. If they don't have a degree in, and produce stats models for a living, they'll be no more accurate. Being a dentist doesn't mean you can suddenly do heart surgery.


no but its not unusual that those people have a few statmath classes under their belt.

Edited by LordNothing, 09 June 2019 - 12:25 PM.


#47 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 09 June 2019 - 12:30 PM

View PostLordNothing, on 09 June 2019 - 12:24 PM, said:

no but its not unusual that those people have a few statmath classes under their belt.


Which I'll point out is not the same as doing it professionally. I make a living getting paid by companies to analyze products they are spending at a minimum hundreds of millions of dollars developing (often billions), so they can make informed decisions as to whether to cancel, change, or move forward with their development plan.

As I listed in the OP, these are healthcare products that directly impact the life, wellness, and happiness of people. All three are endpoints I work with every day, one of those applies to gaming.

#48 LordNothing

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 17,224 posts

Posted 09 June 2019 - 12:36 PM

View PostNightbird, on 09 June 2019 - 12:28 PM, said:


Which I'll point out is not the same as doing it professionally. I make a living getting paid by companies to analyze products they are spending at a minimum hundreds of millions of dollars developing (often billions), so they can make informed decisions as to whether to cancel, change, or move forward with their development plan.


again this is where this kind of analysis belongs, at the beginning of plan. not after the plan has been executed, found lackluster, refactored and to little effect. but the problem was not with your analysis. that was ok, il even admit it was a good read. but with plan of action which was just a one liner.

the problems with this game are the frequent poor design choices and the time wasted on hammering them into a quasi functional state, or abandoning it for a new system. the best we can hope for is a graceful decline of mwo or even a release of the server side once the game has run its course.

at some point the walls between the various modes are going to have to come down and you might even see the matchmaker switch its self off all together in an effort to get timely matches. wait times for fp even qp are starting to get unrealistically long.

Edited by LordNothing, 09 June 2019 - 12:39 PM.


#49 TheArisen

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 6,040 posts
  • LocationCalifornia

Posted 09 June 2019 - 12:36 PM

Excellent post and a great read. Here's hoping PGI sees this.

#50 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 09 June 2019 - 12:39 PM

View PostLordNothing, on 09 June 2019 - 12:36 PM, said:


again this is where this kind of analysis belongs, at the beginning of plan. not after the plan has been executed, found lackluster, refactored and to little effect. but the problem was not with your analysis. that was ok, il even admit it was a good read. but the problems with this game are the frequent poor design choices and the time wasted on hammering them into a quasi functional state, or abandoning it for a new system. the best we can hope for is a graceful decline of mwo or even a release of the server side once the game has run its course.


Thank you.

This thread was written for my own edification. Both for PGI not making the choices that maximize revenue and growth, and for the community for complaining that PGI doesn't listen. They do listen, they do UNFORTUNATELY listen, to opinions not backed up with hard science and when it fails, the parts of the community that suggested something different clamor they weren't listening to them and the cycle repeats.

This is for all the face palms I did PGI....

Edited by Nightbird, 09 June 2019 - 12:50 PM.


#51 LordNothing

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 17,224 posts

Posted 09 June 2019 - 12:51 PM

View PostNightbird, on 09 June 2019 - 12:39 PM, said:


Thank you.

This thread was written for my own edification. Both for PGI not making the choices that maximize revenue and growth, and for the community for complaining that PGI doesn't listen. They do listen, they do UNFORTUNATELY listen, to opinions not backed up with hard science and when it fails, the parts of the community that suggested something different clamor they weren't listening to them and the cycle repeats.

For all the face palms I did PGI, you do one too.


still i wonder of working on the mm isnt a waste of time entirely. id much rather see shorter wait times so im not spending a third of my time allocated to mwo watching the thing spin.

#52 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 09 June 2019 - 12:58 PM

View PostLordNothing, on 09 June 2019 - 12:51 PM, said:

still i wonder of working on the mm isnt a waste of time entirely. id much rather see shorter wait times so im not spending a third of my time allocated to mwo watching the thing spin.


I don't have the data to simulate this, but a fast MM that still makes good matches should wait around a minute and launch when 48/72/96 players are queued up. The best 24 players are put into a game, then the next 24, etc. This way, tiers are basically dynamically generated. The skill rating would need something slightly more complicated than WLR to account for this.

#53 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 09 June 2019 - 08:36 PM

Just a bit of clarification, I do not expect this suggestion to be implemented. If PGI wants to improve things, they can hire a statistician to look at all the data they have but doesn't share with the community, and build the best possible MM that way. Given that MWO is EOL, I don't expect that.

Just as I mentioned in previous posts, the best I expect is some stomach acid churning at the opportunity and revenue missed.

#54 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 10 June 2019 - 02:36 AM

View PostNightbird, on 09 June 2019 - 12:06 PM, said:

This thread is likely the first professional level analysis on the entire forums, with others being degrees of informed (or not so informed) opinions.

LOL ... and here the actual point of this thread becomes obvious.

#55 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 10 June 2019 - 02:40 AM

View PostPhoenixFire55, on 10 June 2019 - 02:36 AM, said:

LOL ... and here the actual point of this thread becomes obvious.


Don't stop there, enlighten us

#56 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 10 June 2019 - 03:19 AM

View PostNightbird, on 10 June 2019 - 02:40 AM, said:

Don't stop there, enlighten us

Not gonna bother.

You know, I've actually spent a lot of time reading your first post (even tho I had a suspicion initially from the iffy thread name), considering the parameters of both models, thinking of arguments I could have brought for and against them. But then, I've read the rest of the thread. So yeah ... Not gonna bother bringing those either.

All I'm gonna say, is that back when PGI announced their own W/L based MM there have been threads almost exactly like this in plenty, claiming that it would be the ultimate answer to all your MM woes. Problem was, and still is, that just like PGI's understanding of MM was lacking, which eventually resulted in utter fail of their W/L-MM, so is yours. Since you probably know a thing or two about medicine you should be able to grasp the analogy ... W/L is a sympthom. You don't build a MM based on a sympthom, much like you don't focus on sympthoms if you want to cure a person. You cure the cause and symthoms go away automatically.

But anyhow ... Don't mind me really. You can surely pat yourself on the back. You've manipulated the models in a way that nearly perfectly illustrates your point but said literally nothing about exactly how far those models are from the actual MM situation. So not only you've done a lot of work for free, but also a lot of work that is neither useful nor relevant. That is what is called a "shperical horse in a vacuum" in scientific circles. But that is the beauty of the statistics and pure math modelling, you can easily manipulate them to pretty much any result you want.

#57 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 10 June 2019 - 03:27 AM

View PostPhoenixFire55, on 10 June 2019 - 03:19 AM, said:

Not gonna bother.


Given that MWO has never used a WLR MM, I have no clue what you're on...

#58 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 10 June 2019 - 03:31 AM

View PostNightbird, on 10 June 2019 - 03:27 AM, said:

Given that MWO has never used a WLR MM, I have no clue what you're on...

Hmmm, and Elo was based on what exactly? ... The gross domestic product of the Austro-Hungarian Empire?

#59 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 10 June 2019 - 03:33 AM

View PostPhoenixFire55, on 10 June 2019 - 03:31 AM, said:

Hmmm, and Elo was based on what exactly? ... The gross domestic product of the Austro-Hungarian Empire?


LOL, you think Elo and WLR are the same, OK... I'm gonna walk away now

#60 PhoenixFire55

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 5,725 posts
  • LocationSt.Petersburg / Outreach

Posted 10 June 2019 - 03:34 AM

View PostNightbird, on 10 June 2019 - 03:33 AM, said:

LOL, you think Elo and WLR are the same, OK... I'm gonna walk away now

Please do, because they are.





15 user(s) are reading this topic

0 members, 15 guests, 0 anonymous users