283 replies to this topic

[Nightbird]

thievingmagpi, on 10 October 2020 - 01:46 PM, said:

I didn't mention Jay Z at all.

But go ahead, keep showing everyone how insecure you are.

lol ok buddy

[Teonas]

I came here for new numbers but all I got was Waldorf and Statler shouting at each other

[Nightbird]

Ball is currently in PGI's court since they promised an update and they have more data than is shared with us.

[Nightbird]

Since we're way past the promised update, here's the 240 games comparison. I missed the snapshot for 240 exactly so I had to use the full season 51 data, which puts us 15 games over.



The accuracy in the fringes are a bit lacking but the main columns in the center are much closer to the prediction than I hoped for. I predicted 44.6% in the center 3 columns, and the result is 44.2%. Even the shape holds true.

If anyone forgot why we're using average games instead of total games, the reason was posted here: https://mwomercs.com...ost__p__6340950

[Natural Predator]

Hey magpie thanks for proving the old adage true. “Better to say nothing and be thought the fool, than to speak and have it proven”

In other words you are defining proof that some people don’t know when to shut the **** up.

https://youtu.be/ks072waMayk

Ggclose

[thievingmagpi]

Cool, add another nobody to the join the fan club.

The results speak clearly.

Those results being: Nightbird failed to present anything of worth and that fact lives rent free in his head.

[Nightbird]

thievingmagpi, on 26 November 2020 - 03:00 PM, said:

Nightbird failed to present anything of worth and that fact lives rent free in his head.

Speaks volumes about you and the group you represent.

[thievingmagpi]

Nightbird, on 26 November 2020 - 03:14 PM, said:

Speaks volumes about you and the group you represent.

Which group is that?

[Sjorpha]

thievingmagpi, on 26 November 2020 - 03:00 PM, said:

The results speak clearly.

They do, and as the results closely match Nightbirds predictions the clear message they speak is that he was right.

[thievingmagpi]

Sjorpha, on 26 November 2020 - 03:36 PM, said:

They do, and as the results closely match Nightbirds predictions the clear message they speak is that he was right.

No one has ever debated "what the numbers say".

Turns out developing an actionable plan takes more than spewing out numbers.

Edited by thievingmagpi, 26 November 2020 - 03:37 PM.

[Sjorpha]

thievingmagpi, on 26 November 2020 - 03:37 PM, said:

Turns out developing an actionable plan takes more than spewing out numbers.

There were several actionable suggestions (assuming we're using "actionable" in the sense of doable), this was one of them. These are all fairly simple ideas after all, none of them that hard to do.

Being actionable isn't very useful if the results are bad though. PGI choose to implement the low quality crowd pleasing suggestion, and so now we're stuck with a dysfunctional matchmaker.

Now to me, given that the results are in and clearly show the system has failed, it's very weird to keep defending a system that doesn't work. We should be working now to make them admit the failure and replace it with a w/l based system, that is in everyones interest and there's no need for any bickering.

[Natural Predator]

Sjorpha, on 26 November 2020 - 05:32 PM, said:

There were several actionable suggestions (assuming we're using "actionable" in the sense of doable), this was one of them. These are all fairly simple ideas after all, none of them that hard to do.

Being actionable isn't very useful if the results are bad though. PGI choose to implement the low quality crowd pleasing suggestion, and so now we're stuck with a dysfunctional matchmaker.

Now to me, given that the results are in and clearly show the system has failed, it's very weird to keep defending a system that doesn't work. We should be working now to make them admit the failure and replace it with a w/l based system, that is in everyones interest and there's no need for any bickering.

Magpie would die on any hill if he thinks hes right. Especially when hes total not. That is just who he is. A sheep.

thievingmagpi, on 26 November 2020 - 03:00 PM, said:

Cool, add another nobody to the join the fan club.

The results speak clearly.

Those results being: Nightbird failed to present anything of worth and that fact lives rent free in his head.

Okay zoomer

[thievingmagpi]

Sjorpha, on 26 November 2020 - 05:32 PM, said:

There were several actionable suggestions (assuming we're using "actionable" in the sense of doable), this was one of them.

A bunch of numbers aren't an idea presented.

Sjorpha, on 26 November 2020 - 05:32 PM, said:

These are all fairly simple ideas after all, none of them that hard to do.

If only it weren't PGI we are talking about.

Sjorpha, on 26 November 2020 - 05:32 PM, said:

Being actionable isn't very useful if the results are bad though. PGI choose to implement the low quality crowd pleasing suggestion, and so now we're stuck with a dysfunctional matchmaker.

PGI chose to adopt a better system than was had previously, which was one of, if not the only system developed with community input from a wide range of people that presented their suggestions in a clear and appropriate manner in the manner and time frame requested. Screaming into the void about how good your numbers are and doing nothing else is not an actionable solution.

It's pretty simple really.

Sjorpha, on 26 November 2020 - 05:32 PM, said:

Now to me, given that the results are in and clearly show the system has failed, it's very weird to keep defending a system that doesn't work.

Except no results show it "doesn't work".

Sjorpha, on 26 November 2020 - 05:32 PM, said:

We should be working now to make them admit the failure and replace it with a w/l based system,

BAHAHAHAHAHAHAHAHAHAHAHAHAHA

come back to me when you've paid the tiniest bit of attention to the last year, if not, 8 years of development.

Trashtier, on 26 November 2020 - 05:47 PM, said:

Magpie would die on any hill if he thinks hes right. Especially when hes total not. That is just who he is. A sheep.

Lmao. Sick burn. Sheep bro. Think 4 urself. Stay off the lamestream media bro. Lmao.

Trashtier, on 26 November 2020 - 05:47 PM, said:

Okay zoomer

Edited by thievingmagpi, 26 November 2020 - 06:41 PM.

[Nightbird]

thievingmagpi, on 26 November 2020 - 06:41 PM, said:

PGI chose to adopt a better system than was had previously, which was one of, if not the only system developed with community input from a wide range of people that presented their suggestions in a clear and appropriate manner in the manner and time frame requested. Screaming into the void about how good your numbers are and doing nothing else is not an actionable solution.

It's pretty simple really.

Not better by a long shot. The last 3 months of retention points made from the last 4 months of new PSR data show that the pop in each of the WLR brackets didn't change. At this same timepoint my dumb WLR model is predicted to have 60% of the pop in the green WLR bracket.




People that can't do math tend to hate on it, I understand.

[thievingmagpi]

Ah yes, the graph that shows absolutely no causality or control for any other variables, let alone any drop in player retention attributable to psr mechanic change. Cool.

Edited by thievingmagpi, 26 November 2020 - 07:24 PM.

[Nightbird]

Don't need any if all I need it to do in this thread is show how the current MM is just as bad as the previous one.

[thievingmagpi]

Except you haven't and never have. Lmao

[Nightbird]

thievingmagpi, on 26 November 2020 - 07:22 PM, said:

let alone any drop in player retention attributable to psr mechanic change. Cool.

It clearly shows a 8-10% drop in Season 48, 49 when the new PSR was first introduced. Then there is a bounce in retention at the same time PGI announced new MWO initiatives. However, as I mentioned in the retention thread, the role of a fair Matchmaker is to move everyone towards 1 WLR, as that means it is measuring people's skill accurately. Every WLR bracket line displays the percent of the pop for that season/month, and there's no notable changes from the old PSR/MM.

Basically, if the new PSR was an improvement, we'd see a large increase of pop in the green WLR bracket line. It didn't happen, as was predicted 5 months ago.

thievingmagpi, on 26 November 2020 - 07:31 PM, said:

Except you haven't and never have. Lmao

To anyone who has learned a bit of math, I offered perfect uncontroversial evidence.

[thievingmagpi]

Nightbird, on 26 November 2020 - 08:06 PM, said:

It clearly shows a 8-10% drop in Season 48, 49 when the new PSR was first introduced.

And you've invented the correlation there. The player population swelled since the beginning of the new year- both in recurring players and in new players with the number of average matches almost doubling. Any further season to season drop you've controlled for nothing. Nada. No disinterest in regrinding psr. No disinterest in soup. No other factors other than the one you want to pump.

Yawn.

Nightbird, on 26 November 2020 - 08:06 PM, said:

Then there is a bounce in retention at the same time PGI announced new MWO initiatives.

Oops, let's just leave out factors that confound your statement.

Nightbird, on 26 November 2020 - 08:06 PM, said:

Every WLR bracket line displays the percent of the pop for that season/month, and there's no notable changes from the old PSR/MM.

Ah yes, the 3 data points for season over season change. Lmao. All of which incorporate the inclusion of soup queue.

Nightbird, on 26 November 2020 - 08:06 PM, said:

To anyone who has learned a bit of math, I offered perfect uncontroversial evidence.

Except you've cherrypicked inconclusive data points throughout and presented nothing but loose correlation.

[Nightbird]

thievingmagpi, on 26 November 2020 - 09:25 PM, said:

snip

I did say "to anyone who has learned a bit of math"


Let me take a different tack. Suppose someone goes to a car mechanic, and insists they have 6 cylinder tires, an automatic engine, and ABS transmission. Sure, these are all car terms, but the mechanic knows the guy is just putting together words without knowing their meaning but wants to sound knowledgeable.

You are throwing some stats terms out there, but to any professional, it is embarrassingly clear you don't know what the words mean. You are that guy.

Edited by Nightbird, 26 November 2020 - 09:59 PM.