Jump to content

- - - - -

Quantified Tier Levels?


18 replies to this topic

#1 BodakOfSseth

    Member

  • PipPipPipPipPipPip
  • Leftenant General
  • Leftenant General
  • 267 posts
  • LocationBay Area, CA

Posted 07 October 2016 - 08:31 AM

I'm a bit of a nerd (go figure) and I've decided that teaching myself Data Analytics via MWO is an excellent use of my time.

To that end, I've been tracking my performance through a bunch of metrics, derived from my pilot stats.

This is just one of the mechs I'm tracking:
Posted Image
(yes yes, I know, not very impressive numbers. Gimme a break, I'm a noob) ;)

Anyway, I realized there is one more metric I'd REALLY like to track and that's my tier score (tracking my tier progression over time), but it seems the only data I can collect on that is the end-of-battle up-down-equal, and the frustratingly numerically absent progress bar on my profile screen in-game.

Am I not seeing where it is available, or is PGI keeping even that basic information to themselves?

#2 Hunka Junk

    Member

  • PipPipPipPipPipPipPip
  • The All Seeing
  • The All Seeing
  • 968 posts
  • LocationDrok's Forge

Posted 07 October 2016 - 08:44 AM

They haven't officially explained it, but there are people who have researched it. I found those threads through google, but I believe the two key metrics are damage output and whether you win or lose.

I started really focused on lights, then came some events where you had to pilot other things and suddenly Mr Bar came to life.

#3 Prof RJ Gumby

    Member

  • PipPipPipPipPipPipPipPip
  • The People's Hero
  • 1,061 posts

Posted 07 October 2016 - 08:46 AM

Tier score is based on match score, with different tresholds for win/tie/loss. Don't remember all the numbers, but i.e. to get a tie you need like <50 points on a win and 350-450 on a lose, or sth like that.

EDIT:
It's near impossible to judge whether a PSR gain is a small, medium or big, but you can check your match score after each match and try to establish where are the loss/no change/gain tresholds for victories, ties and defeats
(I'm 99% sure you cannot loose PSR on a win).

Edited by Prof RJ Gumby, 07 October 2016 - 08:49 AM.


#4 BodakOfSseth

    Member

  • PipPipPipPipPipPip
  • Leftenant General
  • Leftenant General
  • 267 posts
  • LocationBay Area, CA

Posted 07 October 2016 - 08:48 AM

Yeah, I saw a basic explanation for how it works, and in it PGI said they weren't going to share the math -
http://mwomercs.com/...-tiers-and-psr/

And yes, it says exact numbers and formulas will be kept internal - But I was hoping for just a value for the progress bar - even in game or whatever. Something I could use to track my own progression without guessing.

#5 Roughneck45

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Handsome Devil
  • The Handsome Devil
  • 4,452 posts
  • LocationOutreach

Posted 07 October 2016 - 09:02 AM

View PostScottAleric, on 07 October 2016 - 08:48 AM, said:

And yes, it says exact numbers and formulas will be kept internal - But I was hoping for just a value for the progress bar - even in game or whatever. Something I could use to track my own progression without guessing.

Would be nice.

People around here tend to get very upset whenever their in game efforts are quantified.

#6 BodakOfSseth

    Member

  • PipPipPipPipPipPip
  • Leftenant General
  • Leftenant General
  • 267 posts
  • LocationBay Area, CA

Posted 07 October 2016 - 09:36 AM

View PostRoughneck45, on 07 October 2016 - 09:02 AM, said:

Would be nice.

People around here tend to get very upset whenever their in game efforts are quantified.

I guess I can understand that. People are really big into self-delusion, telling themselves or believing they're better (at everything) than they actually are. But how are we to *really* know if we're doing better over time?

I believe they could hide it well enough from people who don't want to see.
Give a percentage complete value on the visual percentage bar that we can already see.
I don't need the formula behind it, or know that (totally making this up) getting a higher Lance in Formation score is more important to that than a higher Spotting Assist score.

However, I feel I have enough self-awareness and can track enough data that to conclude (again, totally making this up), I'm WAY better at playing my heavies as a team than I am with playing a medium as a team.

Look, data can be a great tool. Here's an example from another one of my mechs:
Posted Image

Kills/match spikes on 9/29, but I didn't win a single game. I had focused that day on using my weapons better and killing things, and it worked - to the detriment of my games and (probably) my tier score. Lesson learned: For the good of the team, stick to working together. My performance afterward has been looking better.

Don't get me wrong. I'm still early in my data collection phase, as a new player, learning the mechs I have available to me, the data needs to normalize, but I can already see trends, which helps my performance, which hopefully makes me a better mech pilot.

...I guess I'm ranting.

#7 Aleksandr Sergeyevich Kerensky

    Member

  • PipPipPipPipPipPipPipPip
  • 1,187 posts

Posted 07 October 2016 - 09:54 AM

View PostScottAleric, on 07 October 2016 - 09:36 AM, said:

I guess I can understand that. People are really big into self-delusion, telling themselves or believing they're better (at everything) than they actually are. But how are we to *really* know if we're doing better over time?

I believe they could hide it well enough from people who don't want to see.
Give a percentage complete value on the visual percentage bar that we can already see.
I don't need the formula behind it, or know that (totally making this up) getting a higher Lance in Formation score is more important to that than a higher Spotting Assist score.

However, I feel I have enough self-awareness and can track enough data that to conclude (again, totally making this up), I'm WAY better at playing my heavies as a team than I am with playing a medium as a team.

Look, data can be a great tool. Here's an example from another one of my mechs:
Posted Image

Kills/match spikes on 9/29, but I didn't win a single game. I had focused that day on using my weapons better and killing things, and it worked - to the detriment of my games and (probably) my tier score. Lesson learned: For the good of the team, stick to working together. My performance afterward has been looking better.

Don't get me wrong. I'm still early in my data collection phase, as a new player, learning the mechs I have available to me, the data needs to normalize, but I can already see trends, which helps my performance, which hopefully makes me a better mech pilot.

...I guess I'm ranting.


On a side note, once you get better at the game, I recommend making another account.

If you can dominate a game in a trial mech in your first set of games, your psr sky rockets, where wins gains are astronomical in value! You can literally hit T3 in a matter of a few games.

I think the more games an account has played, there more and more games it takes to gain psr. I think this is the reason many players are now stuck in the underhive... Its their 50k games under their belt weighing them down.

#8 TooDumbToQuit

    Member

  • PipPipPipPipPipPipPipPip
  • The Death Wish
  • 1,539 posts

Posted 07 October 2016 - 10:01 AM

View PostHunka Junk, on 07 October 2016 - 08:44 AM, said:

They haven't officially explained it, but there are people who have researched it. I found those threads through google, but I believe the two key metrics are damage output and whether you win or lose.

I started really focused on lights, then came some events where you had to pilot other things and suddenly Mr Bar came to life.



I bought a BUNCH of lights and some mediums in recent sales. And they are a ***** to do the pilot trees let alone a victory compared to piloting a Kodiak. And I guess playing a different one in each match is not a good idea but they are simply harder to move up in Tier wise.

In a bigger Mech you can do enough damage to go up in a lost, that is harder to do in Lights.

#9 BodakOfSseth

    Member

  • PipPipPipPipPipPip
  • Leftenant General
  • Leftenant General
  • 267 posts
  • LocationBay Area, CA

Posted 07 October 2016 - 10:20 AM

View PostAleksandr Sergeyevich Kerensky, on 07 October 2016 - 09:54 AM, said:

On a side note, once you get better at the game, I recommend making another account.

If you can dominate a game in a trial mech in your first set of games, your psr sky rockets, where wins gains are astronomical in value! You can literally hit T3 in a matter of a few games.

I think the more games an account has played, there more and more games it takes to gain psr. I think this is the reason many players are now stuck in the underhive... Its their 50k games under their belt weighing them down.


That's horrific.
And I'm not sure which is worse, the perception/possibility that PGI has made a system that needs to be gamed to better reflect your current skill level, or that people would abandon their play history to completely restart to get a better skill/rank.

IF that's true (and I'm not saying it is) then it ought to be simple enough to fix, and something easily done behind the scenes.
In their formula, count the records of say, the player's last 1000 (or some other number) matches or so.

#10 Rogue Jedi

    Member

  • PipPipPipPipPipPipPipPipPip
  • Overlord
  • Overlord
  • 4,908 posts
  • LocationSuffolk, England

Posted 07 October 2016 - 10:30 AM

When PSR was first introduced they seeded veteran players based on the last [insert number] (I think it may have been 4) months play.
At the time new players were inserted towards the top of tier 4 and it was literally possible for a new account to hit T3 on the first match. At about the time MWO launched on steam PGI decided T5 was a better fit for new players.

#11 Aleksandr Sergeyevich Kerensky

    Member

  • PipPipPipPipPipPipPipPip
  • 1,187 posts

Posted 07 October 2016 - 10:40 AM

View PostScottAleric, on 07 October 2016 - 10:20 AM, said:


That's horrific.
And I'm not sure which is worse, the perception/possibility that PGI has made a system that needs to be gamed to better reflect your current skill level, or that people would abandon their play history to completely restart to get a better skill/rank.

IF that's true (and I'm not saying it is) then it ought to be simple enough to fix, and something easily done behind the scenes.
In their formula, count the records of say, the player's last 1000 (or some other number) matches or so.


Believe me, you want good players out of T5 and T4 as fast as possible. Imagine the havok that would be caused, if a T1 smurf got to stay in T5 for 100 matches

#12 BodakOfSseth

    Member

  • PipPipPipPipPipPip
  • Leftenant General
  • Leftenant General
  • 267 posts
  • LocationBay Area, CA

Posted 07 October 2016 - 11:33 AM

View PostAleksandr Sergeyevich Kerensky, on 07 October 2016 - 10:40 AM, said:

Believe me, you want good players out of T5 and T4 as fast as possible. Imagine the havok that would be caused, if a T1 smurf got to stay in T5 for 100 matches

As I have no point of reference (beyond watching a dev drop video where Tina Benoit was on twitch with some friends, and the Atlas (or kodiak?) on her team hit 1500+ damage in the mission summary), I really can't imagine that havok...

Look, I have a (probably foolish) basic faith in humanity. People will do what they're doing to the best of their ability (at that time) - this includes the programmers and developers at PGI, the general players I go into matches with, pretty much everyone. Considering the programmers and game designers I've known, I'm really very sure that none of my suggestions above are new or particularly innovative - they've likely included some logic or mechanic that prevents the skill rank/level curve flatten out over time as the volume of performance bogs down changes in the data trend.

I wouldn't at all be surprised if, with patience and perseverance, I will climb my way out of those lower tiers soon enough.
I just want a tool to measure my progression.

If I REALLY wanted, I have the art skills necessary to copy that bar, measure it and derive a number from it - but that's a fair amount of work (and I'm tempted to just to prove it can be done...) UGH.

Dear Aleric,
It's just a game. You have a Pathfinder campaign to complete, blogs and art to make.
Posted Image

#13 Lukoi Banacek

    Member

  • PipPipPipPipPipPipPipPipPip
  • WC 2018 Top 12 Qualifier
  • WC 2018 Top 12 Qualifier
  • 4,353 posts

Posted 07 October 2016 - 11:35 AM

View PostScottAleric, on 07 October 2016 - 10:20 AM, said:

That's horrific.
And I'm not sure which is worse, the perception/possibility that PGI has made a system that needs to be gamed to better reflect your current skill level, or that people would abandon their play history to completely restart to get a better skill/rank.

IF that's true (and I'm not saying it is) then it ought to be simple enough to fix, and something easily done behind the scenes.
In their formula, count the records of say, the player's last 1000 (or some other number) matches or so.


Bear in mind, the starting PSR for players was based on performance metrics. So yes, they may have to play more to get out of a tier, but it's because their historic statistics placed them there.

That being said...it's debatable whether is should even be a player's goal to rise in tiers anyway.

#14 BodakOfSseth

    Member

  • PipPipPipPipPipPip
  • Leftenant General
  • Leftenant General
  • 267 posts
  • LocationBay Area, CA

Posted 07 October 2016 - 11:53 AM

View PostLukoi Banacek, on 07 October 2016 - 11:35 AM, said:

That being said...it's debatable whether is should even be a player's goal to rise in tiers anyway.


Why is that debatable?

#15 B L O O D W I T C H

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,426 posts

Posted 07 October 2016 - 11:55 AM

View PostAleksandr Sergeyevich Kerensky, on 07 October 2016 - 10:40 AM, said:

Believe me, you want good players out of T5 and T4 as fast as possible. Imagine the havok that would be caused, if a T1 smurf got to stay in T5 for 100 matches


Yep, feels like you go full pvp on a bunch of kittens. Guess who i was..
Spoiler


View PostAleksandr Sergeyevich Kerensky, on 07 October 2016 - 09:54 AM, said:

On a side note, once you get better at the game, I recommend making another account.

If you can dominate a game in a trial mech in your first set of games, your psr sky rockets, where wins gains are astronomical in value! You can literally hit T3 in a matter of a few games.


Would not recommend. Reason is simply the low playercount imo.
T3 will be matched up against T2 and T1 90% of all games. You don't want to sit in your only, fresh bought, basic 'mech with no modules and whatsoever VS a team with 2-5 fully kitted out T1 players.

I'd recommend, play really really bad, get your academy rewards and your 25 games Cadet bonus together, buy yourself 2-3 usable mechs and grind your way up to t3. You will have the mechs, c-bills and equipment (engines, some modules) to actually enjoy t1 gameplay (if it happens) rather then getting hammered right from the get go.
I think that is a much more enjoyable experience for new players. If they are up for a challenge, they can go play some Group query or faction play.

#16 Roughneck45

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Handsome Devil
  • The Handsome Devil
  • 4,452 posts
  • LocationOutreach

Posted 07 October 2016 - 12:30 PM

View PostScottAleric, on 07 October 2016 - 11:53 AM, said:


Why is that debatable?

Because of the mechlab and how you want to play.

If you enjoy making franken mechs more than min maxing for effective builds being in lower tiers can be less stressful.

Casual vs. Competitive mind set.

As far as making another account, doing it to reset your win/losses and game your tier is pretty shady and not really worth it. Making another account to get the cadet c-bill bonus again if you regret your purchases can be worth it.

Edited by Roughneck45, 07 October 2016 - 12:36 PM.


#17 Aleksandr Sergeyevich Kerensky

    Member

  • PipPipPipPipPipPipPipPip
  • 1,187 posts

Posted 07 October 2016 - 03:19 PM

View PostRoughneck45, on 07 October 2016 - 12:30 PM, said:

Because of the mechlab and how you want to play.

If you enjoy making franken mechs more than min maxing for effective builds being in lower tiers can be less stressful.

Casual vs. Competitive mind set.

As far as making another account, doing it to reset your win/losses and game your tier is pretty shady and not really worth it. Making another account to get the cadet c-bill bonus again if you regret your purchases can be worth it.


The biggest problem new players have, is they dont have enough c-bills to really find out which weight class they shine in. A lot of people - including myself - regretted the first mechs they purchased. Thats why now, I recommend people make a new account for every weight class so they can explore and tinker more.

The worst would be to loose potential new players early on because they regretted their first purchase. The hunchback almost made me quit my first time around (until some nice pilot taught me the joys of f2p).

#18 Lukoi Banacek

    Member

  • PipPipPipPipPipPipPipPipPip
  • WC 2018 Top 12 Qualifier
  • WC 2018 Top 12 Qualifier
  • 4,353 posts

Posted 07 October 2016 - 04:41 PM

View PostRoughneck45, on 07 October 2016 - 12:30 PM, said:

Because of the mechlab and how you want to play.

If you enjoy making franken mechs more than min maxing for effective builds being in lower tiers can be less stressful.

Casual vs. Competitive mind set.

As far as making another account, doing it to reset your win/losses and game your tier is pretty shady and not really worth it. Making another account to get the cadet c-bill bonus again if you regret your purchases can be worth it.


Rough gets it.

Secondly, with the Tier system being more heavily weighted towards experience vice actual ability (talent, skill, experience all put together), players are moving themselves into tiers where they are not competitive and get shellacked. Additionally, they tend to catch flack from tryhards in chat/on forums for bringing less effective builds or demonstrating less than optimal skillsets, which ruins the experience for them. Should high schoolers play football against NFL players? Probably not, for the very same reasons. It's not helpful to the league, or the retention factor of the weaker player.

This is a business...it needs a customer base. Make it worse for them by making them play in tiers they can't have some fun in will drive off those customers in most cases.

#19 BodakOfSseth

    Member

  • PipPipPipPipPipPip
  • Leftenant General
  • Leftenant General
  • 267 posts
  • LocationBay Area, CA

Posted 07 October 2016 - 08:17 PM

View PostAleksandr Sergeyevich Kerensky, on 07 October 2016 - 03:19 PM, said:

The biggest problem new players have, is they dont have enough c-bills to really find out which weight class they shine in. A lot of people - including myself - regretted the first mechs they purchased. Thats why now, I recommend people make a new account for every weight class so they can explore and tinker more.

The worst would be to loose potential new players early on because they regretted their first purchase. The hunchback almost made me quit my first time around (until some nice pilot taught me the joys of f2p).


I agree.. for the most part. I was lucky enough to join right before the appreciation awards.
Since I came from the WoW game where I paid a $15/month tax to play, I was pretty much all in for a FTP game that gave pretty solid rewards for playing more and being more involved.
"Spend any amount of $$ and get a medium mech"? I can do that.
"Buy any of these mechs and get a heavy mech"? I can do that, too (because BUSHWHACKER).
Once I understood the idea behind achieving mastery of a mech, it became pretty clear my first in-game mech to buy would be a Cataphract, and from there it was just about doing research and figuring out what sounds like fun and is for me.
Thus, I now have an AC20/4x med las brawler, and a 2xUAC5/4x med las mid range.
After that, it's a matter of fiddling with the Cataphracts I have and the trial mechs until I figure out what I like, and collecting exp and cbills while waiting for my reward mechs to come in.
I'm enjoying the crap out of my Cataphracts.
I'm also enjoying the PPC Panther (yes yes, I know it sucks).
Griffin and the Zeus are okay... took some getting used to.
Do like the Grasshopper, though it's a little not sturdy. So I guess I'm going to be looking for something to drop a bunch of Lg Pulse lasers on.

I suppose part of it is that I'm somewhat cautious when it comes to spending money, real or imagined. I do a bunch of research, so that's a thing.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users