11 replies to this topic

[MTier Slayed Up]

So folks have been complaining about this for awhile. Came up with my suggestion (Ranked Que), that probably will not get any attention/traction whatsoever but it's there.

At any rate, I think this is a good way to see where people stand with this. If you feel like you have the end all be all solution, post it here or forever hold your potato.

[Spare Parts Bin]

I like a ranked que. Match Maker is junk currently.

[Agent of Change]

Since PGI is unlikely to redo everything yet again I suggest the following:

https://mwomercs.com...of-skill-level/

It would at least allow them to make it a bit better than a glorified XP bar and keep most of the existing structure in place making it easy to implement.

Is it what they SHOULD do? Probably not, but i think it's more likely than a total overhaul.

[MTier Slayed Up]

For a topic that's complained about a lot, I'd thought that this would raise a bit more eyeburrowz.

[Ravendaark]

Gave my vote. Hard reset. Where's the beers?

[MTier Slayed Up]

[General Solo]

IMO moaar than a hard reset is needed.

Me thinks the same or similar PSR increase/decrease conditions applied for losses need to be applied for wins.
The current system makes it too easy to go up PSR and difficult to go down PSR.
Aka get rid of the upward PSR bias.

Otherwise a PSR reset is just going to be a temp fix for about a year or two.



TLDR: Fix Matchmaker by making PSR change conditions the same for a win and loss.
Currently its to easy to increase PSR on a Win, even if u get carried.

This is wot I wood do.

Edited by OZHomerOZ, 27 June 2018 - 11:27 PM.

[Elizander]

Increase the penalty for bad performance/losing at Tier 2 by 2x and increase the penalty for bad performance/losing at Tier 1 by 3x/4x. Really shaft the players PSR if he has a sub 100 damage, no team bonus, win/loss.

People that are inconsistent and derp around like me will slide to Tier 3 in no time.

Edited by Elizander, 28 June 2018 - 11:29 AM.

[SilentFenris]

DrtyDshSoap, on 25 June 2018 - 05:04 AM, said:

So folks have been complaining about this for awhile. Came up with my suggestion (Ranked Que), that probably will not get any attention/traction whatsoever but it's there.

At any rate, I think this is a good way to see where people stand with this. If you feel like you have the end all be all solution, post it here or forever hold your potato.

My answer depends, what is the point of Tiers?
A) To determine the most skilled pilots in MWO
B) To ensure both sides on a battle are of fairly equal skill?
A and B are mutually exclusive. The current PSR system fails in both respects.

Who is the best player? No matter whether you measure by Kill/Death Ratio, Win/Loss Ratio or Average Match score, a Tier 5 (lowest rank) pilot can have better numbers on the Leaderboard than a Tier 1 (highest rank) pilot. For any of these measures to mean anything, either:
a) Tiers need to be hard-locked out of playing against each other AND given seperate leaderboards.
b) All players go in the same bucket, from the Championship teams last year to the guy playing his first match.

As far as determining a Fair Match - this area in particular was one thing the old ELO system did better than current PSR. Each player was individually rated for EACH MECH they had in their stable. Just cause you can hang with the big boys in an Arctic Cheetah doesn't mean you can cut the mustard in a Panther. To make matters worse, the matchmaker ALWAYS ASSUMES you are piloting a mech you are skilled in.
ex: A individual player might be a Tier 1 comp-team in a Clan Heavy; but what if tonight he is in an Urban Mech or Atlas.
Any system setup to give both sides equal odds of winning would have to take into account how skilled each player is with the specific mech they are playing for that match.

So if you setup the game to find the best player, there is very little to keep new players safe from veterans. Steep learning curves and frequent deaths are not an easy way to maintain or enlarge the player-base. At the same time, if you keep seperate Tiers to soften the playing-field, you pad the stats of lower-tier players and hurt the stats of those playing the higher tiers.

So which is most important to you?

Edited by SilentFenris, 28 June 2018 - 01:48 PM.

[MTier Slayed Up]

The current Tier system accomplishes in separating new players from veterans. What it does not accomplish is separating the 80% from the 90% and the 99% from the 95%. Does that make sense?
For the record, I personally find the Jarl List to be a terrible metric to use anyways since the current system can be abused in a way where you can play 10-20 games a month, get a decent K/D and W/L and get a dramatic stat boost on that leader board.

What does Ranked Que accomplish? Well, for one, there would be ELO or a ranked division rather than a Tier. It can be a simple system to where it could be similar to StarCraft's system (Grand Master, Master, Diamond, Plat, Gold, Silver, Bronze, Paper Weight). The system takes into account wins, match score, damage, kills, alive/dead at the end. I don't think it needs to go further into that, but to avoid the damage padders, it could be a simple way to identify if someones direct fire is on point. Example 4 kills, 2 assists, 1000 damage. 1000/6, you're averaging 166 damage to each of those mechs that you've killed or helped killed and that can be taken into account. This is a barebones idea of what I'm thinking, but already it'd be a dramatic overhaul of "Match Score=Damage-Team Damage+kills+assists True=x .win If(Not= -.lost. (I imagine it looks something like this in terms of code, probably a lot more to it, I'm also rusty/amateur).
In a nut shell, I want a ranked system that takes more in account to individual performance that can help balance the game for better matches long term, rather than a Tier system that would require a hard reset or a change in Match Making to separate veterans from newer players.

How does this benefit match making though for the folks that want to stay casual? Well, the option is there to do Ranked Que or not, and you can still just do casual play (renamed from Quick Play). Let's say Ranked Que has been around a month after it's implemented and the folks that are now ranked want to do casual; Match Maker can identify which of those that are ranked up in GM, M, D, Plat and slot them in the games together, while G, S, B, PW can be mingled together. D, P, G can be mingled together or open it up further.
Folks that have not been ranked will be treated as Silver/Bronze until they Ranked Que in order to avoid higher place ranked people. This would apply to new players as well.

Why would that help new players though that could potentionally be placed into a game against a Gold? For starters, Golds are just "Average," they wouldn't stomp people into the dirt against a team of 12 and vice versa. I think in terms of having more balanced games, this would be an overall better system to have.

Super uber tryhards such as myself would normally be in ranked que. Guys that just like playing the game and cracking open a cold one over the weekend while shooting stuff can just play in the casual group with no stress of being locked into a team or against a team that can contain several comp players or a few people that can shift the balance of a game from just being present.

I probably half assed explained my method of madness, but that's basically it.

Soooooooooooooo
Which is more important to me? Both, equally.

Edit - Disagree with the statement that a comp-tier 1 pilot thats Godly in an Assault will be a complete potato in a light, or just a mech change.
I've personally taken an Urbie out first go and pulled in 700 damage, a few kills and assists. I'm mainly piloting assaults/heavies. These games are roughly my average in QP. I've said this in the past and I believe this to be a truer statement than people give credit; the mech doesn't make the pilot.

Edited by DrtyDshSoap, 29 June 2018 - 12:17 AM.

[SilentFenris]

DrtyDshSoap, on 29 June 2018 - 12:10 AM, said:

Soooooooooooooo
Which is more important to me? Both, equally.

Edit - Disagree with the statement that a comp-tier 1 pilot thats Godly in an Assault will be a complete potato in a light, or just a mech change.
I've personally taken an Urbie out first go and pulled in 700 damage, a few kills and assists. I'm mainly piloting assaults/heavies. These games are roughly my average in QP. I've said this in the past and I believe this to be a truer statement than people give credit; the mech doesn't make the pilot.

Yes, seems a ranked system as you described would be similar to Starcraft. You might actually be able to accomplish both a better player rating system and increase balance of skill in-match.

As far as "statement that a comp-tier 1 pilot thats Godly in an Assault will be a complete potato in a light"- I said

SilentFenris, on 28 June 2018 - 12:57 PM, said:

Just cause you can hang with the big boys in an Arctic Cheetah doesn't mean you can cut the mustard in a Panther. To make matters worse, the matchmaker ALWAYS ASSUMES you are piloting a mech you are skilled in.
ex: A individual player might be a Tier 1 comp-team in a Clan Heavy; but what if tonight he is in an Urban Mech or Atlas.

There is a BIG difference between the meaning of what I wrote and the extreme you have taken it to.

[MTier Slayed Up]

Runner up for ugliest guy on MWO, lets get these votes in.