16 replies to this topic

[farout]

I am enjoying Solaris, but one thing is pretty clear:

If you want to win, your chances are a lot higher if you pilot the division cheese build.
Nothing wrong with that, but it gets stale really quick.

Also, PGI moving mechs around in the divisions, will just produce another top dog for each division.

My idea therefore is as follows:

1. Make divisions purely dependant on weight classes

2. At the end of the match, apply 2 positive or negative multipliers to the ELO changes.

- First multiplier is the personal one : A mech which has a positive W/L ratio gets a negative multiplier, a mech with a negative W/L ratio a positive one.

- The second one works exactly the same, but it looks at the global W/L of said mech.

This system would have the following effects:

1. It would encourage you to play non meta mechs.

2. MOAR EPEEN : if you are doing well in the leader board, everyone would know you got there by using crap mechs !

3. since this system would constantly change depending on your and the global performance of mechs, there would be more builds on the battlefield.

what do you think ?

Edited by farout, 23 April 2018 - 02:43 AM.

[Appogee]

I appreciate the intent of your suggestion, and I think it has some merit.

However, the same Mech may have a meta loadout, or it may have a potato loadout. How would you account for that?

Incidentally:

I believe at the end of each season PGI will probably promote each division's 'meta' Mechs - ie the one or two the leaderboard winners relied on - up to the next division.

At least, that's what they should do if they are actually planning to work to improve the mode.

Edited by Appogee, 23 April 2018 - 02:52 AM.

[Racerxintegra2k]

3 months is way to long IMO you could do it every 3 weeks and it still might get boring.

[calmdawn]

In Quick Play I can increase my rating even when team is loosing.

Why not to do the same in Solaris: you take crap mech - you scratch Annihilator, but he is loosing ELO and your are increasing your ELO, just like in Quick play.

[Racerxintegra2k]

calmdawn, on 23 April 2018 - 03:36 AM, said:

In Quick Play I can increase my rating even when team is loosing.

Why not to do the same in Solaris: you take crap mech - you scratch Annihilator, but he is loosing ELO and your are increasing your ELO, just like in Quick play.

Potatoes should stay in the dirt. I love the fact my ELO drops significantly on a loss. If i don't deserve to be at the top of the leader-board i shouldn't be.

[Dimento Graven]

farout, on 23 April 2018 - 02:37 AM, said:

...

what do you think ?

Nah. Your system would result in certain matches where the person who wins could still end up losing Elo.

[Pariah Devalis]

I'd simplify matters, further. The greater the tonnage imbalance between two mechs fighting, the greater or lesser the Elo gain or loss should be. If a Spider beats a Vindicator, the Spider should get a relatively large boost and the Vindicator a sizable hit to their ranking. If a King Crab beats a Bushwhacker, the King Crab should have a minimized Elo boost, while the Bushwacker should get a very minor Elo hit. Taking into account personal Elo differences, as well, of course.

The greater the personal risk you, as the pilot, take in your mech choices, the faster you might climb the leaderboards. There should be a slow and steady approach, as well as a faster but riskier approach. Taking the biggest mech should not be the absolute best solution in any division.

Edited by Pariah Devalis, 23 April 2018 - 09:36 AM.

[Khobai]

no your mech choice should not affect your elo

that goes against the whole point of an elo system

instead mechs should just be in the right divisions based on tonnage/mech quality instead of being arbitrarily assigned to divisions they have no business belonging in

Quote

I'd simplify matters, further. The greater the tonnage imbalance between two mechs fighting, the greater or lesser the Elo gain or loss should be.

if you want to truly simplify things, just put all the mechs in the divisions they actually belong in. that is as simple as it gets.

Edited by Khobai, 23 April 2018 - 09:47 AM.

[LordNothing]

putting the mechs themselves on a sliding elo system (independant of player elo) will save pgi the trouble of re-organizing the divisions manually each season, just use the last season elo numbers to place the mechs in the right divisions. a little automation script to do what would take pgi 3-6 months to do manually.

[Khobai]

they dont have to reorganize them every season. just one time if they do it right the first time.

a sliding elo system will not help with mech diversity. because instead of players being able to play whatever they want, they will be forced to play the mechs with the most efficient power to tonnage ratio in order to get the best elo gains. no thanks.

what needs to happen is all mechs need to be put in the proper division. And all mechs in a division need to have a reasonable chance of being able to beat all other mechs in the same division.

Edited by Khobai, 23 April 2018 - 09:54 AM.

[Pariah Devalis]

Khobai, on 23 April 2018 - 09:41 AM, said:

no your mech choice should not affect your elo

that goes against the whole point of an elo system

instead mechs should just be in the right divisions based on tonnage/mech quality instead of being arbitrarily assigned to divisions they have no business belonging in

if you want to truly simplify things, just put all the mechs in the divisions they actually belong in. that is as simple as it gets.

Is an Elo system not supposed to ascertain a relative rating between two players, all things being equal?

If so, no. That doesn't go against the concept of the Elo system. You can't call something that is dominating the boards across any given division "equal" to the other mechs it is facing, so you'd need some sort of counter mechanism in a game this diverse in mech capabilities to push back or incentivize things other than "the best mech in the division." Basically adjusting Elo gain or loss for equality's sake.

I do, however, agree mechs should be in the correct division in the first place. However, quite frankly, due to their track record, I don't have faith that those who are in charge of sorting things are knowledgeable enough to accurately do so.

Edited by Pariah Devalis, 23 April 2018 - 09:53 AM.

[Darian DelFord]

Honestly I would be fine with an increase based on the damage done vs the tonnage difference.

Lets face it.

Phract 70 tons

Locust 20 tons

I have to do a hell of a lot more damage to him than he does to me yet I do not get compensated for the vast skill difference win or lose based on those stats.

[Koniving]

Issue of a negative multiplier is it would just remove score or undo score already earned. Now a high and low multiplier, sure. But again as someone else said the mechs may have "potato" loadouts.

[Khobai]

Quote

Is an Elo system not supposed to ascertain a relative rating between two players, all things being equal?

correct. elo is a rating of player skill. its not a rating of mech tonnage.

beating someone in a heavier mech does not necessarily equate to skill either. because there are countless examples of lighter mechs that are better than heavier mechs. tonnage isnt everything. there are plenty of heavies and assaults that are absolutely terrible.

Giving people higher elo gains for using lower tonnage mechs is just going to decrease mech diversity even more. Players will gravitate towards the mechs with the most efficient power vs tonnage ratios.

even PGI understood that. hence why we have different divisions to promote mech diversity. PGI just screwed up putting mechs in the right division. The only way having different divisions works is if every mech in a division has a chance of beating all the other mechs in the same division.

Edited by Khobai, 23 April 2018 - 02:35 PM.

[Pariah Devalis]

Khobai, on 23 April 2018 - 02:29 PM, said:

correct. elo is a rating of player skill. its not a rating not mech tonnage.

beating someone in a heavier mech does not necessarily equate to skill. because there are countless examples of lighter mechs that are better than heavier mechs. tonnage isnt everything. there are plenty of heavies and assaults that are absolutely terrible.

Giving people higher elo gains for using lower tonnage mechs is just going to decrease mech diversity. Players will gravitate towards the mechs with the most efficient power vs tonnage ratios.

even PGI understood that. hence why we have different divisions to promote mech diversity. PGI just screwed up putting mechs in the right division.

Outliers aside, it would be hard to argue, for example, that a bushwacker killing a banshee would take considerably more skill than the other way around. Potato builds aside (as how you outfit the mech chassis could be argued as part of the pilot's skill), it is simply more work to kill something larger than you, and the risk is much higher. There's currently no incentive for the players to either not take the biggest gun for the easiest win, nor for players to take something that takes more pilot skill to get consistent victories out of.

Just the act of going into the fight at less than "the best option" for the division automatically skews the odds in the enemy's favor, and that is before taking player skill into account. There should probably be a mechanism to attempt to compensate for that.

Edited by Pariah Devalis, 23 April 2018 - 02:36 PM.

[Champion of Khorne Lord of Blood]

>play mech with huge elo boosts on it
>win a few matches at low level
>face people who were patient and climbed the ranks with actually good mechs
>get blasted repeatedly by said good mechs in high elo rankings

This seems like a good idea yes.

[Khobai]

Quote

Outliers aside, it would be hard to argue, for example, that a bushwacker killing a banshee would take considerably more skill than the other way around.

its even harder to argue that a bushwhacker should be even fighting a banshee in the first place

that matchup should never happen