8 replies to this topic

[Ryvucz]

Instead of having four ELO scores, make an ELO score for each chassis type.

Locust =/= Jenner

Awesome =/= Atlas

Cicada =/= Hunchback

Catapult =/= Cataphract

It would at least be more accurate.

[Zerberus]

More accurate? yes.

A truly massive increase in DB size and complexity, combined with the necessity to update the database for every chassis to come (i.e. from 4 values the number skyrockets to something like 30, soon to be closer to 40 or even 50 when teh clans come, with no end in the forseeable future)? Yes.

I see teh merit, the idea is generally sound in principle. Maybe when CW is implemented.... but until (at least) then, the workload /reward ratio is abysmally low IMO

Though this will also throw a huge monkeywrench in the Church of Skill^TM´s e-peening and "make Elo scores public so we can arbitrarily decide who knows what they`re talking about" agenda... so yesterday should be soon enough :

Edited by Zerberus, 14 February 2014 - 09:41 PM.

[Firewuff]

Zerberus

its a single table per users, it does not even affect performance when you are using something like NoSQL to store the data. So its not a performance or big data concern in any case. Dont forget they store stats on every chassi and every weapons system for ever user already so 1 more number is insignificant.

This however is a BAD IDEA.... I am a high ELO player (for example), I buy a brand new chassi and drop. What is my ELO? if it starts at zero then I am dropping with noobs who are just target practice until I get the ELO on my new mech up. If you start with an "average ELO" for the chassi you have what you have now..... where is the gain?

[Bhael Fire]

I voted yes, but only on the concept of rating the mechs that players bring to the field.

I'd rather see a revised BattleValue system catered to MWO that rates the chassis, its current loadout/equipment/modules AND the pilot's overall skill (based on their performance over the last 100 games).

Edited by Bhael Fire, 18 February 2014 - 09:18 PM.

[Khobai]

Quote

Instead of having four ELO scores, make an ELO score for each chassis type.

5 ELO categories would be sufficient IMO

Tiny: 20-30 tons
Small: 35-45 tons
Medium: 50-60 tons
Large: 65-80 tons
Huge: 85-100 tons

Then matchmaker should just give each team two mechs from each category then as wildcards also give two random mechs from categories that add upto six (so tiny+huge, small+large, or medium+medium). And try to match ELO for each category. That would give us very balanced teams.

The other option is go all out with a battle value system but thats a lot of work with no guarantee of it even giving us balanced teams.

Edited by Khobai, 18 February 2014 - 09:47 PM.

[BillyM]

I'm ok with this... Minimal impact and hopefully brings out some of the more maligned chassis...

...might actually see a centurion or treb again that way...

--billyM

[Bilbo]

BillyM, on 20 February 2014 - 06:46 AM, said:

I'm ok with this... Minimal impact and hopefully brings out some of the more maligned chassis...

...might actually see a centurion or treb again that way...

--billyM

You'd only see them until the pilot reached their correct Elo level, at which point it would collect dust because there is a reason you aren't seeing them now. Also, the impact wouldn't be particularly minimal to those who endure the drubbings while 'new' chassis are leveled up. People in higher Elo brackets aren't necessarily there because of the mechs they are driving.

[Nothing Whatsoever]

I don't mind being a guinea pig to test something like this out.


Homeless Bill raised a similar idea here.

[ImperialKnight]

what? ELO is not by chasis???? Someone explain to me then why my first few drops in a new mech is always filled with people that sits still and lets me shoot them