Debate on Character Valuation

Alex502

I appreciate that. Could even expand the cast, like I said it could be as wide a net as "Heroes VS Villains" or some combination there in. And just boosting those themed characters would probably be enough, rather than a limited entry list.

I just remember the old Cap V Iron-Man event and that was a fun concept that also kind of fell flat. That happened because of the failed affiliations and imbalance between Cap's team and Tony's. Now, though, Iron-Man would still need a rebalance before he'd be workable, but a Team Cap VS Team Iron-Man would also be an interesting PvP event.

I'm just trying to find solutions that don't take away from anyone's gameplay, but really push for variety. Choosing between two different boost lists to use in VS is, perhaps, a good start.

PS: There's nothing saying the choice for the list should be something as narrow as Mutants VS Inhumans, clearly the Inhuman's don't have the roster to compete there. Combine them. Maybe "Inhumans and Mutants VS Gamma and Symbiote"?

Scofie

Hi @Alex502 , I'm enjoying reading your ideas. Fight4thedream started this thread a while ago to discuss how the PvP game could evolve:

https://forums.d3go.com/discussion/88238/idea-environmental-zones-the-final-frontier-of-pvp/p1

I'm going to try to bring this thread back on the topic of character valuation, though 🙂.

Alex502

To focus that direction, bringing light to the affiliations through any kind of event balances some of the evaluations, yes? I suppose my suggestion would require a careful eye to affilated teams and the balance between those. Combine the right affiliations against ones that can't compare, there's little option to work with that.

So in what way can we properly evaluate characters that breaks new ground in organizing the overall game synergy? Can we utilize Affilations to help with this categorization? And in doing this analysis, would an affilation based boosted list even bring balance if they were paired off in a PvP format?

Scofie

Mod mode off

I think that's right. I think the affiliations is a really good way to create a new model for different types of events. Previously events that had limited character availability haven't been popular but having a wide enough selection should help with that. I'm a big fan of shaking it up a little and I think the effort that has gone into affiliations would suggest they'll become a bigger part of the game going forwards.

So in terms of character values, the meta would get far more complex - each affiliation or combination of affiliations would have its own rock, paper and scissors teams. The blend becomes the important thing, not the individual. Will one affiliation become the meta? Is there a good enough defensive scarecrow in each affiliation? I think this is then where tactics come into play. It's all new so whilst vets will have an advantage with wider rosters and more experience of synergy but everyone will have to adapt to a new play style. And then add supports into the mix it becomes a new ballgame again.

Phumade

I'd feel confident of affiliation (or any themed pvp) based pvps if the devs could manage the basic scheduling issues. Aside from the affiliation idea, we have already had many themed events that had significant restrictions such as Call to Arms, Balance of Power, various class of, Black Vortex (the xman one) etc etc. These have all been relatively successful and welcomed by the community as a good break from the traditional pick 2.

With that said, does anyone feel confident about their ability to do the scripting/development work for this pvp when they have consistently missed on event scheduling? Yes, Demi's game development was tedious and predictable, but at least events ran smoothly and consistently with out errors.

entrailbucket

So they used to run limited-roster events all the time, but in PvE. They called them "Heroic" events. Some of us loved them, because we had everyone rostered. And back then PvE enemies scaled to your full roster, but Heroic events only scaled to your roster of allowed characters.

Everybody else pretty much hated them though. For newer players with lots of gaps in their roster, they might find themselves totally locked out, or limited to 1* and 2* characters (if they didn't have any of the allowed 3* rostered).

They have occasionally run "Heroic" type PvPs over the years, but they're really rare. I'm not sure why that is. There was a Heroic boss event a few months ago, and I don't think that was very popular.

Really, the boosted list pretty much does the same thing, without the unfortunate side effects of locking players out if they don't have a featured character rostered. The boosted list isn't as effective as it could be, though, because some characters are better than the boosted guys.

So we return to their initial problem -- there's no variety because some characters are way too good.

Alex502

Wanted to come back to say that my PvP lists have some variety of like, a dozen characters I'm facing mostly, which is better than the cast of like, only five that I was getting previously.

Still begs the question of how is MMR calculated within the game? Why do I see such a disparity of what I can opt into? Does or should the game weigh characters differently based on solo performance, or some other metric? I still feel like this system is a big wonky. Furthermore, does adding different strengthed Supports into PvP change my MMR?

Also, if we follow the same vein, does an Affiliation bring any weight to the in game value of a character? Namora has only 2 Affiliations total (thank you to @helix72 for their excellent work in listing out Affiliations here). As must as they're trying to work in new dynamics to bring different valuation to characters (retro adding affiliations to work with new released passives, for example) I feel that this requires a shift to the inner valuation systems too.

Furthermore, I'd love to discuss the points you've made about my idea around PvP, but Scofie already suggested we bring that conversation to the thread they quoted, so I'd like to encourage that conversation continue there.

entrailbucket

Matchmaking is really, really simple -- even though it seems bizarre and it can act unpredictably, it always operates under the same simple rules.

It's a series of cascading filters.

• players in your end time who are not shielded
• players with about the same score as you
• players using a team like your roster

They get applied in order. So it always looks for points first, then roster. There are some extra filters for one-offs -- they hide 550s from most people, for example.

In terms of "like your roster," we think they look at the levels of your top 3 guys. Nobody knows for sure, but that seems to account for most of it. Note that this is compared to the levels of the defensive teams those players are using, not unused characters on their roster. So if you used 3* to beat the seed teams, 3* players would see you, but you would see 5* teams.

I don't like "MMR" as a term because it's not a rating, really (it used to be) -- matchmaking is heavily dependent on which other players are unshielded and their scores. Your roster is part of it, but not the most significant part.

For the rest of your questions: no, the game doesn't "know" that some characters are better than others. Wasp and Chasm are treated the same for matchmaking. We don't know what supports will do to matchmaking, but based on the existing rules I'd guess that they won't be factored in. Affiliations are...still sorta lame. Most of them are just small bonuses for certain characters' passives.

If you (or anybody else) has questions about matchmaking, I can probably answer them. It's weird but if you know the rules, they're applied consistently every time.

Bowgentle

@entrailbucket said:
Matchmaking is really, really simple -- even though it seems bizarre and it can act unpredictably, it always operates under the same simple rules.

It's a series of cascading filters.

• players in your end time who are not shielded
• players with about the same score as you
• players using a team like your roster

They get applied in order. So it always looks for points first, then roster. There are some extra filters for one-offs -- they hide 550s from most people, for example.

In terms of "like your roster," we think they look at the levels of your top 3 guys. Nobody knows for sure, but that seems to account for most of it. Note that this is compared to the levels of the defensive teams those players are using, not unused characters on their roster. So if you used 3* to beat the seed teams, 3* players would see you, but you would see 5* teams.

I don't like "MMR" as a term because it's not a rating, really (it used to be) -- matchmaking is heavily dependent on which other players are unshielded and their scores. Your roster is part of it, but not the most significant part.

For the rest of your questions: no, the game doesn't "know" that some characters are better than others. Wasp and Chasm are treated the same for matchmaking. We don't know what supports will do to matchmaking, but based on the existing rules I'd guess that they won't be factored in. Affiliations are...still sorta lame. Most of them are just small bonuses for certain characters' passives.

If you (or anybody else) has questions about matchmaking, I can probably answer them. It's weird but if you know the rules, they're applied consistently every time.

You know the only question most people have about matchmaking is "how can I beat up rosters that are weaker than mine to 3000 points?"
And when we explain, why this doesn't work, and they really don't want that (because they would be the weaker rosters to those above them), they get all flustered.

entrailbucket

Eh, if you don't know what it's up to, MPQ matchmaking is really weird. The one thing I keep repeating on here is the game can only give you players that actually exist. People seem to either forget that or not understand it from the beginning.

Like, there's no 4* players to match you with at 2000 points, because those players don't exist at that score.

HoundofShadow

The dev did say it works using a shifting scale in latest Q&A. And I'm sure I queued 4* players when I hit over 2k in shield sim.

Anyone remember the yearly 1v1 pvp? Okoye and Kitty went from top tier to trash tier, and guess what, Chasm will go from top tier to trash tier as well. The only meta I remember surviving this 1v1 is Apocalypse, BRB, Wanda and Colossus. I can't remember SC was in there.

The environment changes the tier of characters. A simple 1v1 pvp could already drop at least 3 meta tier to trash tier.

entrailbucket

@HoundofShadow said:
The dev did say it works using a shifting scale in latest Q&A. And I'm sure I queued 4* players when I hit over 2k in shield sim.

I don't understand what "shifting scale" refers to here. Again, the rules are very simple, applied consistently and well understood by the top PvP players. If you queued 4* players, there was a good reason for that, that is consistent with the rules.

HoundofShadow

You said no 4* players will be matched with 5* players at 2000 points. I simply pointed out that I queued 4* players after hitting over 2k in shield sim.

entrailbucket

So you're just being deliberately obtuse like usual, but in case your trolling deceives or confuses someone else, let's restate my point more generally:

"In general, one will not be matched with lower-tier players at extremely high point values, because it's hard for lower tier players to score a lot of points."

HoundofShadow

You used superlative word like "no 4* players..." And I quote you:

"Like, there's no 4* players to match you with at 2000 points, because those players don't exist at that score."

You gave a terrible example yourself, and you faulted me for that.

Bowgentle

Also, 2k in Ssim is like breaking MMR in a normal PVP.
Of course you'll see Walrii then - for 5 points.

fight4thedream

Mod note: Removed a few off-topic comments. Please keep this discussion civil and on topic. Thank you.