@Morello and company: will S3 address the "starvation support" issue?

First Riot Post
Comment below rating threshold, click here to show it.

CupcakeTrap

This user has referred a friend to League of Legends, click for more information

Recruiter

10-15-2012

Here's my response to Morello:

Quote:
Originally Posted by CupcakeTrap View Post
Hi Morello! Thanks for responding. I'm glad you're on the same page, and encourage you to keep talking to Support players. I know it's not your role of choice, but I like to think we've convinced you that we're not just "lazy" players who like to passively soak up XP and use heals to make gameplay boring, or lackluster players who want to get carried. (I actually carry harder as Taric than as Veigar, IMO: Supports CAN make a big difference, even if they tend to get ignored. More than once: "go ahead and ini, we don't need Taric, he's just a Support --> ENEMY QUADRA KILL!")

I'd like to continue this dialogue on a few key points. I've summarized some of your statements as bolded headings.

"Synergy is interesting, but it's problematic as a Mastery, because it would require people to spec into a particular tree if they want to do Support."
Agreed: however, why make it a Mastery? Why not make it a default mechanic? Its benefits are naturally limited to non-farming champions. Further, since it would be so crucial to Support, it seems dumb to have it as a Mastery: there's no meaningful choice, as you would ALWAYS want it.

If the argument is "it's a good idea, but it wouldn't work as a Mastery" ... then why not just make it a basic mechanic? Is there a meaningful drawback other than that it would be bad as a Mastery?

"We're not going to create anything like a fifth goldstream."
Then how are Supports going to afford these new Support items?

Ward expenses are significant, but as the graphs show, farm is a crucial infusion of early-game gold. Farm guarantees you at least a couple "real" items for those critical early-mid-game teamfights. gp10 cannot replicate this cash infusion: it's too small in magnitude and it's not timed properly. If these Support items, used well, will scale on teammates' power, then they might still be viable for the lategame (when Supports have started to catch up on gold, as assists/global gold begin to surpass farm as gold sources) ... but they aren't going to be relevant in the early-midgame.

"If we created a fifth goldstream, traditional supports would vanish. You would always take Annie over Sona."
Not necessarily.

First, as has been discussed, there's no reason why Sona can't scale just as well as Annie. (See my distinction between "0-CS" and "Support": Blitz is a 0-CSer but not a Support, while Karma is a Support but not a 0-CSer. "0-CS" is about how well you can survive starvation, while "Support" is about what your abilities do. Of course, like everyone else, I currently use "Support" as a catch-all term for "the person who goes bot with the ADC" ... but I believe these are actually two separate concepts that only sometimes line up.)

Second, even if Sona remains both 0-CS and Support, what's wrong with that? Have it so that Sona, whether fed or not, hovers in power between starved Annie and fed Annie. There would be a meaningful strategic choice there: Annie offers potentially greater gains, but with more risk. Sona offers more reliability in power level but at the cost of lower direct "carry" potential.

Honestly, I think part of why people go "well, if we didn't force people to take Supports, nobody ever would!" is that Supports really HAVE been overnerfed in the war against the passive botlane. How come they're still picked despite having been overnerfed, you ask? Because when the amount of gold flowing in is ZERO, other champs still manage to suck even harder. So create a fifth goldstream to make other champs viable as duo-laners, and then buff the traditional Supports a bit! I don't think that either a buffed Sona or a half-fed Annie would be game-breakingly powerful. I think a buffed Sona, in the hands of a good player, should be at least as significant as a half-farmed Annie. But let's say your 5th player isn't very good at Support. Fine; let them play a half-farmed Annie who just contributes more DPS to the team. Alternately, let's say that your team strategy centers on split-pushing. You might prefer a half-farmed pusher champ or a tanky dps brawler to a Support: Supports thrive in big teamfights, and if you're not planning to have those, you might want someone else.

Let me put it this way: if you're saying that an Annie with 50% farm will always do more for the team than a Sona, maybe that means Sona's been overnerfed and is only hanging on because right now we're at 0% farm. If this is what you're saying, then you've just stated (IMO) that Supports are designed to be inferior champions. I think that sucks.

"Teamcomp will remain static: 1+1+2+J, and the 2 will be ADC + Support."
Without a fifth goldstream, at least, this is the conclusion I see.

I'd rather see things set up so that double-bruiser bot, or carry + mage, or no-jungler, could be viable strategies with strengths and weaknesses, rather than just auto-loss moves.

---

Original post follows

---

I was very pleased to see this post from Morello today:

Quote:
Originally Posted by Morello View Post
No, zero CS is even bigger than this effort from a risk/dificulty standpoint. Supports will still play zero CS (unless you guys find strats that feed supports!)

We are adding a lot of support-oriented items with this in mind, though, and alternate ways to get gold (cool new mastery for supports inc...). I actually don't mind zero CS - I mind that support don't get gold that they can use to buy cool items. That is what we'll fix here.

Zero CS is needed to keep support viable (it's actually always been the optimal way to play, players just didn't have that down pre S2), but they get to do a lot of other interesting things in lane like harass, peel or set up kills. Making the lane fight for CS is inferior to that.
I agree that this is a problem. It's not just bad for Support players, but also (I think) often limits the range of Champions that can be played on any particular team in a boring way.

That said, I have to agree that I share NA1's concern here:
Quote:
Originally Posted by NA1 View Post
I feel like Morello is just going to put a band-aid over the gaping wound.

I'll tl;dr the issue from my perspective.

4 goldstreams for 5 champions means "starvation supports" who don't get to do as much strategic itemization:
  • There are 4 goldstreams but 5 champions
  • Champions scale differently on gold (ceteris paribus, a 100-CS Carry + a 0-CS Support beats a 50-CS Carry + a 50-CS Support every time)
  • Itemization is a critical part of the game, and Riot says it wants to make strategic itemization even more important in S3
  • ... yet Supports are typically so starved for gold (especially early-game farm gold) that they generally get no further than a rudimentary core build every game: Shurelya's, maybe Aegis, maybe Locket. ADC players: imagine if gold were scarce enough that you only ever got to build IE, and never got to make interesting decisions like "do I go for lifesteal or more damage?" or "do I keep going DPS, or do I pick up a Banshee's?"
  • Support players don't get to itemize much, which means they don't get to bring their particular style to a Champion or make interesting choices about how to build their Champion, at least not as often as the other chars.
  • Supports also tend to come into the early-mid game extremely fragile. (Did you see the WE Janna with Boots 1 at 50 minutes?)
  • From a player's perspective, this also just "feels bad": you've been chosen as the one who starves. The team has decided that it will not invest its resources in you. You will be told to cut it out (rightly, from a strategic perspective) if you take the "carry's" CS. Even "pssh, why did you waste that cooldown on Janna? She's a SUPPORT." feels kinda bad.
  • A sign of how bad the situation has become: I used to think of the definition of "Support" as being about what a "Support" Champion could do. A Support, I thought, was someone who used heals, buffs, debuffs, repositions, and other such abilities. I didn't think of Supports as waterboys or cheerleaders, but more like military officers who organized their team and, through clever play, amplified their power and disrupted the enemy strategy. But it seems that now, the main definition of "Support", the sine qua non of being a "Support", is that you're not totally useless even when you're utterly starved from 0:00 onward. A fix to the starvation-support problem could mean that once again Supports are valued for what they can do, rather than what they can survive.


4 goldstreams for 5 champions also severely constrains teamcomp and lane assignment decisions:
  • If you don't have a Jungler, you get only 3 goldstreams for 5 champions (and the enemy potentially picks up your jungle gold!)
  • If your second botlaner is anything but a traditional 0-CS Support, you are making a very risky bet that you'll get enough early kills to snowball. (Else, the second botlaner will quickly become totally irrelevant, because they don't gain as much power from leveling as a traditional support and they don't get enough gold to itemize normally.)
  • There is a strong incentive to make as many lanes solo lanes as possible. This leads to people playing their own little isolated 1v1 game for a big part of the match, which (IMO) is kind of unfortunate in what's supposed to be a team game. You don't get many opportunities to figure out synergistic duos that aren't ADC + Support.

Thus, IMO, LoL really needs a 5th goldstream.

Requirements that any such 5th goldstream solution must meet:
  • It needs to be a meaningful amount of gold. If it's trivial (e.g. 5% of farm revenue), then it will neither meaningfully increase Support itemization options nor enable mages, bruisers, or other champs who aren't trad supports to duo-lane.
  • It needs to somehow be unavailable to those champions who ARE farming a traditional goldstream (you don't want an ADC tapping into a minion stream AND this goldstream.)

Some reply: but if it becomes possible to get gold, then Supports will disappear, because mages/bruisers/etc. are just better! I think this sort of thinking arises from the entanglement of "Support" with "no-farm champion", a confusing of what champions can do with what champions can survive. There's no reason a Support -- someone with a lot of power to amplify teammates' strengths and create vulnerabilities in the enemy team -- can't scale on gold. You would just need to adjust their abilities so that they scale better, something Morello has already talked about doing. I like Support, not because I like having no gold, but because I like helping my team rather than just wading in and wrecking the enemy directly with Tons of Damage.

Example of a Support who is not a 0-CSer: Karma. She has several buff/debuff/heal abilities, but requires gold to be strong; she "starves poorly", and yet doesn't really carry hard enough to warrant mid.
Example of a 0-CSer who is not a Support: Blitzcrank. He can survive on 0-CS, because a grab is a grab, but he isn't really a "Support" ability-wise: he doesn't buff/debuff/heal, he's just a tanky CC machine.

Some possible solutions, not all of which I personally endorse:
  • "Splash" gold mechanic that gives gold to a Champion when a nearby allied Champion scores a last-hit. See, for example, this "Synergy" mechanic suggestion. ( http://na.leagueoflegends.com/board/...php?p=21831209 ) It would provide about 40% of a normal goldstream, but would send the gold only to non-farming champs.
  • Make all gold global, like Tower gold is now. e.g., every time a 20g minion is killed by a Champion, all that team's champions receive 4g. (Pretty radical change; I don't personally like it much, because it would eliminate some interesting strategies based on preventing specific champs from farming effectively.)
  • Create an item with a low resale value that gives massive gp10 but prevents you from receiving any gold from farming. (Would potentially lead to 5-champ bands of roamers, though.)
  • Give assist-gold for minion kills analogous to the assist-gold for champion kills. So if you deal damage to a minion that isn't a last-hit, or buff the killer, you get some bonus gold. (DebosBike proposes this below.)

Morello and company: would you care to say more about this topic? I know that you don't want to talk details until something is ready to go, but perhaps you could say a bit more about the current problems as you see them?

Community: any other ideas for addressing these problems?



EDIT: More from Morello:

Quote:
Originally Posted by Morello View Post
We do have some specific things in there that help supports directly - including alternate gold gain methods that work within the classic support playstyle.
EDIT2: More from RiotVeonneth (thanks, ObscureClockwork!)

Quote:
Originally Posted by RiotVeonneth View Post
I had a long talk with a designer about the problem today - might as well share some of his thoughts...

One of the things that some people can forget is that it's not just the support bottom, but it's the ranged AD carry too. You'll see Ashe, Graves, Caitlyn, Vayne, maybe Corki, maybe Tristana, etc. typically going bottom and farming up. The key thing about ranged AD's is that they scale with gold a lot faster than they scale with levels, and so as a result, bottom lane is perfect for them. The exp loss of the duo-lane doesn't matter as much, but the free-farming matters a ton to the point of ridiculousness.

We've tried a lot of things to solve this one, but ultimately, we haven't found a solution we like yet. One solution we tried was adding in 5-6 new support items that were essentially "must haves" in order to play the role. We tried nerfing support mana into oblivion to force them to buy mana items to have any effect in the late game. But all of them ended up with the same result: get a different champion in bot lane with your AD carry. Because anything that takes away the gold from the ranged AD carry isn't worth it.

The natural next step is to try to smooth out the AD carry curve. Nerf AD items hard, give some AD on level ups, etc. But the end result there is that you make the same decisions, just with another role. Throw down an AP caster that scales really well with itemization, or an assassin that likes the security. In either case, you still make the same investment.

We're not going to solve the support problem by nerfing supports - we've tried that, and the result is people just go to some other "support" that's not really a support (Support Plank) and continue to not farm. The ultimate problem is that you want your AD carry to farm. By choosing support, you're investing in a late-game carry. It doesn't even need to necessarily be an AD carry, but they do exacerbate the problem.

Most players get frustrated by supports when they're in bot lane without a support. It's pretty hard for 2 carries to beat Soraka+1 in lane because of the sustain factor. But we have to handle this situation with care because of how much swing it has over the metagame. Most solutions to this problem are pretty sweeping changes that impact the game pretty hard.


Comment below rating threshold, click here to show it.

Axiom

Senior Member

10-15-2012

I'm excited to see the new Mastery as well. I seriously hope it's something along the lines of your Synergy idea.

I don't think Riot is going to leak any details at this point, but here's too hoping!


Comment below rating threshold, click here to show it.

Ethelon

Senior Member

10-15-2012

I would complete eliminate spreading tower kill gold. That doesn't specifically deal with a 5th gold stream, and it eliminates lane control tactics when for example, you shove your enemy out of lane and then shove your minions into their tower to eliminate the gold.


Comment below rating threshold, click here to show it.

CupcakeTrap

This user has referred a friend to League of Legends, click for more information

Recruiter

10-15-2012

Quote:
Originally Posted by Axiom View Post
I'm excited to see the new Mastery as well. I seriously hope it's something along the lines of your Synergy idea.

I don't think Riot is going to leak any details at this point, but here's too hoping!
Thank you for the support and bump!

I'm hoping to attract the attention of a wild Morello. Or the elegant and noble Xypherous. Or I don't know, maybe Phreak? His Tons-of-Damage Lulu build would benefit from solving this problem.

Quote:
Originally Posted by Ethelon II View Post
I would complete eliminate spreading tower kill gold. That doesn't specifically deal with a 5th gold stream, and it eliminates lane control tactics when for example, you shove your enemy out of lane and then shove your minions into their tower to eliminate the gold.
I don't personally endorse totally global gold either, because I think it might be too radical a change and would eliminate some interesting strategy oriented around starving specific champs.

However, I would point out that it would NOT make pushing minions to tower to deny less effective: if towers kill minions, nobody gets gold; no reason to think that would change under a global gold regime.


Comment below rating threshold, click here to show it.

DebosBike

This user has referred a friend to League of Legends, click for more information

Senior Member

10-15-2012

A really easy way to add a minimal gold stream to support champs?

How about, last hit assists.

Deal damage to a minion or buff the champion who killed it and you get a small % of the normal value of the minion.
This would also help to get people to leash their jungles even in really low skill games.


Comment below rating threshold, click here to show it.

Ethelon

Senior Member

10-15-2012

Quote:
Originally Posted by CupcakeTrap View Post
Thank you for the support and bump!

I'm hoping to attract the attention of a wild Morello. Or the elegant and noble Xypherous. Or I don't know, maybe Phreak? His Tons-of-Damage Lulu build would benefit from solving this problem.


I don't personally endorse totally global gold either, because I think it might be too radical a change and would eliminate some interesting strategy oriented around starving specific champs.

However, I would point out that it would NOT make pushing minions to tower to deny less effective: if towers kill minions, nobody gets gold; no reason to think that would change under a global gold regime.
I typed up a big response and then deleted it because I realized I read what you said wrong. I thought you meant towers killing minions would grant global gold. My bad.

Quote:
Originally Posted by DebosBike View Post
A really easy way to add a minimal gold stream to support champs?

How about, last hit assists.

Deal damage to a minion or buff the champion who killed it and you get a small % of the normal value of the minion.
This would also help to get people to leash their jungles even in really low skill games.
This would be an interesting add. It would make a minor difference that would encourage advantageous behavior for yourself as well e.g. where you need to hit caster minions once so your AD can farm under tower early game.

A "problem" I can see is that it would encourage slight pushing for supports that can't buff the AD, when you might want to zone instead. That could be taken as a balance of rewards and decision making though.


Comment below rating threshold, click here to show it.

CupcakeTrap

This user has referred a friend to League of Legends, click for more information

Recruiter

10-15-2012

Quote:
Originally Posted by Ethelon II View Post
I typed up a big response and then deleted it because I realized I read what you said wrong. I thought you meant towers killing minions would grant global gold. My bad.
Haha, np!

Quote:
Originally Posted by DebosBike View Post
A really easy way to add a minimal gold stream to support champs?

How about, last hit assists.

Deal damage to a minion or buff the champion who killed it and you get a small % of the normal value of the minion.
This would also help to get people to leash their jungles even in really low skill games.
I think this is an interesting idea!

It might even be interesting if it worked to make "missed" last-hits (e.g. you hit but don't quite finish it off; an allied minion deals the last hit) a little less bad. This is reasonable, IMO, because it reflects that you weren't totally zoned out. The drawback I see is that it might make AoEs a little too good.


Comment below rating threshold, click here to show it.

NitroPenguinn

Junior Member

10-15-2012

Interesting. Personally a fan of the synergy, more skill-dependant, and you don't have to fight over last hitting so much (as in, the support getting damage in without stealing the kill).


Comment below rating threshold, click here to show it.

CupcakeTrap

This user has referred a friend to League of Legends, click for more information

Recruiter

10-16-2012

New from Morello:

Quote:
Originally Posted by Morello View Post
We do have some specific things in there that help supports directly - including alternate gold gain methods that work within the classic support playstyle.


Comment below rating threshold, click here to show it.

Frizzy350

Senior Member

10-16-2012

having a considerable amount of gold on a support creates problems in the late game. supports (at least in competitive games) need 2 open item slots just to stock up on wards. each team has oracles so they die fast, but you can't risk getting jumped on out of position by the enemy team.

often times in solo queue we will lose many teamfights because the support is busy building toward an item and can't fit wards in his/her inventory.