Polarity's Guide on Going From 1* to 3* (Updated: 1/22/15)

Trisul

Mau-- wrote:

So can you tank harder nodes then? I know it says you can't tank trivial...but what about throwing a "C" team at a hard/deadly node?

I'll test it next clear.

This appears to be the case, though I tanked on a normal node. On deadly you might be _expected_ to lose (just like you're expected to win a trivial), and so scaling adjustments might be less than for, say, a normal node.

Unknown

While this might apply to what happens to a node once it has begun, the starting enemy levels of the node is most definitely based off of your highest level character. Not entirely based on it, but it has a significant part.

To me, this seems like valuable insight on how a node scales from minute 2 to the last minute. How the starting level of a node is determined is still a mystery with highest level character playing a big role.

NorthernPolarity

Based on the current pve it seems like community scaling matters a lot more than personal scaling. Remember that lower level nodes dont sffect scalig as much so you arent really being punished for running patch laken there.

NorthernPolarity

papa07 wrote:

While this might apply to what happens to a node once it has begun, the starting enemy levels of the node is most definitely based off of your highest level character. Not entirely based on it, but it has a significant part.

To me, this seems like valuable insight on how a node scales from minute 2 to the last minute. How the starting level of a node is determined is still a mystery with highest level character playing a big role.

The only role it plays is that your guys take less damage because theyre more powerful. From my understanding, you have a pve mmr that decays event by event, and starting level is a sole factor of that and nothing else EXCEPT for gauntlet, which indeed is affected by your roster level.

Trisul

NorthernPolarity wrote:

The only role it plays is that your guys take less damage because theyre more powerful. From my understanding, you have a pve mmr that decays event by event, and starting level is a sole factor of that and nothing else EXCEPT for gauntlet, which indeed is affected by your roster level.

I regularly outperform some friends that have a higher leveled roster (and therefore should have a higher PvE MMR), and their starting difficulty is definitely harder than mine. I wonder if IceIX was _technically_ correct in that scaling isn't affected by roster, but base node levels might be.

Unknown

I think we are all on the same page but talking about different things. There are 3 points to PvE scaling.

Starting Enemy Levels (S)
Community Scaling (C)
Personal Scaling (P)

Current Enemy Level = S * (1+C+P) or S * (1+C) * (1+P)

Starting Enemy Level - This is where you highest level characters matter. While highest level character is not the only factor, it is definitely an important one. If we look at a node that was designed to have enemy levels equal to your highest level, someone with 94s might start between 85 and 105 depending on other factors, while someone with 166s will start between 150 and 180 depending on other factors that go into this black box.

Personal Scaling - This would be the scaling that IceIX was discussing, in my opinion. While it is counterintuitive that the amount of damage you take affects your scaling, it makes D3 sense to punish players who prevent damage (through shields, board control, etc) with higher scaling for the next pass.

Community Scaling - I would assume this is a blend of everybody's personal scaling factor or it is derived some other way from community experience.

loroku

papa07 wrote:

I think we are all on the same page but talking about different things. There are 3 points to PvE scaling.

Starting Enemy Levels (S)
Community Scaling (C)
Personal Scaling (P)

Current Enemy Level = S * (1+C+P) or S * (1+C) * (1+P)

Starting Enemy Level - This is where you highest level characters matter. While highest level character is not the only factor, it is definitely an important one. If we look at a node that was designed to have enemy levels equal to your highest level, someone with 94s might start between 85 and 105 depending on other factors, while someone with 166s will start between 150 and 180 depending on other factors that go into this black box.

Personal Scaling - This would be the scaling that IceIX was discussing, in my opinion. While it is counterintuitive that the amount of damage you take affects your scaling, it makes D3 sense to punish players who prevent damage (through shields, board control, etc) with higher scaling for the next pass.

Community Scaling - I would assume this is a blend of everybody's personal scaling factor or it is derived some other way from community experience.

This is what I'm getting as well. Here are some other thoughts:

- I'm pretty sure "personal scaling" stuff actually applies to individual nodes, while the "starting level" stuff is more for an entire sub.

- Anyone who uses tons and tons of health packs actually works against themselves, since restoring health means you'll lose a smaller percentage of it next battle. It's better to start with 5000 HP and lose 1000 (20%) than start with 10k and lose 1000 (10%).

- One thing that makes all of this personal scaling / community scaling stuff make more sense are the goon nodes. Goon nodes are, typically, more often nodes where you either win with almost no damage or just flat lose. Think about the last few events: goon nodes seem like they tend to scale really, really fast, or where the scaling actually goes down a touch. Maybe this is just my memory playing tricks on me, but these nodes tend to have more volatility - which would be reflected if the scaling were based on damage taken. Any node that it's easy to sail through - until it isn't - would experience a lot of volatility.

- On the other hand, the weakest nodes should rocket forward if that were true - so it seems like each node also has its own "rating" for how quickly it can climb, with the harder nodes given a faster rate of increase.

- Some people report tanking does something; others report it does nothing. I tried tanking a normal node twice and it did nothing, but I also hadn't won the node yet, so maybe that's a factor?

- Given that community scaling has gotten better (it's not as bad these days, either because they changed the formula or they're just using more events with shorter subs), "starting level" is very often actually more important than scaling.

- As was mentioned, the scaling stuff "changes from time to time" and often enough that they think it's too often to tell us exactly what it is - so everything we guess could be right or wrong now or in the future.

As for this guide: honestly, my biggest takeaway is that IceIX is absolutely wrong about his own game; specifically, % of HP lost from damage taken 100% DOES mean you should keep your levels low, because that's the only / best way to control how much scaling you receive. Combined with the obviously lower starting levels for enemy nodes, it seems like there's really no reason to ever level your roster if you want to do well in PvE. In fact, if you take into account that he seems to think that being higher level makes you do better, that's 100% the opposite of what you want to do! You want to finish each match with as much damage on you as possible - at least in respect to your own threshold for when you need to use health packs - and the game actually rewards poor play and taking big risks (starting battles with the least health reasonable and still losing more).

loroku

ANOTHER thought: you also want to avoid using boosted characters if you can; these will have higher health values and win matches faster, which works against you! Which means that having a very deep and relatively well-rounded roster is actually optimal, since you want all your guys to be as close to each other in level as possible so you can switch them out for any featured characters. Of course, having a deep and well-rounded roster is optimal in this game in every other way as well - here's just one more reason.

Trisul

Good post loroku, though I think a few of your points might be off.

loroku wrote:

- Anyone who uses tons and tons of health packs actually works against themselves, since restoring health means you'll lose a smaller percentage of it next battle. It's better to start with 5000 HP and lose 1000 (20%) than start with 10k and lose 1000 (10%).

You might be right, but it might also be something like [percentage_of_total_health_before] - [percentage_of_total_health_after], meaning (assuming 10k health max) 5000 --> 4000 HP = 10% total health loss, and 10000 HP --> 9000 HP = 10% total health loss.

loroku wrote:

- One thing that makes all of this personal scaling / community scaling stuff make more sense are the goon nodes. Goon nodes are, typically, more often nodes where you either win with almost no damage or just flat lose. Think about the last few events: goon nodes seem like they tend to scale really, really fast, or where the scaling actually goes down a touch. Maybe this is just my memory playing tricks on me, but these nodes tend to have more volatility - which would be reflected if the scaling were based on damage taken. Any node that it's easy to sail through - until it isn't - would experience a lot of volatility.

This I think is more likely community scaling going crazy, which is definitely specific to nodes. Likely, 90% of those matches being played on goon nodes is zero health loss (which seems to be the primary scaling motivator), which means scaling for that node goes up and up. I think personal scaling affects all nodes, since by tanking one node I eventually reduced levels in another node.

loroku wrote:

- Some people report tanking does something; others report it does nothing. I tried tanking a normal node twice and it did nothing, but I also hadn't won the node yet, so maybe that's a factor?

Tanking definitely works, but takes a lot of time. Not sure if it's really worth it, but I've been able to lower levels fairly reliably (though it doesn't affect that much overall).

loroku wrote:

As for this guide: honestly, my biggest takeaway is that IceIX is absolutely wrong about his own game; specifically, % of HP lost from damage taken 100% DOES mean you should keep your levels low, because that's the only / best way to control how much scaling you receive. Combined with the obviously lower starting levels for enemy nodes, it seems like there's really no reason to ever level your roster if you want to do well in PvE. In fact, if you take into account that he seems to think that being higher level makes you do better, that's 100% the opposite of what you want to do! You want to finish each match with as much damage on you as possible - at least in respect to your own threshold for when you need to use health packs - and the game actually rewards poor play and taking big risks (starting battles with the least health reasonable and still losing more).

I tend to agree with all this. The advantage of higher leveled roster only kicks in when enemy nodes hit the 395 level cap.

loroku

Trisul wrote:

...

Good points, all. Thanks especially that tanking one node seemed to affect others - good to know. Yeah, it's still hard to say how much tanking is worth it, but the fact that it can be manipulated this way is frustrating all the same.

I mean, I get the fact that they need some scaling. You need starting scaling because you can't just have high-level rosters stomp all over low-end nodes and dominate every event (that's what PvP is for ). But I don't understand the personal scaling and community scaling as well.

Community scaling: if everyone is beating the same node easily, why should it go up? What does this accomplish? Everyone still has an equal chance to beat it, right? Making it go up just means it's equally harder for everyone - which means only people who are willing to use limited resources (boosts) can keep up. Is that the idea?

Personal scaling: if I'm tearing up a node too easily, shouldn't refreshes take care of my ability to rank up points too quickly? What does it help to make the node harder (other than require more limited resources)?

I can understand the idea that if I'm dying to a node repeatedly it should go down - that seems fair. Similarly with community scaling, it would be nice if no one could beat a node that it would get easier for everyone. But why the reverse?

All I can think of is that they are trying to throw in another factor so that "starting roster" is not the only thing that affects your ability to compete, since that really would make leveling up 100% a bad idea. I guess they're trying to make the game auto-correct so that if you start too high or too low (likely too low), the "challenge" gets raised to about what it should be - as opposed to simply you started off with a different level roster. BUT this is also sort of counter to how these events actually run - which is, whoever plays at most optimal time periods wins. I don't know: it's confusing because I'm not sure what they are actually going for here. (Either way, nothing said so far leads me to believe that leveling your roster is a good idea for PvE - which is also frustrating.)

Unknown

loroku wrote:

Trisul wrote:

...

Good points, all. Thanks especially that tanking one node seemed to affect others - good to know. Yeah, it's still hard to say how much tanking is worth it, but the fact that it can be manipulated this way is frustrating all the same.

I mean, I get the fact that they need some scaling. You need starting scaling because you can't just have high-level rosters stomp all over low-end nodes and dominate every event (that's what PvP is for ). But I don't understand the personal scaling and community scaling as well.

Community scaling: if everyone is beating the same node easily, why should it go up? What does this accomplish? Everyone still has an equal chance to beat it, right? Making it go up just means it's equally harder for everyone - which means only people who are willing to use limited resources (boosts) can keep up. Is that the idea?

Personal scaling: if I'm tearing up a node too easily, shouldn't refreshes take care of my ability to rank up points too quickly? What does it help to make the node harder (other than require more limited resources)?

I can understand the idea that if I'm dying to a node repeatedly it should go down - that seems fair. Similarly with community scaling, it would be nice if no one could beat a node that it would get easier for everyone. But why the reverse?

All I can think of is that they are trying to throw in another factor so that "starting roster" is not the only thing that affects your ability to compete, since that really would make leveling up 100% a bad idea. I guess they're trying to make the game auto-correct so that if you start too high or too low (likely too low), the "challenge" gets raised to about what it should be - as opposed to simply you started off with a different level roster. BUT this is also sort of counter to how these events actually run - which is, whoever plays at most optimal time periods wins. I don't know: it's confusing because I'm not sure what they are actually going for here. (Either way, nothing said so far leads me to believe that leveling your roster is a good idea for PvE - which is also frustrating.)

So in one bracket all 500 people clear a node, next time out scaling goes up 10 levels. Now only 450 clear the node, scaling goes up another 9 levels. Now only 350 clear the node. Scaling helps create separation between potentially equal players.

jquin94

Hopefully this is the best place to post , but i've been having a roster issue lately. I just found out about this forum and after reading the guide here, I am starting to second guess my team.

Starlord 1/1/1 Level 78
Storm Mohawk 1/0/2 Level 50
Thor modern 1/1/1 Level 45
Squirrel Girl 1/1/0 Level 43
Iron Man 5/5/3 Level 42
Magento NOW 2/5/0 Level 40
Cpt America Steve Rodgers 1/1/0 Level 40
Cpt Marvel Ms Marvel 2/3/2 Level 37
Storm Classic 2/2/3 Level 37
Black Widow Original 1/3/1 Level 25
Hawkeye Modern 2/1/1 Level 25
Ares Dark Avengers 1/4/1 Level 23
Thor NOW 1/2/1 Level 18

Im not too high on ISO at all (4800), but i am also 75 HP away from buying 3 more slots for my roster. Should i continue to get covers for my 2's at the moment or take the 2-3 spots to max out some 1's?

Pongie

Iv only been playing for 40 days, but what I found out recently is that the 1*s are outclassed fairly quickly, even when they are maxed. If you can survive with your current roster now, than I don't see the need to recruit 1*s. Perhaps maybe keep one (the consensus is juggernaut) since there's an event that requires them. But that's about it and focus more on your 2*s. What I find now is that 3*s and fairly useless and eats up a lot of roster space while collecting their covers. Keep a core of 6-8 2*s that will get you through until you can transition into the 3*s.

hurcules

jquin94 wrote:

Hopefully this is the best place to post , but i've been having a roster issue lately. I just found out about this forum and after reading the guide here, I am starting to second guess my team.

Starlord 1/1/1 Level 78
Storm Mohawk 1/0/2 Level 50
Thor modern 1/1/1 Level 45
Squirrel Girl 1/1/0 Level 43
Iron Man 5/5/3 Level 42
Magento NOW 2/5/0 Level 40
Cpt America Steve Rodgers 1/1/0 Level 40
Cpt Marvel Ms Marvel 2/3/2 Level 37
Storm Classic 2/2/3 Level 37
Black Widow Original 1/3/1 Level 25
Hawkeye Modern 2/1/1 Level 25
Ares Dark Avengers 1/4/1 Level 23
Thor NOW 1/2/1 Level 18

Im not too high on ISO at all (4800), but i am also 75 HP away from buying 3 more slots for my roster. Should i continue to get covers for my 2's at the moment or take the 2-3 spots to max out some 1's?

I'd say concentrate on your 2* now as they will carry you through to 3* transition.

Nymr83

1* characters are a waste, once you stop using them you'll never go back.

it may be worth keeping a 5/5 Juggernaut at level 1 (dont waste iso) around if you have extra roster space as he is the best character in the 'balance of power' event. he's also good in 'combined arms' but you need to spend the iso to max him for that one. they recently nerfed the match damage that 1* did as they scaled to 270 but he's probably still the best with the most hitpoints and 2 cheap/relevant abilities making him a force on defense.

if you see one of those events come up though, you could get the jugg covers in a day by playing simulator/opening standard tokens/etc, they drop pretty frequently.

Unknown

Nymr83 wrote:

it may be worth keeping a 5/5 Juggernaut at level 1 (dont waste iso) around if you have extra roster space as he is the best character in the 'balance of power' event. he's also good in 'combined arms' but you need to spend the iso to max him for that one.

Disagree with the end of that statement. There is very little difference between a 101 and a 140 Juggs. In my opinion, it is not worth the Iso for a limited increase in damage offset by a five fold increase in healing time.

As long as he has 10 covers, he needs no iso.

Nymr83

papa07 wrote:

Nymr83 wrote:

it may be worth keeping a 5/5 Juggernaut at level 1 (dont waste iso) around if you have extra roster space as he is the best character in the 'balance of power' event. he's also good in 'combined arms' but you need to spend the iso to max him for that one.

Disagree with the end of that statement. There is very little difference between a 101 and a 140 Juggs. In my opinion, it is not worth the Iso for a limited increase in damage offset by a five fold increase in healing time.

As long as he has 10 covers, he needs no iso.

Fair enough, my thought was more than the extra 40 levels acts as a much bigger defensive deterrent in that event

Unknown

Does anyone has a level cap calculator that's updated with the new level caps? that one goes up to 230 for 4*

Unknown

It seems like the introduction of Deadpool's Daily might significantly change some aspects of transitioning between tiers. The 1* node is probably great for beginners trying to farm iso, but it makes me wonder about the cost/risk of DPD vs Lightning Rounds. It's probably possible to pull off both with reasonably leveled 1*s though, so that may be a nonissue. 1*->2* is so fast anyway, it may be a moot point.

2*->3*, on the other hand, got a huge boost. But the biggest features (Taco Token 2*s and guaranteed 3* covers) are both tied to survival nodes. This obviously makes OBW even stronger for transitioners, but I wonder if it also boosted 2* Cap. Moonstone probably retains her dubious usefulness because of Control Shift's cost, and Bag-Man can't really be helped by anything short of a boot to the tinykitty.

NorthernPolarity

Cammorak wrote:

It seems like the introduction of Deadpool's Daily might significantly change some aspects of transitioning between tiers. The 1* node is probably great for beginners trying to farm iso, but it makes me wonder about the cost/risk of DPD vs Lightning Rounds. It's probably possible to pull off both with reasonably leveled 1*s though, so that may be a nonissue. 1*->2* is so fast anyway, it may be a moot point.

2*->3*, on the other hand, got a huge boost. But the biggest features (Taco Token 2*s and guaranteed 3* covers) are both tied to survival nodes. This obviously makes OBW even stronger for transitioners, but I wonder if it also boosted 2* Cap. Moonstone probably retains her dubious usefulness because of Control Shift's cost, and Bag-Man can't really be helped by anything short of a boot to the tinykitty.

Some help regarding 2* strategies for DP Daily's survival nodes would be great! I'm not convinced that Cap is good regardless just because of how little damage he ends up doing: I'm guessing you can get by with Ares / OBW / random pretty decently.

I'll try to add the "do the DP daily" to the guide sometime in the future.