Analysis of Legendary Heroes and Power Levels

[VeryQuietly]

In the spirit of Black Friday and the massive summon portal in the game right now, I thought I’d take a look at all the 5 star heroes and run some stats. As of this writing, there are 598 legendaries live or in beta. I’m using power value at 4-80 for these metrics, not including LB, CB, talents, or aether power.

Sources:

New Chronological List of Hero Releases

@BlueHair has started to collect the Chronological List of Heroes here: Chronological List of Hero Releases But he is retired from the Forum on 3th of April in 2022. So I taking over this initiative from him. Total released: 658 heroes and 220 costumes We als have 10 one star heroes, and 16 two star heroes. So 892 heroes/costumes in total. I have changed the format a bit compared to what @BlueHair had to better transparency

2017 - 86 heroes

March 2017 - Initial Release (58 heroes) T…

EnPluzz - Data vault (and tools to come at some point in the future)

Hello everyone, To celebrate my first forum anniversary (and my first topic ever on here), I planned to share some more of my data with you all. I am still planning on developing some tools too but life has been busy and I did not have much time for that but it is still my plan to do so. So first and foremost, I think one thing that is of interest for everyone is… heroes database, in CSV format, easy to import in any spreadsheet tool (Excel, Google Sheet, LibreOffice Calc…) and then to add fi…

Based off the info from these two posts and the many beta beat threads along the way, I ran the numbers for all 598 heroes.

Strongest 20

Name	Origin	Release Year	Release Month	Max Power
Jolt	Contest of Elements	Beta		986
Desmond	Contest of Elements	Beta		982
Sparklight	Black Friday - Astral Elves	2023	11	977
Starwalker	Black Friday - Astral Elves	2023	11	975
Ironmaw	Contest of Elements	Beta		975
Farrah	Contest of Elements	2023	11	971
Atlanteia	Untold Tales	2023	11	971
Daemon	Alliance Quest - Clash of Knights	Beta		968
Winnie	Alliance Quest - Clash of Knights	Beta		968
Frond	Covenant Summon	2023	10	968
Tenebrae	Covenant Summon	2023	10	965
Persa	Untold Tales	2023	10	964
Zaria	Contest of Elements	2023	10	961
Fulvia	Tower of Owls	Beta		961
Timius	Tower of Owls	Beta		961
Ommodus	Tower of Owls	Beta		960
Relius	Tower of Owls	Beta		960
Xie Zi Jing	War of the Three Kingdoms	Beta		960
Livia	Tower of Owls	Beta		959
Chang’e	War of the Three Kingdoms	Beta		959

Weakest 20

Name	Origin	Release Year	Release Month	Max Power
Justice C2	Classic - C2	2023	2	755
Marjana C2	Classic - C2	2023	6	755
Richard C2	Classic - C2	2023	9	755
Richard C3	Classic - C3	2023	10	755
Justice	Classic	2017	3	754
Magni	Classic	2017	3	754
Justice C	Classic - C1	2021	1	754
Magni C2	Classic - C2	2022	10	754
Elena	Classic	2017	3	753
Horghall	Classic	2017	3	753
Elena C	Classic - C1	2020	1	753
Horghall C	Classic - C1	2020	1	753
Joon C	Classic - C1	2020	1	753
Horghall C2	Classic - C2	2022	10	753
Elena C2	Classic - C2	2023	8	753
Joon	Classic	2017	3	752
Leonidas	Classic	2017	3	752
Leonidas C	Classic - C1	2021	1	752
Joon C2	Classic - C2	2022	12	752
Leonidas C2	Classic - C2	2022	12	752

Plotting Spread of Individual Hero Power over Time by Group:

Plotting Average Power and St Dev per Release Month:

Probably hard to see all the details, but you get the idea. Heroes stayed in 750-800 range 2017-2021 (so the first 150 heroes or so). Then came the release of the Slayers, which broke the 800 power cap. The 900 power cap was shattered with release of Roz (950) and Xandrella (900) in December 2022. We’re not yet a year later and rapidly approaching the 1000 power mark , with the upcoming Jolt at 986. (Y’all noticing the pattern about Contest of Elements heroes being at a high base power?)

There are still plenty of heroes coming out now that remain at the 750-800 power mark like Classic Costumes, but realistically I don’t see how these can complete anymore. Not when we are rapidly approaching a 1k power hero. And that’s BEFORE any limit breaking.

If you’re after a monster legendary hero, where do you get best bang for your buck?

Let's take a look:

Origin/Family	Average Power
Astral Elves	976
Tower - Owls	960
Event - Opera	951
Event - Goblins	949
Untold Tales	944
Contest of Elements	935
Event - Masked Gods	919
Covenant	916
Alliance Quest	882
Event - Pets	879
3 Kingdom	876
Tower Magic	874
Tower - Styx	866
Season V	862
Event - Gargoyles	843
Tower - Ninja	836
Seasonal	833
Festival II	823
Season IV	816
Festival I	794
Legends	790
Season III	788
Season II	779
Classic	757

Edit - adding some more granularity to the breakdown

Max Power Average and Standard Deviation Breakdown

Plot of the above data, Average Max Power Breakdown

End Edit

So the data is telling us that the best heroes are always the ones in the next portal. Tower of Owls just hit beta 11/23/23. Opera event is coming up end of November 2023. Of the 20 strongest heroes in the game, only 8 are live on 11/23/23. Over half remain in the wings. The pace at which new heroes become outdated and underpowered is getting steadily worse. There is always a shortage of ascension materials so “wasting” them is a real concern. Not even considering limited aethers, alpha aethers, and emblems.

How should you know which hero to max? My opinion at this stage is that I’m no longer at all interested in competing with strongest. I’m after the fun art and the fun or unique game mechanic. I’ve pulled back from a lot of competitive game play for this specific reason. I invest mats and time to level a new hero, and when I’m done, there are at least 15 new heroes out that overpower what I have. So why bother?

More heroes are coming out more rapidly. And those heroes are getting stronger. Harder to get heroes (lower odds or more expensive portal) are stronger.

Though interesting to note that HOTM are changing quite dramatically lately too:

The ones down in the 750-800 range coming up soon are costumes for older HOTM. The January and February 2024 HOTM both break the 900 power barrier.

I’ll close with the following opinion: power creep is real and out of control. Do not chase heroes. Portal odds are terrible and you’ll want the next shiny hero less than four weeks later so even if you do get lucky, you won’t get to enjoy it for long.

Case Study - What healer should VQ level next?

Healers I already have ascended:
Augustus
Prof Lidenbrock (no costume)
Hathor
Alberich+C1
Arco
Raffaele
Four stars that rock with the big boys: Kiril with 2C, Rigard with 2C, Boldtusk with 2C all LB1 +20.

Options:
Lysanor
Thalassa
Hetepheres
Milena
Ariel (no costume)
Darkfeather
Hawthorn
Xnolphod
Toxicandra
Heimdall (no costume)
Cleopatra
Rhys
Furdinand
Vivica+C1+C2+C3

I’ve got plenty of Nature healers, so Toxi and Heimdall are out.
I’m super short of tabards, so Xnol, Hawthorn, and Darkfeather are out. If I had the tabards, Xnol would still be out.

I’ve got enough scopes and darts to do one of each.
Holy: For pure HP recovered in a single turn, Cleo beats out all the others. For other benefits, Furdinand would tip the scales. Rhys is fast mana. Given all these options, why would I consider Viv, even if I have the Toon version? I’m still terrified that Furdinand will get hit with the nerf hammer, which is the only reason I haven’t yet ascended him.
Ice: Milena is out given comparatively low heal % and age. Ariel is out as I lack CB and age. Hetepheres is interesting, but comparatively outclassed on stats and the other part of special is rather niche. So between Thalassa and Lysanor.

And with the mindset being I don’t care if its the “best” but rather if its the most fun for me, I’ll probably pick Furdinand (love dogs) and Lysanor (love the art of all the CoK heroes).

Statistics Appendix for the Curious

Element	Count
Dark	124
Fire	123
Ice	124
Nature	115
Holy	110

Year	Count
2017	34
2018	28
2019	28
2020	56
2021	92
2022	136
2023	186
2024	38

Class	Count
Fighter	73
Wizard	67
Cleric	66
Paladin	63
Barbarian	62
Sorcerer	59
Druid	56
Rogue	54
Monk	52
Ranger	46

Mana Speed	Count
Very Fast	33
Fast	178
Changing Tides	14
Average	222
Slayer	5
Charge	18
Magic	13
Styx	7
Slow	103
Very Slow	5

Aether Power	Count
Attack Up	148
Defense Up	81
Special Boost	61
Regen	35
Special Armor	33
Ailment Immunity	31
Counterattack	29
Damage Reduction	26
Rage	25
Dodge	24
Fiend Resist	20
Heal Increase	18
Bulwark	14
Mana Boost	12
Revive	9
Vampire	9
Taunt	8
Boosted Regen	7
Ailment Reflect	5
Gamble	2
Knight’s Endurance	1

[Elioty33]

Great post. Always feels rewarding to see people make great use of my data vault and that’s the reason I make it, for the people to make their best use of it.

Back on topic, for the origin/average power table, I think you could split the Alliance Quest line to differentiate CoK and Musketeers. Also wondering if it should differentiate Monster Hunters families and the Construct family on the Covenant line. But this one is a bit tricky since the available heroes in the portal are different each time it comes around.

[Mad1max9]

This is now my favorite post on the forum. Thank you to the data collectors and organizers. Thank you for the analysis. And thank you all for sharing!

Yes, CoK differentiated from Musketeers. (And if we’re really getting into the numbers, wave 1, 2, 3…)

Good to see you making magic in the E&P world again, VQ.

[lucky3602001]

Heroes are released so quickly that the heroes obtained by players are replaced by the latest heroes in a short period of time. This is the reason why more people are unwilling to spend money. The heroes I spend money to chase will soon be useless, so why should I chase them?

[Elioty33]

@Mad1max9

And if we’re really getting into the numbers, wave 1, 2, 3…)

If you go that far, it’s going to be complicated because that information is not available from the game data. For CoK, it’s easy because gen1 and gen2 heroes are in different families so easy to filter but for S2 gen2 heroes (Oceanus and the other lady), CF1 gen2 heroes (the new heroes released since the 5 original CE got merged together), old Ninja heroes and newer ones released after the Ninja Tower rework, etc, you have to go through all the heroes by hand to tag them either gen1 or 2. Or script something depending on their origin and release dates.

[Mad1max9]

I’m not able to script, but give me the data and I’ll tag them by hand. I’m off for the next few days.

[Elioty33]

It’s all in my data vault, download the latest hero CSV file, fire it up in your favorite spreadsheet application (Microsoft Excel, Google Sheets, LibreOffice Calc, …), add a column and get to work then share your file with @VeryQuietly if he is interested in this.

[marki]

thank you for your time doing it , apricate it, unfortunately it just confirms our concerns how game goes to fast and losing controls:()

[Mis]

Great post. You may be very quietly, but when you speak, you speak greatly

I’ve just noticed it recently, when my LB +20 Milena had weaker stats & power than the new CoK heroes in beta.

And just a year ago Milena was the best hero in the game (for me at least).

[PlayForFun]

This is a very nice statistics !

Thx for creating it @VeryQuietly !

[CrazyChemist3891]

@VeryQuietly thank you very much for this great analysis!
One question: you used the stats/power level of the heroes as they are now? Not the power level of their initial release, which in some cases might have been modified by Balance Updates, correct?

[Pr0spero]

Great work.

I just noticed this anecdotally when I pulled Neema (hotm) and went to find her in my roster, which is organised by power. Her base stats place her higher up the list than every other 5* Ive got waiting in reserve (of which there are 28).

Lowest is Richard / Lianna / Obakan on 396 base power. Neema comes in at a whopping 462

[Jack.O.Kent]

In 14 years I’ll have all of those heroes in the top 20

[Pr0spero]

At which time they will be called The Bottom 20

[Jack.O.Kent]

Don’t care, I’m going to save for when they’re in FS!

[Pr0spero]

Wonder what the art will be like in 20 years?

Genderless, featureless grey blobs so as not to offend anyone and to keep costs down. Base stats of 8k power. 14 passives.

[VeryQuietly]

Thanks everyone for the feedback. I’ll admit I started this just as a way to “show” myself that chasing the astral elves in the Black Friday portal is pointless, not really intending to post. Thought it looked cool (and so did my SO @Archaeo80) so shared it in hopes it would help someone else.

I’ve got plenty of heroes waiting to be leveled. Got a non-costumed Jabberwock from this portal… with an unleveled power at 408 and since I don’t have a costume, I don’t expect to spend the few tabards I have on him. Which is a shame because there was a time I would have been over the moon for Jabberwock. I know it’s all roster dependent, but I only have enough tabards for one purple hero right now and Hurricane, Darkfeather, or Deadboot will get maxed before Jabby.

I’ve been thinking on how I can show the data in a more granular manner without overwhelming the graphs. I’ll post some stuff breaking it down a bit more.

@CrazyChemist3891 I used the values from @Elioty33 ’s most recent date csv file, which as best as I can tell are current live power values. I manually added release info for the ones missing it and added the most recent beta additions. So in short, values should be accounting for any nerf/buff since release. Having the Black Friday portal with everyone at base stats to double-check has been helpful.

[Oxide]

@VeryQuietly Welcome back - it’s been a while :

And a great piece of analysis too.

For those who don’t remember @VeryQuietly was an early calendar pioneer

[Mikal]

Wish you didn’t make us guess at the horizontal axis, but it was pretty clear. I do hate that you used Power as your metric. Severly flawwed at indicating the strength of a hero. It’s easy and available, so I can see why you chose it. I just wish that someone would come up with a better metric that was as easy to use.

[VeryQuietly]

Thought about that too… I could use attack, defense or health since those are all in the csv, but I don’t know that those would be better indicators. None of this takes into account if a special skill design is OP or weak. If a special skill doesn’t do damage, is a hero weaker? Or glass cannon? That’s why I went with power as it should reflect how the three stats are combined. Can’t use community generated hero grades as those are not being maintained anymore. I opted for 4-80 values since every hero has that and it represents a good baseline. I’m planning to add a personal case study on healer heroes. I recently got Vivica and all three costumes, but I’m on the fence about leveling given the investment.