All Class Sets Ranked by Real Clears

Hello Everyone,

Many of you have debated the direction of patches… were you unimpressed by the S26 theme? Are you sad that your favorite build(s) didn’t get buffed this patch?

Well… this might just be your thread. To help liven the mood, I am presenting a chart that I compiled to show all sets and real clears under 5k paragon. It is a list of hand-picked clears to help show the community and devs how our builds rank. The list takes into consideration item changes from 2.7.3 and excludes clears after historical nerfs.

Majority of the clears are in the 3-4k range – a realistic end of season paragon accumulation for the dedicated. Certain clears just don’t have enough competition behind them – mainly due to nerfs or poor performance compared to those able to clear higher.

The ultimate purpose is to spark enough discussion to prove this topic is where development should be aimed next patch: fixing all underperformers in one swoop via set or legendary item upgrades.

UPDATES
Further down this thread I worked with Rage to create new reports. I went back to the leaderboards and included more competitive data from high paragon players (not exceeding 9kp). Rage supplied a formula to scale these clears to 5kp. We now have a really cool tier list that you can checkout:

Lists to Access

CLASS SET SKILL CLEAR TIME PARAGON
Barbarian Raekor Ancient Spear 142 13:45 3289
Barbarian MotE Earthquake 138 14:53 3673
Barbarian Wastes Rend 141 13:52 3769
Barbarian IK HotA (pre 2.7.2) 131 13:48 3338
Barbarian H90 Frenzy 135 13:38 3560
Barbarian LoN/LoD HotA 144 14:24 4248
Crusader Akkhan/Invoker Bombardment 137 13:19 4000
Crusader Invoker Slash 137 14:01 4304
Crusader Roland’s Sweep Attack 133 13:17 4048
Crusader Seeker of the Light Blessed Hammer 131 14:50 4042
Crusader Aegis of Valor Heaven’s Fury 135 13:02 4000
Crusader LoN/LoD Bombardment 140 14:48 2974
Demon Hunter Marauder’s Cluster Arrow 143 13:56 3810
Demon Hunter Marauder’s Multishot 139 14:04 2155
Demon Hunter Shadow Impale 133 14:17 4261
Demon Hunter UE Multishot 131 13:18 4575
Demon Hunter UE Hungering Arrow 128 13:16 4569
Demon Hunter Natalya’s Rapid Fire 126 12:03 4776
Demon Hunter Natalya’s Hungering Arrow 134 14:23 3262
Demon Hunter GoD Hungering Arrow 137 14:07 3114
Demon Hunter LoN/LoD Rapid Fire 132 14:53 4621
Monk Raiment Crippling Wave 126 13:41 4335
Monk Monkey King Tempest Rush 140 11:34 4320
Monk Uliana’s Exploding Palm 131 14:09 3163
Monk Inna’s Mystic Ally 142 12:58 3657
Monk PoJ Tempest Rush 137 14:22 2853
Monk LoN/LoD Wave of Light 140 14:53 3203
Necromancer Rathma Army of the Dead 138 12:32 4060
Necromancer Trag’Oul’s Corpse Explosion 123 13:36 3602
Necromancer Inarius Corpse Explosion 125 14:23 2812
Necromancer Pestilence Corpse Lance 125 14:01 3187
Necromancer MotBC Bone Spear 133 13:50 4769
Necromancer LoN/LoD Corpse Explosion 138 10:47 2495
Witch Doctor Arachyr’s Corpse Spiders 143 14:32 2862
Witch Doctor Jade Harvester Locust Swarm 130 14:47 3345
Witch Doctor Helltooth Zombie Charger 133 14:32 3393
Witch Doctor Zunimassa’s Poison Dart 137 11:06 3464
Witch Doctor Mundunugu Spirit Barage 139 14:55 3382
Witch Doctor LoN/LoD Poison Dart 130 14:14 2886
Wizard Firebird’s Flame Blades 137 14:43 3800
Wizard Vyr’s Chantodo 138 14:58 4048
Wizard Delsere’s Energy Twister 135 14:27 3001
Wizard Delsere’s Frozen Orb 132 13:31 3532
Wizard Tal Rasha’s Frozen Orb 131 13:29 4549
Wizard Typhon’s Veil Hydra 138 13:05 3458
Wizard LoN/LoD Energy Twister 137 12:52 3669

Sets/builds under GR137 must get buffed.

Do you think any of these builds are not represented accurately? Comment below. If you’re willing to prove it – jump into PTR, set your main stat according to 4k paragon and make a clear. Notify us by commenting in the thread when done.

Share your thoughts below.

POST MOVED TO GENERAL DISCUSSION

18 Likes

I have a question/request. Can you test a Trag’Oul Blood Nova build and share your result?

1 Like

This is great, thanks! I think lots of people want to get this done, so good to see a table! Great starting point!

Where are these clear from? Both PTR and live off-season? You say hand picked, but how did you hand-pick them? Highest clear around 4k paragon?

Overall, it’s not so bad actually. Almost all sets are within 10GRs here.

For strongest sets, didn’t expect to see hota and spooders at 1 and 2. The discussion has been between innas and WoL mostly. Do you think that HOTA and spooders are as strong or even stronger than innas and WoL?

Wizards looks roughly as expected, apart I expected there’d be firebirds and twisters at top, maybe vyrs, and then a few GRs down to typhon and then delsere. Maybe should separate flame blades and explosive blast firebirds?

1 Like

Easiest way is to bring up the bottom rung +15 grs. How? Shiiiiiity builds get to use ethereals! How do we do this? Tie the ethereals to weapon type & passives, bonuses etc to low performing skills, weapon types. Everything else gets taken out of the RNG table.

1 Like

Where do your numbers come from? Which areas ? Off season ?

3 Likes

The data comes from PTR, Non-Season 2021-1 (nerfs excluded), 2021-2 (no Inna’s / Norvald), Season 23 (nerfs excluded). To clarify:

PTR 2.7.3 v2 (season and non-season, ie. no gems in socket)
Non-Season Eras 2021-1 & 2021-2 (NA/EU/Asia)
Season 23 (NA/EU/Asia)

I filtered out AoV & A6I2 Bombs Norvald’s, AoV Nerf, Firebird’s, Inna’s, Bone Spear, Mundunugu, and to some extent what looked like Inna’s Missile Dampening on PTR.

Since no one at 4000 paragon pushed AoV/A6I2, I did quick 3-key clears on PTR at using only 4000 paragon – you’ll see those in this list. Those two clears can be higher, but I needed to fill the holes and show overall capability. The rest are recorded clears from other players on PTR/NA/EU/Asia. Before any item edits are made based on this data, devs should really spot check higher clears to see if it’s going to cause problems.

Add a zero here, let’s turn that 2 into a 9. All kidding aside, they have to be surgical in their approach. For the health of the game, we still want to play skills like elemental arrow and Chakram. An example, Demon Hunter LoN/LoD should be the result of a buff to Fan of Knives. Certain sets do require tuning that goes beyond damage multipliers, like adding DR to Monk’s squishy Raiment. Necro sets are on the redundant side and deserve substantial edits.

Yes, this is probably a smarter path that will bring extra longevity to the game – leaving room for itemization helps all builds (primal 2.0) and seasonal bonuses that doesn’t force the game into time attack mode. In the past I’ve mentioned forcing us back down to GR120 @5k and redoing paragon scaling and the legendary gem system so that nobody feels screwed being unable to clear GR150 for their 150 gems. It’s a much easier sell to pitch the buffs to GR140 for balance, then run a giant across the board nerf bundled with new chases. Heck even a mini expansion would make sense for this raising the level cap to 75 with some new skills or special abilities.

Does anyone with a necro want to volunteer 15-20 keys to see what happens?

It is a strong build, but not the strongest. There is an incredibly talented player pushing HotA and so it presents this way. The UE6 multishot clear also looks this way, GR131 is like 2000+ keys. Most of the non-season veterans are above 5k paragon and their clears are excluded from this data set, but I was very surprised to fill in everything – much thanks to the incentive of playing set leaderboards.

I’ll poke around and take another look at recent Firebird’s boards.

Going by the numbers you’ve shared, it would be far simpler and quicker to bring everything over 137 down.

Great compilation though.

2 Likes

It’s a cool list, very interesting to see, but I wouldn’t necessarily take it at face value for determining what should or should not get buffed, or by how much.

Why? Well, because we don’t really know the background behind any of these clears. I mean, for instance, look at the MOTE Earthquake clear, a 138 with 3673 paragon. But, y’know, if that player didn’t exist, the top would be 133 with 4700 paragon. And, on the other side of the argument, 138 is definitely not the limit there- I can tell you, basically with 100% certainty, that the build can do at least GR 140 at 5k paragon.

The fact of the matter is: not all builds have lower paragon players who push them very hard. I mean, look at that IK HOTA mark. And that’s pre 2.7.2! So, even when handed an extra 6 tiers of power (via the buff to Remorseless), nobody has bettered that 131. That just tells me that nobody at the target paragon range is seriously playing the build. But, at higher paragon (11k), the build has done 149.

If you assume 70k STR at 11k paragon, 5k paragon should be about 40k STR. 70 / 40 = 1.75, which is 3-4 tiers of power. Throw in a couple more for survivability issues at lower paragon, if you like, but the “spread” really shouldn’t be more than 6 tiers, because you could always swap Pain Enhancer, worth about 2 tiers of damage (when you find excellent density, worth a bit less at other times) for Esoteric, worth roughly 25 tiers of defense against non-physical attacks.

So that would be 149 - 6 = 143 at 5k paragon, which is 1 tier weaker than the 144 for LoD HOTA. That mark sounds correct to me.

Bottom line, I guess I’d say that these numbers are only useful when you know the background for that particular clear.

5 Likes

This is very true, I talked about being surprised that I was even able to fill in all of the set data with values that looked like players actually put some effort into their clear. Nothing below 10:00, and some recognizable names behind them.

This list is really is a status check of the game. The fad/stronger builds are played more. These clears are the bottom range of a bigger chart showing a more complete picture of max power vs popularity.

Some builds can’t even equip best in slot items until 7-8k paragon due to survivability –

Like we discussed in that big thread on General Discussion, anyone doing the decision making can scale these numbers up to 5k paragon, then tack on a subjective column indicating popularity or estimated keys to come up with a better guess at GR value.

Decision makers without full knowledge of each class/build should NOT make balance changes based on this data alone, or without consulting an expert. This data should be a discussion topic, not an action item.

2 Likes

Has any scraped the data from all the set leaderboards? If we can get GR, paragon and clear time (ideally across last few non-season eras and PTR, even better across all regions), then we could get a much better idea. Maxroll has overall leaderboards, but not the set ones, and I"m not sure we can download it. I’d be happy to make some analysis and plots if someone can give me the data in any reasonable file format…

It can’t be done like this due to needing to filter out the nerfs. Believe me, it’s MUCH easier to take the top 10 from each board, but that data isn’t clean due to the nerfs.

This is IMHO a better representation of what is going on with build strength – we just need class experts to assess each set/build and forecast where each can do at 5k paragon within 500 keys.

Here is a for instance:

CLASS SET SKILL CLEAR TIME PARAGON EST@5K
Cru Akkhan/Invoker Bombardment 137 13:19 4000 142
Cru Invoker Slash 137 14:01 4304 140
Cru Roland’s Sweep Attack 133 13:17 4048 134
Cru Seeker of the Light Blessed Hammer 131 14:50 4042 133
Cru Aegis of Valor Heaven’s Fury 135 13:02 4000 139
Cru LoN/LoD Bombardment 140 14:48 2974 142
DH Marauder’s Cluster Arrow 141 14:13 2802 144
DH Marauder’s Multishot 139 14:04 2155 143
DH Shadow Impale 133 14:17 4261 138
DH UE Multishot 131 13:18 4575 130
DH UE Hungering Arrow 128 13:16 4569 129
DH Natalya’s Rapid Fire 126 12:03 4776 130
DH Natalya’s Hungering Arrow 130 13:30 4243 134
DH GoD Hungering Arrow 137 14:07 3114 138
DH LoN/LoD Rapid Fire 132 14:53 4621 131

Yes.

Since we both agree that you can’t really use this data without the insight of a “class expert”, I think a more thorough way of determining build power is to work both “up” and “down”.

In other words, I think we do look at these sub-5k clears, and adjust them for paragon (and time), but we also look at the top clears of any paragon, and just adjust those for paragon as well- downward- like I did for IK HOTA in my earlier post.

Yes, there are cases where, at lower paragon, you might need to swap an offensive item/skill for a defensive one. But, that’s where having “the expert eye” clears things up pretty quickly. For instance, one of the most common swaps, for several classes, is “defense ring” <----> CoE. Yeah, we need to take that into account, but CoE is a pretty known quantity. It generally adds 2-3 tiers of damage. There are exceptions to that (like Raekor HOTA, where CoE actually adds 7 tiers of damage), but those are known quantities, too.

So if you’re working “down” from some clear at 10k paragon, you just need to assess where and when any of these sorts of swaps might need to be made.

I think that combining these “up” and “down” approaches would produce a far more illuminating (and useful) picture.

Want to do the barb clears?

Do we have any takers for the other 4 classes?

We’ll need peer reviews too — those who want to chime in and debate the estimated clears can call BS, but must do so with reasonable explanation.

I’m very curious about the Int classes since WD got Shukrani’s, Wiz sets like Delsere’s and Tal Rasha aren’t played much and certain Necro builds used Trag’oul’s Fang before nerfage and were popular before the Follower revamp — it is hard to tell Corpse Lance’s and Nova’s true power.

1 Like

Sure. In fact I’ve probably already done it, somewhere, I just need to find it…

Not sure if Tinne’s “back from paternity leave” yet, but he’s the obvious choice for Wiz. Or if he’s out then I nominate Cascade.

1 Like

BTW, I think we should adjust for time and paragon as follows:

For Paragon, assume 60k mainstat at 10k paragon, and work from there. So that’s 35k at 5k paragon, 30k at 4k, 70k at 12k, etc. And then of course you just compute the difference.

And let’s go to the first decimal place, i.e. if somebody did 130 at 4k paragon, we’d adjust this by saying, 4k = 30k mainstat, 5k = 35k. 35/30 = 1.167 . Log1.17(1.167) = 0.98, and we’d round this to a tenth, i.e. 1.00, so we’re adjusting this clear up by 1.00 GR level, so with the initial clear being 130, here we’d say the paragon adjusted clear is GR 131.0 .

For time, I think we just use the base 1.17x per tier. Yes, there’s all the stuff in the Tier vs Time thread we could use, but that’ll get really complicated, and besides, we never 100% settled that stuff anyway. But, you’re pretty much always going to get at least the value of the excess time, increased by 1.17x per tier, since things like Stricken, Conduit, etc give you better value than that on your time.

So for instance, if we had a GR 135 at 5k paragon, in 12:00, we’d say, 12:00 = 720 seconds. 900 / 720 = 1.25. Log1.17(1.25)= 1.42. And again we’ll round to a tenth. So we’d say this time adjusted clear is 135 + 1.42 = GR 136.4

What do you think- sound good?

Rage will get about +6 GR over most players so maybe he isn’t the best person to do this. We don’t want nerfs.

1 Like

Haha, Apache, you’ve got a problem like that, forget about me and take it up with 엔류. That guy is ridiculous!

Agreed, it’s better to evaluate power by looking at top clears, adjusted for paragon. This tends to reduce the “fishing” variable since top clears are more likely to have been accomplished with a great deal of fishing already.

However it’s not such a straightforward matter to do so. Adjusting based on tiers of damage gained isn’t accurate, mainly because of one gimmick: Conduit pylons.

Finding a Conduit means the player will need to deal less than 100% of the damage themselves, with the remaining % (since Conduit damage scales with tier) unaffected by paragon. The result is that the tier gain from paragon is more than expected from a simple calculation of the DPS increase.

In my experience, about 800-1k translates into 1 tier for 5-10k players (less for 5k- since additional mainstat has diminishing returns).

Hi Rage, I’ve done all of mine based on actual play experience with a mindset of 500 keys.

I downgraded certain clears… ie the Multishot and LoN Rapid Fire. Let’s avoid those 1 in 5000 golden rifts, and not put forward numbers based on the most ridiculous talent. The list I’m delivering is for players like me – I’m above average, I have the muscle memory down, my chars are geared properly, I’ll throw 500 keys a build I really like, but I’m not the very best and I know a freak-of-nature, or luckier player will defeat me.

This is not a list of ‘highest’ it’s a list of ‘realistic’. I might have misled you when I said max clear – my mistake. I think the community will get more out of beating these challenging clears and seeing the list updated. This also lets us collect data much faster on those uncertain clears. (I spent 25 keys and did GR135, the rift and guardian were good, I think I can squeeze +1GR with several hundred more keys).

Please continue to refer to clears made by high paragon players, and use your judgement – I’m sure you likely already know exactly what these guys encountered via a recording, stream, dm, or noise made on this forum or reddit. Imagine yourself with the task to clear with 5000 paragon and 500 keys – that’s what we should deliver.

I’ve suggested a massive game wide stat squish for D3, much like WoW has done numerous times in the past. Bring it back down into the early Torments for all I care so that the game has a challenge to it once again. Would people complain? Sure. But if it’s game wide and not class specific, then no one is at a disadvantage.

2 Likes