Power scaling: What is the standard?

[Nagog]

So all across Gitp and other forums, many players and DMs search for ways to not only bring their desired builds to life but also make them viable options for players. In Interaction and Exploration, the balancing is simple and easy to hit, however, in combat there are a lot of discrepancies and disputes, particularly when factoring in Great Weapon Master, Polearm Master, and Sharpshooter. Anything regarding combat that levels with these abilities is often seen as overpowered, while those who don't compare are often seen as not viable because their relative inefficiency. With that in mind, what should the proper scaling be? Should these feats be included in the scaling, or should they (and perhaps all damage feats) be ignored/excluded from the calculations?

[MoiMagnus]

If there is a standard to be taken, it is the AL rules, since that's the only context where the DM does not have the power to compensate for scaling problems.

Other than that, I would say that a good standard would be feats [because I don't know any table that actually care about scaling, and do not play with feats], with or without multi-classing [doesn't change that much], without any additional book, and with very few magic objects [And by that I mean that the Magic Weapon spell is actually relevant].

[TyGuy]

With that in mind, what should the proper scaling be? Should these feats be included in the scaling, or should they (and perhaps all damage feats) be ignored/excluded from the calculations?
I've come across two suggested changes, both tweaks to the -5 attack for +10 damage powered attacks

First is -proficiency bonus to attack for +double proficiency bonus damage.

The second is a weaker -proficiency bonus to attack for +proficiency bonus damage.

[Kane0]

In Interaction and Exploration, the balancing is simple and easy to hit

I disagree, but that's for another thread.



In combat there are a lot of discrepancies and disputes, particularly when factoring in Great Weapon Master, Polearm Master, and Sharpshooter. Anything regarding combat that levels with these abilities is often seen as overpowered, while those who don't compare are often seen as not viable because their relative inefficiency.

It's all relative, and often more art than science. Crunching numbers only gives you half the story, and there are always extra variables to be accounted for and sometimes cannot be across all situations (such as what enemies the DM throws at you in the case of -5/+10).
But those three are considered outliers, yes. They are all notably good for a couple reasons and other options available suffer by being compared to them. Not the least due to the games focus on combat over exploration and interaction while touting that all three supposedly being equal when making and choosing feats.



With that in mind, what should the proper scaling be? Should these feats be included in the scaling, or should they (and perhaps all damage feats) be ignored/excluded from the calculations?

Personally I think it's fine for combat feats to scale, it's just that some do and some don't. Ignoring the math would probably just make things worse though.

[Bjarkmundur]

I like Treanrmonks method, and I wish it was universal.

He measures damage per action relative to eldritch blast of the same level. I think he uses agonising blast.

So an action that deals 16 damage at level 3 would have a damage rating of 2, because it deals twice as much damage as an Eldritch Blast of the same level.

[Kane0]

I like Treanrmonks method, and I wish it was universal.

He measures damage per action relative to eldritch blast of the same level. I think he uses agonising blast.

So an action that deals 16 damage at level 3 would have a damage rating of 2, because it deals twice as much damage as an Eldritch Blast of the same level.

There are some pitfalls to that (EB is notably strong among cantrips, it uses one or more invocations, riders are not directly accounted for, AoE effects need to be accounted for as well, etc) but it's a reasonable starting point.

[PhoenixPhyre]

I wish people would
A) be more open and clear about their assumptions, especially the CR of the opponents. It makes a huge difference in builds that have different accuracies.
B) care less about high-level builds. Sure, you can do eleventy-billion damage. But until level 18 you're sub par at everything (to exaggerate for effect).
C) test against various styles of combat. Paladins can erase single targets that just stand there, but struggle against hordes or highly mobile creatures. Etc.
D) rate things with and without the feats. I've never had a party where someone has taken those combat feats. Just hasn't happened for whatever reason. They're not banned, just not chosen.That greatly changes the landscape IMO.

[Zuras]

I like Treanrmonks method, and I wish it was universal.

He measures damage per action relative to eldritch blast of the same level. I think he uses agonising blast.

So an action that deals 16 damage at level 3 would have a damage rating of 2, because it deals twice as much damage as an Eldritch Blast of the same level.


There are some pitfalls to that (EB is notably strong among cantrips, it uses one or more invocations, riders are not directly accounted for, AoE effects need to be accounted for as well, etc) but it's a reasonable starting point.

EB is strong, but I think that is the point. As a player and DM who enjoys optimization as a white-room game, any PC I build that can’t meet the EB threshold better compensate via buffing or the non-combat pillars, or it's back to the drawing board.

[Spiritchaser]

Given the power, versatility and adaptability of a kind of normal single class wizard, the bar for optimization is clearly pretty high. Take all the feats you want, MC however you wish, that wizard is going to contribute just as much to the party, most of the time (sure unusual party compositions and or encounters do exist, but...)

[Nagog]

Given the power, versatility and adaptability of a kind of normal single class wizard, the bar for optimization is clearly pretty high. Take all the feats you want, MC however you wish, that wizard is going to contribute just as much to the party, most of the time (sure unusual party compositions and or encounters do exist, but...)

Wizards are specialists though, so a maximized Wizard built specifically for dealing damage will probably deal more damage than the average melee fighter, but said wizard will not be able to innately tank and the like while dealing that damage, similarly to a Abjurer built on tanking won't be able to deal damage to the same extent while maintaining their defenses.

The debate on martial vs casting aside, this is for martial classes/combat and making adjustments or improvements to weapons or styles to fit the curve.

[stoutstien]

I use an standard SnB battle master with duelist as the bar for physical combatants. If they within a reasonable range of damage/damage mitigation thru are fine. If they are below by more than 20% then something is off and if they are above 50% same holds true.

[diplomancer]

I believe Treantmonk's baseline also includes damage from hex. So nine average damage on a hit on level 1, 12 at level 2 with agonizing blast and 16 CHA, and so on.

[PhoenixPhyre]

I believe Treantmonk's baseline also includes damage from hex. So nine average damage on a hit on level 1, 12 at level 2 with agonizing blast and 16 CHA, and so on.

Pinning the baseline to a resource-consuming specific build is...bad. Horribly so. Baselines should be baseline, not top 90%.

[clash]

I use greatsword wielding fighter with no feats or even subclass abilities as an upper baseline and rogue with rapier and basically the same conditions as a lower baseline. It gives a good rough estimate for where baseline damage should fall. Ie level 5:
Fighter with extra attack and 20 str = (2d6 + 5) * 2 ~ 24
Rogue with sneak attack and 20 dex = 1d8 + 3d6 + 5 ~ 20

[Waterdeep Merch]

I find theoretical math to be of limited value. I make some educated guesses using the book's CR and what I know of the players' capabilities (either the exact characters they've built when I'm privy to this information, or how they have performed in the past). Then I prepare some 'test' encounters during the first session, and usually again right after any significant power increase.

First, a variety of fights that, according to what I know, ought to be easy to handle, but each must be handled in a different manner. A brutish solo, a big mob of underlings, a mixed-tactics group that requires careful prioritization of targets, a terrain-heavy fight, and anything else I feel might become a normal sort of encounter for them later. If the players do poorly on one of these, I make a note of it.

Next, one that I *believe* to be right at the perfect place for them difficulty-wise. Not too hard, not too easy. This tells me how close my estimates already are to their real performance.

Finally, a dangerous fight that, by my estimates, should only barely be feasible for them to survive. The odds are intentionally stacked against them. The players aren't always told up front, but this fight usually has a fail-safe in place to prevent them from dying- a powerful NPC, an automatic timer, or some story reason why the villain was never going to kill them in the first place. Even with the worst luck, the players should never die or permanently lose anything significant in this fight. I'll hand wave it if necessary, though I prefer to let them think it's as dangerous as it looks. This fight is to determine the players' breaking point for tough boss fights later down the line.

After a good session of these, it becomes very easy to look at/create enemy stats and know if they're going to overwhelm my players.

[Spiritchaser]

Wizards are specialists though, so a maximized Wizard built specifically for dealing damage will probably deal more damage than the average melee fighter, but said wizard will not be able to innately tank and the like while dealing that damage, similarly to a Abjurer built on tanking won't be able to deal damage to the same extent while maintaining their defenses.

The debate on martial vs casting aside, this is for martial classes/combat and making adjustments or improvements to weapons or styles to fit the curve.

To my mind, dealing damage is, more often than not, the least useful thing a wizard could be doing.

Though of course they can do that as well.

[KorvinStarmast]

Pinning the baseline to a resource-consuming specific build is...bad. Horribly so. Baselines should be baseline, not top 90%.Treantmonk is a bit too 3.5ish for me, but some of the stuff in his guides is useful.

[GreyBlack]

Feats are interesting in that they're an optional rule. In my opinion, if you have that big a problem with feats being overpowered, just don't play with them.

Optional rules don't have to be balanced in my opinion.

[Nhorianscum]

So all across Gitp and other forums, many players and DMs search for ways to not only bring their desired builds to life but also make them viable options for players. In Interaction and Exploration, the balancing is simple and easy to hit, however, in combat there are a lot of discrepancies and disputes, particularly when factoring in Great Weapon Master, Polearm Master, and Sharpshooter. Anything regarding combat that levels with these abilities is often seen as overpowered, while those who don't compare are often seen as not viable because their relative inefficiency. With that in mind, what should the proper scaling be? Should these feats be included in the scaling, or should they (and perhaps all damage feats) be ignored/excluded from the calculations?

My baseline is pretty simple.

I just ask... "Why am I playing this over Cleric?"

Weeds out most charOP.

Similarly as a DM planning for a campain my scale is 1-6 clerics worth of difficulty.

[PhoenixPhyre]

Treantmonk is a bit too 3.5ish for me, but some of the stuff in his guides is useful.

But only from a particular viewpoint. That's an under-reported issue with guides--the underlying assumptions. In most cases, the assumptions swallow the method and dictate the result. People assume styles of play (including campaign styles), encounter styles, etc. If you relax or change those assumptions, the entire result changes. But those assumptions aren't clearly stated.

Someone who brought a GWM/PAM fighter to one of my games would be bored to tears. I don't do big solo monsters or extremely difficult encounters. Instead there are lots of horde encounters and generally "weaker" difficulty. Most things are better solved by talking to them; spells don't do as much as many people seem to think. Lateral thinking and problem solving are much more important, as is being connected to the world and using the information given.

Plus, there's this common assumption that you're high enough level for it to matter. Which, in my experience and in all the evidence, is just wrong.

Take GWM/PAM. Starting V Human, you've got one of those feats and a 16 in your primary stat. By level 4 you can either pick up the other feat or raise your score to 18. You can't max your score with both feats until 8 assuming Fighter. Using my very fast pace of leveling (when # of sessions since last level = min(current level, 5), you gain a new level), that's 25 sessions until level 8. At a weekly pace, that's half a year. And most don't meet every single week, so that's probably 9 months of actual time you've spent before your build comes online. And that's an easy build. Something like a Sorloc spends most of the game below the curve. Most games never get above level 9-11, so you probably spent all the game before your build really takes off. That's a waste, and not accounted for in the guides which focus on high-level play (that few ever really see).

[Kane0]

Bless for attack/save buffs. Agonizing Repelling Blast with optional Hex for attack efficiency. Divine Smite for burst damage. Spiritual Guardians for AoE/DoT. Even a nerfed Healing Spirit for healing (say once per round instead of per turn).

The staples that people typically recommend or fall back to. Those should be the upper limit of balance. Buff useless options like Weapon Master (and I don't mean the situational items with low opportunity cost like rituals on a Wizard, I mean 'why would you ever pick this when almost anything else is a better choice' like Find Traps), and nerf things that go too far (eg surpassing Fireball or Healing Spirit).

[ThePolarBear]

Using my very fast pace of leveling

"My" as in "in respect to other paces i use"?

That's... pretty slow compared to what the guidelines in the DMG give out. To bring into perspective: in the same 25 session timeframe a player would have a character reach level 11 assuming "2 to 3" means "3".

[PhoenixPhyre]

"My" as in "in respect to other paces i use"?

That's... pretty slow compared to what the guidelines in the DMG give out. To bring into perspective: in the same 25 session timeframe a player would have a character reach level 11 assuming "2 to 3" means "3".

2 to 3 means 2. So 6 sessions to level 4, 10 to level 5, then 5 per level after that. The DMG sits at about 4.5 adventuring days per level in T2, but in my experience it's rare to finish a full AD in a session. And for my school groups, the sessions are only about 1hr, 1.5 max. My regular games are 3hrs.

[Zalabim]

2 to 3 means 2. So 6 sessions to level 4, 10 to level 5, then 5 per level after that. The DMG sits at about 4.5 adventuring days per level in T2, but in my experience it's rare to finish a full AD in a session. And for my school groups, the sessions are only about 1hr, 1.5 max. My regular games are 3hrs.
Well, your sessions are probably on the shorter side, but as to the adventuring days, it's more like 2.15-2.33 each level in the tier, assuming there's a source of XP other than combat so that the XP earned matches up with the challenge faced.

[PhoenixPhyre]

Well, your sessions are probably on the shorter side, but as to the adventuring days, it's more like 2.15-2.33 each level in the tier, assuming there's a source of XP other than combat so that the XP earned matches up with the challenge faced.

Not really. The AD table gives adjusted XP, which should be 1.5x or 2x actual XP from all sources. And even that is pretty consistently 4-5 full AD per level in T2. And I doubt most groups are doing full ADs every session. Spend a session doing social stuff or exploring? Very little XP for you, but full for my system.

We ran PotA as a group using XP and milestone awards (so more that expected). After 9 months (probably 20 sessions due to interruptions) we were barely at 8. And that had lots of full ADs that took multiple sessions (clearing out the underground temples).

Edit: after running the numbers, I have to give a bit of a mea culpa here. Turns out you were more right, assuming 1 session = 1 full AD. Even applying a 1.5x deflator for adjusted vs actual XP, 25 AD should run to about level 10 at a pace of about 2.8 AD/level (less before level 3, and a bit more through level 10).

I think I'll adjust my pacing then to leveling up after N = min(current level, 3) meaningful sessions, down from N = min(current level, 5) meaningful sessions.

[ThePolarBear]

2 to 3 means 2. So 6 sessions to level 4. 10 to level 5, then 5 per level after that.

I think you might have misinterpreted my post. What i was comparing your progression to with the "2 to 3" part of the comment, since "2 to 3" is the number of sessions between level ups given in the DMG, not questioning the formula min(level,5), which was clear enough :D


The DMG sits at about 4.5 adventuring days per level in T2, but in my experience it's rare to finish a full AD in a session. And for my school groups, the sessions are only about 1hr, 1.5 max. My regular games are 3hrs.

This helps answering my question, thanks.
This further proves your other points (on which, to stay a bit on topic beside my curiosity, i agree with), however, since the DMG has different stated assumptions for session duration.
And given the various different game experiences and groups i've been in, there's no need to convince me that playstyle influences "how much is done in one session compared to what is an AD".

I don't really agree with your AD calculations, but that's for another topic! edit: missed your last edit,i should learn to read better.

[Zalabim]

For what it's worth, I think the current adventure league rate (and that's playing like adventure league) has characters allowed to level up after 4 hours of campaign play in tier 2, if they haven't leveled already. I don't have the exact wording in front of me, but it's available in the season 9 update. Giving the rate in hours played makes it easier to translate across different session lengths.

[PhoenixPhyre]

For what it's worth, I think the current adventure league rate (and that's playing like adventure league) has characters allowed to level up after 4 hours of campaign play in tier 2, if they haven't leveled already. I don't have the exact wording in front of me, but it's available in the season 9 update. Giving the rate in hours played makes it easier to translate across different session lengths.

AL uses checkpoints (1 per hour in previous seasons, 1-2 per objective in season 9)--

T1: 4 checkpoints (4 hours)
T2+: 8 checkpoints (8 hours, roughly 2 long sessions).

But time spent "off task" isn't supposed to count.

For my school games that's really slow (a level every 4 or 8 sessions) because of the short games. For my "regular" games it's about the same as my revised schedule, but doesn't allow for time spent "off task", which have been some of my best sessions.

I had a great session where the players spent the whole time interacting with a goblin tribe. Not fighting, just helping them gather food, playing with the kids, talking to the adults, etc. And it was wonderful and everyone had lots of fun. It wasn't even "plot critical"--the goblins would have helped them with a much shorter scene. That's one reason (off topic) why I don't like XP and prefer more fiat methods--the one game with XP I played ended up pushing for maximum fight/story progression rather than taking time to smell the flowers.

[diplomancer]

Pinning the baseline to a resource-consuming specific build is...bad. Horribly so. Baselines should be baseline, not top 90%.

Starting from level 5 Hex is pretty much free to a Warlock, though, as long as they can maintain concentration (and they normally don't have better options to spend their concentration on either).

But what Treantmonk means is: this is the easiest way for a beginner to build a character that does good damage. If you are measuring damage and that is one of your goals for your character, anything that falls short of that should be re-considered.

[PhoenixPhyre]

Starting from level 5 Hex is pretty much free to a Warlock, though, as long as they can maintain concentration (and they normally don't have better options to spend their concentration on either).

But what Treantmonk means is: this is the easiest way for a beginner to build a character that does good damage. If you are measuring damage and that is one of your goals for your character, anything that falls short of that should be re-considered.

Wait...don't have better options for concentration? What about the entire rest of the game (including all those non-combat invocations, etc)? Spending concentration like that is a huge cost. Especially since it can be disrupted in turn 1 of fight 1 and then you've spent a good chunk of your resources for nothing.

Setting "optimized" as your baseline, a baseline that most other builds struggle to reach without spending serious investments (I mean standard rogues can't even get there) is a bad place for a system. It's advice like that that "forces" people into very niche builds and makes them think that only damage matters (because if you build to that baseline you have to give up almost everything else). And that's horrible advice that only is meaningful if all you do are single-target big fights once per day. Which is an example of a pathological case that gets used as a standard by many "optimization" suggestions.

Baselines should be baseline, not upper 10%.


Consider a bunch of "normal" builds without feats, all at level 5, all with the same accuracy (so it can be ignored). Assuming a +4 main stat. Also assuming 6 combats, 2 short rests, and 3 rounds average combat.

Hexlock: 2d10+8+2d6 ~ 25 damage per turn. #but has other spells for other purposes*

Barbarian, raging (greataxe): 2d12 + 4 + 8 ~ 25 damage per turn #but can only do this in 6 rounds of combat (2 fights) per day
Barbarian, not raging (greataxe: 2d12 + 8 ~ 21 damage per turn
Barbarian, averaged: 22.33 DPT

Fighter, Greatsword, AS: 4x(2d6+ 4) ~ 44 DPT, #but can only do this 3x/day
Fighter, Greatsword, no AS: 2x(2d6+4) ~ 22 DPT
Fighter, averaged: 25.7 DPT

Rogue, Ranged, L. Crossbow (SA every turn): 1d8 + 3d6 + 4 ~ 19 DPT
Rogue, melee (2x shortswords, SA every turn): 2d6 + 3d6 + 4 ~ 21.5 DPT

Note that only the fighter can reach that baseline, and only by action surging optimally. The rogue falls far behind, and there aren't even feats that can pick them up to that level.


I repeat--using a top-tier, resource-consuming build as your baseline is a bad thing for the game. It suggests that only a few possible builds are even remotely viable, let alone optimal.

The balance baseline for damage (which is the least interesting and least meaningful balance point, but that's a separate matter) should be that you can do resourceless damage on par with a rogue who gets sneak attack most turns. It's nice, simple, and easy to calculate and most basic builds can reach that baseline. Which is rather the definition of a baseline.

[PhoenixPhyre]

Another possible damage baseline uses defensive CR.

Defensive CR == "can survive for 3 rounds on average against a party of 4 PCs of level = dCR".

Assume a standard party composition of Fighter (defensive), rogue (striker), wizard (striker/controller), cleric (support). Assume (for ball-parking the numbers) that the strikers deal about 2x the damage of a non-striker. This means that each striker should output approximately 1/6 [2D + 2*(D/2) = H/3] of the monster's effective (accounting for resistances, regeneration, etc) health per turn.

That looks like (using the numbers from all the mainline published monsters):
Level 1: Average eHP = 30, so baseline damage = 5. Which fits nicely with 1d6 + 1d6 (SA)+ 2, after accounting for accuracy.
Level 5: <eHP> = 114 ==> baseline DPT = 19. Which fits pretty well with the rogue numbers in my previous post.
Level 11: <eHP> = 243 ==> baseline DPT = 40.5
Level 17: <eHP> = 330 ==> baseline DPT = 55
Level 20: <eHP> = 515 ==> baseline DPT = 85.8

Now of course at higher levels, the "non-strikers" will pick up more of the slack, so these are ceilings not floors. But at least at lower levels (where the numbers are easy to check), this matches up pretty closely to the "resourceless rogue" baseline.

Setting the baseline too high means that
1) you lose out on a lot of fun builds
2) you make encounters way easier than they should be...but inconsistently so
3) you encourage an arms race with DMs...which makes worldbuilding, pacing, and general play much less interesting for anyone who is not participating in this arms race.

[diplomancer]

Your napkin math ignores a lot though. A champion with 2-weapon fighting style would have better damage than the warlock without expending any resources. I don't know about a recklessly attacking barbarian, but I imagine it would get pretty close or surpass it even without raging. Those are two resource-free builds that beat Eldritch blast + Hex. Monks are at 24.5 without flurry of blows or any subclass features. Rogues also have ways in their subclass of increasing their base damage, which you also ignore. So why use as the base-line the least powerful of options for damage when you are trying to figure out the standard for damage power scaling(the rogue with no subclass that improves damage at specifically level 5- at level 7 he's already up there with the hex warlock)?

Also, this is somewhat parallel to what I just said, but I find the "average number of rounds per combat" to be 3 very low. I can't remember many combats I've been in that end in 1 or 2 rounds, but I remember plenty that went on for longer than 3. 3, in my experience, is about the minimum (granted, a somewhat common minimum, but not near enough to make it the average).

[PhoenixPhyre]

More rounds makes the fighter worse, not better. And no, a featless champion is not beating the hexlock without action surge. The whole point is to show that if you set the baseline at strong, you're mandating those "broken" feats to even compete. And that's horrible for the game.

Edit: I did the math more exhaustively earlier.

Making various assumptions about working days, all of the martials have resourceless damage roughly equal to the rogue (between 100% and 105%) and spike up to 130% if they burn their resources for damage. This also matches very (suspiciously) well with the numbers based on the real monster math. Which tells me that this is the original design point. Which is the only meaningful baseline. Setting the baseline above this point forces power creep and makes any new published options (that were designed using the lower baseline) to be outright traps or uselessly "underpowered". Which artificially limits the available classes in a harsh way.

[diplomancer]

More rounds makes the fighter worse, not better. And no, a featless champion is not beating the hexlock without action surge. The whole point is to show that if you set the baseline at strong, you're mandating those "broken" feats to even compete. And that's horrible for the game.

I actually meant 2-handed fighting style, sorry. So, without action surge, we are talking about 23.67 points of damage per round, without considering the improved crit range, which I would guess adds around 1 extra point of damage to the average. And at 6th level the champion fighter is at Str 20, easily surpassing the still 18 Cha Hex Warlock, without spending any resources or having any feat support.

If you don't like the hex-warlock as a baseline, (which, from what I understand, means that some builds will be below the baseline, and considered to have bad damage, and others will be above the baseline, and so considered to have good damage), how about a 2-handed fighting style champion? This is a resourceless build, similar to the hex-warlock (better at some levels, worse at others, probably better at all levels once you factor in things like attacks of opportunity).

[stoutstien]

I actually meant 2-handed fighting style, sorry. So, without action surge, we are talking about 23.67 points of damage per round, without considering the improved crit range, which I would guess adds around 1 extra point of damage to the average. And at 6th level the champion fighter is at Str 20, easily surpassing the still 18 Cha Hex Warlock, without spending any resources or having any feat support.

If you don't like the hex-warlock as a baseline, (which, from what I understand, means that some builds will be below the baseline, and considered to have bad damage, and others will be above the baseline, and so considered to have good damage), how about a 2-handed fighting style champion? This is a resourceless build, similar to the hex-warlock (better at some levels, worse at others, probably better at all levels once you factor in things like attacks of opportunity).
Same character with 1hd and duelist has similar damage but is alot quicker with math because of how great sword works with the fighting style and crits.

[PhoenixPhyre]

I actually meant 2-handed fighting style, sorry. So, without action surge, we are talking about 23.67 points of damage per round, without considering the improved crit range, which I would guess adds around 1 extra point of damage to the average. And at 6th level the champion fighter is at Str 20, easily surpassing the still 18 Cha Hex Warlock, without spending any resources or having any feat support.

If you don't like the hex-warlock as a baseline, (which, from what I understand, means that some builds will be below the baseline, and considered to have bad damage, and others will be above the baseline, and so considered to have good damage), how about a 2-handed fighting style champion? This is a resourceless build, similar to the hex-warlock (better at some levels, worse at others, probably better at all levels once you factor in things like attacks of opportunity).

The feat-less, no-resource GWF-style champion is close to the TWF rogue across the wide spectrum of levels (not cherry-picking one where the fighter has an extra ASI). So those are really very similar.

Also note that the melee fighter is...melee, while the hexlock can do so from across the battlefield and can switch targets much easier (since he has very little bonus-action competition). This brings up another reason why sheer DPT is a bad balance point--fights are not infinite-health training dummies in white rooms. Real fights have
* terrain
* lots of mooks (by design solo monsters are rare)
* movement

all of which mean that choosing the most flexible, most adaptable one (ie hexlock) as your baseline is bad design.

[Zalabim]

Consider a bunch of "normal" builds without feats, all at level 5, all with the same accuracy (so it can be ignored).
Broadly, I don't disagree. I don't assume the use of hex for a lot of reasons. However, this is gonna screw up your math time and again. They all have the same attack bonus, but you can't ignore accuracy as a factor. The barbarian has the option to use reckless attacks. If the fighter were a champion, they'd have more bonus critical damage. The rogue can try to hide before shooting, or when using TWF does the majority of their potential damage if even one attack hits. None of these can be accurately represented by simple average damage if everything hits.

[diplomancer]

The feat-less, no-resource GWF-style champion is close to the TWF rogue across the wide spectrum of levels (not cherry-picking one where the fighter has an extra ASI). So those are really very similar.

Also note that the melee fighter is...melee, while the hexlock can do so from across the battlefield and can switch targets much easier (since he has very little bonus-action competition). This brings up another reason why sheer DPT is a bad balance point--fights are not infinite-health training dummies in white rooms. Real fights have
* terrain
* lots of mooks (by design solo monsters are rare)
* movement

all of which mean that choosing the most flexible, most adaptable one (ie hexlock) as your baseline is bad design.

They also have cover which makes it worse for ranged characters. All things considered, whether it is better to be a ranged character or a melee character will depend on a lot of different factors (party composition, battle terrain, CaS or CaW, heck, even player personality!) Assuming that 2-handed style featless Champion Fighters and Hex-Warlocks are very similar in damage for most levels, I don't think that you can claim that the fact that being ranged gives the warlock such a great advantage that, while it is OK to choose the Fighter as a baseline, it is bad design to choose the hexlock.

For myself, I am perfectly willing to suggest the Champion as that baseline.

[PhoenixPhyre]

Broadly, I don't disagree. I don't assume the use of hex for a lot of reasons. However, this is gonna screw up your math time and again. They all have the same attack bonus, but you can't ignore accuracy as a factor. The barbarian has the option to use reckless attacks. If the fighter were a champion, they'd have more bonus critical damage. The rogue can try to hide before shooting, or when using TWF does the majority of their potential damage if even one attack hits. None of these can be accurately represented by simple average damage if everything hits.

Hence napkin math. I've done the bigger math, and the general trends still hold.

Edit @diplomancer--

THe problem is that you're using the majority of the kit for the fighter in that comparison but very little of the kit of the warlock (who still has other spell slots, invocations, etc). So choosing the warlock as your baseline means that it's everything the fighter can do just to keep up (without feats) with a fraction of the warlock's capacity. Thus making the need for the "big combat feats" much bigger. Which is most of the problem.

[diplomancer]

Hence napkin math. I've done the bigger math, and the general trends still hold.

Edit @diplomancer--

THe problem is that you're using the majority of the kit for the fighter in that comparison but very little of the kit of the warlock (who still has other spell slots, invocations, etc). So choosing the warlock as your baseline means that it's everything the fighter can do just to keep up (without feats) with a fraction of the warlock's capacity. Thus making the need for the "big combat feats" much bigger. Which is most of the problem.

Most of the kit of the fighter? Not considering action surge, not considering subclasses that give you more damage when you need it, not considering how much more easier it is to get advantage when you are in melee than when you are at range, etc?

Remember, I am just talking about a damage baseline; the other things that a hex-warlock can do, and which don't really improve his damage, do not enter in that calculation. If they are both very similar (as they seem to be), and if you are ok with the featless 2H Champion as a baseline, I am ok with that as well (consider, if the Champion is NOT getting the "big combat feats", he is getting other feats instead with his extra ASI's... which will probably increase his out of combat utility, balancing it somewhat with the rest of the warlock kit)