PDA

View Full Version : Measuring Class Power



Jack Zander
2007-11-15, 11:47 PM
Since the other thread got unfairly locked for those of us who were not "meta-flaming" as the mods said, I think Jack Smith had some good points there. This thread is to continue the discussion. (Without the flames this time please.)


Individual tasks are also not a good measure - D&D is not (normally) a solo game.

How about this:
You set up a long series of challenges (of varying types - several combat encounters, a few maze encounters, some trap encounters, and so forth) - all level appropriate. The class to be measured is a fifth member to the Iconic Four (Fighter, Wizard, Cleric, Rogue).

The Iconic Four are standard, DMG NPC's for those classes. Exception: Wizard gets to have all spells he could potentially cast in his spellbook from the PHB, but otherwise he gets no equipment beyond spell components, focuses, and spell components pouches (and no, you can't commandeer them) that are only good for him casting his spells. Both Cleric and Wizard prepare the spells they're told to prepare, and cast them when they're told to cast them (and in the way they are told to cast them). The Wizard favors blasting, and will prepare direct-damage Evocations in any spell slot in which he's not told to prepare something specific. Likewise, the Cleric prepares his most effective base healing spell for that level in any spell slot that isn't specified. The Cleric has the Healing domain, and one other of the choice of the class to be tested, although the second domain must be one listed in the Player's Handbook.

The class to be measured is built with the three core rulebooks, and one additional Wizards of the Coast 3.5 of source of choice (classes that do not come out of the Core rulebooks are required to take the source they come from as their one non-core source, for instance, if the class to be measured is the Psion, the one non-core source must be the Expanded Psionics Handbook). The class to be measured must be pure-classed (so if we're measuring a Psion at 15th, it's a Psion-15, not a Psion-10/Metamind-5 or some such; no PrC's, no multiclassing, and so on). Standard Wealth-By-Level, one-use items cost five times normal (other than the Tomes, Manuals, and similar items which are used once and affect the user forever) but are refreshed between challenges.

The challenges cannot be simply bypassed, but can be taken in any order; however, if the challenge is to get past something, then teleportation or stealth works for that challenge unless it is removed from the equation by some means as declared in the challenge (e.g., one of the challenges might be to get past the guard at the gate to a castle. This can be done with Teleportation if the testing level is high enough, stealth, killing the guard, Charming the guard and beating his Charisma check, Dominating the guard, or whatever, unless something preventing such a strategy is written into the challenge - so you can't get by the guard with Teleportation if the area is under a Forbiddance effect (unless you have the SR to beat the Forbiddance), can't get by with stealth if the guard has True Seeing in an area with no cover or concealment (unless you can manufacture Cover or Concealment fit to not be ignored by True Sight), can't Charm the guard if he's got Mind Blank up (unless you can dispel the Mind Blank), and so on). However, killing or destroying a target can be the nature of the challenge, and in such a case, the target must be killed or destroyed for the challenge to succeed.

The class to be tested can declare "next day" between any two challenges at any time, or during a challenge if the nature of a challenge would permit it- however, the strength of the tested class is measured by the time it takes to cover all challenges. Death takes 1 day to undo, and the challenge is reset when this happens. A shorter time makes for a better score.

Testing is to be done at 1st, 5th, 10th, 15th, and 20th (with a different set of challenges for each level, all challenges appropriate to the level to be tested - so no making "Killing the Terrasque in a 100x100 no-cover arena, with a 30 foot ceiling" a challenge for level 1, for instance).

All challenges must be defeatable by the Iconic Four acting without the member to be measured, and without resorting to "cheese" - "cheese", in this instance, being defined by a tactic, spell, or method called "cheese" by three more forum members than contradict that claim. Likewise, the class to be measured cannot use "cheese" by the same definition.

Assuming, for the moment, that the challenges aren't too terribly slanted to a particular class, and are reasonable ones that might come up in a game, and that the DM is reasonable, would this be a fair test for the relative "strength" of classes?

I think this would be a fantastic way to judge classes, except the challenges should be slightly more difficult for the iconic four's CR. They do have a highly optimized 5th wheel with them after all.

Skjaldbakka
2007-11-15, 11:51 PM
Re-starting a Previously Locked Thread Without Permission
Certain threads will be locked by Mods for a variety of reasons, including excessive flaming, inappropriate subject matter, or simple redundancy. If a poster decides to start a new thread on the exact same discussion topic without first gaining permission from the Mods, the thread will be deleted or locked and they will be issued an Infraction—even if the new thread does not, in and of itself, violate any other rule! Which means that if a Mod decides to lock a heated debate over a subject, starting a new thread with a calmer, more well-reasoned point of view is STILL breaking the rules. Re-posting is only allowed if a Moderator gives express permission beforehand.

You really ought to read the forum rules before doing something like this

Jack Zander
2007-11-15, 11:59 PM
While the title is the same, it's most definitely not the same topic. The entire issue has been changed from the original concept to what Jack Smith stated.

Crow
2007-11-16, 12:17 AM
I proposed an idea like this in the Debating Initiative thread. Sort of a D&D Ironman. While arguments will always persist, it can be a useful tool to gauge the capabilities of each class.

Basically, if you cannot bypass challenges, but can decide what order in which to face them, you can assign point values to the challenges, and come up with a score at the end.

brian c
2007-11-16, 12:17 AM
While the title is the same, it's most definitely not the same topic. The entire issue has been changed from the original concept to what Jack Smith stated.

Regardless, you probably shouldn't give it the same title, because that will attract the attention of the same people who were behaving inappropriately before

Jack_Simth
2007-11-16, 12:42 AM
Since the other thread got unfairly locked for those of us who were not "meta-flaming" as the mods said, I think Jack Smith had some good points there. This thread is to continue the discussion. (Without the flames this time please.)



I think this would be a fantastic way to judge classes, except the challenges should be slightly more difficult for the iconic four's CR. They do have a highly optimized 5th wheel with them after all.
Two little issues:
1) Time & Effort. This is not a simple, easy test. We're looking at about 10-20 challenges per level, which need to be actually played out, for each class to be tested. It's comprehensive, and it's probably fair to the nature of the game, but it's a lot of effort.
2) Who is this "Jack Smith" person of which you speak? :smallwink: It's okay - it's deliberate on my part - only about 50% of people notice, if that - and it's always interesting to see who does and does not notice - you're in the "does not notice" category, I'm afraid.

As for the CR of the challenges:
Doesn't specifically matter. It's a time trial. If you're running at 5th and the challenge is "Stop the pair of ogres that have just gone berserk in town from killing people" (an Ogre is CR 3 - two of them is an official CR 5, but I've also been told by my players that an ogre has a CR of "ow, my face" so take that with a grain of salt) then you can measure by rounds to see if adding a Barbarian to the mix is better than adding a Wizard to the mix.

Oh - and be careful with this; technically, you need to PM a Mod before re-opening such threads.


I proposed an idea like this in the Debating Initiative thread. Sort of a D&D Ironman. While arguments will always persist, it can be a useful tool to gauge the capabilities of each class.

Downside of such a proposal: It is unlikely to ever be carried out by those not getting paid to do so.


Basically, if you cannot bypass challenges, but can decide what order in which to face them, you can assign point values to the challenges, and come up with a score at the end.
...

If you assign a "simple" point value to defeating the challenge on a pass/fail system (you didn't specify either simple or pass/fail, but then, you didn't specify), and you have to defeat all challenges, and all challenges can be defeated by the Iconic Four that you're bringing along, everyone should get exactly the same score, regardless of class.

The reason for a time-trial is so that you can measure the effect of each class getting added to the mix - If at 5th level it takes the Iconic Four five rounds to take down those two rampaging Ogres, but two when you've got an extra Wizard, and four when you add a Barbarian instead, then you can say that the Wizard is more useful when taking down a pair of rampaging ogres at 5th. If you put in a lot of different such (e.g., at 5th, we have a pair of Paladin-3 guards at a locked gate to bypass as one challenge, two rampaging ogres that need slaying as another, a self-resetting Cone of Cold trap in a hallway you need to get everyone through, a "Princess" (elven commoner-1; 1 hp) that needs to be rescued from the two leveled dwarves holding her captive (a Wizard-3 and Cleric-3; they stay right next to the princess, and you're not permitted to hurt or incapacitate her) before they can get to the sacrificial ceremony this evening at midnight, a 400 foot wall you need to get everyone over, a slightly paranoid Wizard-7 who holds the artifact the King needs to cure the Plauge that will wipe out the country within three hours of you starting the challenge - and he's not willing to part with it, not at all - and so on) then whoever is faster at it is fairly clearly better overall. Unless, you know, someone says that the challenges are slanted....

tainsouvra
2007-11-16, 01:53 PM
Assuming, for the moment, that the challenges aren't too terribly slanted to a particular class, and are reasonable ones that might come up in a game, and that the DM is reasonable, would this be a fair test for the relative "strength" of classes? Unfortunately, no, it wouldn't be a test of the relative strength of the classes in the first place. You would instead be testing the relative contribution of each tested class as a fifth wheel to an unusually-designed four-man party. While that could be an interesting exercise, it doesn't test what it is designed to test.

Istari
2007-11-16, 02:10 PM
I think that a point based system would work instead of time but you wan the lowest score to win.:smallsmile:

Thinker
2007-11-16, 02:18 PM
Running as a 5th wheel wouldn't work and is somewhat lame. Obviously if the party already has a fighter, rogue, cleric, and wizard anyone who normally fills the role of heavy melee, skill monkey, divine caster, or arcanist is automatically hosed.

I still think the best way to do it is to have a series of challenges, several that each archetype would generally excel at. You would then tally up points for each person based on the encounters they could pass and which ones they could not.

Jack Zander
2007-11-16, 05:29 PM
No mod has said anything and it's been almost 24 hours, but if it makes you guys feel better...

Jack_Simth
2007-11-16, 05:47 PM
Running as a 5th wheel wouldn't work and is somewhat lame. Obviously if the party already has a fighter, rogue, cleric, and wizard anyone who normally fills the role of heavy melee, skill monkey, divine caster, or arcanist is automatically hosed.

Incidentally, that happens to cover very nearly every 3.5 WotC class out there, as very nearly every 3.5 WotC class out there fills one or more of those rolls to some degree or another - which pretty much means everyone's on equal footing, no?

It's far from as bad as you think, though; the DMG standard NPC's are not so grand.

The Wizard's save DC for a 1st level spell at 1st will be 13 (+2 from 15 Int, +1 spell level, +10 base). At 5th, for a 3rd level spell, will be 16 (+3 from 16 Int, +3 spell level, +10 base). At 10th, for a 5th level spell, will be 18 (+3 from 17 Int, +5 spell level, +10 base). At 15th, for an 8th level spell, will be 22 (+4 from 18 Int, +8 from spell level, +10 base). At 20th, for a 9th level spell, will be 24 (+5 from 20 Int, +9 from spell level, +10 base). An optimized Arcanist will outpace him - badly. Plus, as he has no equipment beyond his spellbook and spell components, he's got AC 12, 72 HP. One good Meteor Swarm or Power Word Kill and he's a crispy critter.

The Cleric is in slightly better shape with the save-or spells, as the DMG NPC Cleric in this case is getting Wis boosters - a Peripat of Wisdom +2 at 10th, +4 at 14th, +6 at 17th - which means at 20th, he's got a Wisdom score of all of 26 - vs. the 32 or so an optimized Cleric or Druid will have. Plus only 133 HP and AC 26.

The Fighter, at 20th, has AC 34, 175 HP, and is attacking at +32/+27/+22/+17 (before Power Attack). How high does it get for someone who's actually trying?

The Rogue, at 20th, only has 230 skill points, AC 26, and is only attacking at +26 ranged (his best). Oh yeah - and with 92 HP, Power Word Kill will still take him out.

A reasonably optimized character will simply outshine his counterpart in the Iconics - by a lot.

If you'd prefer, though, we could modify things such that you get three of the Iconic Four of your choice.



I still think the best way to do it is to have a series of challenges, several that each archetype would generally excel at. You would then tally up points for each person based on the encounters they could pass and which ones they could not.
If you send the test character up against challenges on his own, and run a simple pass/fail, you've got three basic problems with the test:
1) D&D is not normally played solo. Such a test is not of the nature you normally see in the game.
2) Any class or build that is focused on other-buffs is simply hosed. The tenth level Bard that gives everyone else in the party +2 To Hit/Damage for the duration of the battle can actually be indirectly responsible for a third of the damage the party deals every round (no joke - depends a bit on the opponents, though; +10% to hit and +2 damage can do that ).
3) A simple "can beat" or "can not beat" leans very heavily in the direction of prepared casters for a couple of reasons.
A) They can prepare for EXACTLY that one challenge. This is not something you generally see in a game - the Wizard isn't sure what's coming, exactly, but may have some idea.
B) They can spell dump to their heart's content (Wizard-20 spending 5 9th level spell slots on a single encounter, for instance - and that's a long encounter at that level) as they are unlikely to run out of their top-tier spells before they beat the one challenge, even going all-out. Each challenge is in isolation, so spending all of your top two levels of spells on this challenge has no impact on the next, nor on your final score. Spell Dumping generally has consequences in an actual game, when the DM keeps to the DMG recommended 3-5 encounters in a day. This also removes a lot of the advantage of the spontaneous casters who get more base spells per day.

Kaelik
2007-11-16, 06:29 PM
A) They can prepare for EXACTLY that one challenge. This is not something you generally see in a game - the Wizard isn't sure what's coming, exactly, but may have some idea.
B) They can spell dump to their heart's content (Wizard-20 spending 5 9th level spell slots on a single encounter, for instance - and that's a long encounter at that level) as they are unlikely to run out of their top-tier spells before they beat the one challenge, even going all-out. Each challenge is in isolation, so spending all of your top two levels of spells on this challenge has no impact on the next, nor on your final score. Spell Dumping generally has consequences in an actual game, when the DM keeps to the DMG recommended 3-5 encounters in a day. This also removes a lot of the advantage of the spontaneous casters who get more base spells per day.

Except that if you look at the other thread (which is of course his thread that he's trying to sell) then you'll see that actually it completely hoses all casters (well everyone at higher levels, but casters more at low levels) because he thinks they should run through hundreds of challenges without any rest.

Jack_Simth
2007-11-16, 06:55 PM
Except that if you look at the other thread (which is of course his thread that he's trying to sell) then you'll see that actually it completely hoses all casters (well everyone at higher levels, but casters more at low levels) because he thinks they should run through hundreds of challenges without any rest.
Ah, the opposite end of the spectrum in "this almost never happens in a game" - okay. Same issue, different direction.

Kaelik
2007-11-16, 07:23 PM
Ah, the opposite end of the spectrum in "this almost never happens in a game" - okay. Same issue, different direction.

Yeah, I just wanted you to know where he was going with it.

I basically said in the other thread that he was wasting his time. He has 30 straight "challenges" with no rest for a single level 1 character. He then further gimps them by saying they can never take 20.

There is absolutely no reason for 30 challenges when every character will be dead before they hit the fifth one.

Moff Chumley
2007-11-16, 07:25 PM
My brief thoughts about spell balance and class fairness: Throw a book at the wizard. It typically gets my point across. :smallamused: