On the Same Game Test's intended functionality.

[Roy]

So I was bored, and reading some forums. And I came across a whiny bitchfest. In the interests of not making this a 'threads that fail' thread no links will be provided, and no names given.

Here is a sample of what the opposition looks like:

The SGT has no bearing on the build of 3.5 dnd because it does not adhere to the mechanics the system places.

If they want to evaluate each class individually against a creature whose CR is the same as the level as the player, that is fine. But it goes against every single balance put in the books.

Actually, a creature of CR 5 is a moderate challenge to a party of 4 (fighter, mage, cleric, rogue) where each member is level 5.

If as you say, same as everyone else, then the people doing SGT are not playing the same game as everyone else. This thread proves that by the overwhelming rebuttals to the results on the SGT. Do over.

Apparently, the SGT does not follow CR rules, because you're supposed to SGT vs a party of 4, or against something 4 levels lower. This of course is completely wrong, and completely defeats the point of a SGT. However, a SGT performed by the intended, by the book standards does exactly what it is intended to do - show you who is performing above, below, and at par with the intended functionality being that you beat yourself 50% of the time and are beat by yourself 50% of the time.

[TarkisFlux]

Ah, the old "everything is fine if you just play with the standard party and the way the designers intended" line. Blah.

I got tired of writing this shit out a while back, so I just wrote the justification once with quotes and everything and saved it on the wiki. It's written with an eye towards some wiki specific ideas, but the first part of it is pretty independent.

It doesn't address why this is a better test than the 4 man test they seem to be praising, but that's hardly a difficult thing to answer.

[Username17]

The 4 man "curb stomp a hobo" test isn't very good because it doesn't tell you anything. Firstly, since you're curb stomping a hobo, it's hard to say if you're batting over or under - since the expectation is that you're going to beat them like you were rolling over Poland. And secondly, it's hard to get any data. Even if the team is batting on par, you still don't know without a lot more analysis whether your aggregate success level is being achieved because one character is too strong and another is too weak.

-Username17

[Maxus]

Here's the rebuttal:

Not all games have the straightjacket of wizard, cleric, fighter, and rogue.

A party of four can shore up its weaknesses, sure. The wizard and cleric alone can do a lot. Enlarge Person is a pretty bitchin' spell to put on a Fighter and gives him some extra punch.

But the fact is, the Fighter by himself wouldn't be able to cut it outside of a narrow cookie-cutter build to hype up one trick. By himself, he is not a threat to the opposition, and in a party he's sort of like attack dog. He's a tool the other players put some gas into and send at the enemies.

Also, the SGT does adhere to the mechanics the system places. Even better than the designers would like to admit. It admits that these do not exist in a vacuum. SRD Fighters and Barbarians live in a world where monsters are far better at fighting than they are, and don't have the chops to survive a constant barrage of save-or-lose spells thrown at them by the spellcasting monsters, leaving them no way to reliably contribute and do something to make the guy playing a wizard say "I'd like to play that sometime." There are second-level spells which can render feat chains obsolete. The melee classes expected to be 'realistic' in a game where you have sixty-foot high dragons and clerics who can fight even better than the Fighter starting at level 7 (Divine Power!), and was shooting honest-to-goodness lasers at people two levels ago (Searing Light). Not for nothing did a player in my first D&D group say he didn't want to be put on Melee Duty.

The WotC designers wrote much of the game like they weren't looking at what else was in the game. Oh, yes, let's make the Rogue bend over backwards trying to be able to hide and still sneak later on, but give the level 3 wizard the spells silence and invisibility, which make putting skill points into Hide and Move Silently irrelevant. Especially since the Rogue also gets UMD and can have those spells put into wands.

They didn't know what they were writing. Skip Smokes Crack, after all. And it's really disheartening to see people taking their word as the Holy Truth when they have been proven wrong on many an issue.

[Lokathor]

Ah, the old "everything is fine if you just play with the standard party and the way the designers intended" line. Blah.

I got tired of writing this shit out a while back, so I just wrote the justification once with quotes and everything and saved it on the wiki. It's written with an eye towards some wiki specific ideas, but the first part of it is pretty independent.

It doesn't address why this is a better test than the 4 man test they seem to be praising, but that's hardly a difficult thing to answer.

Since the only difference between an NPC human with nothing but fighter levels and a PC human with nothing but fighter levels is the person playing them, we will go ahead and assume that the statement holds true for PCs as well.

Actually, the PC gets like 4 times the wealth by level, which means that the PC version should win slightly more often than the NPC version, because of better gear.

[K]

Here's the thing: the tyranny of math has no place in marketing.

People don't want to be judged objectively, and they really don't want their work to be judged objectively. If you take someone's work and find a way to evaluate it objectively, they will fight it.

Skip didn't become The Sage because he's smart, or knows the rules, or is a good designer. He took the job so he could promote himself.

And people who have any stake in something are going market their work as grand, even if it sucks. This is why the SGT offends people.

I'll bet dollar to donuts that people who are offended by the SGT think of themselves as designers, or are sock puppets for designers, or are people who feel the need to defend their hero-worship of designers. These people must trash anything that might put the work of established designers in question because the mythology of greatness is quite fragile. I mean, people make careers on being stars, but that is seldom a reflection on the quality of their work.

But people who care about game design like the SGT because its a simple metric to make sure you are doing quality design. That's a simple truth, and everything else is posturing and theater.

[Roy]

Ah, the old "everything is fine if you just play with the standard party and the way the designers intended" line. Blah.

I got tired of writing this shit out a while back, so I just wrote the justification once with quotes and everything and saved it on the wiki. It's written with an eye towards some wiki specific ideas, but the first part of it is pretty independent.

It doesn't address why this is a better test than the 4 man test they seem to be praising, but that's hardly a difficult thing to answer.

That wiki was already brought up, but the opposition in that case is a Ferrous Cranius.

Here's the thing: the tyranny of math has no place in marketing.

People don't want to be judged objectively, and they really don't want their work to be judged objectively. If you take someone's work and find a way to evaluate it objectively, they will fight it.

Skip didn't become The Sage because he's smart, or knows the rules, or is a good designer. He took the job so he could promote himself.

And people who have any stake in something are going market their work as grand, even if it sucks. This is why the SGT offends people.

I'll bet dollar to donuts that people who are offended by the SGT think of themselves as designers, or are sock puppets for designers, or are people who feel the need to defend their hero-worship of designers. These people must trash anything that might put the work of established designers in question because the mythology of greatness is quite fragile. I mean, people make careers on being stars, but that is seldom a reflection on the quality of their work.

But people who care about game design like the SGT because its a simple metric to make sure you are doing quality design. That's a simple truth, and everything else is posturing and theater.

From the looks of it, the people that are offended by the SGT are offended for the third reason. Apparently 3.5 is an incredibly balanced game, in which fighters don't need a great deal of help to be equal to wizards and that those claiming otherwise should give the designers more credit. And that is why the SGT should be 4 times easier than actually intended (seriously, one of the examples was level 1 Fighter vs Housecat) so as to not humiliate him.

Of course I know that's laughable, everyone here knows that's laughable, and those yelling at the idiots in question knows that that is laughable.

Also, and I already know the answer to this but if you had the choice between a party of Fighter/Rogue/Cleric/Wizard and Cleric/Druid/Wizard/Wizard, which would you pick and why? Ignoring Tome content and such, just talking 3.5. Assume anything goes, book wise (which favors the fighter types).

[angelfromanotherpin]

Actually, the PC gets like 4 times the wealth by level, which means that the PC version should win slightly more often than the NPC version, because of better gear.

I think it was outright stated at some point that NPCs have less wealth because they are supposed to nova off all their expendables like it was 1999. Presumably the rest of their money is in a 401k somewhere.

[TarkisFlux]

Since the only difference between an NPC human with nothing but fighter levels and a PC human with nothing but fighter levels is the person playing them, we will go ahead and assume that the statement holds true for PCs as well.

Actually, the PC gets like 4 times the wealth by level, which means that the PC version should win slightly more often than the NPC version, because of better gear.

I thought 3.5 had wiped that difference actually, but I could be wrong. In any case, if that's the only complaint you have with the reasoning over there I'm basically set. Just give them cash like an NPC for purposes of the test, and follow NPC equipping rules. Which is even more than I want to do for the test, because I don't think it's particularly useful for measure a specific build when you're looking for general balance of a class or whatever.

That wiki was already brought up, but the opposition in that case is a Ferrous Cranius.

I'm more surprised that it got brought up at all than that it got dismissed actually. Yay exposure.

[erik]

Also, and I already know the answer to this but if you had the choice between a party of Fighter/Rogue/Cleric/Wizard and Cleric/Druid/Wizard/Wizard, which would you pick and why? Ignoring Tome content and such, just talking 3.5. Assume anything goes, book wise (which favors the fighter types).

Heh, that reminds me of some of my favorite Living Greyhawk parties.

Usually they try to muster balanced tables, but since it is randomly based upon who shows up with X level characters, it doesn't always work out.

A few times I wound up with a table that was Rogue/Wizard/Wizard/Wizard/Wizard... with varying levels of competency on the wizards.

The distribution oddity was because the mod series suggested some sort of wizard guild reward. I was the odd-man out because my rogue was intent upon becoming an Unseen Seer.

Some people were worried that we had no healing whatsoever. Instead, we rocked every single encounter. Shouldn't really surprise anyone, but damn those combats were easy. It was even relatively low level (4-6 average party level).

In one adventure think we even kicked ass against some bullshit invisible elemental spellsomethinged... ghost(?)... or something retardedbullshittastic out of a retard's guide to stacking templates that make no sense for maximum unfair CR that allegedly raped most traditional parties. The DM was the guy who wrote the mod and since we cakewalked it he felt justified in using such a golgothan tribute. Oh well.

[Roy]

I looked through the thread and saw nothing that matched what you quoted. Links please?

I already said basing the challenge of the SGT on a 1:1 lvl against CR basis, then yes, the caster has a better chance then a fighter.

But I also showed the facts of the 3.5 system and then showed how SGT for 3.5 ignores those rules. It is the SGT team that chooses to ignore the rules and get invalid results.

In response tot he curbing a hobo part:
This suggests the founders of 3.5 SGT think that combat in pnp is easy. Again the DMG clearly states that a single creature of CR X is a moderate challenge to a party of 4 of lvl X (where X is the same). It didn't say it was a one round win, it doesn't even say how long it should take or how much damage the players should suffer BECAUSE it will be different for each encounter.

To the second quite:
I like the idea of an SGT. It is a good measure for something to find a flaw. The only problem is, it is not taking the black and white rules of PnP into consideration and then imporperly running tests on it. In a company, this is a fireable offense. All their results would be wrong because they are not following the rules in the game.

For quote three:
Yes, if you ignore the CR rule mechanic, the fighter will be hard pressed to win against something whos CR is equal to his level. That is why if you want to check combat metrics, you divide by 4. Again, the DMG clearly states that CR is a rating for a party of 4. Strip one out and the CR is too high for the individual to take on moderately.

See, the problem is not in DnD 3.5. The rules there are written, put together by hundreds of people over the course of years, and played by thousands of people-hours to determine and refine the final product.
But a few people say they think Melee is weak and then build a test scenario for it without following the most basic of rules; that is flawed.

Remember that code section up there where I posted monster stats? Use those for a lvl 1 SGT. Give me results.

TLDR version: Yes, melees are hopelessly gimp if you follow the actual rules for a SGT, which in turn means following the actual rules of the game to achieve the intended purpose of a SGT - namely, gauging if things are above, below, or at par. So you should make it four times easier so they have a chance in hell to succeed. Because ya know, false positive results are awesome. And oh yeah, you aren't following the rules at all. Now, show me how the fighter handles housecats.

For the record, quote 1 is Frank, quote 2 is K, quote 3 is Maxus.

And I thought shadzar was completely fucking brain dead. What the fuck is this assclown smoking? Words cannot describe just how big of a complete and utter dumbfuck he is for saying that, and ya know what? Fail pics can't either. He's gone beyond that.

Jesus H Fucking Christ on a cracker, now I remember why I stopped playing D&D. It means sharing the hobby with black holes of intelligence, logic, and reason.

Actually, the PC gets like 4 times the wealth by level, which means that the PC version should win slightly more often than the NPC version, because of better gear.

I think it was outright stated at some point that NPCs have less wealth because they are supposed to nova off all their expendables like it was 1999. Presumably the rest of their money is in a 401k somewhere.

NPCs have less wealth because you kill them and take their stuff. The consumable novas is the logical consequence of this.

[Kaelik]

I might or might not put up a more specific criticism of that wall of bullshit, but it might behoove you to do this:

Take a sample build of level 1 Wizard. Defeat all the challenges with a 100% success rate.

Take a sample build of Fighter, do the same.

Rogue, ect.

Cleric.

Bard, Barbarian, Beguiler, ect.

Commoner. Lightning Warrior, with Familiar.

Once everyone has a 100% success rate, ask him again what that was supposed to prove. Obviously CR-4 is not going to measure anything, because all classes from Commoner to Lightning Warrior have the same win chance.

Oh shit. Looks like it doesn't fucking matter what class you pick because all classes are equally powerful, including commoner.

Or alternatively: CR -4 is too easy to tell the classes apart at all.

[Roy]

If I were to get involved in that discussion directly, it would be to do things illegal in most civilized countries for the good of humanity's continued survival to the guy actually spouting off that monumental bullshit. It would not be to give him the time of day by pretending his has a valid point about anything, on any subject.

After all, this guy seriously thinks that because a level 1 Fighter can kill ordinary cats that there is no problem with melee and the game is perfectly balanced.

I could kill a cat bare handed. I could even kill multiple cats in rapid succession, with no danger of dying. That doesn't mean I'd be able to take on two savage warriors at once with a 50% chance of success unless I were armed with modern weaponry.

[ubernoob]

I could not kill a housecat. I simply love cats too much. Thus, a housecat would always defeat me.

[Prak]

I was thinking about using the SGT as a stable for the opposition in my next adventure so that I can better calibrate my party.

What do I need to take into account, and what do I need to look at, if I'm running a party of four (Lv 5) against the SGT?

[Grek]

You can do a class by class evaluation of a party-based SGT. It's just tedious as fuck. What you do is stat up a 5th level character from each class you want to test, then run through all the possible party combinations through a SGT. Calculate ln[P(party victory|class in party)/P(party victory|class not in party)]. Results near 0 indicate the class is near average strength. Positive numbers are above average, negative numbers are below average. If any result is outside of [-0.33. 0.33] you have objectively fucked up somewhere. The only problem is that it takes a whole lot of time to do. For 10 classes and 10 challenges per test, that's 2100 encounters to run and keep records on. This is of course way, way too many tests to run, so sane people simplify things by doing testing on an individual basis and hope that putting 4 characters with balanced classes into a party will result in a balanced party.

[Prak]

No no, I want to run my party through an adventure where all the challenges they face are SGT creatures, so that I can calibrate future adventures against the results.

[Kaelik]

No no, I want to run my party through an adventure where all the challenges they face are SGT creatures, so that I can calibrate future adventures against the results.

Well you can't do that because the only standardized lists are 5 and 10, and since your party of four level 5 PCs should have a 100% success rate against the first, and 0-10% (according to the DMG) against the second in order regardless of whether or not they actually are well suited to the SGT, you aren't going to learn anything.

If you are instead asking what CR 9 monsters to test a party against, that is a weird question, because as previously established, the SGT works best as a measure of individuals. If you want to know if the party can deal with EL 9 monsters, just pick some random EL 9 encounters out of your ass.

[Whipstitch]

hatsoever. Instead, we rocked every single encounter. Shouldn't really surprise anyone, but damn those combats were easy. It was even relatively low level (4-6 average party level).

Yeah, AD&D wizards are what finally taught me how "balanced" groups often lacked synergy relative to groups that went all-in on particular tactics. For example, blasting was a bit sub-par even back then, but having 3 guys capable of dropping fireballs meant that it was incredibly hard to make cannon fodder matter short of staggering the rate at which they entered the field of battle.

[tussock]

My only real complaint with the SGT is that the examples use levels 5, 10, and 15, which are the worst levels for full-BAB classes given they get a substantial class bonus at 6, 11, and 16.

Well, that and you could consider matters of healing and buffing and whatnot. Higher level casters do have more spells than they need to win, so buffing the mundane party members with low level slots isn't unreasonable.


But it's the same result if one uses EL 3, 8, and 13 as test points, and assume some buffs, and give out heal spells at the top. A fairer test just makes it even more obvious that the grunts are falling further behind as they level up, and drawing more and more resources from the casters. Who notably do just fine without a meat shield.



And because 3e was balanced for the iconic four Fighter, Wizard, Cleric, and Rogue, it's easy when you take four casters and hard as fuck when you take four non-casters. Because the casters are better, against the monsters. Particularly in the ability to just not play the monster's game. Not melee the melee monsters at all.