Grim'n'Gritty - 3d6/2d10 instead of d20?

[kitsune9]

So, soon our low-magic campaign is going to start up, and we're using the Grim'n'Gritty rules 3.3. It makes combat far deadlier, but also a bit too random. To summarize the system, it does these things (mainly):
- Less hit points variations, and less hit points. A first level fighter might have 15 hit points, a fifth level 19 hit points.
- Active defense. Instead of attack roll d20+bonus vs. ac of 10+bonus, it's vs. ac of d20+bonus.
- Difference adds to damage. If my attack roll is 15 and your defense roll 9, I do 6 extra damage.
- Armor as damage reduction.

This makes combat far more deadly as even a high level fighter might die of a stab or two, and there's no risk of surviving a dagger in the back when you sleep - even if the stabber is a first level rogue and you're a 10th level fighter. I like this.

However, I've added a few simple house rules as well:
- Dex determines attack bonus for all attacks, rather than strength for melee. A feat can change this back, like an inverted weapon finesse. This is because it feels more rational to let your chance of hitting be determined by agility and coordination rather than brute strength, for most weapons.
- Strength bonus adds double damage to one-handed and thrown weapons, 3 times damage to two-handed weapons, and single damage bonus to off-hand weapons. This is because dexterity would otherwise be the top choice for all fighting characters.
- Power attack adds +1 strength bonus per -1 ab (so +2 damage for one-handers). Because otherwise power attack would be useless due to damage bonus.

The thing is, the system is a little TOO random, and I'm considering leveling it out a bit by replacing 1d20 as the base dice roll with 2d10 or even 3d6. In that case, I will do it to skill checks and saves as well - because the problem is right there too, with a "very hard check" with DC20 being easily done by a 1-st level commoner, if he can take 20. A DC 15 open locks check can be done in no time by anyone with a single point in the skill. With 2d10 or 3d6, the time it takes will increase severely.

The point of the changes is to make the game bloodier, more dirty, and more deadly while still rewarding skill rather than pure luck. I understand that these are big changes, but most of our group will consist of people who are used to far more advanced systems than D&D. The game will be a bit slower, but on the other hand, with nearly no magic at all it won't be stopped by preparation and spell rule searching.

Could I get your opinion on this? What is preferable, 3d6 or 2d10? I find 3d6 more esthetically appealing, but it's one more dice to count and it makes the game VERY skill dependent. Combats might become too slow, when everyone gets result about 13 all the time. 2d10 looks uglier, but might be faster to calculate and more random (for good or bad).

Also, anything else I should think about?
Thanks a lot for response.

The other posters said 2d10 is probably your best best. I've always enjoyed 3d6 myself, but I'm only saying that simply out of liking 3d6. I have no math skills to back up my opinion. However, let us know how the campaign goes. I remember reading an earlier version of Grim'n'Gritty rules and thought it was a bit too brutal for my tastes.

 

[Ilja]

Well, even without changing Will saves, Wisdom is pretty damn handy in those social situations, as a matter of fact. Sense Motive could save lives, pride, the kingdom! And Perception might keep you out of (now, more deadly) combats in the first place, and negate surprise, ambushes and the like. Or, yeah, traps / snares / minor details / hidden passageways, etc. Handy!

Yes, but the wisdom bonus seems pretty small when a single feat could provide +3 to the skill. But yeah, it might be enough to change linguistics to wisdom. That way, the important thing for social characters would be the ability to read his surroundings and express himself. Feels rational.

But. . . what does Intelligence do for the characters, though? I mean, even before that last change of yours?

Skill points, and some useful skills like Craft are intelligence based (craft will be quite useful in a low-magic campaign). But mainly skill points. This campaign will be quite skill-heavy, so every extra point will do a lot of good. I'm thinking of actually lowering the base skill points of the classes, because Pathfinder puts together many skills into one so now, all rogues have about the same skills.

In my own house rules (based on 3e, mind you) I decided to up the importance of Intelligence, and Charisma. Also Strength, but there are reasons for that (other house rules, actually). The major way I did this was to average pairs of stats for save bonuses.

Average pairs for saves? How do you mean?

Bit of a sidetrack, but it's kinda relevant too, because I found that Int (and Cha, in particular) was under-represented in 3e, other than when your character is a Wizard (or Sorcerer/Bard), basically. Dex and Con can both be uber-stats, and Con especially is just SO important for EVERY character. So I evened things up a bit, with compensation where necessary.

Out of interest, what classes do you make available?

Yes, con is always extremely useful. I don't know how much the effectiveness will be affected by the GnG rules, since on one hand, they don't give as high bonus to HP but on the other hand, the amount of HP is far lower to start with.

The classes I'll allow will be barbarian, rogue, monk (with less magic), fighter, and ranger (with less magic). Also, an improved version of the Expert class that will be useful to any non-rogue skillful character. No multiclassing penalties. For monk and barbarian, I'll probably ask for a pretty good background story too as these aren't very common in the world.

 

[Aus_Snow]

Average pairs for saves? How do you mean?

Basically, it's [Con+Str]/2 for Fort, [Dex+Int]/2 for Ref, and [Cha+Wis]/2 for Will. If one is higher, round up, and if the other is, round down. It all makes sense in the context of the general house rules document. Well, I hope it does!

 

[Ashtagon]

Of course it makes a difference! When you hit on a 10 or more, it won't make a difference, but if you in RAW have AB +3 and your opponent AC 23, you only hit 1/20. Here, he'll have defense +13 and so 50% of the time you will have more than 1/20 chance of hitting him. Also, it makes a big difference since it isn't a hit- or miss situation. Since difference is added to damage, it has a very big impact.

If you attack bonus is +3 and his AC is 23, under raw you'd need a 20 (5%) chance. With this modified system, those work out as you having +3 on your roll and him having +13. A quick excel check makes that a 55/400 = 13.75% chance of a hit. Slightly better odds. This is compensated in an equal way by a reduced chance of hitting in the opposite case, where the target has a very poor AC.

Once you account for the granularity involved as a result of the dice used, the adjusted odds are mathematically related to the size of the variance (1-20 under RAW, 2-40 in your modified system), although it is a bit too complex to describe simply.

To correct my earlier post, an opposed check of (bonuses + 1d20) is mathematially the same as an unopposed check of (bonuses + 2d20 - 10). The only difference is more than one hand is rolling dice, which slows play slightly.

Actually, I can see it being quite a large danger. A dagger does 1d4, x2 on a critical. A 1-st level commoner dealing on an average 5 damage, resulting in a DC 15 fort save or die. A 10th-level fighter with 15 con will have to roll a 6 or higher to survive, not counting any feats or magic.

Past 6th level, normal human ability goes out of the window and we enter the realms of heroic fiction anyway. I find it quite acceptable that Thrud the Barbarian should not be auto-killed in his sleep by a random peasant.

To be fair, let's look at a 3rd level fighter under the same situation. he has about a 50:50 chance to shrug this off. In RP terms, we can describe this as a commoner sneaks up on you in your sleep and tries to slit your throat. But he does such a botch job of it that he manages to wake you in the process. You're badly wounded, just woken up, and he has a knife armed and ready to strike again.

The more likely fiction scenario is a rogue with a few dice of sneak attack bonus damage. Dagger + crit + 3rd level rogue sneak = 2d4 + 2d6, average 12 damage. A DC 22 Fort save to avoid instant death, Assuming 14 or 15 Con, he needs to roll a 17, assuming his hp held out. he's not likely to survive the initial attack.

Especially with your reduced hp totals.

 

[Toptomcat]

The number and magnitude of changes you're proposing to the core system of D&D, over and above the already fairly radical steps of using Grim & Gritty and E6, make me wonder if you aren't looking for a different set of game mechanics entirely- like GURPS Dungeon Fantasy or something.

 

[Ilja]

@Ashtagon: I am a first level commoner, and probably with a strength and dex modifier of -1. If I was to kill a veteran military in his sleep, the risk would be him waking - not me botching the cut. It's not that hard to cut a throat, supposing you've got a decent knife.

@Toptom: Well, I've dropped some of the changes (like dex-based attacks, double strength, power attack change) so what's left is GnGR (instead of GnG3.3) and E6. You'd know, you're in one of the games
The game you're in will still use the d20, and my AFK group will use 2d10. So we'll see how it plays out.

 

[Ashtagon]

1st level commoner (-1 modifiers on all relevant stats) vs. veteran military (let's call him a 3rd level fighter with +1 modifiers on all relevant stats)...

Auto hit, auto critical, as per coup de grace rules. That's 2d4-1 damage, average 4. He probably has 21 hit points (6.5x3, round up), so he'll easily survive the basic damage. He then has to make a DC 14 (10 + damage dealt) Fort save, or die. Level bonus is +3, Con bonus is +1, so he needs to roll a 10 or better to survive - 55% chance.

Way I see it, if he fails that Fort save, you got him good. If he made it, story-wise it means you made some noise in the process which caused him to wake up. The Fort save does not mean the skin on his neck was sufficiently thick and scaly to block the knife from cutting an artery. If you don't want to fluff it as him waking (although I would personally), you can say he rolls over in his sleep at exactly the wrong moment, causing your blade not to strike at the exact right point to instantly kill him.

 

[Ilja]

Yes, and it's that abstraction of hit points as "enough noise to wake him up". Rolling over in his sleep at exactly that time is extremely unlikely, and shouldn't be based on Con. You should probably use a straight percentage chance for that. "Enough noise to wake him up" should according to me be move silently vs I don't know what, a wis check for waking up?

I see that extreme abstraction, that hit points can be really anything at all (I think rolling with the blow isn't such a bad abstraction, but waking in the sleep surely is), as something bad. On the other hand, I'm also used to a system where you except for the skills Blocking, Parrying, and Dodging, and except for having something like 23 different areas of armor coverage and 15 different kinds of armor for each area, has four different hit points meters; tiredness, schock, trauma, and bleeding. That's far too complicated, though realistic, but it's like switching between playing Risk and playing Ludo; sometimes you want something in between when it comes to realism and complicacy (is it called that? Darn my bad swenglish).

Especially when the abstraction coincides with rules that exists in other places, such as a higher level character being more likely to survive a fireball because he can "roll with the damage" when this is exactly what reflex for half is, rolling with the damage.

On that, you have the problem with falling damage and other natural disasters. A first-level commoner (me) falling ten feet would likely be dying afterwards. That's far too deadly compared to reality according to me, seeing as how I've fallen over 15 feet several times and I might have been lucky but the worst thing that's happened is a luxation. A 3th-level fighter who's a bit tougher than the average person (say +1 con) has an average of 20 hit points (not counting maxed for first level, as he's not a PC), meaning he can fall 50 feet and most of the time be able to just stand up and walk away. For a 5th-level one (who's still in the "reality" levels) it's 90 feet. It's the same problem with any natural hazard. Sure, learning to "roll with the damage" might help a bit, but a fighters training isn't mainly acrobatic and few people could stand up and walk away after a 50 ft fall as if nothing's happened.
A solution to this might have been to convert all natural hazards into ability damage, but that's almost as much work as using the GnGR system as you have to recalculate hit points whenever someone falls or gets burned.

 

[Ashtagon]

I agree to a certain extent with the coup de grace rules - it is calibrated toward more of a heroic (possibly even cinematic) scale, where dramatically appropriate trumps realism.

If I were to houserule a "realistic" version of coup de grace, I'd say that any character can apply sneak attack damage to such an attack as if they were a rogue of their character level. Characters who have sneak attack anyway get to apply their normal sneak attack damage in addition to this. Combine that with the Fort save, and pretty much nothing will survive a knife attack in their sleep.

(Check using example from earlier: 2d4-1 + 1d6 damage, average 7.5; 3rd level fighter with +1 Con bonus will have 37.5% survival against untrained wimpy commoner. 1st level rogue (+0 Str mod) doing same attack will do 2d4 + 2d6 damage, average 12; same fighter will need to roll 18+ on the save; 15% survival). This seems closer to gritty, although perhaps still not quite there.

It has occurred to me that the coup de grace rule was written to describe what happens when you make an attack against a motionless person. It seems to describe what happens if you are standing over him with a weapon quite well. But if you walk up to him, turn him over, expose his neck or belly, and then push the knife in, that's a different matter entirely.

I'd say if you spend a full-round (or more if he is sleeping buried under a big pile of blankets) to kneel down and find such a soft spot, the coup de grace rule should be ignored.

Summary: regular coup de grace (you're standing over him etc): You score a critical hit automatically, apply sneak attack damage if you have it, and also apply 1d6 + one die per two character levels above 1st as precision damage. Victim makes a Fort save (DC 10 + damage dealt) or dies instantly.

Throat Slitting: If you spend one (or more; GM discretion) rounds in preparation to kneel down next to the target and expose his neck, belly, or other vulnerable spot, you can kill them instantly with any light piercing or slashing weapon as a full-round action.


Falling damage:

The 1d6 per 10 feet assumes you fall with no slope or intervening tree branches or anything else to help slow your fall.

The Tumble skill allows a DC 15 check to reduce the effective distance fallen by 10 feet. The Jump skill also allows such a check. Than means that if you pass both skill checks, the effective damage from a 20 foot fall is zero (the DMG glossary implies you only get one of these, but the skills imply you could get both benefits).

If the surface you land on is soft (such as grass), one die worth of damage becomes non-lethal too. There's also a sidebar in there which says a DM can make the first 10 feet as non-lethal damage, regardless of other factors

With all these factors combined, it seems entirely reasonable for a commoner to survive a 15 foot fall with nothing more than a dislocated joint.

You're right that the 5th level fighter can survive falls of silly distances and walk away though. I have seen houserules on this forum which would make that end of things realistic. Unfortunately, I can't find them right now. I recall it centred around some short term ability score damage that varied depending on the distance fallen or hit point damage taken.

 

[Ilja]

Yes, giving sneak attack damage might solve the symptom, but I still think the cause is there - that hit points increase too much over time.

On falling: Well, sure, forgot about the land on soft ground rule, though I can't perform a tumble check and my jump mod would be -1. But sure, 15 feet would probably not make a commoner end up dying if it's soft ground below.
Though in my case it was asphalt

Anyway, I prefer the GnG rules (and especially the revised - though I at first thought the others better, after playtesting them a little I'm in love) for several reasons, and it solves all of these problems as well as the "badass until 0 hit points, then dying" problem (which may be an easy system, but to me quite unfulfilling).