Skill / Combat unification mechanic (please help refine)

[Arlough]

Okay, I want a unified mechanic between Skill Checks and Attacks.
Partially, I’ll admit, I want this to simplify my cheat-sheet by reducing the number of tables needed. Another part is because some Skill Challenges I already run as quasi-combats.
[sblock=”Combat Skill Challenge”]
Players make checks against the chosen aspect of a challenge.
The DC is determined by how appropriate the chosen skill is against the challenge.
If they hit the DC, they get 1d6 + (1d6 per 5 points over) to roll against the check.
The challenge is built like an encounter, with a statblock per aspect of the challenge (one aspect per needed success in the classic terms) with initiative, HP and attacks against the players.
There is only 1 standard action per turn. Turns progress until either all aspects of a challenge have been overcome, or until the players have succumbed to fatigue.
[/sblock]
So I was thinking about this. What if:

• Skill Training, instead of giving a +5, just allowed you to add your base (1/2 level) to the skill check
• Skill Focus, instead of giving a +3, allowed you to add inherent bonuses to your check
• Item bonuses to skills instead made the skill brutal (item bonus) So if you had an item that grants a +2 to acrobatics, instead it grants a reroll when you roll a 1 or a 2.
• DCs were calculated as follows
  • Easy = 12+Level
  • Moderate = 14+Level
  • Hard = 16+Level
• Racial bonuses and background bonuses were changed from +2 to +1/tier

What am I overlooking?

 

[Psikus]

What am I overlooking?

I think your patches cover the high end of skill DCs, but the low end (untrained PCs with a low ability score) is basically hopeless.

I've been thinking on this issue for a while, and it requires some heavy changes in order to work well. The DCs you suggest (12+Lv, 14+Lv, 16+Lv) are a good starting point, but you need to make sure that players can reach those DCs from both ends of the skill spectrum.

Basically, you need to meet these requirements:
- All of a character's skills increase by 1 point/level
- The difference between a PC's best skill bonus and the worst one is no higher than ~6 points (regardless of level).

One possible way to achieve this is by:
- Remove ability score increases (needs to patch game math in plenty other places, though)
- Change skill training bonus to +3, replace all other skill bonuses in the game for a non-stacking +2.
- Have all skills increase by 1/level instead of 1/2 levels.

Of course, the part about ability scores is messy as hell. If you don't want to change that, you'd need to:
- For all skills, have a minimum ability mod of 1/5 levels. This allows skills associated with bad abilities keep pace with those with good abilities.
- Change skill training bonus to +3, replace all other skill bonuses in the game for a non-stacking +2.
- For all skills, provide an increase of ~10 points over the 30 levels of the game. A way to do this is have an enhancenment bonus (say, let neck item enh. bonuses apply to skills) and an expertise-like bonus (preferably without feat taxes).

 

[Arlough]

Quick note, inherent bonuses are enhancement bonuses you automagically get at all levels ending in 2 & 7 for attacks and all levels ending in 4 & 9 for defenses.

The current DC table is set up thus

• Easy = 8 + level/2 + (level-1)/20
• Moderate can be calculated two ways
  • Moderate ≈ 12 + level/2 + level/5
    • -1 on all levels ending in 6, 8, or 0
  • Moderate ≈ 12 + level/2 + level/5.35
    • +1 on levels 5, 15, 21, 25
    • -1 on levels 6, 8, 18, 28
• Hard ≈ 19 + level/2 + level/3.65
  • +1 on levels ending in 3
  • -1 on levels 4, 8, 18

All fractions are rounded towards zero before adding

This means that the current setup is designed so:

• Easy checks are 100% successful so long as
  • The attribute linked to the skill is at least 10
  • The subject is trained
  • The subject is focused
  Even if you are untrained, you still have a 60% chance of success.
• Moderate checks are 95% to 105% successful so long as
  • The attribute linked to the skill begins as an 18
  • The player has been putting their +1 bonuses in said attribute
  • The subject is trained
  • The subject is focused
  If you are untrained with an 18, you are still succeeding 75% to 85% of the time
  If you are trained, but not focused and have a 10 in the stat, you still succeed 45% to 65% of the time.
• Hard checks are 50% to 65% successful so long as
  • The attribute linked to the skill begins as a 18
  • The player has been putting their +1 bonuses in said attribute
  • The subject is trained
  • The subject is focused

And this is before item bonuses, racial bonuses, backgrounds, aid another, or anything else that may boost a skill. When you apply a +2 from backgrounds, a +2 from race, a +2 from items, you can’t fail hard checks in your area of expertise except at some levels on a 1.

And with anyone being able to sit out a challenge, you can effectively eliminate the need to even roll on an equal level challenge.

Of course, with skill challenges only allowing for 3 failures, perhaps 100% success at certain things is seen as necessary.

So we either build around that errata, or choose to modify the terms of success.

I think that a 100% success rate for a well-trained artisan is reasonable, but not very useful for adventuring. At most, an adventurer should have only had time enough to adventure, or he or she could be really old (which would also have costs.) They may be good at a few skills, but only a few and only good, as skills represent many related activities and not perfect, as that doesn’t create tension or challenge.

I’ve been pondering this more and more, and I think that there are two aspects to my proposed approach.

1. The expected success rate of skill checks should be about 50%
2. The players should be able to use their highest or second highest stat in most checks.

The first would mean adjusting the success/failure ratio back to the original DMG tables or perhaps even a bit more lenient.
The second would mean allowing for an option rule, that would allow for the players to describe how they do things, and then the DM tells them what stat-skill combination to use. (I’ve been using that lately because it forces my players to think like their characters. ) This means that almost always you will have someone who can perform the skill, unless you have a highly unbalanced party.