I've always thought this about Dark Heresy. And, well, most other RPGs I've played. The idea that people assisting on a test don't even have to touch their dice, and that they just provide a boring flat modifier is really unengaging to me.

What I'd propose (and I haven't playtested this yet, so I doubt it's too balanced yet) is to make it that everyone involved in an assisted test make a roll. Then you calculate everyone's Degrees of Success, or Degrees of Failure where applicable. Add them all together, with each DoF and DoS cancelling each other out. Then, count up what's left. That's the DoS or DoF of the assisted test.

Say for example that the whole party is coordinating to try to get their Chimera repaired and ready to roll out. Each player rolls their test, each applying applicable modifiers for their items, skills, etc.

The first party member scores 2 DoS. The second scores 4 DoS. The third does particularly badly, scoring 5 DoF. And the last party member scores a single DoS.

That's 7 Degrees of Success all up, and 5 Degrees of Failure. The DoF and DoS cancel each other out, meaning the end result of the test is 2 Degrees of Success..

Thoughts?

Feels odd. If half the group on a beer run the odds of passing a test lower than 50% go up.

I can see the simplicity in the current convention: one or two PCs assist (if they possess the skill in question, that is) with +10 or +20 being added to one die roll and no comparison calculations other than whether the test a success or a failure. I don't really like the idea as you've presented it, other than to say it's a fine way of indicating "too many cooks", ya know?

I think having some general examples of what DoS and DoF mean, insofar as what it means to have 1 DoS when attempting to disarm a security system as opposed to having three or four DoS. This treads sticky ground, because it means having, at most, half a dozen charts, and charts eat page space and take time to compile. Didn't DH1 have something like that, though? I know some of the earlier publications, like DotDG, had charts...3 DoS on a relevant skill test meant you knew info X, X, and X. Maybe something for fixing/dismantling tech (cars, tanks, bombs), security systems (even something as simple as a deadbolt lock) and info gathering from tech terminals or encrypted data slates...or not.

The only issue I have with assisted tests atm is a player issue...they never remember how assisted tests work, even though they've done them dozens of times. Celly #1 thinks of disarming the bomb, so he's gonna make the test roll, even though Celly #2 has a better base skill, the test is flubbed, I describe the beginning stages of the explosion, then the players go "Hey, wait. Celly#2 shoulda rolled with me assisting." Duh. Too late.

And even if the test is successful, they always forget the extra DoS for it being an assisted test.

It is kinda lackluster as it is now, but it is simple to use.

As much as I agree with the initial premise and the objective of your suggestions, I fear that this quite easily snowballs as presented, leading to wildly unpredictable results, varying by the competence of those involved (for example, if the rest of the group are dumb as bricks, they'll guarantee a lower mean result than if the Tech-Priest did it himself).

While it is then arguable that "Well, then, don't include the dumb ones.", we face another issue; If two well-qualified tech-savvy gentlemen (say, a Tech-Priest and an Operator, assuming Only War roles) work together, they'll almost certainly simply add a high number (relatively speaking) degrees of success together (depending on difficulty of the initial test).

It is arguable that this is appropriate, but it means that balancing certain mechanics are going to be hard, because they must scale well from anything between 1 DoS up to the 10's.

I would simply add a box about this alternative use of Assisted Tests, telling GM:s to do it when appropriate, with a recommended cap of two people working together, adjusted at his leisure for larger projects.

I like the mechanic, I just don't think it should be the base mechanic, but instead a call the GM could make when it might be odd that people would run the same check twice-but-separately.

While I'm thinking about it...

Insofar as game mechanics, is there a particular reason assisted tests receive the +1 DoS? Typically a skill test is pass/fail. Typically. I could see an extra DoS being groovy when aiding someone in manifesting a psychic power or shooting a monstrous beast, but those aren't skill tests.

Am I missing some one-line rule in the core that differentiates between "pass/fail" and "passing with X DoS is better because reason"?

Edit- Awareness is a skill, and an assisted Opposed Awareness test, like against Stealth, would benefit from the extra DoS. But this is very situational...

My issue with this is that players lack incentive to assist on a test unless they have a greater than fifty percent chance of success, as they are otherwise equally likely to hinder things as help them. It kind if defeats the point of a group of lowly acolytes working together to achieve something beyond one of them. An exemplar example would be a strength test. Shouldn't two people of equal strength be able to accomplish almost twice as much in a test of strength?

What about some way of combining characteristics when determining the target number? For example, finding a way to add both strengths of 30 to get a final target number of 60? This is obviously unbalanced to me, but I think that if it's given some proper restrictions it could be a cool mechanic. For example, when testing everyone's awareness or having everyone search a room/detect a lie, you could combine all of the individual rolls into a single roll and maybe base the number of individuals succeeding on DoS and go in order of highest characteristic. This would also cut out the problem where you'll have the tech priest fail a tech roll and then the next highest tech use goes, and then the next , and finally the poor -20 feral world we tries rolling below a 6 to help. In essence, you'd eliminate some of the gaming of the dice system that serves to pull people out if the narrative.

As much as I agree with the initial premise and the objective of your suggestions, I fear that this quite easily snowballs as presented, leading to wildly unpredictable results, varying by the competence of those involved (for example, if the rest of the group are dumb as bricks, they'll guarantee a lower mean result than if the Tech-Priest did it himself).

While it is then arguable that "Well, then, don't include the dumb ones.", we face another issue; If two well-qualified tech-savvy gentlemen (say, a Tech-Priest and an Operator, assuming Only War roles) work together, they'll almost certainly simply add a high number (relatively speaking) degrees of success together (depending on difficulty of the initial test).

It is arguable that this is appropriate, but it means that balancing certain mechanics are going to be hard, because they must scale well from anything between 1 DoS up to the 10's.

I would simply add a box about this alternative use of Assisted Tests, telling GM:s to do it when appropriate, with a recommended cap of two people working together, adjusted at his leisure for larger projects.

I like the mechanic, I just don't think it should be the base mechanic, but instead a call the GM could make when it might be odd that people would run the same check twice-but-separately.

Felenis Seal of Approval. Everything you say now has my vote

I wonder how much the system I proposed would be fixed by keeping in assisted test modifiers, but having them apply to everyone. So, if you've got three people collaborating on a test, each person gets +20. This means there's a lot more chance of the assistance helping, but you still run the risk of an underskilled participant ruining everything.

The insane amount of DoS is a good point, though. Maybe put a hard cap of 10 or so on it?

Anything beyond 10 would be ridiculous overkill anyways. Beyond 6 even

Anything beyond 10 would be ridiculous overkill anyways. Beyond 6 even

There are cases where over 10 is considered (and sometimes likely), like RT's Hit and Run.

Assisted Tests could always be a probability modification: You can all roll to do it, and all characters pool their DoS and DoF for the final result. (This could be combined with a bonus as well, though.)

That of course introduces the chance that an assistant may cause it to fail-- but then that is fairly plausible. (And may be a more dramatic choice on whether to request help or not.)

There could also be a 'limited assistance' option, where the bonus might be less, but failure is limited or removed. But that may not be necessary. It is the grim darkness of the far future after all.

Yeah, I'm actually a fan of unskilled or unlucky people worsening the result. It makes it less a case of everyone just piling in and hoping for the best, and creates more in the way of hard choices for the players to make.

We could give +5 per DoS or per Skill bonus (Characteristic bonus would get odd).