Why I Hate Skills

[Bill Zebub]

That is exactly what I had in mind.

Yeah somehow I wasn't parsing what you expressed perfectly clearly. :-/

The play loop in those examples don't meet my criteria, but I do understand what you meant now.

 

[Thomas Shey]

Absolutely. But if you show them they rolled a 2, they'll know there's a very good chance they're missing important information.

Which is why I don't have them make the actual die roll until the information is relevant and would probably be obvious in its lack.

 

[Thomas Shey]

Sure...and, in my opinion/experience, that sort of "lumping" tends to shift skills from being "buttons to press" or, worse, "buttons that need excuses to be pressed", into something that lends itself more to players devising plans.

Eh. I'm finishing up a 13th Age campaign, and 13th Age uses Backgrounds rather than skills (which is to say its pretty big and lumpy) and I can't say I've seen any real difference; its just a case of bigger hammers.

 

[Thomas Shey]

In my opinion that's great in theory but really hard in practice. The GM secretly rolling and telling the player what they think they know doesn't really replicate the feeling of "thinking you know" something: the players themselves don't have any sense of certainty; they know it's just the result of a random roll.

Maybe that works for some, but the times I've tried it the results have been pretty unsatisfying for everybody.

Again, this tends to be less pronounced when the system involved doesn't have massive swing in it. You also can frame things about the obscurity of the information to give them a better idea of how reliable the information is likely to be. And of course if the information isn't at least a little obscure, there's no reason to roll in the first place. But I think most people have an idea that the roll they know is going to be 90% is more reliable than the one that's liable to be 50%.

(I've argued before that knowledge skills should usually not be expressed as rolls anyway; they should be thresholds, unless the information is so obscure that even high end experts in the field might not know it).

 

[soviet]

(I've argued before that knowledge skills should usually not be expressed as rolls anyway; they should be thresholds, unless the information is so obscure that even high end experts in the field might not know it).

Yes. The development of skill systems has led to a lot of things being accepted as normal when they are actually bad procedure. 'Roll to see if you notice anything' and (particularly) 'roll to see if you know anything' are bad play and should be eliminated.

 

[VHawkwinter]

Ah, right. I had to go back to see what Micah had written, and realize I must have somehow mis-read it the first time. That all makes sense, of course.

That's exactly what I was trying to convey as to why some checks are secret checks.

 

[Thomas Shey]

Yes. The development of skill systems has led to a lot of things being accepted as normal when they are actually bad procedure. 'Roll to see if you notice anything' and (particularly) 'roll to see if you know anything' are bad play and should be eliminated.

I don't agree about perception, but then you probably already know that. I think there are a few cases where information is obscure and whether a given character should know it is as well handled by a die roll as anything else, and where being particularly knowledgeable in the general subject probably should put a thumb on the scale, but I think a lot of it should just be "If you have Rank X in Skill Y you'll know this." (I can see cases where these interact, where someone below Rank X might just happen to know the piece of information anyway because they've crossed it in their learning Skill Y at all, but above a certain level of training there are things in most fields you just will know).

 

[VHawkwinter]

Again, this tends to be less pronounced(I've argued before that knowledge skills should usually not be expressed as rolls anyway; they should be thresholds, unless the information is so obscure that even high end experts in the field might not know it).

I have been setting things up like that for my game. Knowledge DCs are set by how complex they are to understand followed by how obscure they are. And if it's complex but not a secret (like professional knowledge an architect would employ) I ask what your take-10 is unless you're trying to recall something under pressure in combat. And then there's also a procedure for library research. Good to hear someone else came do similar conclusions independently.

 

[Thomas Shey]

A take-10 style mechanic can, indeed, handle much of this problem as long as two things are true: 1. People set difficulties with the actual thought of what an expert in the field will have as an expected bonus rather than getting carried away with said target number, and 2. can get away from thinking of all Take 10 resolution as being only about things that are trivial.

 

[Pedantic]

A take-10 style mechanic can, indeed, handle much of this problem as long as two things are true: 1. People set difficulties with the actual thought of what an expert in the field will have as an expected bonus rather than getting carried away with said target number, and 2. can get away from thinking of all Take 10 resolution as being only about things that are trivial.

This is most easily resolved by just making perception/knowledge defenses. Gated information can have an obscurity modifier the GM rolls against the PC's defense, and then you can just report everything they know outright.