Help me nail down this 'take 10, take 20' nonsense

[dcollins]

Well, for what it's worth, I basically agree with "knifespeaks" on this issue. I've said it before, but the non-randomness of the Take 10 and Take 20 rules make for some really degenerate probability plateaus.

A smith wants to make a complex item (DC 20). At one level, he's got a +9 bonus and fails 50% of the time. He gets one single more skill point and he fails 0% of the time due to being able to Take 10. I want the success to curve smoothly over increased skill, not jump cartoonishly like that.

Particularly with regard to scenarios that are dangerous and have established Challenge Ratings -- such as Search, Disable Device, etc. Those things really feel like they need to be dicey, like combat, not metagamed away. I want a good Rogue in my campaign to have a 95% chance of Search success, not either 50% or 100% due to the Take 10 rule.

As DM, I started backing out Take 10 and Take 20 for CR-related things like Search, and consider that justified under the existing rules as one of the rolls the DM can take over. But myco-DM liked the mechanics even less than I did, and persuaded me to scrap them altogether in our campaign.

For something like Search, where there's no clear "success at finding nothing", the PCs shouldn't be able to know when further searching does no good. In addition, if it's a trap-like Challenge, it should resemble the combat mechanic (re: rolling probabilities, not talking about auto-success/fail).

(Also, regarding the ship's rigging, everyone realizes you only need skill +5 to avoid ever actually falling, right? Easily accomplished by a fit 1st-level Commoner?)

 

[Lamoni]

So, the fact remains - a take 20 is the same as saying 'roll a 20, success is guaranteed'.

The fact remains that you are misinterpreting the rules and had to come up with a house rule to fix a rule that isn't there. Taking a 20 will give the same result as a die roll of a 20. Are you saying that if you are trying to jump up and grab a limb that is 60 feet high... that you could say, "I'm taking a 20!" and then automatically grab it? Sorry, that isn't how it works. Yes you can take a 20 on vertical jumps if done outside of combat when there is no consequence of failure, but if you try to jump for something that your character isn't capable of reaching, it doesn't do any good except to waste 2 minutes while not being any closer to your goal.

Edit: I apologize for posting the above without reading all the pages of the thread. I'd like to change it to add a few examples of what equating 20 to success would mean.

A 1st level player could:
jump 80 feet high to get out of the pit (who needs to climb?)
Escape from masterwork manacles
pick a lock to the bank's safe
Find a grain of salt on a sandy beach
Break through the safe's wall.

I am quite happy 20 doesn't equal success. Enough examples have been given why rolling a 1 shouldn't equal failure. It is amazing how many pages this thread went on for since the topic is pretty straightforward.

 

[Patryn of Elvenshae]

(Also, regarding the ship's rigging, everyone realizes you only need skill +5 to avoid ever actually falling, right? Easily accomplished by a fit 1st-level Commoner?)

Until you house-rule away the ability to Take 10 and replace it with the fail on one mechanic, as both you and knife are talking about.

As soon as you do that, you end up with a mortality rate of 1 death per 2 man-minutes in a ship's rigging (assuming that a fall from the rigging will be enough to kill an Expert 1).

Given a crew of twenty people manning the topsails, you'll likely kill them all in about 6 minutes, and with near certainty in 8 minutes.

 

[dcollins]

Until you house-rule away the ability to Take 10 and replace it with the fail on one mechanic, as both you and knife are talking about.

C'mon, read what I wrote. This was immediately prior to what you quoted:

(re: rolling probabilities, not talking about auto-success/fail).

At no point did I suggest making a natural 1 an auto-failure. I said just the opposite. It's not like lack of Take 10 logically implies fail-on-one.

 

[Patryn of Elvenshae]

Actually, I was responding to this:

I want a good Rogue in my campaign to have a 95% chance of Search success, not either 50% or 100% due to the Take 10 rule.

[Edit: In conjunction with this:

(Also, regarding the ship's rigging, everyone realizes you only need skill +5 to avoid ever actually falling, right? Easily accomplished by a fit 1st-level Commoner?)

]

The only way you can have a 95% chance of success on anything is if a roll of 1 is a failure.

And if a *good* Rogue only has a 95% chance of success, that means he's autofailing on a 1 (barring the off chance that he's looking for something with a Search DC of 11, and he's got a Search skill of +9).

 

[Deadguy]

[snip]But myco-DM liked the mechanics even less than I did, and persuaded me to scrap them altogether in our campaign.

You have a fungus for a DM? Cool!



Bet shriekers and myconids are a lot nastier in his game!

 

[Ranes]

Yeah, and he's a fungi. Excuse me, I think that's the bad joke police at the door.

 

[knifespeaks]

Ok - it's now 6.00am here, I have just woken up and feel like starting all over again!!

In all seriousness, help me out here:

For what is mundane, take 10/20 is fine. But, player's skill levels notwithstanding, there will be times when I want failure (or success) possible - however unlikely.


Consider this example (really, this is about the only issue I have with taking 20):

There is a secret door, on the first level of a dungeon. It leads to the treasure room on the lowest level of the dungeon. How do you make it POSSIBLE that a player of skill 0 to find, yet prevent them from taking 20?

Assume that once the denizens of the room/dungeon level are killed, they are safe enough to search to their hearts' content, although the odd wandering monster is still possible.

It could easily be I am missing an elementary mechanic - my only thought thus far is to specify how the secret door operates, and allow a circumstance bonus if the players get close to the mechanism - seems like alot of work though....

 

[Scion]

If there is a hidden door that is very easy to find (anyone with a skill of +0 could find it eventually) and they have unlimited time and the desire to search everywhere then there is no chance* of failure. With or without the take 20 mechanic. The take 20 mechanic simply lets you know how long it is going to take without having to roll all of the time and it helps keep people from getting frustrated by what should have been trivially easy (I search all over) into making them roll hundreds, or thousands, of times.


*With rolling there is always a finite chance of never finding something so long as there is a number on the die where it could fail, but the chance becomes vanishingly small rapidly.

 

[knifespeaks]

Well, thats a massive oversight in the rules then - I don't consider a dc 20 secret door easy to find. That may be mere semantics, but still...

In fact, it's utterly ridiculous that there is NO mechanism for that pretty straightforward scenario....surely, there must be SOME way.