Median based RPG...

[simon_hibbs]

But failure in this case can't be defined as crash and burning.  Forgetting to use a turning signal.  Waiting too long to turn the headlights on when it gets dark.  Driving too fast in the rain.  Following too close on the highway.  These are all "failures" to drive properly.

You expect your players to roll their driving skill every time they make a turn?

Surely success and failiure are measured against goals and consequences. What are my character's goals? What are the consequences of failiure?

Is this the roleplaying game of trailing the vilains car to his secret hideout, or the roleplaying game of passing my driving test?


Simon Hibbs

[Christoffer Lernö]

Is this the roleplaying game of trailing the vilains car to his secret hideout, or the roleplaying game of passing my driving test?

I think you're missing Ralph's point Simon. What he is admirably trying to do is to make the d100 mechanic work. If you, like Ralph suggests, grade the results and make the failed roll not a failure but a "below average success", you have a system that might be used on any scale, from trivial tasks to crazy car chases.

The traditional (in my view quite broken) approach in d100 is to say that "don't roll for trivial tasks" without mentioning on how to judge this transition, as well as ignoring the huge drop in chance of success that usually comes from introducing the game.

Let me take an example from an old school BRP style game. We have the skill of "Jumping". According to the description in this game, you use this to jump further than usual. It says that in general you can jump twice your height without any problems, as long as you do that you don't have to roll.
The skill have a basic chance, which is you dexterity (measured on a 3-18 scale) %. In other words, if I am 1.80 m, I can jump 3.60 m without having to roll... automatic successes. But if I want to jump 3.65 m, 5 m, 10 m or 100 m, I have to roll. And in all those cases I have Dex% of success. Ignoring the extreme distances, the fact that bumping my require jump by say 5 or 40 cm suddenly slashes my chance of success from 100% to maybe 11% seems pretty problematic.

An in this case there actually is a criteria on where the transition goes. Most skills don't even have that.

Ralph's suggestion is trying to bridge that problem. It's a very real concern no matter the focus. It's about transitioning from one scale or one area to another. If the shift is to big it's gonna feel weird (at best).

[Valamir]

Pale Fire touches on part of what I was going for Simon.

The other part would be that what Blackguard was going for was a system that was grounded in how things function in reality.  So, what I was giving was a way to justify a 50% chance of success on a routine task by how we define success.

To be clear my idea was not about rolling for every turn.  A single turn would not be an "average" task.  It would be a trivial task with a very high chance for success.  What would be an "average" task would be the trip as a whole, where a failure could be interpreted as a failure to use a turning signal, or someone flipping you the bird for cutting them off, to getting a traffic ticket, to causeing an accident depending on how bad a failure it was.

Now in practice you wouldn't roll for routine things any more than you do in any other game.  You don't make a dex roll for crossing the street (although in a very extreme form of simulation you should make some sort of check...people do get accidentally killed while crossing the street) why would you make a driving roll for going to the grocery store.  Whether or not you make such a roll in practice is entirely seperate from whether or not its sound logic to build the mechanic on.

What I would probably do in practice is as Black Guard suggested a "take 10" roll.  If my skill is 40%, I add a default 50% (the average result of a % roll) and I can default to 90%.  That means when people ride in the car with me they view me as a 90% driver.


Now, this brings up a related issue for me, which if I were Mike Holmes I might immortalize as a "standard rant" which would be called something like "why I hate %tile systems".  The 3 reasons in heavily abbreviated form would be.

1) As Walt mentioned, the linear curve.  Which I would expand to include the wide range of numbers on a linear curve.

2) The (IMO) unnecessary perception of accuracy.  I don't believe that increments of 1% have any real meaning when comparing a persons capabilities outside of a controled laboratory environment.  I'd say that its hard for most folks to even recognize a 5% difference in average ability between 2 people.  Just look at any list of who the "best baseball pitchers" are.  If it were easy to determine with precision, the list should be pretty firm with few dissenters.  In fact, however while you can seperate (with less dissention) the "stars" from the "utility" players getting more fine than that is more arguement and bias than rigorous evaluation.

In otherwords, %tiles use a level of granularity that I think is LESS realistic rather than more.  We can't say Schilling has a 93% skill while Johnson has a 92% skill.  Its just way too fine to judge.

3) The need to roll two dice which have distinct identities (i.e. keeping the 10s die and the 1s die seperate) severely limits the variations you can do.  It is extremely difficult to do a percentile die pool, for instance.


So, after performing this excersize on an average person's abilities etc, which I actually like what Blackguard's trying to do...what I would do is this.

First:  Switch to a d20...where a d20 is nothing more than a %tile dice with 5% increments.  In other words treat the system EXACTLY as it is now...just divide all the numbers by 5.  IMO if a modifier isn't at least a 5% increment its 1) too small to care about, and 2) not likely to be accurate from a simulationist perspective anyway.  Multiple small modifiers are going to add up to a 5% modifier.

Second:  Roll 3d20 and use the median roll to give the game a more normal distribution.

Any how...how's that for cramming two posts into one.

[Blackguard]

I think that Valamir got quite succintly what I was aiming for.  I'll give a few examples for clarity, though:

a) Journeyman smith makes nails.  Does he actually fail at making nails - an average task in a smithy?  No.  If his skill is 0 and he rolls a 60, 60% of them come out just fine but the other 40% have flaws in some fashion, though they may be useable as nails.   The master smith, upon reviewing his work may be perfectly OK with this result, given that *he* notices that the batch has subpar items in it but his customers likely won't.  As the journeyman smith progresses in experience, his percentage of flawed nails decreases to the point where even a discerning customer would be perfectly satisfied most of the time - at which point he should be a master.

b) Master Smith asks Journeyman smith to come in early to get the smithy set up for the day.  Journeyman smith notices once he has the forge fired up that there are people in the yard of the smithy.  He peers through a crack and sees deadly enemies of the master smith from his previous profession as a thief.  He needs to spike the door!  There are no nails around and he has to whip one up while the enemies are looking around and dismounting (This can be done, I assure you because I have done it myself).  At this point, his skill roll determines actual success or failure - did he make the nail in time.  We all know that if he had all day, the nail would be a piece of cake.

c)  The door gets spiked OK.  The baddies are pounding on the door, but he knows it will hold for at least a while because he whipped up a few more nails in the meantime and pounded them in.  Now, he needs something to defend himself with.  A half finished sword head lies in the corner, waiting for the last round of hammering and quenching, then polishing and grinding and slipping on the grip over the tang.  He has about a half an hour to do it, so can he get it done?  This is more difficult than just pounding a nail, but the time constraint is exactly the same.  His roll is harder, say with a 20% penalty only because most of the work is done (grip and knob are waiting).  Failure means he didn't get it done in time.  

If he had all the time in the world, he may still make a sword out of it 100% of the time, but the master coming in would look at it and blow up if he failed.  That last pounding and quenching, improperly done, can ruin the edge of the sword forever, wrecking all  the master smith's work even though it worked fine for dispatching the baddies on this occasion.

[Blackguard]

BTW, thanks Andrew for that link.  I found the text to be very educational.