Re: Jock-o-Trades [sic] (and MegaTraveller Tasks)

> From: (Charles Martin) [email removed]

> I don't think much of the MT method for handling JoTs.  Allowing retrials
> up to the level of JoT is an immensely powerful advantage.  Let's look at
> the numbers (you have to do it sometime!):

Charles has given a very persuasive and (in my opinion) compelling
argument for doing away with the MT method for handling JOT.  However,
I'd be curious to see how the numbers crank out for one change in the
set of assumptions.

The change is this: Charles has assumed that any die roll (+DMs) may be
rerolled if it has failed, even in the cases of extreme failure and
fumble rolls (extreme failure is when you miss the throw by more than
two; ex: you need an 8+ and you throw a 5-; fumble means rolling a
natural 2 on two dice, and always fails: whether it's exceptional or
normal failure is determined normally, I think).  The rules imply this
assumption.  What if you only allow JOT retries if the failure was
normal? This way fumbles and exceptional failures terminate your chances
at JOT-style re-rolling.

Okay, let's crank out some numbers:

> In JTP's example, the hot pilot gets +4 to dodge asteroids (Pilot-3, Dex-
> terity 7).  In a Routine task (7+), odds are 35/36 (97%).

> The total novice (Dexterity 7) gets +1; dodging is Difficult (11+), so
> the odds are 6/36 (17%).

> The nimble novice with JoT-3 now gets [up to] 3 additional trials!  What are
> his odds?

They are 1 - (1 - odds for first trial) * (1 - odds for second trial) *
...  * (1 - odds for fourth trial).

Here's where we get tricky.  The odds for the first trial are 17%, as
above.  The odds for subsequent trials are 17% * (1 - odds for
exceptional failure or fumble in previous trial).  Since exceptional
failure or fumble occurs on a roll of 3 less than Difficult (11+) with
DM+1 (Dex 7) or less (natural 7-), the odds for exceptional failure are
21/36 = 58%.  Subsequently,

    odds for first trial = 17%
    odds for second trial = 17%*(1 - odds for failure) = 17%*(1-58%) = 7.1%
    odds for third trial = 17%*(1 - odds for failure)^2 = 17%*(1-58%)^2 = 2.9%
    odds for fourth trial = 17%*(1 - odds for failure)^3 = 17%*(1-58%)^3 = 1.2%

Hence,

	First Trial (unskilled): 1-(1-17%) = 17%
	Second Trial (JOT-1): 1-(1-17%)(1-7.1%) = 23%
	Third Trial (JOT-2): 1-(1-17%)(1-7.1%)(1-2.9%) = 25%
	Fourth Trial (JOT-3): 1-(1-17%)(1-7.1%)(1-2.9%)(1-1.2%) = 26%

Which is much better than with the non-restricted reroll:

>     1-(1-17%)**4 = 52%.

Here, JOT-N + unskilled is nowhere near as good as Skill-0.  Keep in
mind that this is a Difficult task.

> But wait!  What if our the novice had picked up Pilot-0 along the way?
> Then the task would be Routine (7+), DM+1, and the odds are 26/36 (72%).
> But he still gets four trys at it!

Okay.  The odds for the first trial are 72%, as above.  The odds for
subsequent trials are 72% * (1 - odds for exceptional failure or fumble
in previous trial).  Since exceptional failure or fumble occurs on a
roll of 3 less than Routine (7+) with DM+1 (Dex 7) or less (natural 3-),
the odds for exceptional failure are 3/36 = 8%.  Subsequently,

    odds for first trial = 72%
    odds for second trial = 72%*(1 - odds for failure) = 72%*(1-8%) = 66%
    odds for third trial = 72%*(1 - odds for failure)^2 = 72%*(1-8%)^2 = 60%
    odds for fourth trial = 72%*(1 - odds for failure)^3 = 72%*(1-8%)^3 = 55%

Hence,

	First Trial (Pilot-0): 1-(1-72%) = 72%
	Second Trial (JOT-1): 1-(1-72%)(1-66%) = 90%
	Third Trial (JOT-2): 1-(1-72%)(1-66%)(1-60%) = 96%
	Fourth Trial (JOT-3): 1-(1-72%)(1-66%)(1-60%)(1-55%) = 98%

Wow, slightly better than Pilot-3! In general here, JOT-N + Skill-0 is
slightly better than Skill-N, for a Routine task.

> But wait, there's more!  Suppose they are trying to fix the jump drive
> simultaneously.  The task is now Difficult (11+).

>     Pilot-3: DM+4, odds 21/36 = 58%.
>     Pilot-0, JoT-3: DM+1, odds 6/36 = 17%, after four tries 52%

We get after four tries 26%, which is better than the 17% Pilot-0 only
person, but not as good as Pilot-1 only (which is 27%).  Here, JOT-N +
Skill-0 is at least Skill-0 but not near as good as Skill-1; again, this
is a Difficult task.

With Charles' assumptions:

>     JoT-N is NOT EQUAL TO Anything-N; however JoT-N and Foo-0 is FAR
>     SUPERIOR to Foo-N normally, and ALMOST AS GOOD under stressful
>     circumstances.

With my restrictions (no retry on exceptional failure or fumble):

	JoT-N is NOT EQUAL TO Anything-N, unless the task is routine or
	simple; JoT-N and Foo-0 is about Foo-N for simple and routine
	tasks, only a little better than Foo-0 for harder tasks.

In this way, Jack-o-T can still be used even if you have skill in some
area, reflecting the general resourcefulness of these people.  It helps
anyone out alot for simple or routine tasks, but when the task becomes
difficult or formidable, the effect of Jack-o-T becomes much less.  I
feel that Jack-O-T should be useful in all situations a character should
encounter, not just those in which he is totally unskilled.

However, even with this (close) interpretation of the rules, JoT is
still overpowering for simple and routine tasks.  I would still not
relish refereeing a character with more than JoT-1.  They would be quite
a Mr.  Bond (James Bond).

> I've only seen MT very briefly.  How do the rankings go?  (What is the
> throw for Formidable?)  And what do people think of the "interrupt" idea?
> I couldn't figure out what they had in mind.  Anyone want to explain how
> interrupts work?

The basic skill levels are:

	Simple		3+
	Routine		7+
	Difficult	11+
	Formidable	15+
	Impossible	--

No accumulation of DMs may exceed +/-8.  Skill level for a task usually
increases one level where there's a total lack of skill, unless the task
is marked "(unskilled OK)".  Lots more detail that I can't remember.

I haven't looked at interrupts that much, but as I understand it, at the
beginning of each combat round, each "side" has the ability to make one
interrupt.  Individuals on each side expend action points one-at-a-time
in strict order.  If the opposing side makes a single-action-point move
that is unfavorable to your side, you may expend your interrupt by
delaying the opponent's action while you make an action.  Hence you can
get a stack of interrupts.

This is supposed to be useful in "covering" a comrade; your comrade
moves out into the hallway, expending an action point.  Opponent
declares he's shooting at your comrade.  You declare an interrupt to
shoot first.  You shoot, THEN opponent shoots.

This doesn't always help: your comrade moves out into the hallway;
opponent declares he's shooting; you declare an interrupt to shoot
first; opponent's comrade declares an interrupt to shoot you.
Opponent's comrade fires, THEN you fire, THEN opponent fires.

As usual, I don't have my books with me, so this is from memory.

James