-	-	-	-	-
Dragon 133	-	-	-	Dragon

Balin, the fearless svirfneblin, moved
quietly down the corridor. Somewhere
ahead lurked his foe, a drow elf. Balin
carefully edged around a corner and was
suddenly face-to-face with his foe. . . .

Player 1: "Okay, Balin surprises on 9 in 10. Is he surprised?"
Player 2: "Hey, wait a minute! I've got a
drow elf who's only surprised on 1 in 8.
How can Balin surprise my drow on 9 in
10?"
DM: "Well, uh, um. . . ."

With the expansion of player characters
into new racial types and character classes
with unusual surprise values, the time-honored
method of determining surprise
can quickly become a headache for many
DMs. A scan of the DMG (pages 61-62) or the PH
(pages 102-103) reveals what to do in
those cases where both the surprised
value (How often am I surprised?) and the
surprising value (How often do I surprise
the other guy?) can be expressed using a
6-sided die. However helpful this discussion
may be when a ranger encounters a
wererat, it doesn't answer the question of
what happens when you have an encounter
between a drow elf and a svirfneblin.
The solution lies in converting die rolls
into decimal percentages. For example, a
"normal" surprise roll is 2 in 6, or approximately
33%. The value for a party that is
surprised on a 1 in 6 is approximately
17%. A reduction value of 17% is constant
in all surprise conditions not using 1d6
rolls; the DMG (page 62) illustrates this by
showing that a 5-in-6 chance to surprise is
reduced to 4 in 6 when the party being
surprised is normally surprised only on a
1 in 6 (4 in 6 = 67%, 3 in 6 = 50%; therefore,
67 - 17 = 50%).

So, if the character who is surprised on
a 1 in 6 has a constant reduction of 17%,
then a character who is surprised on a 1
in 8 should have a constant reduction of
21%. If the svirfneblin surprises normally
on a 9 in 10 (90%), then he would surprise
the drow elf 90 - 21 = 69% of the time.
For those who claim that this number is
too low (after all, Unearthed Arcana does
say that the deep gnomes surprise 90% of
the time), counter with the position that
the drow elf picks up 21% more clues
(sounds, odors, that little prickly feeling on
the back of your neck when someone is
watching, etc.) than the average individual.
For those who say it is too high (since
drow are surprised only on a 1 in 8),
repeat that a svirfneblin normally surprises
90% of the time and that the DMG (page
62) firmly states that a party surprised on
a 1 in 6 has only an additional 17% (1 in 6)
in their favor "and not a 50% better chance."

Table 1 lists a matrix of surprise conditions.
Since monks (from either the PH or OA)
lower their chances of being surprised for
every level above 1st level, they should use
the normal 2-in-6 line and subtract their
surprise bonus from this value. Also, since
the application of silence (as per the
second-level cleric spell silence 15? radius
or a magical item) or invisibility (by whatever
means) adds an additional 1-in-6
chance of surprising each (as per the
Players Handbook, page 103), characters
using such powers should increase the
values shown in Table 1 by 17% apiece.
If the player rolls less than or equal to the
listed percentage value, his character is
surprised. If both or neither party is surprised,
then the encounter progresses normally.
However, if one of the parties is
surprised, the number of surprise segments
lost to this party must be determined. This is
done by cross-referencing the same roll used
to determine surprise for each party and its
surprise factor using Table 2. The surprised
party subtracts its factor from the surprising
party?s factor. Treat all negative results as
zero.

Consider the following example: Balin
the svirfneblin encounters the drow elf.
Balin is normally surprised on a 1 in 12.
However, the drow, being an elf and not in
metal armor, surprises on a 3 in 6. According
to Table 1, the drow elf has only a 25%
chance of surprising Balin. Likewise, the
elf, who is ?normally? surprised on a 1 in
8, has a 70% chance of being surprised by
Balin. The drow rolls 15 and Balin rolls 37.
This means that the drow is surprised.

Checking Table 2, we find that a roll of 15
yields a surprise factor of 1 and a roll of
37 gives a surprise factor of 3. Therefore,
the drow is surprised for 2 (3 - 1) segments.
Of course, this result may be modified
by the PCs' dexterity reaction scores,
as per page 11 of the PH.

Changing surprise values from straight
fractions of die rolls to percentages does
cost a little in terms of complete accuracy.
However, it also allows the DM to handle
better those surprise conditions which
cannot be translated into fractions of 1d6
without compromising game play. This
enables a DM to tailor surprise conditions
and encounters to the circumstances, and
not fudge the surprise factors.

Table 1: Chances of Being Surprised
                                                 Surprises on a

Surprised on a	2/6 (33%)	3/6 (50%)	4/6 (67%)	6/8 (75%)	8/10 (80%)	5/6 (83%)	7/8 (88%)	9/10 (90%)
1 in 20 (5%)	5%	22%	39%	47%	52%	55%	60%	62%
1 in 12 (8%)	8%	25%	42%	50%	55%	58%	63%	65%
1 in 10 (10%)	10%	27%	44%	52%	57%	60%	65%	67%
1 in 8 (13%)	13%	30%	47%	55%	60%	63%	68%	70%
1 in 6 (17%)	17%	33%	50%	59%	64%	67%	72%	74%
2 in 6 (33%)	33%	50%	67%	75%	80%	83%	88%	90%

Notes
Monks use the 2-in-6 (33%) row. Subtract 1% if the monk is 2nd level or ((2% x (level -2) + 1) if 3rd level or greater.
If the encountered creature is silent, add 17% (1 in 6) to the value in Table 1.
If the encountered creature is invisible, add 17% (1 in 6) to the value in Table 1.

Table 2
Surprise Factors

Surprise roll	Surprise factor
1-17	1
18-33	2
34-50	3
51-67	4
68-83	5
84-00	6

-
 

1. "The solution lies in converting die rolls into decimal percentages." - Dragon 133.74
2a. Key concept: CR = Constant Reduction (17% for being surprised 1 in 6, 21% for being surprised 1 in 8, etc.)
2b. Key concept: CI = Constant Increase (as above).