Erom

2009-12-06, 06:01 PM

From my notes while building a 4e world:

I rejected initially the fact that all town guards are 1st level fighter NPC's. So I decided to do some catgirl-killing and run the numbers. Now, I'm not trying to simulate reality here, I'm just trying to see what falls out as a result of using estimates from the published 4e material. This is versimilitude, not realism.

Assumptions:

Guards start at lvl1 and advance like adventurers (Fighting in 4-5 person groups.)

Using the 1 guard to 100 civilians ratio from the DMG, a village will have about 10 guards.

Defending a village and enforcing law generates 3 1st level and 1 2nd level encounter per year.

A guard has a 10% chance of biting it in a given encounter. Dead guards are replaced by new 1st level guards.

A guard who has been in the guard for 10 years (40 encounters) and survived will retire, being replaced by a new 1st level guard.

The results:

A given guardsman has:

* a 34.9 % chance of surviving from 1 to 2

* a 28.2 % chance of surviving from 2 to 3

* a 22.9 % chance of surviving from 3 to 4

* a 1.5 % chance of surviving till retirement

A given village therefore probably has:

* 6 lvl 1 guards

* 3 lvl 2 guards

* 1 lvl 3 guard

Which is actually pretty close to the normal approximations people use.

I'd like to improve this model. Right now we're assuming a 10% chance to die in each encounter. For one, as shown above, that's kind of high lethality(who would volunteer to be a guard if only 1 or 2 in 100 live to retirement?), and for two, it needs to scale by level, both of guard and of encounter.

Does anybody have any good estimates or sources of data for death rate given party level X and encounter level Y?

I rejected initially the fact that all town guards are 1st level fighter NPC's. So I decided to do some catgirl-killing and run the numbers. Now, I'm not trying to simulate reality here, I'm just trying to see what falls out as a result of using estimates from the published 4e material. This is versimilitude, not realism.

Assumptions:

Guards start at lvl1 and advance like adventurers (Fighting in 4-5 person groups.)

Using the 1 guard to 100 civilians ratio from the DMG, a village will have about 10 guards.

Defending a village and enforcing law generates 3 1st level and 1 2nd level encounter per year.

A guard has a 10% chance of biting it in a given encounter. Dead guards are replaced by new 1st level guards.

A guard who has been in the guard for 10 years (40 encounters) and survived will retire, being replaced by a new 1st level guard.

The results:

A given guardsman has:

* a 34.9 % chance of surviving from 1 to 2

* a 28.2 % chance of surviving from 2 to 3

* a 22.9 % chance of surviving from 3 to 4

* a 1.5 % chance of surviving till retirement

A given village therefore probably has:

* 6 lvl 1 guards

* 3 lvl 2 guards

* 1 lvl 3 guard

Which is actually pretty close to the normal approximations people use.

I'd like to improve this model. Right now we're assuming a 10% chance to die in each encounter. For one, as shown above, that's kind of high lethality(who would volunteer to be a guard if only 1 or 2 in 100 live to retirement?), and for two, it needs to scale by level, both of guard and of encounter.

Does anybody have any good estimates or sources of data for death rate given party level X and encounter level Y?