I thought I'd post the house rules I'm using for weather. These are designed to reduce the fluctuations caused by dice rolling, so that you are quite certain to get something close to the expected number of Fair, Poor, and Severe months in each weather zone, and you always get weather that could have been obtained using the normal rules. It will just be more "average" than usual. The motivation is to reduce the significance of good or bad weather rolls, which otherwise might have too much importance in deciding the course of the game, at least to my tastes.
To use these rules, you follow an algorithm, that I'll describe in detail. Unfortunately, the algorithm inevitably seems very complicated when described in detail, so you'll think I'm crazy, but it's really not complicated to implement once you understand it. If such a long description is too frightening, I'm sorry, but I think the benefits of a detailed description are worth it.
Each weather zone (Warm, Mild, and Cold) will have a triplet of 3 "adjustments" for each of the 3 possible weather types (Fair, Poor, Severe). For example, the Mild weather zone might have the adjustments (+1, +1, -2), which means fair weather and poor weather currently will be a little more likely than usual, and severe weather will be somewhat less likely. The adjustments for a zone must always be integers that sum to zero, and they begin a scenario at (0, 0, 0).
For each triplet of adjustments, in each weather phase of each month you first "decay" the adjustments for weather types that are impossible in a "decay" sub-phase, than roll for the weather using adjusted odds in a "determination" sub-phase, and then update the adjustments based on the rolls that occurred in an "update" sub-phase. We'll begin with the decay sub-phase.
When the decay sub-phase of a new weather phase begins, you first go to the printed weather chart, and for each zone, check which weather types are impossible during that month. Incidentally, in this system, you will never obtain a weather type for a zone that is impossible according to the printed chart, which is a nice feature--the sequence of weather types you get is always one that could have possibly been rolled for. For those weather types that are impossible according to the charts (e.g. during Jul-Sep, the poor and severe weather types in all zones), you first decay the adjustments so that they are no more than +2, and no less than -2. (If they are already between -2 and +2 inclusive, nothing happens.)
The precise way you do this depends on the number of impossible weather types in a zone. If the number of impossible weather types is two (e.g. during July for every zone), every increase or decrease in the impossible weather types is compensated for by a corresponding decrease or increase in the one possible type. Thus, if the adjustments in one of the zones were (+6, -3, -3) going into July, they would be (+4, -2, -2) after this initial decay sub-phase, while if they were (+2, -4, +2) going in, they would be (0, -2, +2) after the decay sub-phase.
If the number of impossible weather types in a zone is one, the adjustment for every increase or decrease in the impossible weather type is alternately compensated for by the other two weather types, starting with the type that is most similar to the impossible one, where Poor is most similar to Fair, Severe is most similar to Poor, and Poor is most similar to Severe.
Thus, for example if Severe is impossible for a zone, and the adjustments for that zone are (-3, -2, +5) before the decay sub-phase, the resulting adjustments after the decay sub-phase would be (-2, 0, +2).
Note that the sum of the three adjustments in a triplet *must* always be zero.
After the decay sub-phase, we enter the determination sub-phase. For each zone, we first take the printed chart, and add the number of "raw" chances for each weather type to the adjustments to get an adjusted number of chances. For example, in December in the Mild Zone, the printed chart says that Poor weather results from die rolls of 1-4, and Severe weather results from 5-6, so the raw chances are (0, +4, +2). If the adjustments after the decay sub-phase were (-2, +1, +1), then the adjusted chances would be (-2, +5, +3).
At this point, one takes all those weather types that are possible according to the printed chart and have a positive number of adjusted chances, and rolls a die to find out which type occurred, where the chance of each type was the adjusted chance. In our example, there would be 5 adjusted chances of poor weather, and 3 adjusted chances of severe weather, so we need to do something like roll a 10-sided die, saying it's poor on 1-5, severe on 6-8, and re-rolling on 9 or 10.
OK, now that we've determined the weather types, we need to update the adjustments in the "update" sub-phase. The way we do this is increase each adjustment by the number of *raw* chances it had, and then reduce the adjustment of the chosen weather type by 6. Thus, in our example, if we rolled for severe weather, we would first increase our adjustments from (-2, +1, +1) to (-2, +5, +3) [because the raw chances were (0, +4, +2)], and then since we rolled for severe, we would reduce the adjustment for severe by 6 to get an adjustment of (-2, +5, -3).
If you work through an example, you'll see that weather types that are chosen become less likely in the future, so that you inevitably get something close to "average" weather. The decay sub-phase is there so that for example the results of the winter season of one year do not excessively influence the winter season in the next year.
I'd be happy to answer questions or give more examples, if anybody cares.
- Last edited Sun Dec 14, 2014 6:50 am (Total Number of Edits: 2)
- Posted Sun Dec 14, 2014 6:30 am