Slitherine

As I understand it, the weather prediction is not 100% accurate. I'm curious what the probability of being accurate is, and does it vary for example to any other factors?

22. Next day's weather is not 100% accurate. (Intended feature)

Rudankort wrote:Yep, this is intended. Unlike PG, where forecast was arbitrary, in PzC it has good accuracy and a fixed chance of error (right now 20%). So, it is good enough to base your decisions on it, but sometimes you will get surprises - just as in real life.

Personally, I think because it's not 100% accurate, the player's should be informed of that fact, in addition to other tooltips such as somewhere that states in plain text exactly what the weather is for the current day/turn.

We need to add tooltips the first weather icon on the left is actual weather, the other on the right is your weather forecast. A forecast by definition is a prediction of course, there is a small change the prediction is not spot on and I think it is variable between 80% and 100% being accurate.

Just like in real life weather forecasting during WW2 was good, but not as accurate as today with all our satellites and more powerful radars, nothing is as unpredictable as the weather.

We will also redo the icons so it is more clear if you can expect sunny or clouded but clear weather, rainy muddy conditions or snowy winter grounds. Probably best is to make them animated.

A forecast by definition is a prediction of course, there is a small change the prediction is not spot on and I think it is variable between 80% and 100% being accurate

Hmm, so that means there isn't a 4-to-1 roll die, but in fact a 0-through-20 roll. So the forecast could hit 99% accuracy?

Correct info is what Kerensky has quoted from my earlier reply: forecast is 80% accurate, period. This means that I roll 1-100 every turn, and if it is 1-20, I pick random weather which is not correct one, but which is possible in this scenario. All such "wrong" forecasts are equiprobable.

Note: this is done for weather only. Ground state is extrapolated from the current state and the new weather prediction. It can still be accurate, even though it is based on wrong weather prediction. So, for ground state error chance is probably less than 20%.