Random event triggering: A small note

[MichaelM]

I'm posting this because it has come up on the forums and on Discord several times.

At game startup, the game schedules a random event for each country (excluding non-"real" countries like Pirates). These events are scheduled to happen roughly 1-3 years after game start, and their triggers must be valid. When a random event's scheduled time hits, the event is either fired or not fired depending on the trigger, and the game then schedules the next one for 1-3 years later regardless of whether it fired or not. And on and on through the game.

The important point, which I don't think I've seen published even in Havard's guide, is that a random event's triggers must succeed BOTH when it is scheduled AND when it fires 1-3 years later. That means that a random event with a "random = 10" trigger does NOT have a 10% chance of firing in a given cycle. It has a 10%*10% = 1% chance of firing.

Similarly, any trigger targeting a volatile property of a country, e.g. a NOT = { stability = -2 } trigger requiring stability to be at -3, has a much lower chance of firing than one might believe by looking at the event code. Not many countries will be at -3 stability for long enough to fire such an event.

Random events are best when targeting properties that change rarely. Domestic policy, religion, culture, country size, and geographic location are examples of good triggers. Use others with good judgment and at your own risk.

 

[Trmdrg]

...

This is terrible news for me. Yes, I am a slowpoke. Such slow. Such a poke.
For the past few months, I have been working on a "revolutionary" random event system that uses "random = xx"% and a lot of triggers for each event, trying to use all the "new" engine features to achieve a new level of FtG event coding. I wanted random events to depend on the current game situation and not be as out of place and out of time as in the original FtG. I was inspired by one post on this topic here on this forum.
Now that I have learned about this crazy weird system, I realise that it is almost impossible and, unfortunately, makes little to no sense.

Edited: After much thought and with a cool head, I say that this is a dubious and strange, but interesting system. This, of course, makes sense, but the system is very difficult to work with.

 

[MichaelM]

This is terrible news for me. Yes, I am a slowpoke. Such slow. Such a poke.
For the past few months, I have been working on a "revolutionary" random event system that uses "random = xx"% and a lot of triggers for each event, trying to use all the "new" engine features to achieve a new level of FtG event coding. I wanted random events to depend on the current game situation and not be as out of place and out of time as in the original FtG. I was inspired by one post on this topic here on this forum.
Now that I have learned about this crazy weird system, I realise that it is almost impossible and, unfortunately, makes little to no sense.

Edited: After much thought and with a cool head, I say that this is a dubious and strange, but interesting system. This, of course, makes sense, but the system is very difficult to work with.

You can still use random triggers, but the math will be different. E.g. if you want a 70% chance for a particular event given some condition, you have to set the random trigger to the square root of 0.7, or "random = 84". Then 84% when scheduling and 84% when triggering gives a total chance of 70% (or 70.56% if we want to be precise).