Does anybody know the exact odds of a particular entry in randomevents.txt being triggered in, say, a given month?
-
We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
It depends on various settings of your country, your size, the game year, many other factors. The more random events you qualify for, the less likely it is that a particular one will happen.foolcow said:
I guess I wouldn't be much off if I said that this probability is usually somewhere around 1% per year (for those events you qualify for, of course. Otherwise it's zero). I don't think anybody can give you the exact probability.
robin74 said:
Couldn't we try to calculate it?
If we know how often we got a random event that would be a start. I made a little try to calculate that once. I simply analysed the history log in my save file. IIRC it turned out to be slightly less than once a year. If we then count the number of possible events we will come up with a figure that should be close to the truth. But of course, the probability for some events depends on the DP settings.
From my guide:
From the complete list of random events, your nation can only get a sublist due to religion, DP positions, and date requirements for many random events. If you are under the impression that you get the same events over and over, it is probably because your sublist has become relatively small. The way it works, after one year, a check is made to see if you are to receive an event. If you do, an event will be chosen and checked, if it doesn't apply, a new event is chosen and checked, and so on.
It is said that random events have a counter of over a year, so if you save and reload every 11 months you don't get them.
This info was collected from the forum. I cannot guarantee it is 100% accurate, but it does agree with my observations. Hope it helps. If it is wrong I would like to know.
From the complete list of random events, your nation can only get a sublist due to religion, DP positions, and date requirements for many random events. If you are under the impression that you get the same events over and over, it is probably because your sublist has become relatively small. The way it works, after one year, a check is made to see if you are to receive an event. If you do, an event will be chosen and checked, if it doesn't apply, a new event is chosen and checked, and so on.
It is said that random events have a counter of over a year, so if you save and reload every 11 months you don't get them.
This info was collected from the forum. I cannot guarantee it is 100% accurate, but it does agree with my observations. Hope it helps. If it is wrong I would like to know.
We could but:Daniel A said:
Now, it's a problem, because the number of possible events changes as the settings change. It will be different in 1450 and in 1750. It will be different if you're full serfdom and narrowminded, and different if you're free subject and innovative.
The way I would go about it would be to pick the event that is completely random and doesn't have any triggers (like Excellent Minister for example), and check how often it happens in any given game. That's really what you want - you're always eligible for this event, the question foolcow's asking is essentially "how often should you expect it in a normal game". Well, you could take several completed games and simply count it. True, the probability will differ for different settings, but if you want to get just one number, then the rough estimate obtained this way would be as close as one can get.
But my point is - we might try to estimate it (my off-the-top-of-my-head estimate would be 1% per year), but it would only be a rough estimate. Since foolcow was asking for the exact probability - I don't think anybody will give him that.
You can calculate a maximum and minimum probability. But within that range it will change constantly as events are taken in and out of your list of possible random events.
The exact probability for a given moment can also be calculated, but it would be a pointless exercise
The exact probability for a given moment can also be calculated, but it would be a pointless exercise
I tried to calculate.
And I abandonned. It's just too much pain keeping track of all the saves, all the random events, etc.
I agree with robin74 : the closest approximation is 1 event per year (from 10 to 15 months between two events, generally), and any country usually qualifies for a hundred events, so an average of 1% for any particular event.
Off course, when you have the same event duplicated in the file, like the Explorers events, you get more chances to get one (4% approximately for this one, if you qualify for all triggers).
And I abandonned. It's just too much pain keeping track of all the saves, all the random events, etc.
I agree with robin74 : the closest approximation is 1 event per year (from 10 to 15 months between two events, generally), and any country usually qualifies for a hundred events, so an average of 1% for any particular event.
Off course, when you have the same event duplicated in the file, like the Explorers events, you get more chances to get one (4% approximately for this one, if you qualify for all triggers).
Re: % chance of random event
Thanks for the replies, everyone.
I had assumed that a separate probability check (this is the number I was looking for) was made for each event in the list, and then a check was made for whether that country qualified for it. From Fodoron's reply, it seems that isn't the case.
Thanks for the replies, everyone.
I had assumed that a separate probability check (this is the number I was looking for) was made for each event in the list, and then a check was made for whether that country qualified for it. From Fodoron's reply, it seems that isn't the case.
lawkeeper said:
I took a bunch of data from a save game a while back. My best estimate of the interval between random events is near 400 days - or call it 13 months.
The interval is not purely random. By which I mean that the probability of getting a random event on a particular day is very much related to the time since you last had a random event.
The distribution of the intervals isn't flat either, you're more likely to get an interval near the average than far from it. I think I used a gaussian distribution that fit reasonably well. But the data were too sparse to be sure what the distribution actually is. The standard deviation was somewhere near 30 days.
I have run tests that prove that the number of random events you get is independent of the number you are eligible for.
It sounds like there's a reasonably constant chance of getting *any* random event - something like once every 12-16 months - and then it's just a matter of the number of events you are eligible for and which one of those the game chooses.
Yes, I concur, I've noticed the same gaussian curve, most (80%) intervals were in the 12-14 months. 10 and 15 were just the more extreme intervals I got.Isaac Brock said:
Lawkeeper,
Regarding the 1% chance for a specific event.
IMO the best event you can get is the exceptional year. I always count them for myself in my games. I therefore believe I have som vague kind of knowledge how often you get it and I'd say it's closer to once every 50 years than to one every 100 years, at least in the first 200 years. I almost always get at least two during the 15th century. Does the number of events rise quickly when you enter the latter part of the game?
I will check some savefiles tonight if I remember.
Regarding the 1% chance for a specific event.
IMO the best event you can get is the exceptional year. I always count them for myself in my games. I therefore believe I have som vague kind of knowledge how often you get it and I'd say it's closer to once every 50 years than to one every 100 years, at least in the first 200 years. I almost always get at least two during the 15th century. Does the number of events rise quickly when you enter the latter part of the game?
I will check some savefiles tonight if I remember.
There're several events.Daniel A said:
One without any trigger.
One with years 1400-1549 ; another for years 1550-1649 ; and a last for years 1650-1799.
So, that makes 2 available events at all times (except the last 20 years). Counting an average of 100 events available, that makes 2%, or an average of one every 50 years.
lawkeeper said:
I am not so sure about that. Once all your sliders are at 10 or 0 and the date is late (usually goes together), you qualify for a lot less events. At the last years is when I notice that events tend to repeat a lot more. This might indicate a chance significantly higher than 1%. I am not going to check it though, so if you disagree I will concede.
From a very quick check, I can say the minimum minimorum is 73 (with 2 as margin of error) events, provided you're at peace, at stability +3, nor christian neither muslim, with less than 30 provinces, in the 20 last years of the game, and have moved your DP-sliders to trigger the less numerous events (which isn't very wise when you look at LAND, CENTRALIZATION, INNOVATIVE, and especially the combinations). (you could probably get less by cheatingFodoron said:
your DP-sliders at the beginning, but even so you won't be able to go under 50)At 73, the chance is 1.36%. So, this is not much difference, as often you'll qualify for more than 100 events.
And like I said :
I spoke of an average, and usually qualifying for 100 events.myself said: