random events guide

z0wb13 · Mar 13, 2009

update to 3.17

comments welcome. i like random events and always play with them on. for those that dislike random events due to barbarian uprisings, i address this concern in the barbarians section.

the gambits, for the most part, are common enough events that, with a little forward planning, you can expect to contend for most them in about 80% of your games.

furthermore, good and bad events have been separated for convenience, as have events dealing with unit promotions.

credits to moderator ori, et al, for mining all of this data, ShannonCT who knows more about maths than i do, and PieceOfMind for helping me to think some of this stuff out.

if you need more general information on how all of the active/weight scores works, scroll down.

gambits

Spoiler :

barbarians

Spoiler :

z0wb13 · Mar 13, 2009

updated to 3.17 to include Solver's events.

good

Spoiler :

bad

Spoiler :

units

Spoiler :

z0wb13 · Mar 13, 2009

quests

Spoiler :

civ-specific events and disabled events

Spoiler :

Soirana · Mar 14, 2009

Spoiler :

Kesshi · Mar 14, 2009

z0wb13 said:
Event77
Ancient Olympics
Prereq: POLYTHEISM AND State Religion is NOT Judaism OR Christianity OR Islam
Obsolete: MACHINERY or FEUDALISM or MUSIC or PHILOSOPHY or CIVIL_SERVICE or THEOLOGY
Active/Weight: 75/400
Result:
1.Nothing
2.pay 20 gold AND gain +1 attitude with all neighboring AIs

z0wb13,

Do you know how this is affected by the "Choose Religion" custom game option?

troytheface · Mar 14, 2009

Like the idea, random events are just as important as espionage in so far as getting the ball rolling.

Which tech path allows for the most random events to occur, or adds to upping the odds.
(Sailing, animal Husbandry are two i can think of)

and too- "the best example of why you should keep at least 10 gold on hand at all times"
..i say always have 100 you don't touch. (for more choice)

z0wb13 · Mar 14, 2009

@ Kesshi: i don't think that it is affected at all.

oyzar · Mar 14, 2009

Why is this in this forum?

PieceOfMind · Mar 14, 2009

oyzar said:
Why is this in this forum?

Are you trying to suggest it shouldn't be here?

IMO it's definitely worthy of being called a discussion on strategy, as some random events are good enough to alter your strategy.

Or maybe you're implying that random events should not be part of any strategy. Should combat and GP births be part of strategy, in that case?

@z0wb13,
Thanks for making this thread. I'll have a closer look when I have more time, but I think the way you have categorised the random events is exactly what I was looking for. :goodjob:

.

oyzar · Mar 14, 2009

I ment shouldn't it be in the strategy article section...

PieceOfMind · Mar 14, 2009

Actually, @z0wb13, just one thing...

The active number (eg. 45%) means the event is active or possible in 45% of games. It does not mean the event will happen or occur in 45% of games. This would be obvious especially if your games lasted an average of only ten turns.

The distinction is very important, so I thought I'd mention it.

oyzar said:
I ment shouldn't it be in the strategy article section...

Oh, maybe.

Bleys · Mar 14, 2009

I keep ori's legendary post on a bookmark, but I like the way z0w is sorting them here.

z0wb13 · Mar 14, 2009

PieceOfMind said:
Actually, @z0wb13, just one thing...

The active number (eg. 45%) means the event is active or possible in 45% of games. It does not mean the event will happen or occur in 45% of games. This would be obvious especially if your games lasted an average of only ten turns.

The distinction is very important, so I thought I'd mention it.

Oh, maybe.

i understand. the events in the gambit spoiler are mostly high active score events.

so, like the federal reserve has a 90% active score; it is going to appear in 90% of my games. but that's like almost every single game that i play. i understand that there is a 10% chance difference between always and 90%.

but in the case of the federal reserve: many players run free market , most research corporations at some point in their game, however, very few players keep 1000 :gold:

on hand for many turns. these players have a 0% chance of triggering this event in any game that they play because they don't meet a rather strange prereq.

so but then you say, well corporations is an industrial era tech, you only get a 6% chance of this popping, and that chance is spread out between all of the other active events, so the chance that it will trigger is like less than .01%. and that's true on any single turn. but the game lasts hundreds of turns. i like to think about the odds of happening like the way a :nuke:

half-life works. if you wait for a certain amount of turns, and that all depends on the weight, eventually there will come a point where you can say something like: well if, i keep 1000 :gold:

in the bank for around 25 turns, there is a 50% chance that it will trigger the federal reserve. my math is just totally pulled from thin air here, i just wanted some numbers to illustrate my point.

so there is definitely a few things that i haven't worked out yet, like what the average total weight is in any given game. some of the better math kids could help me with this. i think that you might need to know statistical regression. but i think that the most important number is going to be active score; if something has a low probability of being in the game at all, you could hope to trigger some event over many games and never have it happen.

so like, the experienced captain event provides a unique opportunity to pay for a military academy and it has a 95% chance of being active. i want that event to trigger, so i fulfill the requirements- a drydock and 7xp naval unit in target city- and all i have to do is wait a certain number of turns and it will very likely happen. but there is more too it than that, because you also say that after 10 turns, there is a 20% chance that experienced captain will trigger, after 20 turns, there is a 50% chance it will trigger, and after 30 turns there's something like a 75% chance to get it. after a certain number of turns it becomes statistically impossible that the even won't trigger if it is in that particular game at all, so say you fish for 50 turns or something and don't get it, you can know to stop fishing. again, i need help with my math.

it's like anything else that gives you a bonus when you are the first to acquire it. think of liberalism or physics or every single wonder that you build. so, instead of spending :hammers:

to create an edge over your opponents, you spend turns meeting a few prereqs. all i did was take ori's list and slice it up for the convenience of newer players.

Bleys · Mar 14, 2009

z0wb13 said:
these players have a 0% chance of triggering this event in any game that they play because they don't meet a rather strange prereq.

OK, this is going to sound crazy, but I actually agree with Attacko on this. I think its worth it to make sure to qualify for as many positive events as you can on purpose. The Fed. Reserve event is just one example of it, but troy has mentioned others in his ramblings. Actually, the concept of Random Events is what drives half of Attackos chaos concept. The concept is that its more likely that stuff WILL happen. So up the odds to add as much chaos (events) as possible.

PieceOfMind · Mar 14, 2009

z0wb13 said:
so like, the experienced captain event provides a unique opportunity to pay for a military academy and it has a 95% chance of being active. i want that event to trigger, so i fulfill the requirements- a drydock and 7xp naval unit in target city- and all i have to do is wait a certain number of turns and it will very likely happen. but there is more too it than that, because you also say that after 10 turns, there is a 20% chance that experienced captain will trigger, after 20 turns, there is a 50% chance it will trigger, and after 30 turns there's something like a 75% chance to get it. after a certain number of turns it becomes statistically impossible that the even won't trigger if it is in that particular game at all, so say you fish for 50 turns or something and don't get it, you can know to stop fishing. again, i need help with my math.

With all respect, I don't really agree with the way you are describing things. Firstly, "statistically impossible" is a phrase that doesn't sit well with me. If something has 0.1% chance of not happening (i.e. 99.9% chance of happening) you can still see it, and this is clear when we hear people whine about how they just lost a 99.9% battle unfairly.

I can appreciate that even if an event has, say 1% chance of happening each turn it becomes more and more likely to happen the more turns there are in the game, but one has to be very careful here not to fall for the gambler's fallacy. In other words, be careful not to assume that just because you have not seen an event in the previous turns that it is more likely to appear in upcoming turns. A consequence of this is that as you progress through a game, an event that you might be hoping to see actually becomes less and less likely to appear rather than more likely.

An event I've found to appear in some games several times but never in other games is the discovery of a resource on a mined hill. This event is actually independent of the random events code as far as I know, but it works in a similar way. The chance of it occuring in game increases if the game is longer or more mines are worked during the game. But it is not a given and you certainly cannot expect to see it necessarily in every game, even if it is "statistically" probable (maybe 75% chance).

With the random evenets, the probability of an event happening will depend on its weight and the sum of the weigts of the other active events which have their prereqs met. Even an event that is active in 95% of games could be extremely unlikely if there are many other high-weight events eligible for most of the game. My understanding of the way random events are picked is that first the game somehow decides a random event is going to occur. This part of the process is the part I'm unsure about, in terms of how it works. But once a random event is going to happen, that's when the weights come into it.

So even if an event has a huge weight relative to other events, and is very likely to occur, it won't happen almost every turn - it will happen almost every turn where an event is about to occur. But it's quite possible that they programmed random events to happen more often when there are more events eligible. In other words, I'm guessing the probability of an event occuring is mainly dependent on the sum of the weights of the eligible active events. This information is probably in ori's article.

Bleys said:
OK, this is going to sound crazy, but I actually agree with Attacko on this. I think its worth it to make sure to qualify for as many positive events as you can on purpose. The Fed. Reserve event is just one example of it, but troy has mentioned others in his ramblings. Actually, the concept of Random Events is what drives half of Attackos chaos concept. The concept is that its more likely that stuff WILL happen. So up the odds to add as much chaos (events) as possible.

I agree. Attempting to be eligible for as many positive events as possible, and as few negative events as possible, is the main thing to take away from this discussion. Obviously one should not care too much about positive events that have little gain or are near 0% chance, or bad events that are almost always eligible. But several events appear to be worthy enough to adapt into one's strategy in the hope of giving one greater odds of getting a positive event. The Federal Reserve event was the first one I ever thought about doing this with, because I figured the reduction in inflation was a huge advantage for late-game corporation abuse. But it's now clear there are many events like this, including the Experienced Captain one, where it is not difficult to fulfill the prereqs, and doing so makes other events (both good and bad) less likely to occur. I guess one could describe it as stacking the cards in one's favour.

z0wb13 · Mar 15, 2009

thanks for your comments, PieceOfMind.

PieceOfMind: With all respect, I don't really agree with the way you are describing things. Firstly, "statistically impossible" is a phrase that doesn't sit well with me. If something has 0.1% chance of not happening (i.e. 99.9% chance of happening) you can still see it, and this is clear when we hear people whine about how they just lost a 99.9% battle unfairly.

true, .1% is a statistically significant chance. i wouldn't take .1% and say that it is impossible, but that is only on a single turn. if you cheated and reloaded and waited one turn to make that attack, all other things being equal you would, statistically certainly, win that battle. so i know that my argument is kind of convoluted, but try out these two math problems. they are basically the same question, but:

1+1/2+1/4+1/8+1/16...=?

and

4/1-4/3+4/5-4/7+4/9-4/11+4/13...=?

Spoiler :

PieceOfMind: I can appreciate that even if an event has, say 1% chance of happening each turn it becomes more and more likely to happen the more turns there are in the game, but one has to be very careful here not to fall for the gambler's fallacy. In other words, be careful not to assume that just because you have not seen an event in the previous turns that it is more likely to appear in upcoming turns. A consequence of this is that as you progress through a game, an event that you might be hoping to see actually becomes less and less likely to appear rather than more likely.

where am i going with this? the gambler's fallacy is something like: if you flip a coin 100 times, and it comes up heads 75 times, there is a 75% chance that on the 101st coin-flip it will land on heads. so but i'm not saying that, e.g., on turn 10 of fishing for an event you have a 15% chance of triggering it, but then on the next turn you have a 16% chance.

what i'm saying is that if you fish for 20 turns, you might get a 50% of triggering an event any of those turns. it is not more or less likely that it will occur on the first or last turn. if you then waited 40 turns, for example, your odds would have increased to 75%, supposing that an event is in your game at all. but there is no more likelihood that the event will trigger on the first, middle or last turn than any other turn.

i think that maybe i have fallen for something like the inverse gambler's fallacy by believing that any event is any game at all. but, if you know about what your odds and such are, you can figure that if a event doesn't occur after a certain number of turns that it isn't in your game. the gambler's certainty: vegas wasn't built on winners.

PieceOfMind: With the random evenets, the probability of an event happening will depend on its weight and the sum of the weigts of the other active events which have their prereqs met. Even an event that is active in 95% of games could be extremely unlikely if there are many other high-weight events eligible for most of the game. My understanding of the way random events are picked is that first the game somehow decides a random event is going to occur. This part of the process is the part I'm unsure about, in terms of how it works. But once a random event is going to happen, that's when the weights come into it.

i think that i can clear some of this up, although this is all just speculation based on speculations from the boards.

active: at the start of the game, the computer rolls a 100-sided die for each event. if the die comes up more than an event's active value, then the event is excluded from that particular game.

weight: this works kind of like gpp. say it's time for a gpp to pop, and the chances are like 95% scientist and 5% artist. i don't know how, but the computer decides to give you a great artist. whatever mechanism the computer used to decide this, though, is very likely the same as the one that it uses when picking which events to trigger. so let's say that you know, although you couldn't, that a random event was going to trigger on some turn. the computer would take the weights of all the events that have met their prereqs on that turn, throw them into a hat, and pick one. the weights have about as much influence as they do when choosing great people.

here is an actual example of what i'm talking about:

ShannonCT: But let's say that you know for sure that Man Named Jed is a possible event (no way to know this until it happens), that you park your scout on oil, and that the sum of the iWeights for the other eligible events is 3000. When a random event occurs, the chance of it being Man Named Jed is 5/8. In the classical era, there is a 2% chance of a random event occuring each turn. So there would be a 1.25% chance of Man Named Jed occuring each turn. We want to know what value of t makes (1 - .0125)^t < 0.5. The answer is 56.

this is telling you that after 56 turns, there is a 99.5% chance of this event triggering during one of those turns, if it is in the game at all. this particular event is not really a practical example, but it is illustrative.

on every single game turn, there is a chance of triggering any active random event. the chances increase with the advance of ages, like this:

ancient- 1%
classical- 2%
medieval- 4%
renaissance- 4%
industrial- 6%
modern- 8%
future- 10%

z0wb13 · Mar 15, 2009

troytheface said:
Like the idea, random events are just as important as espionage in so far as getting the ball rolling.

Which tech path allows for the most random events to occur, or adds to upping the odds.
(Sailing, animal Husbandry are two i can think of)

and too- "the best example of why you should keep at least 10 gold on hand at all times"
..i say always have 100 you don't touch. (for more choice)

also: polytheism, calendar, radio, being buddhist and reaching the industrial era. NOT researching astronomy.

Phil725 · Mar 15, 2009

Just to comment on a few events:

Event139
Partisans
Prereq: Emancipation civic AND one of your cities was razed
Obsolete: None
Active/Weight: 35/0
Result:
1.a number of drafted units (equal to city's culture level -1) spawn at the city site
2.half that number of drafted units spawn in the capital

this one is weird because it has a zero weight, but i'm pretty sure that i've seen it happen. i don't seem to recall this event happening more than twice in a single game. but note that the ai can get this event, making city razing more hazardous. the computer usually takes the first option just to harass you.

I believe that this event is glitched (coded wrong?) In that it was intended to occur in 35% of razed cities in every game. However, it seems to occur in 100% of razed cities in 35% of games. When this event is active, it becomes impossible to raze more than a few cities because of the sheer amount of units that build up.

Regarding barb uprisings; I've never heard of someone being wiped out by HA's, by then the whip and any kind of reasonable defense will save you. If anything, HA's can hit at an inopportune time, and maybe raze a poorly defended city or two that you just settled. The Vedic Aryans are the ones that is basically a game over. The problem is that the event can come so early that it is literally unstoppable. On monarch+, everyone starts with archery, and chances are, some zealot will research poly as their first tech. You could theoretically; settle on a hill, build a warrior first, keep your initial warrior in defending (I believe that is the most possible defense you can have,) and still be wiped out by the Vedic Aryans if they attack on turn 12 or whatever. If you play a game normally, there is no way to stop them if they attack soon enough.

The AI will also often get killed by the Aryans, although it isn't guaranteed since they have archers. In a recent game I played, this led to a 30+ city Darius. If this event had, say, BW as a prerequisite, I doubt anyone would have a problem with barb uprisings.

r_rolo1 · Mar 15, 2009

One thing that you skipped in your speech in OP ( and below as well ) is that events scale badly with the number of civs: my favourite example is the plane crash. In a standart map it will happen often but as long as you get more civilizations in game ( by using a bigger map, for example ) the event prereqs almost force that the event will happen much more often..... and with a event that gives a option to gain +1 in diplo, that is game-unbalancing if you push it to the extreme.....

So, rule of thumb: events that can be triggered by a civilization having X tech will happen much more often when the number of civs in game goes up, especially in games with AI having lots of vassals ( as those huge maps tend to be.... ). That takes a lot of the "reserve event" power that the player has in a game with 4 or 5 alive civs.......

Calouste · Mar 15, 2009

Quest9:
Crusade
Prereq: State religion but you do not control the Holy City
Obsolete: None
Active/Weight: 50/1000
Aim:
Declare War on the owner of the Holy City, stay in war until you conquer the Holy City
Result:
1.get 0.5*default number of players for this world size+1 conscript units (4 for standard)
2.build the shrine for your state religion
3.spread your state religion to default number of players for this world size cities (7 for standard)

even with its relatively high probability, this quest will not trigger in many games. i don't know why not, but i have never seen it. however, it has a rather good payoff, and capturing a foreign holy city is almost always a good idea. if you have a stack of units that you've been itching to use, the crusades will help you decide when to attack.

Wouldn't this also have the requirements that the owner of the holy city has to have a different state religion? That would make more sense historically, and also explain why you have never seen it in your games.

random events guide

undead

undead

undead

Emperor

Emperor

Deity

undead

Have quit civ/forums

Drill IV Defender

Have quit civ/forums

Drill IV Defender

Deity

undead

Deity

Drill IV Defender

undead

undead

Warlord

King of myself

Deity

Similar threads