Anyone ever have 3 losses at 96% in a row? Just happened to me.

[AspiringScholar]

I know it's theoretically possible, but within 5 combat rounds on the same turn, 3 of them averaged 96% for me, and I lost all of them. That should be a 0.0064% chance (not counting the other two, which would change the math, but I don't know how to do that, honestly). This is sort of a frustrated rant, but I'm also curious how many people can relate to that kind of terrible luck frequency.

 

[Kaitzilla]

Luck streak feels like

Good luck

[BTS] - Extremely Unlucky Battles

Edit: I did more tests with more battles and concluded that the AI did not cheat. I was just suspicious because of extremely unlucky RNG immediately after raising the difficulty level. I was defending a city from a surprise attack by the AI, and noticed that I lost almost all the battles. I...

forums.civfanatics.com

Bad luck

Post a (tough) Deity map for me to play

@Lain Around 11:30 in your Mansa part 1 you are researching fishing and you've met Shaka who obviously has the tech, since you see the clams are netted. However, you do not gain any bonus:science: since the bonus for "1AI knows the tech" is only 4%, thus you'd need to make at least 24:science...

forums.civfanatics.com

Anyone else lose a Great General at 99.9% or 100% odds of victory?

 

[lymond]

Anyone else lose a Great General at 99.9% or 100% odds of victory?

Yup...more than once

 

[Fine_Distinction]

I know it's theoretically possible, but within 5 combat rounds on the same turn, 3 of them averaged 96% for me, and I lost all of them. That should be a 0.0064% chance (not counting the other two, which would change the math, but I don't know how to do that, honestly). This is sort of a frustrated rant, but I'm also curious how many people can relate to that kind of terrible luck frequency.

yep, has to happen to somebody Thanks for taking one for the team! BTW, even with 96% odds, chances of losing at least one of these 3 fights are pretty substantial - ~12%. Take care

 

[AspiringScholar]

It is interesting to consider the psychology of our cognitive biases WRT things like this. When you have more than 1,000 combats in a game, I guess something with 1 in 1,000 odds happening at least once shouldn't be a surprise, but when we see even something far more likely such as a losing a GG unit on 98% odds, it seems impossible and like the game is cheating.

 

[6K Man]

Everyone who has played Civ4 for a while has experienced this.

 

[D64]

Almost. I lost four consecutive 90.9% battles just today...

 

[NickGabben]

I beat a spearman with an immortal yesterday at 1.3 % odds.

 

[Kiwirob]

i recall a post long ago that said that the game balances against your performance - fall far behind the AI and you start getting lucky, and outperform and bad luck follows you.

i saw a superb Deity game that Henrik posted on YouTube several months back that seemed to back that up - Henrik made an awesomely good start (the best I’ve seen anyone do) and was absolutely rolling the AI. Then suddenly he got three sub 1% losses almost in a row in key battles. Three sub 1% rolls in close proximity is difficult to come by randomly. To have it happen when the human player was doing so well seems more than coincidental.

Just a suspicion on my part, since I don‘t know the inner mechanics of the game, and at my standard of play I’m in no great danger of getting those stacked RNG outcomes…

 

[Fine_Distinction]

i recall a post long ago that said that the game balances against your performance - fall far behind the AI and you start getting lucky, and outperform and bad luck follows you.

i saw a superb Deity game that Henrik posted on YouTube several months back that seemed to back that up - Henrik made an awesomely good start (the best I’ve seen anyone do) and was absolutely rolling the AI. Then suddenly he got three sub 1% losses almost in a row in key battles. Three sub 1% rolls in close proximity is difficult to come by randomly. To have it happen when the human player was doing so well seems more than coincidental.

Just a suspicion on my part, since I don‘t know the inner mechanics of the game, and at my standard of play I’m in no great danger of getting those stacked RNG outcomes…

Civ is open source, we would know by know if the RNG was "cheating". It is just that we remember much more such "unjust" situations due to strong emotional reactions. Winning at low odds also happens sometimes more often than at average expected frequencies but nobody complains

 

[AcaMetis]

For the record a more recent Henrik game showed the opposite example (though I'm assuming it's an example of RNG being RNG moreso than anything else). Specifically

Spoiler :

his NC 328 series had absolute god-tier combat RNG against the barbs. Though I'm not sure if that pendulum swung back the other way later on, admittedly .

 

[ArchGhost]

Played a bit in a Mansa game where two different barbs had 25% (Warrior) and 26.1% (Archer) odds to win against my jungle/river crossing and entrenched jungle/hill warriors within about 10 turns of each other.

With random seed on reload, the warrior won 4 times back-to-back, and the Archer won 6 times back-to-back before I ragequit the game. So he could have been beating those odds even more than that, who knows.

Throws the idea that First Strikes work better when you have the higher power mod right out the window. With the hidden +10% barb combat, that warrior was defending at 4.2...things like this make me start looking at the Civ5/6 combat where it's actually difficult to kill units in one attack with favor.

 

[Jivilov]

YouTuber Grimith has experienced this sort of thing numerous times in his games. His potty-mouthed verbal ejaculations and other aural histrionics are something to hear. Methinks it's more than just random chance but can't prove it---sorta like Elite or just-promoted units in Civ3 which lose disproportionately against weaker units but nobody wants to believe the fix is on.

 

[Bibor]

I don't know what the official name of this phenomenon would be, but from the development of my own company (and also too much civ4), the problems with "chance" is that you "built that axeman" for a very specific reason, at a very specific time (turn), while sacrificing i.e. not building building something else, like a worker or granary. "Having that axeman" is probably a key part of your strategy that you already built up around "that axeman" existing. If you lose "that axeman" to whatever odds, it's bad.

Thus, from an investment (cost) perspective, and it also applies to civ4 units, the "safe" way of doing it – if you already stack odds in your favor – is by "counting on it costing one and a half". This is also because it's very hard to retroactively ask of anyone (in business from a paying client, in civ4 from a city) to change their own plans. If the odds are *not* in your favor, the multiplier is x3, rather than x1,5.

Thus, if you *need* two axemen, you *must* build three. Even at favorable odds, you are planning on losing one.
If odds are *not* in your favor (lets say enemy axe vs your flatland archers), you need 3 archers. One to die, one to be badly wounded, and one remaining at full health after the event (thus continuing to serve its role without need to retreat).

The trick is to make investments in a way that guarantees their utilization even if it turns out that excesses are not needed. For example, deliberately lacking a city garrison in a nearby city, so the surplus axeman can fill that role if needed. This way, you "lost" only 20 hammers (since 15-hammer warrior is a valid militia, and an axeman costs 35). What you gained, however, is far more valuable, and that is that the rest of your empire can go as you planned, with no need for panic-whipping and the like.

 

[Jivilov]

I don't know what the official name of this phenomenon would be, but from the development of my own company (and also too much civ4), the problems with "chance" is that you "built that axeman" for a very specific reason, at a very specific time (turn), while sacrificing i.e. not building building something else, like a worker or granary. "Having that axeman" is probably a key part of your strategy that you already built up around "that axeman" existing. If you lose "that axeman" to whatever odds, it's bad.

Thus, from an investment (cost) perspective, and it also applies to civ4 units, the "safe" way of doing it – if you already stack odds in your favor – is by "counting on it costing one and a half". This is also because it's very hard to retroactively ask of anyone (in business from a paying client, in civ4 from a city) to change their own plans. If the odds are *not* in your favor, the multiplier is x3, rather than x1,5.

Thus, if you *need* two axemen, you *must* build three. Even at favorable odds, you are planning on losing one.
If odds are *not* in your favor (lets say enemy axe vs your flatland archers), you need 3 archers. One to die, one to be badly wounded, and one remaining at full health after the event (thus continuing to serve its role without need to retreat).

The trick is to make investments in a way that guarantees their utilization even if it turns out that excesses are not needed. For example, deliberately lacking a city garrison in a nearby city, so the surplus axeman can fill that role if needed. This way, you "lost" only 20 hammers (since 15-hammer warrior is a valid militia, and an axeman costs 35). What you gained, however, is far more valuable, and that is that the rest of your empire can go as you planned, with no need for panic-whipping and the like.

That's sound advice. However this situation is especially critical in the early game when, say, you've got a few Warriors and Barbs start coming at you. Losing three Warriors in a row at 96% odds or better could mean the whole ballgame right there. But such is life.

 

[Bibor]

That’s why BW (the so called “emergency whip”) is so important. It’s the only tool in the game (until democracy) for “banking” production - in form of excess population.