Your experience with combat odds...?

[Tobiyogi]

At the moment, I am a bit tired of the randomness of combats in Civ 4. Although "randomness" is not the right word, I would call it one-dimensional.
I am really tired of losing every single 60% fight, half of my 70% fights and at least one third of my 80% fights. Ok, one could claim that you shouldn't rely on such bad odds. But I play lots of AW these weeks and here you have to take those fights quite often, because coward AI enters your territory and camps on a hill without ever attacking. So you have to get rid of them.
What I hate even more is the completely different outcome from City Raiding. When 2 AI catapults hit my city, my LONGBOWS are half-dead and the city is taken by chariots afterwards. When I hit an AI city with 5 catapults, the units are in much better shape and I still lose city raiding units.
When I chose to play on a difficulty above monarch, I can live with slower research, with more units for the AI player, with high city up-keep etc...because I don't want it to be too easy. But to lose very good fights for nothing, that is really the most annoying part of the game. Because it is pure math, probability calculations and I would like to rely on those. In that case, I would rather see my "real" odds adapted to my level. That would be fair and one could know what to expect.
What are you experiences?

 

[Gumbolt]

Unless the odds are 99% you are not guaranteed a win.Same for odds of 30%. I often find you can win those combats.

On always war you want CG3 longbows. That should slow down catapults. Albeit I prefer to go on the offensive instead of waiting for Ai to attack my city.

 

[Tobiyogi]

That's why I lose many fights on AW because I play offensively too. It works well to pre-damage with catapults but you don't always have them. My observations here are more general and not only from AW games.
I thought it would be more mathematical at all. So higher "guarantee" to win at 80% than at 60%.

 

[sampsa]

I am really tired of losing every single 60% fight, half of my 70% fights and at least one third of my 80% fights.

What I hate even more is the completely different outcome from City Raiding. When 2 AI catapults hit my city, my LONGBOWS are half-dead and the city is taken by chariots afterwards.

What are you experiences?

My experience is that even many intelligent people have a hard time understanding probabilities and variance.

 

[earthy]

Its easy to forget all the wins and remember the losses. This is a big thing in poker as well. Probability distribution is just so volatile over small samples that youre never going to win all the fights you "should win".

 

[Fippy]

While all true, there have been 100% odds fights lost (i saw at least one myself).

 

[Tobiyogi]

Actually, I don't expect any wins in a single situation, once in a while you lose a 99% fight (100 never seen) and that's okay. Maybe it's the problem that I also teach math... 60% means for me that over the long run, I expect to win 60 over 100 fights. And the reality is faaar from that.

Is that linked to the difficulty? I dont't remember playing noble or prince, maybe it is better on those levels.

At the same time, I see that my thought is a bit stupid because the game cannot "remember" what odds it gave me in the past, it can only judge every single situation. Well....

 

[dankok]

I don't think odds of winning battles has anything to do with difficulty level. I'm pretty sure you always have a chance of winning that is displayed in the game. That said I'm often frustrated by a bad run of RNG.

 

[Tobiyogi]

Unless the odds are 99% you are not guaranteed a win.Same for odds of 30%. I often find you can win those combats.city.

I just remember one situation where you can win 30% fights quite regularly. It is when attacking garrisoned archers with Cho-ko-nus. It has certainly to do with the fact that CKN, unless archers, have 2 first strikes.

 

[AcaMetis]

As far as I've experienced/witnessed Noble does not skew RNG results in the player's favor any more than Deity does (or doesn't, as the case may be). Well, there's the two Mulligan fights against barbarians you get on Noble, but that's obviously a different case.

 

[sampsa]

Actually, I don't expect any wins in a single situation, once in a while you lose a 99% fight (100 never seen) and that's okay. Maybe it's the problem that I also teach math... 60% means for me that over the long run, I expect to win 60 over 100 fights. And the reality is faaar from that.

First of all, how do you know that the reality is far from that? As a maths teacher, you probably also understand the problem in your claim "60% means for me that over the long run, I expect to win 60 over 100 fights". You should expect to win 60 out of 100 fights on average, but it doesn't mean you will. So that expectation will leave you disappointed quite often. As Fips earthy already mentioned, many people won't even notice when they get luckier than expected.

At the same time, I see that my thought is a bit stupid because the game cannot "remember" what odds it gave me in the past, it can only judge every single situation. Well....

Indeed. Having poor consecutive rolls doesn't mean you will get good consecutive rolls later. It's possible just to be very unlucky for your entire cIV-career, but of course it's unlikely.

 

[Tobiyogi]

That moment of "OMG, how could i win that fight?", I remember it quite well, but it is very rare. Most of the cases, it is "how could I ever lose that fight"?
I do not note consecutive odds/outcome so I cannot prove bad numbers, it is just a clear feeling that in my games, statistics are far from being balanced.
Actually, probability in real life is quite balanced. For exemple, long term studies have shown, that over very long periods, one number in lotto or roulette shows up exactly as often as other numbers. That is maybe what you mean by "on average"? I know that I cannot expect anything in a single situation, that's why everyone loses in lotto except one person.

 

[Tobiyogi]

Btw, it is not about portraying myself as the most "unlucky" guy who is screwed by the RNG. I wonder if other players have same issues but they are used to it or can compensate with their skills. Lain for exemple has really bad luck in most of the videos I was watching, that's why I was thinking that this is the extra naughtiness from playing on deity. He loses regularly his >95% fights. He can compensate of course.

 

[sampsa]

For exemple, long term studies have shown, that over very long periods, one number in lotto or roulette shows up exactly as often as other numbers

Source? Exactly as often? That's very unlikely. Are you familiar with poker and the mathematics of variance in it? I feel like it's closer to the topic at hand than lotto numbers.

Lain for exemple has really bad luck in most of the videos I was watching, that's why I was thinking that this is the extra naughtiness from playing on deity.

Slightly unlucky here and there and he certainly remembers to whine like a baby every time it happens. When he gets lucky he pays much less attention to it.

 

[Tobiyogi]

No, not at all familiar with poker. I will look for a source. It is surely framed to chaos theories and on the other side, to the hidden invisible order that reigns over the universe.

Possible that Lain knows how to complain and whine but in that case, it is MY observation, not a reaction to his comments. I am seeing that quite often in his games.

 

[AcaMetis]

Lain for exemple has really bad luck in most of the videos I was watching, that's why I was thinking that this is the extra naughtiness from playing on deity. He loses regularly his >95% fights.

As someone who's rolled over plenty of maps with Cossacks on Noble, I can guarantee that low odd losses still happen plenty of times even on lower difficulties. Hell, I've lost two Cossacks to a single Barb Sword once. Really wanted to hire the guy afterwards since he obviously was more of a Great General than the others I got, but alas, the game did not give me that option .

 

[Tobiyogi]

Yeah, some barbs are really brave. But also stupid like the conditionned dog.
I once won a fight with a wounded warrior against a barb axe. The axe was attacking over river, big mistake...

 

[Tobiyogi]

I am still reading the LotteryCodex.

Quote:
"Playing the lottery is like a war.

It is vital that you know the enemies, strategize your plan and execute the attack to win the war.

No one describes it better than Sun Tzu:

“The general who wins the battle makes many calculations in his temple before the battle is fought”
End of quote.

It is a lot about combinations / patterns. Some of them are more likely than others. Like 3odd/3even is more likely than 2/4 or 1/5.
I didn't get to the number section yet. There seems to be a secret too.

 

[Tobiyogi]

Source? Exactly as often? That's very unlikely. Are you familiar with poker and the mathematics of variance in it? I feel like it's closer to the topic at hand than lotto numbers.

That is still from the Lottery codex, written by Edwin Hiltner. If it wasn't holidays, I probably wouldn't read it

"Mathematically speaking, hot and cold numbers don’t exist.

For example, in a few draws of the lottery, you will notice that some numbers tend to be drawn more frequently than others. However, as the number of draws increase, all the balls tend to even out. Then some of the numbers that were left behind, later on, catch up.

This event is described in mathematics as the Law of Large Numbers or LLN.

The law simply means that the frequency of each ball tend to get closer and closer as more draws take place to infinity.

To illustrate, let’s take a look at the actual results from the Canada Lotto 6/49 From 1982 to 2018."

Now some graphs follow (that I cannot paste) where you can see, that over the years, the drawn numbers get closer and closer to each other so that in infinity, they would nearly even out.

As a sort of conclusion, it has no meaning at all what number to chose, because they are not only probably but even statistically even.

Does this help us for Civ 4? However, I find it interesting.

 

[sampsa]

However, as the number of draws increase, all the balls tend to even out. Then some of the numbers that were left behind, later on, catch up.

However, this does not mean that if one number is left behind there would be some force that makes it catch up. Of course, with infinite number of lotteries it will happen, eventually. But in practice, with a limited sample size, balls don't tend to even out.

Also, and this is probably crystal clear for you, many books are just nonsense.