boonekim24
Farmer
11 attrition gained in 39 battles is 20 percent.
-
Dear forum reader,
To actively participate in our forum discussions or to start your own threads, in addition to your game account you need a forum account. You can REGISTER HERE!
Please ensure a translation in to English is provided if your post is not in English and to respect your fellow players when posting.
Discussion Inno math
- Thread starter boonekim24
- Start date
-NinjAlin-
Emperor
Nothing wrong with the math, that is how probability works.
I had rounds in which I had 563 fights with 113 attrition and in another day had 700 with just 107 attrition, all on 80% reduction sectors.
NEVER sample size the percentage within few fights. Do for a month or so an average of your fights and attrition taken and You will see fighting only on 80% reduction sectors is in fact ¬20% attrition taken.
The REAL downside here, and I dislike it, it is when all the attrition increases faster from 0 to 80-90 attrition and after 90 attrition is slowing down, that makes me lose more units and get hurt much often than increasing the attrition slower from 0-80 in the other days, but that is how probability works after all.
I had rounds in which I had 563 fights with 113 attrition and in another day had 700 with just 107 attrition, all on 80% reduction sectors.
NEVER sample size the percentage within few fights. Do for a month or so an average of your fights and attrition taken and You will see fighting only on 80% reduction sectors is in fact ¬20% attrition taken.
The REAL downside here, and I dislike it, it is when all the attrition increases faster from 0 to 80-90 attrition and after 90 attrition is slowing down, that makes me lose more units and get hurt much often than increasing the attrition slower from 0-80 in the other days, but that is how probability works after all.
boonekim24
Farmer
And now it’s 14 attrition in 48 battles. That’s even worse! That’s almost 30 percent.
Deleted User - 244520
Guest
The title of this thread should be: boonekim24's math.
Deleted User - 57457
Guest
There's every victorious battle / successful negotiation a 20% chance to increase the attrition. It's not an every 5th victorious battle / successful negotiation attrition increased by 1.
On average after something like 1.000+ battles/neg. attrition is to be expected on average 200 but no guaranteed. Real world numbers likely deviate from 200 attrition around 1.000 battles/neg. for example ranging from 180-220 attrition. Deviation is always to be expected with RNG's. At 10k battles/neg. the deviation from 2k attrition is likely relatively speaking smaller but more than likely none the less present.
It's frustrating when you'll end up at the unfortunate side of the deviation, great if you'll end up at the fortunate end of it but deviation is always to be expected. Same goes with BG. It often deviates, variating from favourable to unfavourable deviation. Generally speaking averaging out closer to the promised chance.
On average after something like 1.000+ battles/neg. attrition is to be expected on average 200 but no guaranteed. Real world numbers likely deviate from 200 attrition around 1.000 battles/neg. for example ranging from 180-220 attrition. Deviation is always to be expected with RNG's. At 10k battles/neg. the deviation from 2k attrition is likely relatively speaking smaller but more than likely none the less present.
It's frustrating when you'll end up at the unfortunate side of the deviation, great if you'll end up at the fortunate end of it but deviation is always to be expected. Same goes with BG. It often deviates, variating from favourable to unfavourable deviation. Generally speaking averaging out closer to the promised chance.
Deleted User - 57457
Guest
I'll gonna roll a d6 dice and bet at 6, I'll be doing it 20 times. The dice rolled only once 6. The dice is rigged, cause it didn't roll the average 3+ times a 6 after 20 attempts.
Another day I'll doing the same. The dice rolls 5 times a 6 after 20 attempts. I'm joyfully surprised but keep quiet. Convinced the dice is biased (something that can happen at roulette) towards 6. In actuality it isn't biased at all and I'm just lucky this time. All other times I'll complain of rigged dices.
Another day I'll doing the same. The dice rolls 5 times a 6 after 20 attempts. I'm joyfully surprised but keep quiet. Convinced the dice is biased (something that can happen at roulette) towards 6. In actuality it isn't biased at all and I'm just lucky this time. All other times I'll complain of rigged dices.
Publius Cornelius Scipio
Marquis
Cheating with everything and with attrition, of course.
What I noticed lately is, that I acquire very quickly attrition and than make a lot of battles with 113%, 125% and 133% attrition. Obviously the idea is to make loose a lot of unities...
Restrictions, obstacle and cheating, I start really to be pissed of...
Have the sensation, that someone feells happy by making the game digusting.
I don`t understand that.
This must be a game and not a tool of a constant torture.
Don`t get it, really...
What I noticed lately is, that I acquire very quickly attrition and than make a lot of battles with 113%, 125% and 133% attrition. Obviously the idea is to make loose a lot of unities...
Restrictions, obstacle and cheating, I start really to be pissed of...
Have the sensation, that someone feells happy by making the game digusting.
I don`t understand that.
This must be a game and not a tool of a constant torture.
Don`t get it, really...
Sl8yer
Emperor
Lose your alu hat. That might help.
Pericles the Lion
Emperor
Please share some data. Every day I hit 20% sectors (because there are plenty of them). When I finish I check (1) the number of battles and (2) the amount of attrition. My day is not yet over but I'm at 438 battles and 83 attrition. 18.9%. I've been checking my attrition daily since the new GBG rolled out on live and have always been in the 19%-21% range.
The given attrition percentages are correct. The rest is just math.
Publius Cornelius Scipio
Marquis
I don`t follow the precise number of battles, but I know how much I do approximately every day.
In the days, that I`m talking about and today was one od these days, i made something like 150-160 battles less. That usualy happend the first day of the season. Two seasons ago was identic.
I mean, I always do the same things and fight always with maximun SC support. And such a number of less fights is too high.
The fist time when I noticed it, I wrote to the support and, as you can imagine, there was no an answer till the end of the season. Than I saw, that someone had ansewerd, but I didn`t even see, because it was all over. The fact, that this is repeating means, that the answer was the usual blabla of the guy...
Publius Cornelius Scipio
Marquis
That`s not true.
Pericles the Lion
Emperor
Maybe you should start tracking the precise number of battles then compare this number with the exact attrition that you accumulated during these battles before you say that the math is wrong (and use a decent size number of battles while you're at it). "I don't follow the precise number of battles" and "approximately" and "something like" are not the basis for any conclusions about the math.
Deleted User - 57457
Guest
1st 50 fights, 18 attrition gained. With a 60% chance of attrition. I should've gotten 30 attrition, not 18. Th RNG and chances are clearly broken. Biased in my favour.
.Chris
Baronet
Can we PLEASE once and for all stop with the accusation that RNG is rigged without having an adequately large sample size?
Facts don't care about your feelings; gather data and present it. Precisely, not approximately.
FFS
Arch1e
Marquis
It’s very easy to check daily, vastly helped by mostly everything being 80% boost. If it isn’t, you will know what you have hit anyway. It averages out fine, like others, my chance of getting attrition hovers around the 20% mark, give or take some on variance daily, but it’s fine.
Deleted User - 57457
Guest
Could be quite interesting to start collecting data on the subject. Noting down per 100 fights at X% chance no attrition. Probs per season it's easy to get 10-20 instances of this. Doing it for 5-6 seasons should at a consistent X% chance no attrition increase, should give roughly data for 10k-ish fights.
Looking to the total average % of attrition and total average deviation. Alongside the mean deviation per 100.
After this collect data once more, every 100 fights for the same X% chance. Apply the mean deviation to averagely expected attrition gains. This should give a more consistent and fairer expectation.
Let's say for example the mean deviation turns out to be 15%. Using this with 20% risk at attrition per 100 fights: 20-attrition average, deviation of 15% is 3 attrition. So, a range between 17-23 attrition is to be expected most of the time, after 100 battles. The average is rarely achieved exactly. More often there's a range. This is just a hypothesis. It's likely that real world deviation is smaller at higher numbers vs lower numbers. For example 1% deviation at 5k battles with 20% attrition risk gives us a range between 950-1.050 attrition estimate.
This is all just a hypothesis that I think is worth exploring. To look deviation at 100 fights, 200 or even higher numbers. To create a more accurate attrition estimate to work with.
Looking to the total average % of attrition and total average deviation. Alongside the mean deviation per 100.
After this collect data once more, every 100 fights for the same X% chance. Apply the mean deviation to averagely expected attrition gains. This should give a more consistent and fairer expectation.
Let's say for example the mean deviation turns out to be 15%. Using this with 20% risk at attrition per 100 fights: 20-attrition average, deviation of 15% is 3 attrition. So, a range between 17-23 attrition is to be expected most of the time, after 100 battles. The average is rarely achieved exactly. More often there's a range. This is just a hypothesis. It's likely that real world deviation is smaller at higher numbers vs lower numbers. For example 1% deviation at 5k battles with 20% attrition risk gives us a range between 950-1.050 attrition estimate.
This is all just a hypothesis that I think is worth exploring. To look deviation at 100 fights, 200 or even higher numbers. To create a more accurate attrition estimate to work with.
UBERhelp1
Viceroy
With 20% attrition gain, 100 fights, the 95% confidence interval for effective attrition gain is between 12% and 28%. (SE 0.04).
With 1000 fights, between 17.4% and 22.6%. (SE 0.013)
With 10000 fights, between 19.2% and 20.8%. (SE 0.004).