Thank you Segrain for your calculation.

BTW .... i dont know if anyone of you might have noticed, that up to turn 20 (i then was visibly defeated) ALL of "Master-Chief" s attacks from a

4 on a 2 (attacking a 2-troop territory with 4 attack troops)
5 on a 2 (attacking a 2-troop territory with 3 attack troops)
3 on a 1 (attacking a 1-troop territory with 2 attack troops)

worked.

I took myself the freedom to count the number of those attacks:

13 attacks 4 on a 2 (88 % success rate)
9 attacks 3 on a 2 (68% success rate)
3 attacks 2 on a 1 (89 % success rate)

I took the success rates from the "Analyzse Function" of a 75% luck game/ 60% offensive rate / 70% defense rate (default settings)

By multiplying this i get a number of about .004 = 0.4%

THATS WHAT I CALL REAL LUCK :D

oops, typo:

in my last post i wrote in line 5 "5 on a 2", what should read as "3 on a 2".

his cumulative offense luck ended up being +6.3, end game..


and if you take any players game, and multiply his *successful* attacks, you'll end up with a low probability..
I had begun calculating out all his attacks, but then got interrupted.. and think I botched my data-gathering nonetheless..

I did notice however that he had won a 5v4 attack, and the most notable loss was a 4v0 *0 being the number he killed.. idc how many were defending*

if you take the number of free throws any basketball player makes, then figure out the odds of them making that many, it is pretty thin.. unless you include his failures as well..

chief only failed, by the time I stopped watching, turn 20-21 or something.. 18-9 4-3 4-0 11-8 14-7 8-4 17-11 4-2 1-1 3-1 3-1... again, the format is attacking armies-killed armies

Perrin,

i checked your stats and indeed, you didnt do too few games :D. So try to remember one game where you took about 20 neutrals in row with about 10 each from a 4 or 5 territory. And this without exception.

Good luck finding one :D.

to be more precise: i am talking about a game with the same luck settings.

I don't think I tend to take such risks in 75% luck games, lol

Yeah, my plan is better served by just reducing or eliminating luck entirely. I'm of the opinion that nobody should ever lose entirely due to preventable luck, and nobody should ever win by the same.

I think that adding an option to rubberband the luck still allows for some odd outcomes of individual battles, but will tend to correct over time. The worse someone gets hit with bad luck, the more likely they will be to recover over time. Something to note about this idea, is that it still won't guarantee that future outcomes are "fixed" compared to previous, just another factor that influences the numbers.

that is a faulty arguement duke..
0 cumulative luck is not balanced..

if I get lucky first turn, and get my bonus a turn before you, that means I get 4 more armies the first turn, which means I get my 2nd bonus a turn before you.. which means I'll have a cumulatie 8 armies more then you by the time I get my third bonus.. a turn, maybe two before you..

you winning that 2v2 against me on turn 26 is not going to make up for me capturing a bonus with a 2v2 on turn 1

and as it's been stated previously.. variances are realistic.. reliability is meta.. 16% or 75% are the most common luck factors for both of those outcomes.

As I said, I would pick the low luck, because it serves my desires much better. I was simply musing about a way to not mind the high luck as much. Generally I create games with less than 20% luck and avoid joining games with more than that.

I find it tiresome lately to play any game higher then maybe 40% luck, but will play them if invited

Duke, I also dislike games with such high luck, but it seems that the majority of open games are made by non-members. Not sure if this is a fact, or just my perception. But I would imagine that most members, being slightly better than the average bear, tend to stick to ladder games or games created for a select group of players, rather than games with open spots.

Okay, to turn this subject back to the original game.... what's the talk in the chat about fizzer paying people to play here to bump up stats and make it more marketable?

How is this a return to the original topic? And where did you hear such rumors? I don't think he would do that...

(R) John. I just read through that too..
it might be a founded rumor if warlight was incorporated, and in turn would then release information showing how many players actually play the gam, extra traffic is unneccessary unless people are actually playing, in which case it's just called advertising..
It sounds more like mosq is just spouting drivel to me..

Afaik Fizzer only has one account, and has no need of another account, save for perhaps one to test how features appear to non-members

Just a question adressed at WL experts related to this context:

Does the Analyze Function round the probabilities to the nearest integer? I get a 100% success rate indicated in a 75/60/70 default game if i for example attack a 1 neutral with 5 troops. Is this really a 100%?

i think its better know as "Analyse Tool" than as "Analyze Function"

mosq... the analyze tool has been discussed pretty heavily in the forums in the past. I recommend you do a search if you want the in-depth story.

The short story of it is that it reports results of 1000 attempts. Just uses brute force to determine the percentages and as such can stray by 1-2 points in either direction depending on the results.

That was a quick reply, thank you! I ll try to find that discussion.

u can''t trust it sometimes it has failed me manytimes

Richard, I believe it was also confirmed by fizzer, that at exceptional percentages, it tends to just consider it complete.. IE like 98%+ or 2%-, will show as 100% and 0%

can't remember the exact range it uses, but it's relatively small..