Thanks Gnullbegg for the info

Yet I see it to make things more complicated since it stops you from advancing rapidly in some of the most very good starting poins.
Also wastelends tend to change locations from game to an other thus making it very hard to grab a style and change only from time to time according to the oppnents.
It makes it more luck based instead of strategy based.
And im an unlucky guy :)

I fought off the last cancer, but this line gave me a remission:

This shows: 15+6+0.81+20+20+20 = 81.81/600 = 0.13635% chance of happening

I thought IT studies always have an obligatory probability course. Anyway please read at least on binomial distribution so you can easily calculate the probability of defense and offense kills. Also my general advice, stick to games with low luck setting.

Yet I see it to make things more complicated since it stops you from advancing rapidly in some of the most very good starting poins.
Also wastelends tend to change locations from game to an other thus making it very hard to grab a style and change only from time to time according to the oppnents.

Yes. That's exactly why wastelands make the game more interesting.

hmm although that was a bad way of working it out, I just wanted to give the idea
Probability works a bit different
if i remember correctly from school its:
15+6+0.81= 21.81/100 * 20/100 = 0.044% chance
Though I could be mistaken since it was a long time ago
Thanks to this nice discussion i'm gonna go check this out heh.

geez u guys are active heh, didn't even give me a chance to correct my mistakes heh

wah worng again
man u need an edit in this forum heh
to work out the probability of getting anything in 3 turns is highly complex

it would take too much to list all the factors even worse work it out

at first i thought they were possible in every turn thats why I added them yet there are so many factors involed that adding them won't make sense apart that a * would be better to describe the relation of getting them in 3 turns like:
15/100 * 6/100 * 0.81/100 =0.00729 = 0.729%
yet this is saying that you get them in this order which is not the case
this result would give a lower number then the right result
also no one is saying that these guys cannot occour in the same turn (1 of the 3)
things start to get even more complex here

so to cut short here
the point is that its increadibly difficult to get all those 3 in 1 game, in 3 turns and in the very first 3 tuns.
(20/100 * 20/100 * 20/100 = 0.8%)
20% (first 3 turns) of the 20% (only 3 turns) of the 20%(same game)

The final probability number must be multiplied by somewhat a 0.8%
(which btw it should be less)

hmmmm

what is the percentage of winning a 7 vs 5?
what is the average & minimum losses for a 7 vs 5?


http://warlight.net/MultiPlayer.aspx?GameID=2975756

While WL seems to have, at first glance, a lot of similarities to Risk, you'll notice there are quite big differences once you start to play it a little more.

For instance, in Risk you attack with at most 3 attackers at a time, versus at most 2 defenders. In WL on the other hand, all attackers and defenders fight at the same time.
One of the consequences is that the number of attackers is completely irrelevant when calculating how many attackers die; the only thing that matters is the number of defenders and the defence kill ratio. (Well, of course the number of attackers which die is capped at the total number of attackers.)
This also means that, in Risk, it is possible (unlikely, but possible) for 2 defenders to kill 3 attackers. In WL (even taking everything into account: all possible luck and kill ratio settings, the rounding mode, "the roll of the dice", everything) the maximum number of killed attackers is equal to the number of defenders. Likewise, the maximum number of killed defenders is equal to the number of attackers. (By the way, it will usually be lower.) Attacking with fewer attackers than there are defenders could make tactical sense in some cases, but it will never capture the territory you're attacking.

There's actually a pretty good description of almost everything about how WL gameplay works on the Wiki. I think you should start with http://wiki.warlight.net/index.php/Combat_Basics. Further "suggested reading" is listed in the "See Also" section of that page.


My apologies if this sounds a little harsh, but (as the example szeweningen replied to shows), you are not very good at statistics. I would very much encourage you (both for WL-reasons and for real-life-reasons) to brush up on basic probability theory. For instance, combining two chances is done by multiplication, not addition. If we have a 1-in-4 chance (0.25) and a 2-in-5 (0.40) chance, the probability of both events occurring is 0.25 * 0.40 = 0.10. Quick check: 1-in-4 and 2-in-5 is 2-in-20, which is the same as 1-in-10, which indeed equals the 0.10 we found.

However, that might take a little time; in the meantime, you could try low-luck games (such as the Strategic 1-vs-1 Template). In those games the effect of luck is strongly reduced, so that outcomes are always fairly close to the average, expected outcome (which you will find easy to get a feeling for, once you understand the combat system).

Of course, even if you are (or become) good at statistics, you might still prefer low-luck games anyway, because you find them more enjoyable. But if you can get very frustrated or annoyed by "unlikely" outcomes I think it might be better for you personally to stay away from 75% luck games, at least until you've done a statistics-refresher.

---

ps. English is weird; the opposite of "defence" is "offence", but even though "offenders" is a word, it isn't the opposite of "defenders"... :p

Whoops, forgot to reply to this question:

what is the percentage of winning a 7 vs 5?
what is the average & minimum losses for a 7 vs 5?

With 60% offence kill rate, 70% defence kill rate, 75% luck and weighted random rounding mode (the default, but all of those can be changed, WL is very configurable), that works out to:

Chance of capturing the territory:

38.22%

Minimum possible attacker losses:

Floor ( ( 5 * 0.70 ) * 0.25 + 0 * 0.75 ) = 1

Maximum possible attacker losses:

Ceil  ( ( 5 * 0.70 ) * 0.25 + 5 * 0.75 ) = 5

Minimum possible defender losses:

Floor ( ( 7 * 0.60 ) * 0.25 + 0 * 0.75 ) = 1

Maximum possible defender losses:

Ceil  ( ( 7 * 0.60 ) * 0.25 + 7 * 0.75 ) = 7

I'm sorry, but the full calculation for that 38.22% success rate is too complex (and too long) to show here. However, you don't want to do that calculation anyway... and you don't have to!
Please note the "Attack" dialog window has a button labelled "Analyser". Here you can get a rough idea for success chances. One important detail to keep in mind is that it doesn't calculate, it simulates. When it tells you something has a 100% success chance it really means to say "I tried this a hundred times and it worked a hundred times", but in reality it could still be a 99.9% chance (and therefore have a very small chance of failing).


For illustration, 60% offence kill rate, 70% defence kill rate, 100% luck and weighted random rounding mode gives:
Chance of capturing the territory:

41.99%

Minimum possible attacker losses:

Floor ( ( 5 * 0.70 ) * 0.00 + 0 * 1.00 ) = 1

Maximum possible attacker losses:

Ceil  ( ( 5 * 0.70 ) * 0.00 + 5 * 1.00 ) = 5

Minimum possible defender losses:

Floor ( ( 7 * 0.60 ) * 0.00 + 0 * 1.00 ) = 1

Maximum possible defender losses:

Ceil  ( ( 7 * 0.60 ) * 0.00 + 7 * 1.00 ) = 7

And 60% offence kill rate, 70% defence kill rate, 0% luck and weighted random rounding mode gives:
Chance of capturing the territory:

20.00%

Minimum possible attacker losses:

Floor ( ( 5 * 0.70 ) * 1.00 + 0 * 0.00 ) = 3

Maximum possible attacker losses:

Ceil  ( ( 5 * 0.70 ) * 1.00 + 5 * 0.00 ) = 4

Minimum possible defender losses:

Floor ( ( 7 * 0.60 ) * 1.00 + 0 * 0.00 ) = 4

Maximum possible defender losses:

Ceil  ( ( 7 * 0.60 ) * 1.00 + 7 * 0.00 ) = 5

Thanks for the reply
it did help a lot

Hey, guys.

I hope a lot of you remember me.

szeweningen, I thank you for the link to this documentary and I aploud your passion to understand and even explain things to others.

I really love probabilities so my current position is gambling games mathematician.

Buuuuut....
This post is not about probabilities.

It is about retiring. I understand why anybody would like to stop loosing time on this fun game.

It starts to take way too much time.

It becomes obligation to just continue to struggle in unfunny games on huge or boring maps, just to not get booted=ruining the game for your mates.


Well- my advise:
Don't retire! Just join very few games.
Only those with your favorite settings.

And definately- avoid noobs that will ruin your experience.



Advise to anybody- if you join more than 20 tournaments, you will be really, really unhappy from the lost time. If you temporarily need more games, you can even try real time games, but having to play too much (sometimes with noobs, or booted noobs replaced by AI) will ruin your experience.

sorry for the oftopic. It was needed