Hey

Some of my team games don't fill - I don't think the filters are too strong I usually use a 40 or 45% win rate in 1v1 and 2v2 or 3v3 depending on the size of the game, a 9% max boot rate and 200 games played.

The settings I usually use are 16% (sometimes 0%) Luck Manual Distribution Warlords occasionally Full Distribution on bigger maps and no cards.

In some games it has been known that people say my luck settings are too low on one occasion I was told this was bad as higher luck needed more skill.

I'm wondering what settings you prefer for games - obviously they will vary from map to map but I think there are settings that people will tend to prefer and what filters you use? I want a game to fill - but on the other hand I don't like games that are over too quickly.

16% luck. full manual distribution. 4 picks ( 5 if europe 1600 - 3vs3


45% winning rate is quite high when there are too many people around (and doesnt prevent noobs entering in tha game).

Luck: Are you using 16% weighed or straight? 16% Weighed gives you an 80% chance of a 3v2 success, which is the ladder setting, thus making it the most playable and acceptable setting. It adds a bit of luck but isn't as obnoxious as the 75% or 16/0% straight which kind of eliminates luck entirely.

Filters: The way I do it is as follows: I have one template with 55% 1v1, 50% 3v3, 10 or 15% boot rate and 400 games played. This will allow for average players to join making it almost always a fairly good game.
In case you are online during a pretty "dead" period, I suggest making another template with not so harsh filters.

Settings: I already mentioned luck, but try using 3 starting spots of 5 armies, manual warlords on regular Europe. This usually makes for great games where each player has to choose carefully not making the mistake of being eliminated early on. But then again, this is entirely up to you.

Just my three cents.

Best,
Doggy

I feel so discriminated with my massive 15% boot, gotta play more victories. I usually meet all the skill and experience tequirmenta, but the boot thing filters me :P

45% overall win- check
200 games played- check
40-45% in 1v1 2v2 3v3- check
8% boot- :P

Ladder luck I∂ƒ∂oggY obviously I can amend the filters and do so at certain times but my average is 40 - 45% wins on the filter.

Your filter is stronger than mine - how long do your games take to fill up?

I'd like to hear from non-members about the luck you guys prefer as well - I have an idea though I find it difficult to understand! That higher luck may actually be preferred by a lot of people?

I really don't understand the concept of luck, so I usually don't mind whether its 75 or 16.
On some of my 'non- noob' games, i set at leat 300 games, and 40% 1v1 win.

Nike, luck works as follows.

In standard settings (60/70 O/D kill rate) and army of 21 attacking a defender of 10 will kill 12 (20*.6) and lose 7 (10*.7). That is your expected outcome.

The luck factor determines what your possible deviation from this expectation is. In 16%, that range becomes 10.08 to 13.92 (so you are guaranteed to kill 10 and at best kill 14) while losing 5.88 to 8.12. In 75% those numbers change to ranging from 3 to 20 attacking and 1.75 to 10 defending. (Multiply the expected value by 1-luck and 1+luck to determine these values)

Obviously the second scenario is far more unstable and makes matters far more difficult to predict.

Thanks for the explanation, Richard. Iv won against a few members who have, after and during the game, complained about 75% luck. I tried asking them about the difference, but they didn't do a good job at explaining it.

What are your usual filters for your games, Richard?

@Ska:
My games usually take about 10-15 minutes, very rarely does it take more than that.

I understand your dilemma of win ratios - all you can do is for an entire week try stat boosting to about 50% each... HHH's games even have a 60% so that's tough; A lot of good players face the same dilemma but unfortunately there is nothing that can be done about it if you don't want noobs joining your games.

@Nike:
I remotely remember you saying in a thread a few days ago or so, that whenever you lose, you just leave instead of surrendering making your boot rate so high. Forgive me if that was someone else, but if it was you, then this is your own damn stupid fault for being such a douche :D

Yes I said that, but I do that less now to decrease my boot ratio, and I have been getting better and winning more in this month. It is quite annoying to wait for your surrender to be accepted before you can leave :)

Multiply the expected value by 1-luck and 1+luck to determine these values

Are you sure Richard? Because for 100% luck, the number of kills could be anywhere from zero to the number of armies.

---

@Nike:
The way I prefer to explain "luck" is as follows.

Warlight calculates two things for each attack:

• "Certainty": The expected number of kills (armies * kill rate)
• "Luck": Flip a (weighted) coin for each army, to determine whether it will kill an opposing army

These two numbers are then combined into "the result" based on the luck setting.
The outcome of the "Certainty" calculation is multiplied by 1-luck (default 1-0.75=0.25), the outcome of the "Luck" calculation is multiplied by luck (default 0.75) and those two numbers added.

Example:

8 armies attack, 60% kill rate:
"Certainty"     "Luck"
8 * 0.60 = 4.8  Army 1: 0
                Army 2: 1
                Army 3: 1
                Army 4: 0
                Army 5: 1
                Army 6: 1
                Army 7: 1
                Army 8: 1
                       --+
                        6

4.8 * 0.25 + 6 * 0.75 = 5.7

Note that (in extreme cases) the "Luck" part could be anywhere from 0 to 8. On 100% luck that would mean the result of the battle could be from 0 to 8. However, on 75% luck, you gain a little certainty:

• 4.8 * 0.25 + 0 * 0.75 = 1.2
• 4.8 * 0.25 + 8 * 0.75 = 6.75
• You'll kill between 1 and 7.

On low luck settings (let's use 16% for this example), uncertainty is almost entirely removed:

• 4.8 * 0.84 + 0 * 0.16 = 4.032
• 4.8 * 0.84 + 8 * 0.16 = 5.312
• You'll kill between 4 and 6.

I don't actually want to read your explanation though Rvw ... don't take this the wrong way I'm just not a maths person and your post just looks too much like homework to begin reading it. Though I'm sure it is on the money and an excellent description I'll stick with Richard's :)

"Are you sure Richard? Because for 100% luck, the number of kills could be anywhere from zero to the number of armies."

Well if luck is 100% that means the range is 0 to double the expected value which by definition is greater than your available armies.

I dont play with 40% or 45% win rate games, too much chances get into team noob :/

Unfortunately HHH I cannot see your games - a mix of bad play when I begun and playing against myhand too many times I think!

But out of curiosity which settings and filters do you use?

At now i playing these games:
http://WarLight.net/MultiPlayer.aspx?GameID=3033009
http://WarLight.net/MultiPlayer.aspx?GameID=3032974
Dont play with myhand, he is total noob :(

I can't understand why anyone would want to have a high percentage of luck. I understand the "gamble" of not being sure your 3v2 attacks are going to work out, but I don't understand why you should sometimes be punished for executing a 13v2 attack (14v2, on 75% luck, is the minimum amount of armies required to always take the territory). That involves no luck at all. Really? The game is telling me that I get a bonus a turn later because there's a 0.000001% chance of it happening? Higher luck settings require more skill? The opposite. I frequently have games battling the luck factor more than I'm battling my opponent. This is not showcasing your skill. It is showing how much of an impact the luck factor really has on games.

So, yeah, luck factor at 16% to at least involve the risk factor but not allow almost any attack to fail due to luck (or have a big army still lose much more armies than a much smaller army when attacking). Offensive/defensive kill ratio I don't care too much about, as long as the former is lower than the latter and the difference isn't enormous. Manual distribution is a must. Cards are okay, but the more you include, the more luck you can expect.

http://WarLight.net/MultiPlayer.aspx?TournamentID=2720

This guy ^^

Well if luck is 100% that means the range is 0 to double the expected value which by definition is greater than your available armies.

Only if the kill rate is 50% or higher. Also, it shouldn't be greater, it should be equal to your available armies.

So, while I think your formula is wrong, I have to admit not investigating whether it's a good approximation; it very well might be. (I prefer my calculations to be spot on and perfectly accurate, even at the cost of them being multiple pages in length. Not everyone will agree with me on that.)