"Luck" in those graphs doesn't always mean what you think it should. If you're on 0% sr, and attack with a 3v2, you'll get +.2 luck even though there's absolutely no luck in winning in that situation, since you "really" should have only killed 1.8 armies.

Edited 6/25/2014 19:52:23

I had this discussion with Piggy before. It's generally true that games "won because of luck" are because one player decided to play risky and the other player decided not to push luck.

Not pushing luck means not relying on getting 1st order/last order. Not relying on 3v2/6v4 working for your expansion to be good. Not relying on guessing where the opponents are.

You play in a way that would be good for the 3 scenarios above to be true, but not to end the game if it doesn't work. It's hard for all 3 scenarios to work in a game, and in those particular games where all 3 of the above work for someone, that game is decided on luck.

"1 in 335,544
I think you dropped a few digits there"

I think he didn't.

0.000002980232238769531%

Yeah, looks like 1 in 33,554,432, due to max being 100, not 1 when dealing with percents.

when I refer to 'luck', I'm talking about the cumulative luck graph under Statistics. Yes, there is also the luck of the initial territory distribution, luck of the cards, etc.

I check the cumulative luck (which I presume is combat luck only) after every game - it seems to be the prime determinant when playing against anyone who is at least half decent.

luck is far from the primary determinant. It very well could be in your case but as Lolowut mentioned if that is the case for you then you are likely playing to risky and relying on luck to win. If you rely on luck to win then you are at the mercy of luck. In other words you are creating a self fulfilling prophecy.

you have a 60% win rate so are you saying that on avg you are 10% luckier than the avg player?

If you picture the distribution of win percentages as a bell curve, I am saying that bell curve is kinda tall and skinny, perhaps due to luck.

a small standard deviation, to put it mathematically.

http://images.flatworldknowledge.com/shafer/shafer-fig05_006.jpg

"1 in 335,544
I think you dropped a few digits there"

I think he didn't.

0.000002980232238769531%

Yeah, looks like 1 in 33,554,432, due to max being 100, not 1 when dealing with percents.

1 in 335,544
1 in 33,554,432
He definitely dropped a few, unless you're saying 33 million and 330 thousand are the same

Edited 6/28/2014 18:22:06

Even with 75% luck, games aren't decided by luck. All luck does is define a range of possibilities. A good player is aware of this spread and makes tactical decisions.

For example, if it's very important to take a certain territory defended by 2 neutrals, then a smart player will commit more than 3 armies to take said territory.

Unfortunately I have to disagree Taxi. Yes if all things are equal luck determines the outcome. BUT all things are not equal. People have distinctly different skill sets and this can work to the advantage or detriment of another player. I would not say it is luck that prevents me from beating ottoman but rather his particular skill sets play out as an advantage against mine. There are far to many factors at play to suggest luck is the singular determining factor.

not the singular determinant, but the primary one.

Taxi, by your reasoning your record has a probability of 0.000004%.

I mean, if luck is the primary determining factor between victory and defeat in a 1v1 then the game is a simple binomial probability with odds of 50%. A coin toss if you will. The odds of getting at least 429 of 714 falling the same way is on the order of 1 in 2.5 million.

So no, luck is not the primary determining factor.

It'd be extremely interesting if Fizzer could provide some stats for that. For example make a sample out of all 1v1 ladder games and show the relation between cumulative luck and winning %. Suggestions:
- win% for players with ending luck differential set between x and y, where x and y will be 0,1; 1,2; 2,3 etc.
- win% for the same thing, except we look at luck on turn 1
- same, but we look at turn 2
- same, but we look on turn 3

I'd expect to see a higher correlation when we look at early turns than late game.

I'd expect to see a higher correlation when we look at early turns than late game.

γηράσκω δ᾽ αἰεὶ πολλὰ διδασκόμενος.


Game changing luck is not only the simple math of armies clashing, but who gets the first order in the critical moment decides plenty of games by itself.

Yeah, but i can't think of easy methodology to gather relevant data to measure the importance of that.

myhandisonfire WarLight Member

Game changing luck is not only the simple math of armies clashing, but who gets the first order in the critical moment decides plenty of games by itself.

finally someone who made some sens

luck is a primary factor.

The problem lies with complexity.

Skill vs luck vs experience(in the map/settings) vs mood vs time available.

It has been proven that multi day games show better moves then real time games.
So i bet no disagreement there that time available is a factor.

Mood is also a factor, if one is drunk then it is likely to perform differently then usual when playing, so if he is tired or with a headache.
So i bet no argument there.

Now some may disagree here but it is clear to me that skill and experience in a particular map are 2 different things that do work hand in hand.

You can be the best player of warlight with regards to skill but if u never played a map you wouldn't be able to know every thing there is to know about it.
eg what is the best positioning to be in at turn 20 or so. One can try to guess but in the end you need a couple of games with different scenarios and different players to really grasp the best starting positions of the map in the long run.
Even different settings may trow all your skills out the window and change the game completely.

all these are quite strait forward, the better players has better chances of victory.

when it comes to luck and playing against an average or above average player luck may decide the game.
So looking at stats of players with different skill levels is useless, don't bother asking fizzer to waste his time.

the problem with luck is that luck factors are just too many in this game:

1) loosing your first and getting bad counters
2) missing on neutrals with 16% luck
3) your enemy not missing on neutrals with risky moves thus getting superior expansion rate.
4) first moves decide games
5) last moves decide games
6) unlucky on guessing the enemy tactics(agressive/defensive)(not when done on purpose but just because he was drunk or something), resulting in your not best move.
7) guessing correctly the amount the enemy reinforces is mostly luck.(in equal probability scenarios)
8) in equal probability scenarios of attack/defense on multiple fronts, guessing the right place to defend/attack.
9) bad luck of not having the right card in time could decide the game.(if i had that blockade 1 turn before)
10) luck on having Intel on enemy locations by sheer chance which gives you an edge on the enemy.
11) not loosing your first but loosing other some other vital picks and their counters.

and it goes on like this....

The main point here is that while the other factors have levels(not drastic advantages), the luck factors are just too effective(very drastic advantages) in the game and too many.

So if i had to give a ratio of how much all these luck factors effect a game of 2 equally skilled players with equal experience, all i need to do is looking at some of their games and see that most of the time if not all the times, they win or loose because of a luck factor mentioned above.
On very rare occasions one would completely outsmart the other.

this means that luck is the main factor when comparing 2 people of the same skill and experience.

but in warlight we don't have everybody with the same skills and experience, we have a jungle of variety.

so skills and experience, tactics and styles have an effect but they are in now way even close to the size of the luck factor when above average players are playing.

so my ratio would be:
50-60% luck factor which includes all those mentioned and more
20-30% skills, time, mood and styles
20% experience in the map

That is why you see skilled players winning glorious games against pros then later you see them loosing a game to an average player.

This wouldn't happen if the game wasn't around 50-60% luck based.

I'm not sure how much credibility I can give to someone who doesn't know the difference between lose and loose.

I encourage you to look up the definition of expectation value, mean and standard deviation. If you understand those concepts, the luck setting in a game simply requires you to adjust your strategy. A person who understands statistics and adjusts his strategy to the luck setting can make that luck work for them.

really, he writes more than 670 words and you don't give him credibility for a simple spelling mistake? I suggest you look up the definition of credibility.

---

Some people like to play purely luck (lottery), while some prefer as little as possible. What's so great about warlight is that each player can play on the kind of settings they prefer.

Also I'd like to point out something, In chess there's arguably no luck*. While in warlight there's, how come that in both games a much weaker player can beat someone who's presumably much stronger? It seem like most of you tend to think the reason for that is luck, yet in chess (with no luck) the weaker players also sometimes beat the stronger ones. Luck is not why you "see skilled players winning glorious games against pros then later you see them loosing a game to an average player".


*I'd like to argue there's luck in chess too, but that's a whole other discussion. nevertheless the point that a good player lose to much weaker players in games where there's definitely not 50% luck based, proves Metatron's last statement "This wouldn't happen if the game wasn't around 50-60% luck based. wrong, because it still happens.

A person who understands statistics and adjusts his strategy to the luck setting can make that luck work for them.

It seems you understood nothing of the list of luck factors i listed since they are all independent of the luck settings adaptability.(which is part of skill)

In chess there's arguably no luck*

If you have anyway to support this claim, please do so.
there is luck in chess, but it is no where close to 50-60%
I play chess and the ratio of losses vs average players is very very low compared to warlight.

Warlight is a much more complex game then standard chess with different luck factor, so you cannot really compare the 2.

"This wouldn't happen if the game wasn't around 50-60% luck based."wrong, because it still happens.

It doesn't happen in chess and as I said, they cannot be compared.
In chess in most cases the better player wins, its very rare that a much lesser average player wins against a pro.
assuming an average player reads 3-4 moves ahead and a pro 6-10 moves ahead with a bunch of contingency plans.

Edited 7/1/2014 23:02:54

a lot of discussion. I'll respond to the original subject. Luck isn't everything. You're a noob.

taxi driver = boston