<< Back to General Forum   Search

Posts 1 - 26 of 26   
bad luck on the rise?: 5/11/2012 17:22:02


TRUMP 
Level 59
Report
5 beats 4 in iceland, turn #1: http://WarLight.net/MultiPlayer.aspx?GameID=2647315

7 beats 6 in austria, turn #1: http://WarLight.net/MultiPlayer.aspx?GameID=2651392

I seem to be seeing this more. Wonder if Fizzer changed anything in the way luck works?
bad luck on the rise?: 5/11/2012 17:56:22

The Duke of Ben 
Level 55
Report
I wondered the same thing, since I've seen more 2v2 and 1v1 succeeding lately. I've also noticed more 3v2 winning, but that could all just be my perceptions.
bad luck on the rise?: 5/11/2012 19:41:00


TRUMP 
Level 59
Report
interesting. correction on my first post, it's turn #2 in first game, not turn #1
bad luck on the rise?: 5/11/2012 20:06:38


Honey Badger 
Level 48
Report
I have also seen some bizarre fluctuations in luck of late. Hopefully it's purely subjective.
bad luck on the rise?: 5/11/2012 20:08:42

Heyheuhei 
Level 57
Report
i saw 1 year ago, when 10 beat 10
bad luck on the rise?: 5/11/2012 21:34:39


PinkiestPink
Level 52
Report
I saw 1 beating 50 some years ago, but it might have been Jackie Chan, I'm not sure
bad luck on the rise?: 5/11/2012 21:50:52


Ironheart
Level 53
Report
i once saw 2 beat 3 once.
bad luck on the rise?: 5/11/2012 22:03:58


Richard Sharpe 
Level 59
Report
Ironheart - no you didn't.

Everyone else - I wager it's simply bad luck and coincidence, nothing more.
bad luck on the rise?: 5/11/2012 22:17:27


Ironheart
Level 53
Report
i did i really did
bad luck on the rise?: 5/11/2012 22:21:43


Richard Sharpe 
Level 59
Report
No, you really didn't.

|>"Each army that attacks has a 60% chance at killing one defending army"
bad luck on the rise?: 5/12/2012 03:29:30

RvW 
Level 46
Report
@Iron:
You might have *seen* it, sure, but it just didn't *happen*. Possible causes for seeing it anyway include just not paying enough attention (it was a 3vs3, one attacker got killed, you saw 2 armies move in), sleep deprivation or any number of chemical "goodies" some people (no idea whether you are among them) subject their body to.

According to the game rules, even at 100% luck (or 0% luck and 100% offence kill rate), in order to beat X defending armies, you will always need, at the very least, X attacking armies; if the number of attackers is lower than the number of defenders, the attack fails, *by definition*. (For completeness' sake: there are lots of cases where even X vs X attacks cannot possibly succeed either.)

As you might remember from the [What are your most improbable failed attacks?](http://warlight.net/Forum/Thread.aspx?ThreadID=3153&Offset=20) thread, I worked on it quite a bit (arrggh, should finish it... :s ), so I really do know what I'm talking about.

---

@Everyone:

Human intuition and statistics are a terrible combination. A great example of just how badly human intuition stinks when probability is involved is the [Monty Hall problem](http://en.wikipedia.org/wiki/Monty_Hall_problem). Or, as the (pen and paper) role players amongst you know: "A fifty-fifty chance never happens; a one-in-a-million chance is a sure thing!".

There are a number of reasons why you might *think* more attacks are succeeding that "should've failed". The most important one is observation bias combined with psychology.
Let's say you toss a fair coin 16 times. On average there "should be" one string of four heads in a row. Your feeling tell you this is weird (it has a 0.5^4 = 6.25% chance of happening when you toss a coin four times in total; for comparison, at 75% luck, 60% offensive kill rate, 7vs6 has a 9.98% chance of succeeding). However, you might well shrug this off as "weird things happen", or simply not notice it happening at all (probably more likely with weird attacks than with coin tosses).
On the other hand, if you toss a fair coin 64 times, you expect four strings of four heads in a row. But, if that happens, you are (1) far less likely to "shrug it of", you can't accept that something that weird should happen "that often" (that's psychology for you; in a sample size four times as large you don't accept "unlikely events" will happen four times as often) and (2) you are far less likely to overlook the occurrence in the first place.

*(Disclaimer: I just got back from an all-night film marathon; it's 5:30 AM and I'm tired as hell. The above prose is probably rather badly written and I wouldn't recommend copying any of my calculations onto a statistics test... However, I'm pretty sure I didn't mess up so badly that the general idea I'm trying to convey is invalid.)*
bad luck on the rise?: 5/12/2012 06:57:20


MilitaryManiac 
Level 57
Report
I think there is something to this conspiracy, I had a game yesterday where a 10 killed 8 with 7 remaining armies. It was the deciding factor in the game :/
bad luck on the rise?: 5/12/2012 16:16:49


TRUMP 
Level 59
Report
played a game with my hand today, he also thinks that something has changed. he saw 4 kill 4 2x in 2 days.
bad luck on the rise?: 5/12/2012 16:39:36

Dr. TypeSomething 
Level 3
Report
This is a testable hypothesis. Set up a practice game, and keep having say a 10 attack a 7 (doesn't really matter, but large enough to make a purty histogram by the end). Keep doing it and record it each time how many armies those 10 killed. Repeat maybe 100 times. Make a histogram of the results. Compare it to what the analyzer says should happen. Should be close. If not, the great luck conspiracy caper of 2012 will be well underway.
bad luck on the rise?: 5/12/2012 16:43:20


Ⓖ. Ⓐrun 
Level 57
Report
i had a case today where 3 armies killed 2 with all two spare
i don't know whether that is anomaly or just cos there was a fair chance of it happening though
bad luck on the rise?: 5/12/2012 22:27:08

RvW 
Level 46
Report
|> Make a histogram of the results. Compare it to what the analyzer says should happen.

Unfortunately that is utterly pointless. The analyser uses the exact same code as the game engine itself (this has been stated on multiple occasions by Fizzer). The analyser comes up with its **estimates** by simply simulating the attack a lot of (1000??) times and simply counting how often the attack succeeds. In other words, what you're doing is essentially manually simulating a run of the analyser. Even if the calculation / code / whatever *did* change, this experiment will not confirm it.

If you want, you can use my analyser (link below) which actually calculates the success chance. When I first wrote it, it either agreed with the WL built-in analyser fairly closely, or it disagreed (but only in cases where the built-in analyser was known to get it wrong; the reason I even started that little coding project).
If it suddenly starts disagreeing with the built-in analyser now, that would be an indication something has changed. Note, if you find such a disagreement, please post number of attackers, number of defenders *and* offence kill rate and luck percentage (if it happens to be an X-vs-X attack, equal numbers of attackers and defenders), you should also include the *defence* kill rate.

***Warning:** as pointed out by Fizzer the last I posted this link, you're downloading executable code; essentially you are blindly trusting me to not have put any viruses or other nasty stuff in there.* (If you have even a tiny bit of programming experience, you don't have to trust me; [here](http://warlight.net/Forum/Thread.aspx?ThreadID=3153&Offset=19) is the important part of the code; feel free to copy-paste it into your own program.) If you choose to trust me, you can download my analyser [here](http://dl.dropbox.com/u/12477348/AnalyserGUI.exe).

---

@Arun:
Do you mean three attackers kill two defenders and only lose a single army themselves? That is a combination of 3-killing-2-(or-more) and 2-killing-1. Assuming 60% offence, 70% defence and **100%** luck (that's a bit easier to calculate :p ) that works out to:

60% * 60% * 40% * 2 nCr 3 = 0,432
60% * 60% * 60% * 1 nCr 3 = 0,216
70% * 30% * 1 nCr 2 = 0,42
( 0,432 + 0,216 ) * 0,42 = 0,27216

For **0%** luck (also easy to calculate) the result is:

3 * 60% = 1,8 -} 80% chance of killing both defenders
2 * 70% = 1,4 -} 60% chance of killing one attacker
0,80 * 0,60 = 0,48

So at 100% luck that's better than a 1 in 4 chance and at 0% it's virtually 1 in 2; either way not exactly "extraordinary" if you ask me. (A 1 in 4 chance is flipping a coin twice and having it come up heads both times, a 1 in 2 chance is flipping a coin once and having it come up heads.)
bad luck on the rise?: 5/13/2012 05:04:23

[REGL]Nighthawk30
Level 7
Report
Well, using the warlight analyzer, with the settings from your games (75% luck, 60 and 70 kill rate on offense and defense respectively) a 5 has a 27% chance of taking the territory from a 4. A 7 has a 10% chance of taking the territory from a 6. Not really that unusual, when you take the probabilities. For the 5,if you did it 100 times, 27 of those times, it would take the territory (just using the percentage). For the 7 over 6, it would happen 10 out of every 100 times.

We would probably see 5's beating 4's, and 7's beating 6's (and the like) more often if we actually attempted them. We just know instinctively that most of the time it would not work,and that we would need more troops for it to have a better chance, so thats why it seems like such a surprise when it did happen.
bad luck on the rise?: 5/13/2012 07:55:44


I'M RETIRED
Level 5
Report
You know that saying: an army of donkey led by a lion is better than an army of lions led by a donkey?
Don't forget that the armies were led by me :P:P:P:P:P


On topic: as the guy b4 me said, 10 percent chance. Unlikely but not impossible.
bad luck on the rise?: 5/13/2012 08:00:50


Ⓖ. Ⓐrun 
Level 57
Report
i meant that the 3 didn't lose any armies
bad luck on the rise?: 5/13/2012 09:00:48

RvW 
Level 46
Report
|> i meant that the 3 didn't lose any armies

Seriously, the math isn't that hard (especially since you only need to slightly modify what I gave above).

60% offence, 70% defence, 100% luck:

60% * 60% * 40% * 2 nCr 3 = 0,432
60% * 60% * 60% * 1 nCr 3 = 0,216
30% * 30% * 1 nCr 2 = 0,18
( 0,432 + 0,216 ) * 0,18 = 0,0317447424

~= 1 in 30

60% offence, 70% defence, 0% luck:

3 * 60% = 1,8 -} 80% chance of killing both defenders
2 * 70% = 1,4 -} 0% chance of killing zero attackers
0,80 * 0,00 = 0,00

Can't happen. Then again, since you didn't specify otherwise, I assume you were playing at default (75%) luck, where (I assume) the value would be much closer to the 100% luck value than it is to the 0% luck value.

I don't know about you, but I do quite a few 3-vs-2's, since if we only consider the chance of the attack succeeding, ignoring how many attackers survive, it isn't all that bad (68,40%). Doing something a lot means you will sooner or later run into it working out in an unlikely way, that's inevitable.
bad luck on the rise?: 6/17/2012 08:24:52

splinteriam
Level 2
Report
anyone knows how much luck do single player game has?

many times i attack with 5 (send 4) and lose against 2
and even sent 5 (having 6) and lost against 2
and i not sure but i think it also did happen sending more (6 sent).

cause i read some info that the 1 vs 1 template has 16% luck and u have 4vs2 win guaranteed (btw i think they mean 4 sent but anyways not only that was not guaranteed for me but also as i told i sent 1 more and even 2 more and lost
thanks in advance
bad luck on the rise?: 6/17/2012 09:57:47


Muppet
Level 12
Report
You can check the luck in the settings tap within each game. (On the lower left).

Level 1:25%
Levels 2-4:75%
Levels 5-6: 50%

You are correct that all of your examples can fail to win (4v2,5v2,6v2) and even higher. When making an attack, use the analyze tool (in the little window that pops up to control how many armies attack) to see the graph and percentages of success. Note that even when attacks appear to have 100% certainty, sometimes it is only 99%+, meaning that there is still a chance that the attack can fail.

What you're seeing is normal.
bad luck on the rise?: 6/17/2012 13:40:39

RvW 
Level 46
Report
Please keep in mind the analyser works by simulating the attack a lot of times. If it reports a success chance between 1% and 99%, that means the attack is neither guaranteed to fail, nor guaranteed to succeed; its actual success chance is (with very high likelihood) close to the reported number. If it reports 0% (or 100%) the attack is not actually guaranteed to fail (or succeed), you should treat it as a very, very small (or large) to succeed.

Assuming you're using a luck modifier smaller than 100% (default is 75%) it is possible to have guaranteed successes (such as 4vs2 on 16% luck), however there is no possible way to tell the difference between "very high likelihood" and "guarantee" from the analyser.

cause i read some info that the 1 vs 1 template has 16% luck and u have 4vs2 win guaranteed (btw i think they mean 4 sent but anyways not only that was not guaranteed for me but also as i told i sent 1 more and even 2 more and lost
thanks in advance

Yes, it is talking about the number of armies actually sent out to attack, not the total number of armies on the attacking territory. There are at least two easy ways to remember this. First, a territory with a single army on it cannot launch an attack, yet 1vsX attacks do exist. Second, if that same territory performs two attacks (let's say it has 8 armies on it and launches an attack of 3 armies to one neighbour and an attack of 4 armies to another neighbour), including the extra army which stays on the territory suddenly becomes very confusing.
bad luck on the rise?: 6/17/2012 15:19:25


Perrin3088 
Level 44
Report
4v2 is a guaranteed victory at 16% luck with 60% O kill rate..
4*0.6*0.84=2.016
bad luck on the rise?: 6/19/2012 20:18:38


[REGL] Pooh 
Level 60
Report
10v9 wins :(

http://WarLight.net/MultiPlayer.aspx?GameID=2820669

Turn 7 in Rivendale Bonus
bad luck on the rise?: 6/19/2012 22:06:32

RvW 
Level 46
Report
Considering the name of this thread, I do think it's worth pointing out that, from "The Germanator"'s point of view this is incredibly good luck (1.61% chance of succeeding by my calculation).
Posts 1 - 26 of 26