<< Back to Ladder Forum .GrayedOutTextBox { color: Gray; }

 Posts 1 - 23 of 23
The Current Ladder System and the Future of it: 1/2/2015 23:40:17

Benjamin628
Level 59
Report
The 1v1 and 2v2 ladders are arguably messed up. I propose that the solution to this is switching from the ELO System to the True Ranking System, and changing up the settings

The ELO System is the system used in the game of Chess, although it has its advantages, it is not optimum for Warlight Ladders.
In Chess one will always have 1 of 2 starting positions, in binary this is a 1 or a 0, carrying 1 bit of information, while in Warlight one will always have 1 of 2248091 possible starting positions (1v1 ladder), carrying 2248090 bits of info.

The randomization factors make the ladder games influenced by luck so turning down the luck to the new "0% luck option" would make games less of a gamble and actually show true skill. Here are two scenarios:

Lets imagine Players A, B, C and D are all on the 1v1 Ladder. Player A is first, Player B is second, and so on.

Then they all play games against each other:

With Current Settings, we can simulate:

Player A Beats C & D but loses to B
Player B Beats A & D but loses to C
Player C Beats A & B but loses to D
Player D Beats B & C but loses to A

With this scenario, although it is very unlikely it looks exactly like this it can skew the settings. since player C beat the top two ranked players, he gets the most points, although being the "third best player"

With No-Luck-True-Ranking, we can simulate:

Player A beats B, C & D
Player B beats C & D
Player C beats D
Player D loses every game

Then the ratings correspond to the player skill level.

Due to the law of large numbers, the second Scenario I showed must be true, because if Player A is truly the best, then he will beat the other players more often than they beat him.

Edited 1/3/2015 03:38:05
The Current Ladder System and the Future of it: 1/2/2015 23:51:46

Thomas 633
Level 56
Report
*looks around solemnly*

We did mate, we did.

P.S. You are absolutely right. Also I kept getting games in real-time AFTER I said leave ladder.

???
The Current Ladder System and the Future of it: 1/3/2015 00:22:42

Richard Sharpe
Level 59
Report
Tactical Booting; the top player has not moved in position since November 26th and I looked back at his games and most of the time he won by booting.
No such thing. Autoboot means Warlight does the booting on the ladders, not players. Can't blame a player if his opponents abandon the games instead of surrendering.

Usage of Money
Frankly, who gives a shit? If some player is so pathetic that they feel the need to BUY their way to the top of the ladder then let them. As the saying goes, a fool and his money is soon parted.

Not developed for stats like average income per game, total armies deployed,
Nor should it be. Income and total armies are utterly meaningless. Any moderately experienced player could have told you that.

If you are very good at starting in Africa
If you are very good at starting in Africa and only starting in Africa then chances are you are not a very good player to begin with.

Note, these are just the most glaringly obvious holes in your comment. There are many, many others that I can't be bothered to address

Edited 1/3/2015 00:29:04
The Current Ladder System and the Future of it: 1/3/2015 01:04:25

Benjamin628
Level 59
Report
Tactial Booting is a major problem in many games, most noticeably diplomacy.

Warlight should have a prevention system for cheaters as I have already pointed a cheater out.

My example of Africa is just that - an example.

Statistics makes things more interesting. Here is the ladder Homepage (no better word):

1 v 1 Ladder using the Strategic 1 v 1 settings.

2 v 2 Ladder using the Strategic 2 v 2 settings.

Seasonal Ladder which uses different settings each 60-day season.

Real-Time ladder which uses different settings for each game.

Utterly Boring for most.
The Current Ladder System and the Future of it: 1/3/2015 01:13:31

Richard Sharpe
Level 59
Report
Sorry, but booting is NEVER cheating. It may be poor sportsmanship but it is by no means cheating since the boot time is a construct of the game.

Statistics only make things interesting if they have some meaning. Placing more armies or killing more enemy armies is meaningless... it just means you take longer to kill your opponents (or to die). It is by no means indicative of player ability or skill.
The Current Ladder System and the Future of it: 1/3/2015 01:21:54

Benjamin628
Level 59
Report
Again I am giving examples

But because you clearly do not understand that, here are somethings that could be tracked on the ladder and are correlative to skill:

Average Time in turns it takes to win a game
Win:Loss Ratio
Average Income at the End of the game
Local Statistics: like % of the time beating people in your country or such

And other things like that. If the Ladders tracked those things it would be an interesting experience.
The Current Ladder System and the Future of it: 1/3/2015 01:35:09

Richard Sharpe
Level 59
Report
Sorry, but neither W/L ratio, Average income or local statistics are correlative to skill in any fashion. The latter two are utterly meaningless while the first (along with average time) is far too dependent on quality of opposition to be useful in determining individual ability.

I have to say, too many of your comments/points illustrate a basic lack of understanding the game and its core concepts
The Current Ladder System and the Future of it: 1/3/2015 02:16:21

Benjamin628
Level 59
Report
Actually, I do understand the game. I have been playing RISK since the age of 5 and understood that on the Risk Map starting in Australia/South America is a good idea. So yes, I do understand the core concepts of the game, and if win:loss ratio is not skill correlative then ELO sucks for warlight.

https://www.warlight.net/Map/2-Small-Earth

The Current Ladder System and the Future of it: 1/3/2015 02:38:03

Level 60
Report
"I have to say, too many of your comments/points illustrate a basic lack of understanding the game and its core concepts"

I have to agree strongly with Richard on this. Ben, I actually looked at your last few 1v1 ladder games. You make picks of 4, 7, and 9 to start off with. You do not even understand the most basic concept of making 6 picks when each player gets 3 starting spots. Take a look in the mirror and try to improve you play, instead of advocating to change the system that you do not yet fully comprehend.
The Current Ladder System and the Future of it: 1/3/2015 02:47:56

Gundisalvus
Level 58
Report
As a player of the game of chess, I found this thread to be nonsense.
The Current Ladder System and the Future of it: 1/3/2015 02:48:47

Benjamin628
Level 59
Report
I do understand the system. According to Richard, there is no need to draw attention to the unnecessary. I pick territories without thinking about the numbers, I stop when I think it is impossible for the opponent to force my to randomly pick, and the game where I picked 4 I got all 3 picks.

I have been trying to improve my play, I have not been at my peak, but I sure do understand how the game operates. I know how ELO works, as I said in my original comment, I play Chess.
The Current Ladder System and the Future of it: 1/3/2015 03:13:06

Richard Sharpe
Level 59
Report
Actually, I do understand the game. I have been playing RISK since the age of 5

Shame this isn't Risk. It's a Risk-like game but the actual mechanics of the game are dramatically different.

Not understanding something so simple as how many territories to select is proof that you haven't actually paid attention to the game mechanics. The fact that you seem to ALWAYS triple pick (and triple pick poorly at that) tells me you don't understand the strategy of the game either.
The Current Ladder System and the Future of it: 1/3/2015 03:38:12

Benjamin628
Level 59
Report
Edited. I understand the fallacies in my original

Edited 1/3/2015 03:38:27
The Current Ladder System and the Future of it: 1/3/2015 04:25:33

Richard Sharpe
Level 59
Report
Nice... I like the approach of changing your entire opening argument, thus making all ensuing conversation look completely off-topic and out of place.

As for the new one, you have a slight problem...
Player A Beats C & D but loses to B
Player B Beats A & D but loses to C
Player C Beats A & B but loses to D
Player D Beats B & C but loses to A

You have every team at 2-1, giving an aggregate record of 8-4. Not quite possible.
Of course, that happens when A beats C but C beats A and when B beats D but D beats B...

As for the discussion regarding the ladder ranking system, perhaps you should review past forum posts as it has been discussed ad nauseam by people far more familiar with the game mechanics and ladder than you or I.

Edited 1/3/2015 04:31:43
The Current Ladder System and the Future of it: 1/3/2015 04:51:51

Benjamin628
Level 59
Report
I understand that has a loophole. I, frankly do not care. My whole point, regardless of the delivery is still the same - ELO Rating does not clearly represent the skill level
The Current Ladder System and the Future of it: 1/3/2015 14:05:47

Ollie
Level 62
Report
42% 1vs1, 32% 2vs2 and 29% 3vs3... Sorry guys but i dont even understand why there needs to be a discussion on how much he understands of this game. Let him live in his dreamworld where he will become a chess worldchampion and beat everyone to become #1 in the ladder. Sooner or later he will fall back down on earth..
The Current Ladder System and the Future of it: 1/3/2015 14:32:23

Hennns
Level 58
Report
ELO Rating does not clearly represent the skill level
And what, exactly, makes you think it is supposed to? you say you understand how ELO works, as you play chess. In chess ELO does not represent skill level, what makes you think it does in Warlight? ELO represents the relative skill between players, that is a higher rated player will on average win more than 50% against a lower rated player, the win ratio is greater with greater difference in rating. Not only is Richard correct in that you seem lack a basic understanding of Warlight, to me it seems like you do not understand ELO either..

here are some things that could be tracked on the ladder and are correlative to skill:
Average Time in turns it takes to win a game
Win:Loss Ratio
Average Income at the End of the game
Local Statistics: like % of the time beating people in your country or such
And other things like that. If the Ladders tracked those things it would be an interesting experience.

While I do think it would be fun to look on statistics like that, they are completely pointless with regard to skill. Since you said you played chess, lets use that as an example again. let's say player A beats played C, the game last 82 moves, player A have 9 Queens left. Player B also beats player C, but he "only" have 1 queen left, the game lasts 45 moves. Now who is best of player A and B?
of course there's no way to tell, in warlight there wouldn't be either, whether you use avg number of turns/income or some other stat.

Win/Loss ratio is a bit different, but except for the very top/bottom the ladder will try to make the players w/l 50%, because then you'd be at your skill level.

Edited 1/3/2015 14:32:35
The Current Ladder System and the Future of it: 1/3/2015 15:33:22

Benjamin628
Level 59
Report
9 Queens left
Chess

I'm not even bothering to defend my point but this is clearly not an argument
The Current Ladder System and the Future of it: 1/3/2015 15:41:14

Hennns
Level 58
Report
No, it is not an argument, that's an example; to prove a point. (the point being that you cannot use income at the end of game etc. to determine skill for the same reasons 9 queens doesn't show more or less skill.)
The Current Ladder System and the Future of it: 1/3/2015 15:53:59

Benjamin628
Level 59
Report
income at the end of game

If your income at the end of the game is 0, you lose. If your income is over 90 you win almost certainly, so if your average is 50, chances are you win most of your games. If you win games in less moves, you are usually better, because some clans do not recruit players if they don't have a gold star on the tutorial levels and such. The whole point of the new coin games is to not have any luck, so bringing it to the ladders would show more skill in first place.
The Current Ladder System and the Future of it: 1/3/2015 16:04:33

Hennns
Level 58
Report
If you win games in less moves, you are usually better
No, because it depends more on your opponent than you. just because the opponent surrenders earlier doesn't make you better.

If your income at the end of the game is 0, you lose. If your income is over 90 you win almost certainly, so if your average is 50, chances are you win most of your games.
You cannot have 0 income on any of the ladders... an income above 30 usually is enough that you've won (with exceptions ofc). and average income is more an effect of how long you keep going, eg if you opponent doesn't surrender early, and you've 5 more turns with +20 that'd drag the avg up.

The whole point of the new coin games is to not have any luck, so bringing it to the ladders would show more skill in first place.
This is the only thing you've said this far that isn't plain stupid (whether i agree or not doesn't matter). However it is completely irrelevant to your other posts.
The Current Ladder System and the Future of it: 1/3/2015 16:53:19

Ollie
Level 62
Report
are there seriously clans who recruit people only if they have gold stars on the tutorial levels? pls tell me which clans that are and remind me te never join those.. If you think a gold star tells anything about skill then you even understand less of this game then you showed so far. Getting a gold star on the single player means only you know how to beat the computer fast and ai plays dumb and predictable. Imho people with a gold star on each level have either to much time on their hands or are scared to meet real opponents
The Current Ladder System and the Future of it: 1/4/2015 04:41:32

Benjamin628
Level 59
Report
Imho people with a gold star on each level have either to much time on their hands or are scared to meet real opponents AIs are predictable. Humans are very predictable. The whole point of AI is to mimic a human, so your argument is invalid. Also it took me about 45 seconds to star level 1 for an achievement or whatever.

If you think a gold star tells anything about skill then you even understand less of this game then you showed so far

Stating that clans that apply with stars as a factor exist is nothing similar to saying stars show excellence.
 Posts 1 - 23 of 23