PDA

View Full Version : The % by our names in lobbys.


JesusYo
1 Sep 2007, 16:09
What does the % mean? everytime i win or my enemies quit it seems to go down. Currantly at 71%.

highheaven
1 Sep 2007, 19:55
rank i think

jb.jones
2 Sep 2007, 02:33
So the lower the number the better then right?

highheaven
2 Sep 2007, 16:08
I think so, but I don't know what he % is for

Bgretydews
3 Sep 2007, 08:47
It's the Signel Strength.

Shirdel
3 Sep 2007, 09:51
duh I knew that ever since the 1st time I connected to the Internet with my PSP.

parsley
3 Sep 2007, 10:04
It's the percentage of the last 100 games started that the player has finished.

Bgretydews
3 Sep 2007, 10:08
It's the percentage of the last 100 games started that the player has finished.

Don't think so, i see alot of 99% and im sure they havent played 100 games already.

parsley
3 Sep 2007, 10:29
Well, I implemented it... :D

As a low number is indicitive of bad connection or bad behaviour, we could hardly of let newcomers start with 0% - no-one would play with them.

To ameliorate this, newcomers are 'gifted' 100 completed games when they create their account.

(What we're really counting here is 100 minus the number of the last 100 games the player failed to complete, but "percentage completed, higher it better" is a much more user friendly definition and close enough for jazz.)

Bgretydews
3 Sep 2007, 10:52
Well, I implemented it... :D

As a low number is indicitive of bad connection or bad behaviour, we could hardly of let newcomers start with 0% - no-one would play with them.

To ameliorate this, newcomers are 'gifted' 100 completed games when they create their account.

(What we're really counting here is 100 minus the number of the last 100 games the player failed to complete, but "percentage completed, higher it better" is a much more user friendly definition and close enough for jazz.)

Hehe, sorry diden't see the Team17 Staff under your name ;)

Luther
3 Sep 2007, 11:14
Dont apologise to him. He'll just take it as an excuse to use words like "ameliorate" in his reply. There's no telling where it'll end.

JesusYo
3 Sep 2007, 11:57
So wait... Everytime i'm ***ing a game and my foes simply flick the switch and leave. That counts as a disconnect for me and them. Not a Forfeit for them/win for me. That's incredibly wrong.

Daz555
3 Sep 2007, 12:18
I have yet to finish a multiplayer online game. They always end with disconnects. :mad:

JesusYo
3 Sep 2007, 12:26
Yeah serious, 99% of ppl on wormnet have no chance against me. The l33t3st skills i've seen from a foe online are the basic ninja rope / dynamite tactics that my 4 year old neice can do. I'm blocking all the players that just leave when their down to 1 low hp worm instead of trying again on the 2 other rounds.
That's another thing, you should be updating stats at each round, people rarely do all 3 rounds especially after losing 1 in an awful manner.

parsley
3 Sep 2007, 12:56
So wait... Everytime i'm ***ing a game and my foes simply flick the switch and leave. That counts as a disconnect for me and them. Not a Forfeit for them/win for me. That's incredibly wrong.

In anything other than a two player game, you'll continue and the winner will win and everything's covered in lovely.

Unfortunately, in a two player game, there's no way of detecting who's being rude, so yes, you'll both be marked as failing to finish a game.

Your reliability may not be 100%, but then no-one else's is either. What constitutes well-behaved and ill-behaved will eventually be set by the standards of the community.

If you're a well behaved player, you'll seek out other well behaved players in order to protect your reliability and they will seek out you for the same reason. If you frequently fail to finish, your reliability will be significantly below the community's expectations and well behaved players may start avoiding you.

So, although you may think it's 'incredibly wrong' it does, in fact, work to the advantage of well-behaved players and provides strong encouragement to be well behaved for the others.

JesusYo
3 Sep 2007, 13:08
I'm afraid your idea isn't working in practice. I'm in the top 20 on many of the leader boards but i've got a very indesirable 68%. Simplying because 1/3 of my games arent against my buddys, and those games are 99% garuanteed d/c because no middle level player wants to spend 30 mins being bannanbombed up the worm hole. You should have it so when a game ends in a d/c whoever is still connected to wormnet/didn't choose the ingame quit command doesn't lose %. The player that actually has a failure of an internet connection or is just a bad sport should be the player being punished. Not the decent player.

parsley
3 Sep 2007, 13:33
The fact that 68% is undesirable is your own interpretation. The community will eventually work out what level is or is not undesirable. If others are running at 68%, they're not going to find yours any worse than their own, are they?

The "connected to wormnet" idea simply doesn't work: firewalls.

There is simply no reliable method to detect which side of a two player game has failed.

Domovoi
3 Sep 2007, 14:26
There is simply no reliable method to detect which side of a two player game has failed.

The one who is doing significantly worse than the other player? It may not be 100% reliable, but surely that's a pretty good indicator most of the time?

On the other hand, that might cause cheating. If you get lucky and kill two of your opponent's worms, you could just disconnect for an ensured victory.Tricky.

But, when your opponent disconnects, your game notifies you of said fact. Why can't your version of the game then send a message to the server that "I, [player1], was playing against [player2] who gave up. I am still online." If the server didn't receive an "I, [player2], was playing against [player1], and I gave up. I am still online." message from the other player, then that means a disconnect from player2.

parsley
3 Sep 2007, 14:33
We did think about that, but it's really, really, unsuitable. You've identified the first problem: it magnifies an advantage in the game into a guaranteed win. The second problem is one of definitions: what does 'significantly' mean in measurable terms. (Imagine the situation where the expert places himself at a short term disadvantage, knowing that his superior strategy will win in the long term?)

At the end of the day, if the community decides that it's a rubbish statistic, then everyone will ignore it and there's no harm done.

Domovoi
3 Sep 2007, 14:39
We did think about that, but it's really, really, unsuitable. You've identified the first problem: it magnifies an advantage in the game into a guaranteed win. The second problem is one of definitions: what does 'significantly' mean in measurable terms. (Imagine the situation where the expert places himself at a short term disadvantage, knowing that his knows his superior strategy will win in the long term?)

I was thinking in terms of "Player A has three worms with 100 health, and one with 56 health. Player B has two worms, each with 15 health. Oops, disconnect." You won't catch them all, but at least the extreme cases will be covered. I doubt that a player's short term strategy would encompass wiping out 92.5% of his team.

Anyway, you guys obviously gave this much more thought that we did, but it's still an interesting problem.

AndrewTaylor
3 Sep 2007, 14:56
I was thinking in terms of "Player A has three worms with 100 health, and one with 56 health. Player B has two worms, each with 15 health. Oops, disconnect." You won't catch them all, but at least the extreme cases will be covered. I doubt that a player's short term strategy would encompass wiping out 92.5% of his team.

Anyway, you guys obviously gave this much more thought that we did, but it's still an interesting problem.

As Parsley said, any player who's winning can just quit, then, and remove any possibility for the other player to recover the match. Worse still, how about player A has three worms with loads of health, but they're all very near the water. Player B has one worm with 15 health, in a hard-to-reach position, and stacks of homing missiles and airstrikes -- his frugality with weapons is probably why he's lost so many more worms. Player B is almost sure to win, but player A can disconnect, clock a win, lose Player B some reputation, and save himself a defeat.

It's too easy to exploit.

parsley
3 Sep 2007, 14:57
Yup, we've thought a lot about it and I know it's a long way from perfect, but it's a very, very tricky problem, very easy to get wrong.

If I had implemented the message-to-the-server algorithm, the only advantage would have been that quitter could quit with impunity. (I don't have time to go into it right to, but think firewalls.)

Anyway, catching extreme cases in a singular network situation is not the right way to approach it. Whatever the best solution is, it must be directed towards catching the majority of cases (to make it cost-effective) and must also be easy enough for average users to understand (to make it perceived to be fair). Note that it doesn't have to be short-term fair, only that long-term fairness wins out.

Domovoi
3 Sep 2007, 15:07
As Parsley said, any player who's winning can just quit, then, and remove any possibility for the other player to recover the match.

As stated above, I'm aware of that. I was responding to his "how do you determine who is on the losing hand" question.

Worse still, how about player A has three worms with loads of health, but they're all very near the water. Player B has one worm with 15 health, in a hard-to-reach position, and stacks of homing missiles and airstrikes -- his frugality with weapons is probably why he's lost so many more worms.

I find this scenario to be a tad unlikely. ;)

If I had implemented the message-to-the-server algorithm, the only advantage would have been that quitter could quit with impunity. (I don't have time to go into it right to, but think firewalls.)


Too bad, because I don't understand what the problem is. :p Oh well.

AndrewTaylor
3 Sep 2007, 15:13
Yes, but it a system can be exploited then it will be. If you have a system where disconnecting can lead to a win then that will be used as a strategy by some players. You could, say, abondon all positional play, inventory economy and long term planning, pick off a couple of easy targets early on with your most powerful weapons, and then unplug your router. Chalk up one free win. Such players would end up with a very good win/loss ratio and an equally good 'reputation' for playing 'til the end.

Is that unlikely?

Domovoi
3 Sep 2007, 15:33
Yes, but it a system can be exploited then it will be. If you have a system where disconnecting can lead to a win then that will be used as a strategy by some players.

Again, I'm aware of this. I'm responding to the "How do you determine who is losing" question. I'm not saying that it's a good way to prevent disconnecting.

Is that unlikely?

I didn't say that the 'disconnect when you are at a slight advantage' scenario was unlikely. I said that the "opponent worms at full health all in very precarious situations with your own worm at low health holed up with tonnes of weapons" scenario was unlikely.

AndrewTaylor
3 Sep 2007, 15:42
I didn't say that the 'disconnect when you are at a slight advantage' scenario was unlikely. I said that the "opponent worms at full health all in very precarious situations with your own worm at low health holed up with tonnes of weapons" scenario was unlikely.

Well I think it would be pretty likely if the game scored unfinished matches by judging who was winning, for the reasons I just said. People would engineer it, to get the win.

They would. They're just that sad.

Domovoi
3 Sep 2007, 15:52
Well I think it would be pretty likely if the game scored unfinished matches by judging who was winning, for the reasons I just said. People would engineer it, to get the win.


How do you engineer your opponent to hole up with loads of weapons while at little health, and then position all your own healthy worms in precarious situations? Wouldn't it be easier to just blow that single low-health worm away with your three healthy worms?

AndrewTaylor
3 Sep 2007, 16:20
How do you engineer your opponent to hole up with loads of weapons while at little health, and then position all your own healthy worms in precarious situations? Wouldn't it be easier to just blow that single low-health worm away with your three healthy worms?

You do what I just said: you blow all your powerful weapons as soon as you can to pick off easy targets and cause maximum health loss. It'll put you at a disadvantage in the long run because you won't be able to use those weapons when you really need them, but that's okay because you're going to disconnect before that happens. Assuming your opponent is conserving his weapons, the early use of a dynamite and a couple of homing missiles should be able to swing the balance far enough for you to quit and score a "win" at some point.

Domovoi
3 Sep 2007, 16:31
You do what I just said: you blow all your powerful weapons as soon as you can to pick off easy targets and cause maximum health loss. It'll put you at a disadvantage in the long run because you won't be able to use those weapons when you really need them, but that's okay because you're going to disconnect before that happens. Assuming your opponent is conserving his weapons, the early use of a dynamite and a couple of homing missiles should be able to swing the balance far enough for you to quit and score a "win" at some point.

If you manage to get your opponent that damaged without enduring more than a reasonable share of damage yourself, it's probably easier to just win the conventional way.

AndrewTaylor
3 Sep 2007, 16:37
Not really. He can return the favour at any time just by using his dynamite and so forth. If he waits for an opportune moment then he can probably do more damage with them than you did.

Don't get me wrong, I'm not suggesting it'd be foolproof, but it'd probably do better than playing fair.

Of course, you could fix it to a point by setting the bar for "looks like Player A is winning" ridiculously high, but that would achieve nothing except making people quit when they were losing a bit instead of when they were clearly doomed.

chili69
6 Feb 2009, 23:20
Well, I implemented it... :D

As a low number is indicitive of bad connection or bad behaviour, we could hardly of let newcomers start with 0% - no-one would play with them.

To ameliorate this, newcomers are 'gifted' 100 completed games when they create their account.

(What we're really counting here is 100 minus the number of the last 100 games the player failed to complete, but "percentage completed, higher it better" is a much more user friendly definition and close enough for jazz.)

I'm puzzled now... my score's dropped over time, but it always seems to drop after a sore loser disconnects just before/after the final killing blow but before the game has ended. A bug / loophole in the system I wonder?

robowurmz
14 Feb 2009, 11:32
Did you even read the statement of the GAME DEVELOPER up there?!

dapope
14 Feb 2009, 19:52
Ive been playing worms for a long time now and the only explaination of the % sign that I figured is that it is the Chance of winning against that person out of 100%. You'll have to do math yourself to find out the actual % compared to another experienced player. The begginers start off with 100% because they have no experience. The higher you look in the leader board you'll notice the lower the % are.

In the forum you'll find out that the lower the % means you've been quitting or people you played against were quitting. You can get your percentage back high by playing games where nobody quits.

I try to never play against people with low percentage coz they are probably quitters.

Chrrayden
15 Feb 2009, 12:40
Yeah, I noticed that. I waz just talking out my butt, not expecting anyone to read what I posted. Thats why I deleted it!
I never quit unless someone causes the game to lock up. Usally I'll wait a good ten mintues to varify the lock up.

Ive played you before!