Hey, so when are these changes scheduled to be implemented?

Weird idea: could it be done double way?
I mean: the RR menu works like "everyone"(default), "at least E", "at least D" ... "only A": single way looks logical here.
In 'site-experience', things look different: a Veteran may want to exclude all Newcomers but a Newcomer may want to exclude all Veterans! or to play only with Newcomers!
So what if the options were:

- Only Veterans
- At least Practit
- At least Rookie [= no Newcomers]
- Everyone (default)
- No more than Practit [= no Veterans]
- No more than Rookie
- Only Newcomers

Or, slightly different but more intuitive, how about 4 checkboxes? Default is "all checked" (everyone allowed), and you can un-check the box(es) of the rank(s) you don't want into the game.

Thoughts?

I like that, Guaroz. But what determines a noob versus a veteran. As I pointed out in the other thread. We have veterans here with few turns under their belt and who won't likely have some because they don't play lots of games at a time and like longer phases. But you can get noobs who jump into every live game and manage to survive cause they don't know a survive is worse than a defeat and rack up the phases quickly. There needs to be a balance between time on the site and turns or maybe have a flag the mods can turn on that says someone is a veteran even though they haven't got the numbers yet. Kind of like recognizing the sister/parent site status.

Yes, YCHTT, you're making a good point, but I believe this is a different matter. Oli explained it well, some post above:

"It's just to make it easy to set a certain minimum-site-experinece for your games, not for people to feel special. That's what the DPoints are for. :-)"

So any player's quality you want to measure must be based on some fact, or some stat, you can gather here and elaborate automatically.

So:
- Skill-Level and Rank are based DPoints;
- RR is based on NMRs and CDs;
- site-experience is based on phases played.

Nobody should think that a Newcomer is a noob only because he joined this site yesterday. Perhaps it's 30 years he plays Diplomacy and he played 1000 games on other sites. But he joined THIS site yesterday and this is a fact.

It's important that those quality-mensurations are automatic. Only if they're automatic and based on facts thay are fair for everyone.

So YCHTT, if you want good stats on this site...you should play on this site! :-)

Now please play here and let us see how good you are! ;-)

To be clear: I don't want introduce a new stat.
I just need an easy way to set the minimum required phases for a new game (if someone want to define this during gamecreation). It's very obvious that this is no indicator for experience. Maybe a single checkbox for 50+ phases played is enough. There are so many variants here that I think it's not good to make joining a game extra-hard.

@Guaroz - that was my point. Although looking at my phase count, I'm already up over 40. So maybe I will hit 100 pretty quick. But I was just suggesting that WebDip players are a "special" group in that they are not only experienced in Dip, but in the interface in general, so maybe a flag that could be set to indicate they have the experience despite the lack of phases should a mod/Oli be able to confirm they are who they say they are.

@YCHTT. Yes, I got what you was suggesting. The point is that there are may sites sister to WD. I played about 40 full press games on Facebook Dip before I got here. And when I got here there were maybe less than half Variants that are now. People may come from webdiplomacy.it, and God only knows how many sister sites there are on the web. So, since this feature has already a rather limited usefulness, why should we make it more complex than we need?

We could remove the RR-grade till someone has played 20 phases. He won't be able to join any games that require a special grade, but most of the games have no limitations anyway...

Well, the complexity I was talking about was ycthtt's proposal of making a further group for who comes from WD.

It's hard to call. 20 phases are too much for someone who comes from another site and too little for someone who has started Diplomacy 2 weeks ago.

The first one has probably already played tons of phases somewhere else.

The second one hasn't even completed a game in his life and probably doesn't even know what all these buttons and features are for. Or like the one who created his first game (1066, anonymous gunboat), advertised it on the forum (debatable), and when the game started he posted on that thread: "Yippiee yah yeaaahhh! I'm England!" :-)

Been thinking about this again. I do tend to come back to the question: "Will this player miss any phases when i play a game with him?"

Just calculate this with: #phases missed / #games playing. Don't use # games finished to prevent a division by zero for new players. In the case of a player that has not started or finished any games you can't give a reliability rating anyways.

If you tend to make it more fair you could use the average number of phases a game takes (about 13 by what i can see on the variants page, adjusted for variant popularity). Then the formula becomes: # phases missed / # phases played * 13

So what to do about leaving games? Because that's what really terrible. Well, make it obvious. For instance: If you've never left a game your reliabilit rating is shown in green. If you've left a game the color is more yellow (pay attention!). If you've left two, three or more games color it orange and bright red respectively.

You see a red reliability rating? Get outta there. Is it green but at 90%? Watch out, this person is pretty forgetful. It's yellow but at 100%? The player might tell you that he/she thought leaving was no problem. I would have that conversation anyways if a players profile shows even just one 'left' game.

I dunno, but seemed a more kiss solution (http://en.wikipedia.org/wiki/K-I-S-S) and certainly not overburdened by complexity. You can still have the A, B, C etc ranks if you want.

In the second to last paragraph i meant that a player might have thought leaving was no problem when he/she just started out. Until you've given it some thought you might not see it's pretty terrible behaviour.

Green v Yellow in the computer world is a nightmare to the colorblind.

Oh, the hard life of the colorblind ;( So what colors would work better?

We could also use a bold or underlined typeface or something.

I like the existing way. Sqrg's way makes mine slightly lower (still an A though)

Lose the things that use a single primary color difference, especially red or blue. So no blue and pruple, green and brown, yellow and green, red and orange. Instead, use Red, Yellow and Light Blue or White and alter the intensity.

@gman: Everyone their own preferences. Personally I think you're not the moest reliable player with 4 missed phases ;D But again, to each their own.

Personally I like the idea of simply applying the game reliability rating:
'I only want B or higher players in my game', or
'I only want A+ players in my game' (IMHO Amby's idea of A+ for 98% higher players is a good thing)

as for somehow mod-crossing from webDip to vDip... well what can I say? it's a different site that is much more interested tweeking/evolving the original code than the parent website.

I'd like this idea of A+ for 98% higher players, but... many people liking it... isn't it just the symptom that RR is not harsh enough towards NMRs? Then just divide the missed phases by games played (instead of phases played) and/or change the p parameter.

I tried to summarize in some graphs the current situation. I hope it helps to understand the issue.
Anyone interested in the topic please have a look at this:
http://wiki.vdiplomacy.com/index.php?title=Some_stats,_27-03-2012

Those graphs look good, the only other interesting stat I would like to check is:
x-axis=time
y-axis=# of games with an unbalanced CD

x-axis=time
y-axis=# of 'open' games

It would be interesting to see if the situation is getting better or worse...

I'm not great at reading graphs, but what I got out of it is simply that the new players are much less reliable than the players who've been here longer. It seems that once a player's committed enough to hit 600 phases on this site, reliability stops being much of an issue.

@fasces349
that would be nice to see how the situation changes with time...
...but I don't have this kind of data available.
By now I'm fine with checking the #open games.
It's still about 40 (more or less the same number as 2 weeks ago):
perhaps the majority of the users still doesn't know or doesn't realize the benefits of a Take Over.

@gman314
Perfect.
That's THE information those graphs are supposed to show.

yeah, its a pitty, I suppose I could keep checking the open games tab.

All you need to do to check on a couple of stats relating to the site is scroll down at the bottom and find out how many starting, open, active and finished games there are at one time. You can also find out about the number of players registered, playing and logged on. If you want, you can then graph some data over time.

Except they could be reflectingbthe fact that quarters don't hang around to hit 600 phases.

Good lord the autocorrect demon runs rampant. i was trying to say that quitters don't don't necessarily hang around 600 phases so that may be what the graphs reflect.

Yes, and that is why players with 600+ phases are generally reliable. By the time you get to the category with that much experience, the quitters have been separated from the rest.

I agree with gman314, and I wonder if the current system shouldn't be improved. If I'm not wrong, I believe that now a player, who's just signed up the site, can join 2 games. If he doesn't miss any (of the 2) starting phases, he can join a 3rd game. After 6 phases there are no more restrictions, so maybe it takes a couple of days.

This system's goal is to prevent curious people, who maybe don't even know the game, to join too many games right after the sign-up and then disappear because they're not really interested in Diplomacy, or because they are but they realized this site speaks a language they don't. For this purpose it works perfectly.

But, looking at DL's stats, I wonder if it couldn't be improved making the steps a bit longer for newcomers. Many of them are like children who get tired of a new toy after a week or two. Perhaps there's a way we can do it without discourage honest people willing to play properly.

We'd have 2 advantages:
1) Give to newcomers the time to understand how this site works, esp the Reliability Rating thing. If it took 2 or 3 weeks before they could join an infinite number of games, we'd have so much less NMRs in our games and so much less newcomers destroying their RR.
2) People with a low RR would try to recover their RR instead of just open a new (limited) account.