Take a look at the new variant-page...
http://www.vdiplomacy.com/variants.php

Love it Oli!

Good improvement, that will help in the orientation for the choice of a variant.

Just 1 notice, I wouldn't report it normally but it's about the 1st ranked variant:
"Classic Pick your countries" has a users-dependent number of players (mostly used for 1vs1), not necessarily 7, looking at the tab in
http://vdiplomacy.com/stats.php?variantID=42
the correct number to insert is the sum of the fields, 545.
Thus the actual ranking falls down to the 7th place, currently "Modern Diplomacy II" is the most popular map.

Ohh... I love stats, my congrats to both!
Thanks to Decima Legio for his 'popularity rating' idea: simple and meaningful.
Thanks to Oli for the quick improvement and for the 'sort by' option: fantastic!

Would it be possible to add the 'map size' (total territories & seas) of each variant?

Nice!

Modern, Modern, Modern... also really enjoying this new Abstraction III variant. It's a fantastic expansion of Classic, making the map even more playable, and some of the weaker powers more formidable. A couple of minor tweaks, and it will be perfect.

fixed...

Ohhhh... more than I could hope...
Thanks Oli, fantastic as usual! :)

Question about avg. Duration.
The "turns" are intended to be diplomacy phases, right? So 2 per year, mm?
I ask because, if I want a forecast on how long a game will last before I join it, the correct computing would be (say Classic 24h/ph):
15 turns = 7.5 years
7.5 years x 5ph/y = .ca 38 phases
38 phs x 24h/ph = .ca 38 days
So, if I'll leave for a vacation on next May 1st (50 days from now), I can join a Classic 24h/ph today with excellent chances to play it till the end. Am I right or am I missing anything?

Right.

Some players are clearly aware about the time duration of a game.
I mean, they are aware BEFOREHAND.
I'd like to propose a way to expand this kind of awareness to all the v-dip community:
data (duration per variant in turns of play) are already gathered, let's just use them!
On the replies above there's a simple way to let the system estimate the duration in time units.
In my opinion, the best place to put this information is the page "are you sure you want to join this game"?
That's right after you click the "join" button on a new game, right before you are in.
The aim of this initiative is to prevent some of the "giving up" and some of the "force pause requests".
(the first issue is related to CDs, the second is related to talkings on forums about ongoing games).

i.e. the message should look more or less like this:
"You are joining a Modern Diplomacy II 36hours/phase game:
our stats say that this game will commit yourself for a duration of about 84days after the joining line is filled up. If you join, please be sure you can connect at least once every 36hours and carry on until the end.
Are you sure you want to join this game?"

The number of days is to be taken as an indication, an averaged estimation.
I think 1 significant figure is enough: the former value in the example message should be truncated to 90days.

I know the developers are rather busy. It's just a proposal.
It's up to the users to discuss and up to Oli to decide whether this "change" is worth or not.

That's GENIUS. My +1000.
This would be a smart way to use stats in our fight against CDs.
I've always figured lots of kids that, in some boring rainy afternooon, join 3-4-5 games without realizing how long they could last. Then, when Easter's holidays come, or Thanksgiving, or anything, simply say "Oh boy...I have to leave...I just can't finish this" and go CD.
And not only the kids. Some absent-minded user, too. Looks like we have a lot.
Also, this would TEACH, to new users & beginners, that committing in a game until the end is what this site's Administrators and this site's Community is expecting from them.
You are expected to play a game you're joining EACH TURN UNTIL THE END.

Perhaps I'd use the real date instead of xxDays, so: "this game will end approximately around April 30, 2012." What do you think?

Fantastic idea, DL. I hope Oli will consider it, as he got time enough.

Cool idea.
But how should be calculate the duration? The software only keeps track of the turns played, not the phases. And a turn can have 1 to 3 phases...

Well, first of all, it's not essential to have an accurate computing. It's just a forecast based on averages, the exact real duration is impossible to guess. So this is not the point.

That said, each year has 5 phases (Spr Dip, Spr Ret, Aut Dip, Aut Ret, Aut Bui). Yes, sometimes they will be only 4 because nobody will have to retreat, but it's rather rare. Further, sometime the very first turn could be repeated because of a CD. So, with a good approximation, a little rounded up, we can assume it's 5 phases/year. That is to say that it's 5 phases every 2 turns. So, if our stat says AvgDuration=15 turns, we could say:
Phases = AvgDuration*5/2
Phases = 15*5/2 = 37.5
Now, looking at the game pace (hours/phase), it easy to turn phases into hours and then hours into days.

Players would be warned that this is just an estimate.

The Point is that, since many CDs depend on Real Life issues, this way beginners, unexpert players and compulsive joiners would all have a Real Time forecast on how long their committment will be.
I've met beginners who told me "Hey, will this ever end? I thought it would have lasted 10-days or a couple of weeks...". Instead it was a couple of months...
So the point is not if 60 days is a forecast more or less correct than 70 days. The point is that:
- Many players (beginners, mainly) would discover that a game lasts longer than they believed and perhaps they won't join it if they realize they can't finish it (practical purposes).
- Any user would know that this Community expects that they _commit_ playing games until the end without NMRs (teaching purposes).

This is good.

Well, I don't have much explanations to add over what Guaroz already mentioned.
The above formula is mainly an overestimation:
it's based on both the conservative assumptions that 1year contains all the 5phases, and every single phase lasts until the time deadline.
If you want to find a fault in my proposal you shall say:
"changing the VCs will let the data gathered about average turns played lose some of their meaning".
I know, but IMHO I think it's ok this way; at least it's better than the current warning message.
I don't pretend to guess the real duration of a game.
I pretend every single user to be aware that the game he is going to join will last 1week-instead-of-1month-instead-of-6months, though.
Furthermore, a remind about the connection frequency you are "supposed" to respect (due to the game speed) should be mandatory.

Considering that a phase usually does not take the full phase-time it's really hard to calculate a reasonable timeframe with only an average turns-stats. I think I might start monitoring the real duration of a game (create a record of the start and end-time) and base an estimation on such a stat. But this will take some time...

Well Oli, If you want to add some new stats...I'm happy! I love stats!
But I feel like this is an unnecessary work to do. And, being you pretty busy, perhaps this is not what you want.

In my experience, I got all players finalized before deadline only when most of the players are defeated and only 3 or 4 players are still alive. Even then, there's often someone who likes grey 'V's more than green ones. In 80% of games I play there's someone who saves without finalizing since start to end. Perhaps is just people who joined a 'borderline' paced game (IE: I can log in only once a day and I joined a 20hrs/ph game. Then I always 'save' so that the game doesn't run too fast)

Each new stat you start is welcome, I love stats. But I call to your attention that accuracy is not essential for the purposes of Decima's idea. It's not even important.
Even if you had exact and reliable stats on Duration, the information you'd give would not be complete. Imagine we say _everything_:

"You are joining a Modern Diplomacy II 36hours/phase game:
our stats say that this game will commit yourself for a duration:
min: 73.2 days
average: 79.1 days
max: 89.7 days
after the joining line is filled up. If you join, please be sure you can connect at least once every 36hours and carry on until the end.
Are you sure you want to join this game?"

Now... Nobody knows how long the next game will last.... pauses and/or extensions could be granted.... and it could last 100 days, in spite of every forecast.
So it doesn't matter having accurate stats or formulas.
What does matter is that nobody will join that game thinking it's a 4 weeks committing, when once it lasted 3 months.

Decima said he'd have his "84days" value truncated to 90days. This matters.
3 Months is clear enough. Most of people DON'T like stats as you, Decima and I do. They find stats bothering and annoying.
So, wouldn't it be so much clearer this:


"You are joining a Modern Diplomacy II 36hours/phase game:
our stats say that this game will commit yourself for an approximate duration of:

- 3 MONTHS

This is just an extimate, the game could be something shorter or even a bit longer. If you join, please be sure you can connect at least once every 36hours and carry on until the end.
Are you sure you want to join this game?"


and, if doable, typing "- 3 MONTHS" in a bright color big fonts?