I don't know that weakening WTA is necessary or a good idea.

WTA seems to be a minority game type on this site, and I'm guessing those who swear by it only, will care about their pure WTA standings, separate from the overall rankings and not really pay attention to any PPSC rankings. Just a thought.

Can we see the algorithms compared with only an adjustment to PPSC solo victories? WTA should still be 100% of SC's for a solo imo.

I know I am very late to the game here, but on PlayDiplomacy.com they have a concept called a "friends" game as opposed to a "rankings" game. You click a checkbox upon game creation specifying what kind of game it should be. The friends games don't count toward site rankings and perhaps that would appeal to some people who are new and getting the hang of it, playing with real life friends, or just want to play casually.

Apologies if this was already suggested in the previous 19 pages.

I also think that a solo victory should be the ultimate achievement in a Diplomacy game !

For those who think so, or for those who are simply interested by an additional ranking, we could also have a "Solo Ranking" which takes into account only the solo results of a player.

It could be calculated, for example, very simply like this :

x is the total number of opponents that a player has played against in all his games in which he has achived a solo.

y the total number of opponents that a player has played against in all his games.

A player's score is simply x/y.

For example, a player has played 3 games :
- 1 World War IV game (35 players = 34 opponents),
- 1 Classic game (7 players = 6 opponents) and
- 1 1vs1 game (2 players = 1 opponent).
That player has achieved a solo in the Classic game.

-> His solo score is 6 / (34+6+1) = 14,6341 %.

We should of course take into account all the necessary digits after the decimal mark in order to rank players with precision.

That system could of course be more complicated to weight more or less such or such parameter but I think that even in its simpliest form, it would give good results.

In order to avoid some absurd results (such as a player being ranked N°1 with a score of 100% because he has only played 1 game which was a 1vs1 game that ended in a victory), we could simply state that a player's score is only valid if y is at least equal to 35 (which makes sure that he has played at least 2 games : for example 1 World War IV game and 1 1vs1 game).

What do you think about it ?

I didn't follow much this thread, but I believe I can say something useful about PPSC/WTA.

Since I started playing Diplo online, I had pretty clear the difference between these two Variants.
In WTA, any result different from win or draw means nothing, zero, total defeat.
In PPSC, there's an addictional result, "Survived", and what it means depends on SC count.

So while an almost suicidal action in order to stop a solo and get a draw makes sense in a WTA (because you can expect the others will work likewise), it doesn't in a PPSC because it's very likely that you'll get stabbed by someone who is content with a good survive. So, unless the solo is easy to stop, you'll have a propensity for defend your achieved result.

Almost every player has a preference with one of the two, but I won't get into this argument, since everyone is allowed to join games of the variant he likes the most.

I've never played a PPSC as if it was a WTA (and this doesn't mean I just looked for survives) and I've never played a WTA as if it was a PPSC.
Can you imagine someone playing for a "good survive" in a WTA? It would make no sense.
Likewise, there are things you can't imagine in a PPSC because they would make no sense.

So. My point is that if someone's desire is that a solo victory trump all survives, then he should play WTAs.
And Leif, "If the community's desire is that a solo victory trump all survives" then we should cut out the PPSC option and play only WTAs.

A rating system treating the PPSC games as if they were all WTA (or viceversa), would be hugely flawed because it's not "how the game have been played", like Leif himself said.
I'd have nothing against some VERY SMALL corrective if you think it's needed, but the spirit of each variant should stay on, IMHO.
I wouldn't like people starting playing WTAs as if they were PPSCs or viceversa. This is my concern.

Just my .02, and sorry in advance if I missed something important, I didn't follow much this thread.

Thanks Oli for the very interesting addition. I like the feature of being able to see everyone's game history in a concise, chronological table.

I do agree with solo ranking, if that is necessary

"I like the feature of being able to see everyone's game history in a concise, chronological table."

Sequential (per individual) is probably the more correct word. There isn't enough chronological information to compare to another player's game history to compare skill levels at a certain point in the past.

I believe the game history table is listed in chronological order based on when the games were completed.

so will one get ranked higher if he has more SCs in the end or is just the outcome (survive/win etc...) important?

I don't entirely agree with solo ranking - this encourages smaller maps + WWIVs, and discourages games such as Haven where solos are extremely rare. Obviously the system of having how many people you're against helps this a little, but if this is the case WWIV needs to be excluded entirely or be reduced to give a much smaller number of points, as it is much easier to solo than, to take my above example, a Haven.

The same goes for ImpDip2, as that's pretty much impossible to solo by any of the smaller powers and the larger powers are fairly likely to do so, so that would need to be excluded or the resultant points downsized for larger powers and upsized for smaller ones. This is not such an issue in the current system - as draws are perfectly feasible in this variant for smaller powers - but a system that focuses only on solos would need this excluded.

Also, it would be much, much faster to win 34 1v1s than 1 Chaos. This system would promote 1v1s much more than any other variant as they would be faster to solo and a solo would be more feasible.

Added the CDs too. They score 0:1 against all other (non CD) players. The HoF is not recalculated at the moment, but you can see this if you check the games...

@Captainmeme :

Of course, no system at all will be totally correct.
In my opinion, EVERY system will pervert the way that some players will handle their games.

Of course, variants like Imperial II seem to cause problems to any ranking system. So, that problem is not specific to a solo ranking.
----------------------------------------------------------------------------------------------------------------------
As a reminder, I have proposed that :

x is the total number of opponents that a player has played against in all his games in which he has achived a solo.

y the total number of opponents that a player has played against in all his games.

A player's score is simply x/y.
----------------------------------------------------------------------------------------------------------------------
1° You wrote that this encourages smaller maps.

That's wrong : if you win one 1vs1 game, x becomes x+1 and y becomes y+1
So, in my example of a player whose score would be 6 / 41 = 14,6341 % ; his score would become 7/42 = 16,6667 %.
But of course, don't forget that you can lose a 1vs1 game. In my example, if the player loses, his score would become 6 / 42 = 14,2857 %.

On the other hand, if the same player plays a fourth game with a Classic variant (instead of a 1vs1 game) and :

a) wins, his score would become 12 / 47 = 25,5319 %

b) loses, his score would become 6 / 47 = 12,7660 %

-> The conclusion is that the more players a variant has, the more points a player could win compared to the points that he could lose. But of course, you have less probability to win a game with more players.
So, why do you say that players would be encouraged to play 1vs1 games ? Have you only made one single calculation before asserting such a thing ?

And also, what kind of mentality is that ? Are players going to select their variants, their opponents, their game options, and such and such parameter in order to improve their ranking so that they could believe that they are strong players ?
----------------------------------------------------------------------------------------------------------------------
2° You wrote that it is much easier to make a solo on a WWIV map than on an Haven map. Really ? The current statistics show that :
11 out of 30 WWII games ended in a solo, that's 36,67 %
5 out 20 Haven games ended in a solo, that's 25 %

That is indeed a difference. Is that such a big deal ? Do you really believe that players will mostly select variants with the highest percentages of games that ended in a solo ?
By the way, that would be a miscalculation !

Indeed, for example :

11 out of 30 WWII games ended in a solo but only 1 player out of 35 wins a game that ended in a solo. So, 11 / (30 * 35) players who have played a WWIV game have made a solo with that variant. That's 1,0476 % of the players.

5 out of 20 Haven games ended in a solo but only 1 player out of 19 wins a game that ended in a solo. So, 5 / (20 * 19) players who have played a Haven game have made a solo with that variant. That's 1,3158 % of the players.

Ha ! Ha !
First of all that difference is tiny but, in fact, you have more chance to make a solo in a Haven game than in a WWIV game.


And anyway, don't you think that players will play variants because they like them or because they want to discover them ? Do you really think that players will choose a variant because they *think* that such or such variant is supposed to be easier to win ?
Yes, some players could wrongly be encouraged to play such or such variant because they *think* that it will improve their ranking. But as you can see, who really knows which variant is best to play in order to improve one's score ? Is there really such a variant ?
----------------------------------------------------------------------------------------------------------------
------
----------------------------------------------------------------------------------------------------------------------
By the way, I would like to say that it is my opinion that any ranking system will be flawed because they compare things that have nothing to do together, even though they are all called "Diplomacy games". Indeed, in my opinion, Diplomacy is a communication game. So comparing in the same ranking system a gunboat game with a full press game is, in my opinion, TOTALLY RIDICULOUS. Same thing with a 1vs1 game : where is the diplomacy in such a game ?
----------------------------------------------------------------------------------------------------------------------
Some people might be interested by a solo ranking. That's why I have proposed one.

bozo,

In response to:
"I believe the game history table is listed in chronological order based on when the games were completed. "

This is partially true, but, I can't compare the games in your list to the games in my list using a game ID or a timestamp and thus plot our two score histories together on the same axis over time. Hence it's not completely chronological, but rather individually sequential.

Captainmeme,

Retillion's proposal (which I think is a useful system in that it adds additional information) would be implemented alongside the ranking system that Oli has already implemented, not in place of. While I prefer Oli's system, I still think there is value in the simple rating that Retillion has proposed.

Hi again, is there some sort of working system in place now? If so, is it viewable by the public? I tried reading this thread but there is a lot of text.

@ Jimbozig : yes, please click on "HoF" in the upper right corner of this page, right under (Log off).

I won a game today (gameID=12539), but that is not shown on my history table (http://vdiplomacy.com/halloffame_elo.php?userID=790) and i did not get any points for that...why? :(

I don't the the automatic calculation part is live yet.. Is that true Oli?

I don't *think* the...

Since when is 'the' a verb?

oh ok :)

I need to add the code for the updating.
I tried not to mess with the original code till we have a working prototype.
Will add this soon now.

The code is now working and the HoF does update after each game.
I will remove the top-link in a few weeks, if there are no problems.
More features (like different pot/variant/presstype) will be added slowly.

If you want another rating added you need to make a detailed post on the wiki and open a discussion thread here. If the algorithm is final I will add this as I have the time.

Sometimes I look like a boring nerd, I know.
But sometimes I’m also convinced that I’m right, so…

There’s a correction to apply to the game value “gV” formula:
the formula was meant to be monotonically increasing with respect to the “numberCountries” n.

The current gV formula in the algorithm instead provides a local maximum around an hypothetical variant of size 16, namely World Diplomacy IX or Known World 901.


IMO you should be able to accomplish what you were looking for simply deleting the multiplicative factor
* (1 - (numberCountries / 100))
from the gV definition, but of course this is not the only way.

This operation will increase some rating oscillations due to the large variants games with respect to the current calculations.

We can't see the old Dpoints rankings in the HoF.

Is there a reason ? I thought that the new system would be in addition of the old one and not a complete replacement.

In my personal opinion, I think the list needs a little bit of cleaning up... I honestly don't feel it's accurate to have people in the top 150 that haven't been online since 2011... Isn't there a way to set it up to only show those who have been online in say, the last 6 months or at least year?

You can see the Dpoints on the same page as before (http://www.vdiplomacy.com/halloffame.php). I had them both on the same page so you can compare them more easy. Now that we have settled on a rating there is no need to compare both all the time.

@butterhead:
This might be a good idea for an additional filter-setting.
Maybe a default of 1 game played in the last 12 (or 6 or 3) month.
Also a min. number of games played is a good filter, but I do not have much programming-time at the moment. Family keeps me busy (in a good way).

Thank you, Oli.

But unless one knows exactly this address (which I didn't know) :
http://www.vdiplomacy.com/halloffame.php

there doesn't seem to be a way to access that page.
------------------------------------------------------------------------------------------------------------------------
About a filter about a minimum number of games, please be very cautious !

Indeed what is "1" game ? Some players can play 10 live or 10 1vs1 games in a few days while it could take maybe a year for some players to finish 10 games !

For example, I have just finished yesterday my 7th game after almost 8 months of DAILY (without one single exception !) playing on this site. I have only played the variants with many players (except one 8-player game) and always with full press. I wouldn't find it fair at all that, after all that time spent playing diplomacy here (and THOUSANDS of game messages), I am told that I haven't played ENOUGH to deserve a place in the HoF !!

I will remove the HoF link at the top soon and put the link in the help-page beneath the DPoints link. It's just at the top, so I do not forget about the work left to do.. :-))

Retillion,

I would imagine that submitting a phase in the last 3 or 6 months would be an adequate filter. I highly doubt that we would consciously accept a filter that eliminates players simply because they've been playing an 18 month WWIV game that hasn't finished yet.