Reposting what I posted in global:

"Captainmeme (Moderator): Hi everyone,

I cannot figure out how to get this to uncrash. I tried everything I did last time, plus a bunch more, and none of it helped.

I have looked through all the orders, though, and I can confirm that Turkey would have won this phase, losing no centers and gaining the following supply centers:

Murmansk
Munich
Morocco
Lubango
Zimbabwe
Okinawa
Taiwan
Hong Kong
Guam
Nauru

This would put him on 128 SCs, giving him the solo victory.

Since this is a WTA game, that result is the same as if everyone else voted to concede to Turkey. Therefore, I am going to put up a concede vote for every other player, giving Turkey the victory here."


Unfortunately, it seems that whatever the bug was, it also stopped the game from ending properly when the concede votes were in - so it sent out messages saying everyone was defeated, but nobody actually was. I'm still working on fixing it.

Uhhh... It seems to be constantly trying to resolve the situation and marking everyone as defeated, then failing and trying again. This will mean your homepage will be spammed up with messages saying "<country> was defeated" over and over again.

Nothing I can do about it until the game properly crashes so I can access it again... That might take a little while.

Thank you Captain

I can't look at the global chat, because the game is crushed, so thank you for your respond Captain! Gg, well played matthew!

Oli fixed the issue :) Congratulations Turkey!

Thank you very much Captain, and Oli. By the way, is 128 a Diplomacy record? I am curious

Even if it's not, you now have the most vPoints of anyone in the game...

On this website, even...

I don't understand why I got so many? It was 529 right?

Nevermind

Thanks to this solo in WWIV, you are now the top player on this site. Congratulations on your richly deserved victory.

Damn. 1283 awarded for a single game. No offense to 00matthew, as entering that many moves every phase is quite a slog and obviously it was a great victory, but that seems like a bug in the formula!

I don't think it's a bug - it's just a side effect of the way the formula works. vRanking takes into account your rating and the rating of the other players in the game, calculates your likelihood to win, and rewards you based on how unexpected the win was. If you play 5 7-player games in a row and win all of them, your rating will be going up each time you win, meaning that you'll gain much less for the next win because your higher ranking means you're considered more likely to win. If you beat 35 players at once, however, all of them are compared to your initial rating, which means that you'll gain a huge amount.

It's not ideal, but it's really difficult to have an elo-like rating system balanced out across variants with different numbers of players, different scoring systems, and also account for draws.

So, I took over United Kingdom as a replacement player,but when I lost my rating went down. Wasn't there some rule about this not happening, to make it easier to find replacement players for hopeless positions?

bigrob and Mapu make excellent observations here in the aftermath of this game. Good for Matthew collecting a solo, and demonstrating the devastation that a solo victory should bring. Yet, one game should not make or break a player when compared to dozens of games or years of play. The Elo ratings system has needed an overhaul for a long time, and this game only serves to amplify that. Furthermore bigrob is penalized for his nobility here in trying to help the community by taking on a hopeless position. We need a system that identifies mercy poistions and absolves players from points penalties (and/or rewards possibly). Tracking replacement position contributions separately seems a better solution.

Some of the Dip communities that I have played in previously have awarded badges/medals/martyr points for various outcomes of games, which track seperately from strict performance. Examples are this might be loyalty (# of games played), reliability (NMR-free), mercy/other replacements, and so on. The idea here is to create incentives and rewards for players to engage in play that benefits the community.

There's probably a long list of items that need fixing whenever the site gets the administrative assistance it needs, but it would be nice if someone would keep a list of things like this for that time. If the moderator team is too overwhelmed to take this on, I would volunteer to maintain such a list.

To clarify, I didn't take over the United Kingdom in a hopeless position, and I was optimistic until very late I would be part of a draw in this game. But I am curious what the situation is regarding ranking points and takeovers.

RUFFHAUS, I'm sure it wouldn't hurt if you maintained such a list even if mods track it as well. Redundancy + more squeaky wheels = better chance of attention and resolution. I've come to loathe the rating system and how it has impacted the choices of good players and what games they will join, among other flaws and unintended consequences. Matthew, quit now while you're ahead, you may reign as the vDip champion for all time.

I was in a game that crashed when everyone (except me) conceded. When someone fixed the game up, after a month or so I think, the game concluded and I won, but I lost 46 vPoints, even though I was up against good opposition and probably should have gone up 80 or so...

So, it is possible there is some weirdness with scoring when games are uncrashed.

But if months had passed, randomizer, then the differential between you and your opponents at well have changed do to other games ending. It's the rating at game end that determines the change.

I'm not randomizer. :-)

Anyway, I decided to look for more info. If you look at my Hall of Fame history, I scored -46 for the win. http://www.vdiplomacy.com/hof.php?userID=6557 (game Build Anywhere Classic Boating), but if you look at the score allegedly calculated for the game, I should have gone up 102. http://www.vdiplomacy.com/hof.php?gameID=24405

I don't really think the differential differences would have come into play to that degree. Most of my solos with 7+ players earned 80+ at least and even the "little" ones keep me in the positive.

Maybe I'll get this fixed, now that I've found a discrepancy. :-)

But it does look like the numbers for the original game in question match up. Was just putting it out there.

Sorry, stupid phone auto corrected.

And yeah, a solo should never go negative. That does sound like a bug.

That's... really weird. I'll see what I can do.