"Of course cirumstances differ case by case"

This in itself raises the workload of the mod team infinitely. And I doubt they have the time to commit. Its an unfortunate facet of the game that we'll just have to deal with. I don't think there's anything that can be done.

I'm not sure who it was but in a discussion about our current point system I heard a great suggestion.

First of all, fuck the point system. The idea of it was to limit how many games new people can join in order to cut down in the CDing. That obviously doesn't work.
The suggestion was to simply have a cutoff of how many games someone can play at once until theyve finished x number of games. After that, if someone CDs, you take away their right to play in that many games until they again clear their name by finishing a few games.

Joker: yes, that's why, I'm suggesting to make such action automatic only in those cases when it's surely not an NMR/CD because of losing the game, or the very end is clearly in sight (like last turn of a WTA game)

I think any NMR in the first 3 years cannot be justified with anything, also any NMRs above 3SCs when there's another Builds round coming. (Meaning if there's another builds after, then the end wasn't in sight)

My point is, if we can reduce the number of NMRs substantially then we have reached a goal. The objective is not to have NMR free games, the objective is to reduce their ratio.

Why are NMRs in bad positions more justified than NMRs in good positions?

Yonni: the point system is good for not having beginners in high pot games. Usually NMR do not occur above 1000 pot size. All that said, sadly they do occur sometimes.

Rules without punisments and enforcment are no rules at all. This applies in every aspect of life. I think, when you joined a game, you have undertaken the responsibility to play according to that game's properties (phase length, etc)
If this responsibility is not enforced, then it's not a responsibility. On the otherside it wastes a lot of time from 6 other people's life.

Personally, seeing NMRs and continuing games for a couple of phases until it ends somehow make me angry only, and definiately takes away the joy of game. The best way to avoid those players in the future to take away some points from them.

Or to ban them for a few days/weeks to join ANY games. I'm open to any solution, but I think the behaviour must be address some way.

CSteinhardt:
They aren't, but there are some cases when you just give up, because it's clear someone will win in 1-2 turns, and it makes useless to enter moves at all.

But, because we do not want to make extra work for Mods, we can punish only those NMRs which are clearly not because of the reason above. Also, when you have 1 unit left, and all you can do in each turn is to hold, then why should a NMR punished in that case?

Anyway, my point is, we can punish automatically only those ones, where there's no question about that, it has ruined the game. All the other cases should be investigated, for which noone has the capacity.

I like the system on vDip where you get a reliability rating that shrinks every time you NMR or CD, and you can set games to not let anyone below a certain reliability join. I'm not too fond of the rest of vDip though, so I'm not sure how it works in practice. I can see people simply creating new accounts to get around it.

I guess you're talking more about live games which is a whole different beast. I rarely play high pot games but, for the most part, can avoid CDs by playing passworded games.

Vaftrudner hit on an important part of the process of handling any account-tied issues: that people can simply create a replacement account and abandon their old one(s) if they feel the need. Whatever solution is implemented would necessarily have to restrain new accounts' ability to join games in order to prevent chronic quitters from getting new accounts and continuing the process.

vDiplomacy has an excellent feature which I think is extremely well-implemented and thought-out that would probably help this site out immensely in handling the CD issues - the aforementioned Reliability Rating.

The Reliability Rating tracks players for how often they NMR or CD and allows game creators to filter low-rated players out by preventing them from joining. The best part (imo) is that players who have messed up previously and want to repair their RR receive bonus points to the RR if they pick up a position in CD. That not only sharply disincentivizes quitting, but incentivizes people to pick up games which have been marred by quitting. The RR also restricts the number of games people can join at once, so that will also make it more difficult for chronic quitters to ruin a bunch of games.

Another big advantage is that we can completely rethink our understanding of the points system. As it is, points exist to serve two functions: filter players by quality of play (if the entry point value is set high enough), and limit the number of games new players can join. We know that the first function is working fine because of how common it is to see all those high-stakes games getting organized. We know the second one isn't, because we're having this discussion. Implementing the RR would render the second function of points irrelevant, allowing us to refocus points simply as a means of setting up higher-quality-of-play games (where they've succeeded) instead of expecting them to prevent chronic quitters from ruining games (where they've failed).

The drawback, of course, is that new players can't join as many games as they can right now, even if they turn out to be reliable players. I don't think this will be a huge issue, though. In practice, what will probably end up happening is that the established players who want quality games will set the RR higher than the default entry RR for new players. All those new players would simply make games at their RR level. Those games might get hurt by quitting new players joining up, but the players who stick through them will see their RR increase and thus soon be able to play higher-RR games themselves.

And, finally, the RR is already coded and ready for use, so implementation would not be a particularly tedious task.

So to conclude, I think that implementing vDiplomacy's RR system would go a long way toward solving NMR/CD issues while also refocusing points toward a task for which they're far more efficient.

There are three more "features" that deal with the NMR/CD issues.

1. You can set your games not to process the turn if there is a NMR. Instead the country is send in CD and the turn is extended for a new player to join. (During gamecreation you can set hown many turns this should work, and how many phases it should delay the processing)
2. If you can't access the site for some days you can vote to extend the phase for 4 days.
3. If you have a longer break you can send your game to another player to sit this game for you. You do not need to give away your password, and you can give different games to different players.

@rokakoma: I feel the opposite; a power that CDs in a strong position is paying the price of losing a strong position, so it's the ones that are already losing and feel like leaving is costless that need to be more actively dissuaded.

CSteinhardt: maybe you are right. For me, it's irrelevant, if it was up to me I would punish any kind of NMR in the very same way.

Reliability Rating seems a very good idea, especially when it can be set at game start, how reliable players you want to see. Still, it doesn't punish the current NMR action, only prevents future games having NMRs. Which is good on the long run, but as a player in an ongoing game, I'd like to see an immidiate punish, such as taking away a bet size number of points.

It can be implemented easily, and through bet size it prevents the player to risk the same amount of money again. Of course new registrations aren't solved by this, that's why RR seems a better solution.


Question: who is able to make the programming for it? Also, as I know, NMRs and CDs are stored in the database, so such rating could be generated for the past as well, also resign ratio could be used as well at game creation.

Also experince (number of finished games) could be a filter as well at game creation. Setting it to 25-50 finished games would be quite a good filter I assume.

@rokahoma: CD's carry a really heavy penalty in RR. If someone leaves a losing game they can forget about joining A-rated games until they make up for it by joining and playing a CD.

@Gobble: sounds great, quiestion is, whom should I contact to implement it here?

Punish COs, fine. But NMRs can be a valid tactical move. One may wish to order all holds. It is legal and as long as they show up to view the page they won't CD so they are within their game rights to do so.

I am with CS on this. The COs that need to be punished are the quitters who give up just because they are losing. They ruin through game. Strong position COs usually have a reason behind them and are punished by the loss of the position.

As far as NMRs go... Only punish a true NMR as in they didn't even look at the turn that phase. You cannot punish someone for issuing valid orders and all holds is considered valid orders.

@Draugnar: NMR is never a valid tactical move. Submitting all holds? *That* is valid, albeit probably stupid. Not submitting any orders? Not a valid tactical move. If you think all holds is your best set of orders, then set all holds and hit ready.

@Draugnar: NMR is not equal as holding everything. NMR is when the player didn't even bother to press save/ready at least once in the phase. NMRs should be punished, all holds is totally fine, especially when you have 1-2 units left, but that's not NMR.

@CS - None is ever obligated to hit ready so leaving the default of all holds is perfectly valid. If we want to differentiate between no orders and all holds, we should change the code so your units come up with something like "no orders" as the default drop down value. Then all holds would require submitting the hold orders. Or we should allow save to go out and record the holds. Right now it doesn't and obligation a player to ready or be penalized is complete and utter bullshit.

But honestly as long as they didn't record an actual missed turn (the true definition of an NMR on this site) does it really matter?

@rokakoma - you can't press save and have it record anything if nothing on the page has changed. So pressing save doesn't do shit with all holds.

By your definition and the fact that you can't just come onto the page and president save, all holds is the same as NMR.

@Draugnar, no orders and saving holds are differentiated already, otherwise the site wouldn't know whether to show "!!" or check mark. As order both are treated as hold, of course, but the database knows whether they were saved or not.

You can't save all holds without changing *from* all holds and saving first. I've seen the code and I write code for a living so I know of what I speak.

@Draug: okay? And? let's change the code! The whole topic is about, to change the code and put some punishment system into working. This is the 3rd time I'm asking, whom should I contact with the issue, to change the code?

If needed I'll write myself, I'm living from php/mysql coding, I doubt it would take much time.

The "intentional" NMR tactic isn't counted by the RR, iirc. I know, for instance, that one of the strongest Italian openings in live gunboats is an intentional NMR in S01 followed by a last-second dive into Trieste in A01 (the idea being to fake Austria into thinking you're not there and thus to vacate Trieste). If you view the page but don't submit orders, I don't think it counts against you for RR. (Admittedly, I haven't had a chance to test for that, though, because usually when people NMR they don't check the page, period; intentional NMRs are rare on vdip.) So the tactic would be preserved under the RR system.

@PE, diplomacy is a boardgame, originally played facetoface, because of that, you can't do intentional NMR, because the others are there with you, and they see you.

If we assume webdiplomacy is the implementiation of a face2face board game, any kind of NMR cannot be part of your tactic.

Has this forum thread thoroughly gone into Reliability Index, as used on the Vdip site?

It has.

@roka - But do we? Our options are to stick with a purist implementation of f2f or to accept some mixture of that implementation with online features. There's not really precedent for the former, since we have a points system, transcripts of all press we've sent and received, and the ability to, say, carry on six conversations and turn in orders for five-minute deadlines. I'd say we're already doing the latter, in which case I'm not inclined to rule out a new tactic which doesn't break any original f2f rules or any of the house rules for the website, and the intentional NMR passes that test.

@Roka - So change it. Go to the development forum and discuss the change then do it. There is no one programmer or team who does the work. Anyone who wishes can make the changes. Kestas will then test and decide if he wishes to implement it. But anyone who knows PHP and mySQL can do the coding.