Blitz Queue Trial (Closed)

[Ether]

So, uh...

I've been talking about running my Chosen (which is a 9-player game) with a 72-hour Day 1 and 48 hours thereafter, since only one mod seems to have gone below 72 so far. But would you guys actually want to deal with that?

[pisskop]

yes. I havent because Im terrified of the turnout, but I dont see why a 2:1 phase ratio is obscene.

[Rob14]

The biggest problem with running <72 hour games is the activity requirements. If you set a 24 prod timer in your setup, that only guarantees one post each day phase (one after the first 24 hours, then let time expire). I imagine you'd have to go higher on activity requirements (18 hours? 2 posts guaranteed; 12 hours? 3 posts guaranteed). How many players will be able to meet 12 or 18 hour prod timers with regularity?

At that point, it's better to just run a marathon. Which btw is an interesting possible use of this queue.

@Admin/list mod team: Is it allowable to run a game with extremely short phases (<1 hour) in this queue?

[Xtoxm]

We can lower the prod threshold without making it an automatic prod. ie, allow prods by in thread player request after 12 hours.

[Ythan]

What if there's nobody around to request a prod?

[Xtoxm]

Then it can't be that important for the absent player to be prodded?

[fferyllt]

The biggest problem with running <72 hour games is the activity requirements. If you set a 24 prod timer in your setup, that only guarantees one post each day phase (one after the first 24 hours, then let time expire). I imagine you'd have to go higher on activity requirements (18 hours? 2 posts guaranteed; 12 hours? 3 posts guaranteed). How many players will be able to meet 12 or 18 hour prod timers with regularity?

At that point, it's better to just run a marathon. Which btw is an interesting possible use of this queue.

@Admin/list mod team: Is it allowable to run a game with extremely short phases (<1 hour) in this queue?

I went with an 18 hour prod timer in my (72 hour day) game, and gave the prodded player 6 hours to post. Feedback in that game indicated 24 hours would be preferable to the prodded players, but I think 24 hours plus the waiting period before replacing would drag too much on the game pace. With a 48 hour game day, 24 hour prod timer would be worse on the pace.

[Rob14]

I'm a fan of the idea of 18 hours prod timers, personally, but player response to that idea has been poor when I've brought it up. To me, it makes perfect sense. Post when you wake up and before you go to sleep. Apparently, a good amount of players can't regularly fit that into their schedule (or don't wish to) based on the response I got to the idea.

@Xtoxm: I generally consider it undesirable for players to have to request prods. Moderation should take place in the background and players should be able to focus on playing, not checking the activity overview for which people need prods.

[Ether]

I think old Westeros games used to have 24 hour deadlines, and just modkilled you if you missed a day. It seemed to work out okay for them.

(Although this probably gets less feasible the longer the deadlines are. 48 hour days take twice as long.)

For a nine-player game, I don't want to modkill. But I might just force-replace after the first missed 24 hours without bothering with prods, or at least start looking for a replacement and let them jump back in until I've found one the first time, and take no excuses the second time.

Replacements are also a fun issue.

[Ether]

I'm also leaning toward gating signups to only allow people with 4 ongoing games or fewer through, but ditching that requirement for replacement purposes to make it easier to find people quickly.

Not sure if that'll push signups down too much--hopefully for a small game, it should be workable, but I'll rethink it if things drag on too long. If I completely sabotage the trial and destroy any future hopes for a blitz queue, uh, I'm apologizing in advance here.

[fferyllt]

I'm a fan of the idea of 18 hours prod timers, personally, but player response to that idea has been poor when I've brought it up. To me, it makes perfect sense. Post when you wake up and before you go to sleep.

Apparently, a good amount of players can't regularly fit that into their schedule (or don't wish to) based on the response I got to the idea.

@Xtoxm: I generally consider it undesirable for players to have to request prods. Moderation should take place in the background and players should be able to focus on playing, not checking the activity overview for which people need prods.

Then the commitment of a blitz game may be a bit much, I guess.

[Ythan]

Different strokes etc

[fferyllt]

Different strokes etc

Obviously.

There's a whole site full of games with longer deadlines for people who can't or don't like to post more frequently than at least once every 18 hours.

Or, maybe 96 hour days makes more sense for blitz queue at MS. 24 hour prod timer and 6 hours to replace wouldn't mess that schedule up too badly.

[Klazam]

Tangential discussion regarding rob's idea for this queue supporting marathon games

I vote NO

We have marathons very infrequently for a reason.

[fferyllt]

That said, I needed 2 replacements out of 14 players in Blitz 2, and only one of them was an activity-based replacement. The level of commitment and posting was mostly quite good. I've seen worse on sites that always run blitz-length-ish games.

[Ythan]

Tangential discussion regarding rob's idea for this queue supporting marathon games

I vote NO

We have marathons very infrequently for a reason.

Different strokes etc

[zoraster]

@Admin/list mod team: Is it allowable to run a game with extremely short phases (<1 hour) in this queue?

So far as I can tell we've not made any rules about minimum deadlines. If you want to run something with 1 hour deadlines, that's okay for now, but (1) you still need to have a reviewer and (2) you still need to run your game through the queue.

If we get a lot of these games, we'll probably have to institute a rule on it though.

[Rob14]

I have no interest in running such a game, just curious.

[tn5421]

I'm thinking that RealTime games might become a hit if there's enough activity. (Day/Night tied to the realtime day/night cycles)

[saulres]

I'm thinking that RealTime games might become a hit if there's enough activity. (Day/Night tied to the realtime day/night cycles)

Different people in different timezones would really screw with that though.

[zoraster]

[tn5421]

Just be upfront with what timezone you're using and it shouldn't be too much of a problem.

[zoraster]

The trial is now extended until January 31st.

Does this mean the trial has been successful?
On some metrics, yes. Last playerslot census, it has roughly the same number of playerslots as Micros, Mini Normals, Mini Themes and Large themes (~50).

But on many it has not shown itself to be particularly great.



Right now, our largest queue is Newbies, which is roughly at the same level as it was in Oct. 2014. All other queues have seen reductions, none more than Large Themes (losing 97 slots since August).

Right now, it seems like we are lacking a supply of games, particularly large theme games.

---
So why keep the trial running? I'm not sure what the cause of this across the board reduction is. Is it due to a very temporary decrease in supply and interest in non-Newbie queues? Or is it something that speaks to the corrosive nature of these games that reduces overall participation in mafia?

I'm not sure, but I think it's worth another 45 days to test it out.

We're still looking at the following things:

(1) Are the games of sufficient high quality that MS is offering something that other sites are not providing it?, (2) How many games were played?, (3) How often was there no mod supply?, (4) Has it affected other Queues?, (5) Was the interest in playing consistent, particularly toward the end of the trial or did it taper off?, (6) Were there a variety of games provided?, and (7) How often did people end up having to replace out of games?

At the end of the trial everyone will get a chance to weigh in. Please try the queue out! But also if you're a mod, please don't hesitate to get other types of games running.

[Ythan]

Could you clarify the problem? You haven't explained it clearly.

[zoraster]

The problem is that despite being as successful as any other queue in isolation, there is a possibility that is driving down our numbers overall.

Playerslots are defined as any slot in a game that is currently alive of the census. So a unique player could account for many different playerslots.