Mini Theme Queueing data

[Isis]

I am interested in actual and perceived demand for games, and decided to pick some data to see if I can learn anything, try to pick characteristics and see if they're a predictor of whether a game will be wanted. The normal queue has really strong demand, all the time. I'm interested in doing similar thing for Open Queue, maybe. Micros pretty much always fill, so that's less important to study (Note that a lot of these games in this data scrape dip below 13p and they very clearly enjoyed an advantage as fill rate, I don't feel like anyone will even attack that thesis. The excitement that a game might fill soon makes joining more likely). So I looked at Mini Themes.
I counted how many unretracted /ins per day a game got from the time it joined the queue until it was filled or pulled. I didn't count /ins that overfilled the game.
Mod identity/reputation is a huge factor, but it's hard to categorize or crunch that, so you should be aware of the bias. In particular I noticed that every single moderator with a fully open game that filled has (imo) very positive reputation factors and so I can kind of perceive the ability to get a fully open game to fire as a "privilege" of well-liked mods in the current ecosystem.
I'm a little too weary about just the collecting to draw conclusions but I maybe want to pull this into excel and see if there's any statements you can make. I know the sample size is small and any statement is weak.
"Essentially instant fill" games were all marked as "playercount.0", even though they may have filled in less than 24 hours or one of them might have been 36. It's just not particularly important to look at anything beyond the game essentially being an instant fill. Potentially anything above 3 should be lumped together as "essentially instant fill"
I worked by hand, I'm sure there's lots of mistakes, if you played a game and new it was queued bastard or actually open or etc let me know.

Fully Open : The game was either open or semi-open, you could mod the game yourself using only the /in to queue if you needed to.
Open Mechanic: The game was closed, but a special game mechanic was guaranteed to appear in the game
Closed: The game was closed. I didn't differentiate on how much information was given about what types of roles to expect, some gave you a general notion of what might appear, but did not have an outright open mechanic.
Upick: The game was a upick. This is a kind of closed game.

Oh also this was January 1 2019 to present day

Name	Type	Bastard	Rate
Mod is Mafia: Plat Edition	Open Mechanic	Yes	.29
Unstable Mafia	Closed	Maybe	1
Secrets Anuket Topaz	Closed	No	1.6
Rapier Upick	UPick	No	.93
Giffy Q	Closed	Maybe	1.08
Access Point	Open Mechanic	No	1.63
factory adella	Fully Open	No	.47
What's your Emergency?	Closed	No	.81
Timeshift Mafia IV	Open Mechanic	No	.68
Madness at Port Arthur	Closed, RM	No	.72
Restless Spirits	Open Mechanic	No	1
Any nondead person can post	Open Mechanic	Maybe	.59
Cult vs Jester	Closed	Yes	.59
Guns and Roses I	Fully Open	No	5.5
Vengeful Ghosts	Open Mechanic	No	1.18
Pokémon Ruby	Closed	No	2.17
Surprise Mafia	Open Mechanic	No	.65
Clue Master Detective Mafia 3	Fully Open	No	.1
Zero Escape Mafia	Closed	No	3.0
Purge	Closed	No	1.08
CYS Atmosphera of Chaos	Upick	Yes	3.25
The Search	Open Mechanic	Yes	2.6
Undertale SemiOpen	Fully Open	No	1.2
I don't remember the name of this game	Closed	No	1.18
Make Me Reget This Upick	Upick	No	1.18
It's a Metalocalypse thing	Closed	Yes	.54
NFL Mafia	Open Mechanic	No	.06
Democracy Mafia	Open Mechanic	No	.3
Extremely Bastard	Closed	Yes	.9
Star Trek Mafia	Closed	No-ish	.3
Undertale Semi-Open 1.1	Fully Open	No	11.0
Assassin Mafia	Open Mechanic	No	.87
Magia Record	Upick	No	.5
I hate Mafia	Fully Open	No	.59
Skyrim Mod Upick	Upick	No	.43
Thread upick	uPick	Maybe	.47
Grand Idea Gpick	Upick	Yes	.21
Undertale S Open 2.0	Fully Open	No	2.75
Spooky Mafia	Open Mechanic	No	.38
Intrigue	Closed	No	0.0
Your Waifu is Trash uPick	Upick	Noish	13.0
Franklin's Expedition	Closed	No	0.19
Grand Idea Upick IV	uPick	Yes	0.65
Intrigue	Closed	No	0.0
Fairly Special	Open Mechanic	No	0.72
VinePick	Upick	No	0.19
Everything is Completely Normal 2	Closed	No	0.5
Anime uPick: Shuffle	Upick	Noish	4.3
Furret's Bizarre Adventure Part 2: Tax wars	Fully Open	No	13.0
XP Mafia II	Open Mechanic	no	3.25
Ginngie picks for you	upick	Yes	13.0
City that Never Sleeps	Fully Open	No	2.2
Pandemic!	Fully Open	No	0.14
Partition	Fully Open	No	1.18
Gacha Mafia	Fully Open	No	4.5

[Isis]

I thought of something I missed
will correct this over time.

[northsidegal]

it seems like being fully open is a serious advantage? this just makes me sadder about the state of the open queue, really. i think the low sample size might be misleading here, though. not all fully open games fill super fast, but almost all of the games which have filled super fast have been fully open.

[northsidegal]

if i were to look at this data and post-hoc rationalize it, i might say that it's logical that open games fill faster, because people can see what they're getting into and thus decide to join the game easier, whereas a closed game has to do something else to convince people to join.

that being said (and keep in mind that this is just my non-empirical opinion and not actually what the data says), it seems to me like you're correct in that reputation has a huge effect. almost all of the games which had an instant fill rate were mods of very high reputation, and from memory many of the games which failed to fill were from mods with less of a body of work. like i said though, that's just my non-empirical feelings, and actually looking again i can see some entries which contradict that – mods of high reputation whose games failed to fill and lesser known mods having their games fill very fast.

[Isis]

Open games are more likely to be smaller than 13p, enabling a fast fill.
Schadd_ has a highly positive reputation and failed to fill a fully open game within this data set. As though some force that impacted the less reputed mods was even too strong for schadd. That contributes to my coming to a viewpoint that fully open minithemes are perhaps unwanted.
This could be because the experience and open game provides isn't one people expect or are looking for when they go into that queue

[Isis]

Anyway I think the instant fillers are less important/less interesting than, "what's the difference between games with a .9 and .5 fill rate" and I'd like to see if I can look and get some kind of proscriptive nudge.

I still haven't reviewed the numbers yet