Feedback on final AMULET CONTEST PAGE

joshwithguitar · 2015-05-03 00:45:58

.. wrote:

joshwithguitar wrote:
Theoretically, with 12 amulets the choosing process could go on for up to 121 days.
No, supposing that one iteration happens per day (Jason sends out emails and everyone responds), the maximum is 11 days, because at least one amulet is awarded per iteration, and I assume there's no point asking about the other prizes (do you want $50 or $10?). That worst case happens if player 1 came first for all 12 amulets, player 2 came second for all 11 that player 1 didn't want, etc, all the way to player 12.

Sure, each day at least one amulet has to be awarded, but who that amulet ends up with is not set in stone. There is nothing stopping the person awarded the amulet from being given the choice of a different amulet later on, allowing them to put the other amulet back into the system. Overall, each person awarded an amulet on the first day theoretically can reject all but one of the amulets and each day could see only one amulet given up.
So the process can last for:
((Number of amulets - 1) * number of amulet winners on the first day) days
as you need at least one amulet passed on to the next day, the maximum number won on the first day is: (Number of amulets - 1)

And so the theoretical maximum number of days the process could last for is:
(Number of amulets - 1)^2

So for 12 amulets this is 11^2 = 121 days.

Here is an example with 4 amulets - labelled 1234 and four players labelled ABCD:

Ranking order of each amulet:
1:ABCD 2:ABCD 3:BCAD 4:CABD

Day 1 - A12: A chooses 1
Day 2 - B23: B chooses 2
Day 3 - C34: C chooses 3
Day 4 - A14: A chooses 4
Day 5 - B12: B chooses 1
Day 6 - C23: C chooses 2
Day 7 - A34: A chooses 3
Day 8 - B14: B chooses 4
Day 9 - C12: C chooses 1

(The second column shows the player with multiple amulets and which ones they hold)

So as you can see, with 4 amulets the choosing can last 9 days (4 - 1)^2.

joshwithguitar · 2015-05-03 00:56:58

In this comp assuming that players will always choose Gold > Silver > Bronze the maximum number of days should be:
(3-1)^2 + (4 - 1)^2 + (5-1) ^2
= 29 days

Cobblestone · 2015-05-03 09:18:28

Is a player who win multiple days choosing strictly the amulet, or the amulet and the prize money?
I really want TwoClaws... so if I win that day and gold on another day (highly unlikely, but pretend), could I choose the silver amulet and still get the higher dollar value? Or are they a pair?
If they are a pair, why is choosing an option? I want TwoClaws, but I don't want it enough to sacrifice $100. You could just award the higher value and not involve the players.

joshwithguitar · 2015-05-03 12:11:51

Whoever wins two claws will most likely be happy to swap it for a gold amulet Cobble. So you'd probably just both have to let Jason know of the arrangement before he sends the amulets off.

.. · 2015-05-03 18:17:32

Oh, I wasn't complaining that there are only 12 amulets. I'm just saying that awarding 4 prizes per day and then reassigning prizes at the end means you can't tell what you'll win without consulting 12 scoreboards and making assumptions, in practice it'll be complex and confusing. It would be nice if the scoreboard for each amulet could mark any prize already won previously by a player, although this would have to be a guess.

Compare to the alternative of giving out one amulet per day, plus 36 cash prizes to the top 36 players (by total score, or possibly by maximum score on any day) who didn't win an amulet. In that case there's one scoreboard for the 36 prizes, but you still don't know where you'll rank: you have to wait 12 days to find out, during which you can both go down or up the scoreboard (unlike daily prizes, where your prize can only improve). The only thing wrong with that is if it means you'd have to play as many games as possible over the whole 12 days to compete. However there might not even be 48 players considerably better than the average cabal member, so you might actually not need to play too many games to win a prize.

There are clearly advantages to resetting every day, and I can't come up with a suggestion that's a clear improvement (also it's damn close to the contest to be making large changes)

I was surprised to find out how little half a troy ounce of silver is worth! (Half what it was 3 years ago, but more than double what it was 10 years ago.

BTW, I hope you can earn interest on player deposits. There's going to be a lot of money sitting in accounts forever, either because they never withdraw it when they stop playing or it's less than $3. Interest could possibly make the game profitable longterm without a rake (inflation is irrelevant).

EDIT: didn't see page 2.

joshwithguitar wrote:

Here is an example with 4 amulets - labelled 1234 and four players labelled ABCD

Alright, I worked through your example and admit defeat.

Last edited by .. (2015-05-03 18:33:49)

LiteS · 2015-05-04 15:23:54

jasonrohrer wrote:

What I really don't want is 4 "usual suspects" swamping the whole thing and making everyone else feel like they have no chance.

With the current leaderboards system, if there are 4 exceptional players, they will show up at the top of the leaderboards each day to fight for the amulet regardless of their eligibility to win the 2nd-4th prizes. Marking people with an asterix as having already won a prize would be imperfect, but better than nothing. As an outsider, I'd feel more encouraged to play/win if by day six half the leaderboard was visibly intelligible for a cash prize. Otherwise you're implementing an anti-usual suspects system that is invisible to the casual observer.

Cobblestone · 2015-05-04 16:51:10

That wouldn't work, because they wouldn't actually be ineligible for the prize. If someone wins day 1, they can still win day 6 and then day 1 goes to second place. If you're playing day 6 and that player has an asterix on them for winning day 1, the asterix is misleading. Also, allowing winners of two to choose their prize makes an automated system like you're suggesting unreliable.

Edit: Whoops! I missed the part where you're only talking about 2nd-4th place.

Last edited by Cobblestone (2015-05-04 17:26:28)

AnoHito · 2015-05-04 17:40:03

So the contest is starting this Wednesday, correct? I was wondering if there was any sort of plan in place to help publicize the game before the contest starts. If you search for Cordial Minuet in Google right now, most of the top results were things published last year. I'm just worried that the contest will start before word has enough time to spread that the game is officially launching.

jere · 2015-05-04 18:20:45

AnoHito, I was thinking the same thing. I made a mention on the last page. Seems kind of odd, but this isn't Jason's first time at the rodeo. I'm sure he has something planned...

The contest lasts an unusually long time. I just realized that the quality distribution of the amulets is meant to hit a peak this weekend, cool off, and then ramp up to a final climax (2 golds). Makes sense. Most people will probably not want to jump into a new game until the weekend. So if there is some announcement later this week, there will be tons of opportunities left to win and simultaneously the feeling that you can start right now (as opposed to: this game is shipping in 3 quarters).

Dan_Dan84 · 2015-05-05 03:22:43

jere wrote:

AnoHito, I was thinking the same thing. I made a mention on the last page. Seems kind of odd, but this isn't Jason's first time at the rodeo. I'm sure he has something planned...
The contest lasts an unusually long time. I just realized that the quality distribution of the amulets is meant to hit a peak this weekend, cool off, and then ramp up to a final climax (2 golds). Makes sense. Most people will probably not want to jump into a new game until the weekend. So if there is some announcement later this week, there will be tons of opportunities left to win and simultaneously the feeling that you can start right now (as opposed to: this game is shipping in 3 quarters).

I have the same worries, but I was thinking the same thing as jere (right down to the "rodeo" idiom!).

I know originally Jason was thinking of having two amulets up for grabs per day, but now it's going to be one per day, stretching out to 12 days. That is a long time. The Steal Real Money TCD contest ran for a week, if I'm not mistaken. That was a long time both to defend your house and keep stealing and adding to your vault. Here, you can play for as many or as few days as you want, but your score only counts for that day.

I'm sure there's a good reason he's gone for 12, not 6.

In any case, I've tried to start promoting the game itself through Facebook. As I expected, "that weird gambling game Dan's been playing" is a tough sell (sandwiched between photos of my friends' lunches, which get far more attention: the 'ancient dreams of the Internet being squandered' indeed).

jasonrohrer · 2015-05-05 16:29:53

Yeah, there is a plan today.

The game launches at 10am PDT today, which is 14 hours before the contest starts.

Kotaku, RPS, Dtoid, and Eurogamer are writing articles about it today, hopefully.

I will also tweet, which will hopefully get retweeted.

I will also email my 20K subscriber email list.

I will also email the 10K non-Steam people who bought TCD.

So, why no buzz yet? This is my 18th rodeo, as you said, and over the years, I learned that you only get so much buzz. Any buzz that happens before the game is actually live is wasted, because the traffic flowing to the site doesn't turn into players. It's not wasted if it's months ahead of time, because that's building awareness. But if it's a few days ahead of time, it's wasted, because those outlets are only going to cover your game once this week. If it's only once per outlet, it better be AFTER the thing goes live.

As far as the prize algorithm, it may end up being confusing on the leaderboards for the cash prizes if the same 4 people have '*' at the top of every board. In that case, I may write a bit of new server code to apply the automatable portions of the algorithm for the cash prizes and put the '*' marks in more accurate places. We'll see if that's necessary.

By "automatable," I mean that I don't even have to ask you if you prefer a $40 over a $20 prize.

Regarding the amulets, they ARE bound to their cash prize. But maybe I'll rethink that if someone really wants copper over the gold that they won. We can cross that bridge as necessary at the end.

computermouth · 2015-05-05 17:00:17

I may have already sent an email to kotaku's tipline

Cobblestone · 2015-05-05 17:09:32

I've also been talking the game up to a lot of people, and I think I'll probably be posting on Facebook and other social sites today. I have several friends that are interested in the game but just haven't taken the plundge yet. Hopefully the contest will entice them! Although one friend did say he didn't want to be learning a new game during a contest, which I thought was interesting. I wanted to try to set up a party themed around the game with a few stations and a local server, but I just haven't had the time to get that going. I even got these awesome coins they can bet with if they really want to try it out risk free! I really wanted to do that before launch, but it'll happen eventually and hopefully they'll want to join afterward.

If I have time tonight, I'll make a video or something about the game to put up somewhere. Then again, today is everyone get drunk day, so I'm not sure that'll happen.

Edit- DOOR is OPEN:
http://www.rockpapershotgun.com/2015/05 … -launches/
http://www.eurogamer.net/articles/2015- … is-out-now

Last edited by Cobblestone (2015-05-05 21:03:10)

.. · 2015-05-05 21:41:17

I was surprised to see the number of active players suddenly exploding (has been amazingly slow) and yet not more than a couple new registrations. Now I see the email set to existing players.

So the asterixes would mark players who have already won a better prize and won't be winning anything according to the current standings? (It would be nice if the scoreboards for previous days also show the latest information.) But if there's even one person who wins two coppers uncertainty could ripple on, so "definitely has already won a better prize" could become false in a lot of cases. I don't think there is a good way to summarise the state of the awards tables for this contest.

Here are some FAQ questions:
How many cabal members are there?
Are you told after playing a cabal member? Even if your score remains the same?

jasonrohrer · 2015-05-05 22:54:50

Well, currently, the top 4 players on each leaderboard get a *, and it's not more sophisticated than that. Each day operates in isolation.

The point is, if we have a "usual suspect" problem across multiple days, those *'s will be misleading. If it's a huge problem, I may do something about it mid-contest.

.. · 2015-05-06 06:58:54

OK, waiting and seeing makes sense. I might actually put together a scraper and make a combined scoreboard myself, if there's any point doing so.

Just logged in to the server, a few minutes before the start of the contest. Server has been packed all day, but I'm the only one on; madness. Was the server restarted?

jasonrohrer · 2015-05-06 07:16:24

No, there was no restart. But it is very late almost everywhere in the world except the far east!

.. · 2015-05-06 08:16:44

Also nobody wants to play CM when they could instead wait 10 min and then score points on an amulet.

We had 4 cabal members turn up right after midnight (the majority of active players), but I think they were putting in token appearances.

Last edited by .. (2015-05-06 08:19:13)

computermouth · 2015-05-06 08:35:44

I've played ~4 games in the past hour. No amulets, but games is games. Come at me, previous-directory-up.

.. · 2015-05-06 09:24:58

How do you know you haven't?

I find it amusing that everyone-but-one is on zero points because the cabal are taking all the winnings.

We derailed this thread pretty heavily, should start a new one for contest discussion.

jasonrohrer · 2015-05-06 16:15:13

Yeah, I saw all those zeros too.

I actually gave up some points last night, but only because my spouse called me to help with something and I timed out.

Dan_Dan84 · 2015-05-07 00:36:18

jasonrohrer wrote:

Yeah, I saw all those zeros too.
I actually gave up some points last night, but only because my spouse called me to help with something and I timed out.

Ah, I was wondering what happened to my opponent in that game...

LiteS · 2015-05-07 15:11:08

joshwithguitar wrote:

.. wrote:
joshwithguitar wrote:
Theoretically, with 12 amulets the choosing process could go on for up to 121 days.
No, supposing that one iteration happens per day (Jason sends out emails and everyone responds), the maximum is 11 days, because at least one amulet is awarded per iteration, and I assume there's no point asking about the other prizes (do you want $50 or $10?). That worst case happens if player 1 came first for all 12 amulets, player 2 came second for all 11 that player 1 didn't want, etc, all the way to player 12.
Sure, each day at least one amulet has to be awarded, but who that amulet ends up with is not set in stone. There is nothing stopping the person awarded the amulet from being given the choice of a different amulet later on, allowing them to put the other amulet back into the system. Overall, each person awarded an amulet on the first day theoretically can reject all but one of the amulets and each day could see only one amulet given up.
So the process can last for:
((Number of amulets - 1) * number of amulet winners on the first day) days
as you need at least one amulet passed on to the next day, the maximum number won on the first day is: (Number of amulets - 1)
And so the theoretical maximum number of days the process could last for is:
(Number of amulets - 1)^2
So for 12 amulets this is 11^2 = 121 days.
Here is an example with 4 amulets - labelled 1234 and four players labelled ABCD:
Ranking order of each amulet:
1:ABCD 2:ABCD 3:BCAD 4:CABD
Day 1 - A12: A chooses 1
Day 2 - B23: B chooses 2
Day 3 - C34: C chooses 3
Day 4 - A14: A chooses 4
Day 5 - B12: B chooses 1
Day 6 - C23: C chooses 2
Day 7 - A34: A chooses 3
Day 8 - B14: B chooses 4
Day 9 - C12: C chooses 1
(The second column shows the player with multiple amulets and which ones they hold)
So as you can see, with 4 amulets the choosing can last 9 days (4 - 1)^2.

This 121 day process lasts only 1 day if the first email sent out by Jason asks all potential winners their ranking order preferences.

Edit Note:
Only the top 12 players on any given amulet will have a potential to win, so if Player A is in the top 12 for only 3 amulets, that player only has to express preference for those three.

There's some simple optimization that can be done here to minimize emails sent as well, e.g., if player A is the second place player for amulet 2, and player A is in the top 12 for only one amulet, no player in the rankings for amulet 2 below player A will have to be notified of their preference, since A will take the amulet by default.

Edit Addendum Note:
Also, as mentioned before, we should consider the case in which a player prefers 2nd place gold to a Copper Amulet, since the cash prize in that case is higher.

Last edited by LiteS (2015-05-07 15:33:43)

Dan_Dan84 · 2015-05-09 07:05:51

Cobblestone wrote:

computermouth wrote:
Day 4 looks kind of like a SICK CAT, that WILL BE the one I GO FOR.
That's the ONE I WANT. For exactly the SAME REASON.

The day of TWO CLAWS has ARRIVED.

GRAB IT.

Cobblestone · 2015-05-09 07:32:24

I WANT IT I WANT IT I WANT IT

Last edited by Cobblestone (2015-05-10 07:12:09)

CORDIAL MINUET ENSEMBLE

#26 2015-05-03 00:45:58

Re: Feedback on final AMULET CONTEST PAGE

#27 2015-05-03 00:56:58

Re: Feedback on final AMULET CONTEST PAGE

#28 2015-05-03 09:18:28

Re: Feedback on final AMULET CONTEST PAGE

#29 2015-05-03 12:11:51

Re: Feedback on final AMULET CONTEST PAGE

#30 2015-05-03 18:17:32

Re: Feedback on final AMULET CONTEST PAGE

#31 2015-05-04 15:23:54

Re: Feedback on final AMULET CONTEST PAGE

#32 2015-05-04 16:51:10

Re: Feedback on final AMULET CONTEST PAGE

#33 2015-05-04 17:40:03

Re: Feedback on final AMULET CONTEST PAGE

#34 2015-05-04 18:20:45

Re: Feedback on final AMULET CONTEST PAGE

#35 2015-05-05 03:22:43

Re: Feedback on final AMULET CONTEST PAGE

#36 2015-05-05 16:29:53

Re: Feedback on final AMULET CONTEST PAGE

#37 2015-05-05 17:00:17

Re: Feedback on final AMULET CONTEST PAGE

#38 2015-05-05 17:09:32

Re: Feedback on final AMULET CONTEST PAGE

#39 2015-05-05 21:41:17

Re: Feedback on final AMULET CONTEST PAGE

#40 2015-05-05 22:54:50

Re: Feedback on final AMULET CONTEST PAGE

#41 2015-05-06 06:58:54

Re: Feedback on final AMULET CONTEST PAGE

#42 2015-05-06 07:16:24

Re: Feedback on final AMULET CONTEST PAGE

#43 2015-05-06 08:16:44

Re: Feedback on final AMULET CONTEST PAGE

#44 2015-05-06 08:35:44

Re: Feedback on final AMULET CONTEST PAGE

#45 2015-05-06 09:24:58

Re: Feedback on final AMULET CONTEST PAGE

#46 2015-05-06 16:15:13

Re: Feedback on final AMULET CONTEST PAGE

#47 2015-05-07 00:36:18

Re: Feedback on final AMULET CONTEST PAGE

#48 2015-05-07 15:11:08

Re: Feedback on final AMULET CONTEST PAGE

#49 2015-05-09 07:05:51

Re: Feedback on final AMULET CONTEST PAGE

#50 2015-05-09 07:32:24

Re: Feedback on final AMULET CONTEST PAGE

Board footer