Fourth test contest with MONEY prizes

jasonrohrer · 2015-04-08 18:29:25

This contest will run from 9am to Noon US PDT time (Los Angeles time) on Thursday, April 9, 2015.

Several cabal members have been chosen and contacted in secret. They will play just like they normally do. They are in the running for prizes in the contest too and can score points if they play games against other secret members of the cabal.

Players score 1 point for each net coin that they win in games against cabal members and lose 1 point for each net coin that they lose in games against cabal members. Leaving a game with 103 coins means you won 3 net coins, while leaving with 95 coins means you lost 5 net coins. Games played against non-cabal players have no effect on the contest.

At the end of the contest, the top three players will win prizes:

$100
$50
$25

The leaderboard for the contest will appear here:

http://cordialminuet.com/gameServer/ser … name=test2

After the contest ends, prizes will be paid as direct payments into the game balances of the winners.

In the case of a tie for a given place, the player with the oldest account wins the tie.

NOTE: this contest is intentionally short. It is not intended to be fair to people who work daytime jobs or live in other timezones. It is simply meant to test the new code for bugs.

joshwithguitar · 2015-04-09 13:21:27

Good luck guys, I'm off to bed.

claspa · 2015-04-09 14:04:46

Thanks, Josh. I am looking forward to it.

Cobblestone · 2015-04-09 16:56:17

Well, I was only able to squeeze in one game before work and it wasn't against one of the chosen ones... it was YOU claspa!! Hope you fair better than I did!

.. · 2015-04-09 19:09:37

I'm curious, how many cabal members were there? I played 14 games, I think, but only one against a cabal member. Often I'd refresh the leaderboard after 20 minutes and nothing would change, especially in the last hour. If any cabal members make it to the top of the scoreboard themselves then they have an incentive to stop playing; I wonder how significant that was.

I think that for a launch contest that runs for a long time the cabal system would be OK as long there are enough people in it that everyone can expect to play plenty of games that count. It's clearly not a very good system for a short contests if only a tiny fraction of games count, but obviously this was just a test contest. I guess I could also be used for some kind of 'verified' highest-profit contest which is short if a majority of contestants are verified as 'probably not a colluder or bot'.

Also, is there any meaning to the amulet symbol displayed after you play a contest game?

Last edited by .. (2015-04-09 19:14:45)

CaravanDisturber · 2015-04-09 20:15:19

I'd be interested to know which cabal member scored the most coins off people in non-cabalVcabal battles.

jasonrohrer · 2015-04-09 20:33:05

There were a total of four cabal members this time.

Of course, I cannot say whether any of them won a prize or not.

It seemed to work better than the contest where it was just me, as there were 16 names on the leaderboard as opposed to only seven last time, with roughly the same number of people trying. So, clearly more cabal games took place this time, which is good!

Yes, there is meaning to that amulet symbol... everyone sees the same symbol, so it doesn't have gameplay meaning. Just thematic meaning. It actually now has a double meaning.

jasonrohrer · 2015-04-12 18:44:28

Sorry for the delay... caught up in other stuff, like carving wax amulets...

All prizes have been paid.

Professor Chin · 2015-04-13 07:44:48

This format seems to work fine, it will be very important to have an sufficient number of cabal members.

One note: of course it is fair to remove points for lost coins, but compared with awarding only "positive" points for coins won I think it will discourage newcomers: people who make a "false start" might think they will never make it as they are already too far behind when they start to have a grasp of the game.

jasonrohrer · 2015-04-13 16:22:16

Yeah, we should probably discuss this point. Getting negative on the leaderboard is pretty harsh.

The problem is, you have to have some way to discourage reckless play....

CaravanDisturber · 2015-04-13 17:32:11

1 Point for Positive coins won. Negative half point for coins lost? That way winning is always valued more.

I would also like to see a 10ish bonus for taking all coins from someone. Same could be added for loss.

Last edited by CaravanDisturber (2015-04-13 17:33:15)

LiteS · 2015-04-13 23:59:21

Losing is an important part of the game, and I like the way contest coins can be taken away almost as easily as they are earned. Bonus coins are interesting, but probably unnecessary. Sure, you can only gain 99 coins from one game and risk losing 100, but that's how the game treats money. You put up a game for 1.00, the most you can leave with is 1.99.

It is a little troubling that quick decisive games ending with a score of 199-0 against the cabal pay out the maximum 99 coins, whereas longer, more cautious games will pay less. But that seems mitigated by the reckless and (perhaps) unsustainable nature of winning 199-0, along with how winning two games from slow play will beat winning two and losing one from more reckless play. The top two winners here won multiple games against cabal members, so relying on a few lucky breaks likely won't be a winning strategy.

In this contest the cabal was more skilled than the average contest player, as shown by the fact that more coins were won by the cabal than lost. If similarly skilled cabal members are selected for the launch contest, they'll likely shy away from the newbie penny stakes, since they will likely be able to use their skills at the higher tables to try to net some cash for themselves.

One thing we might not have considered is a new form of the dreaded collusion. The test contest wasn't long enough for this to be a problem, but imagine this scenario: I've lost to a cabal member and now have a score of -100. Regarding the contest, I now have a very strong incentive to create a new account than to continue playing. On top of that, if I didn't create a new account I'd have to win against two cabal members to break even. Perhaps allowing for negative scores is a bad idea after all.

Dan_Dan84 · 2015-04-14 04:56:20

I'm back, and glad to hear that it sounds like we've hit on a launch contest idea that will work. Congrats to the winners of this contest. Now all I have to do is actually get good for the real contest...

One piece of input I will give: while clearly it is better to have a "cabal" than just "The Creator" to create special matches, as more people will encounter these matches, the concept is not quite as exciting. The idea of "This might be a game against Jason Rohrer, and if I win, I score special points" is very powerful. Playing against a faceless cabal... not quite as interesting.

I suppose in the MARKETING, this could be framed differently. But when I got the e-mail for the third contest, I was really excited about the idea. For the fourth contest... it didn't grab me as much.

As for others' input...

jasonrohrer wrote:

Getting negative on the leaderboard is pretty harsh.

It does look kinda harsh, but hey, CM is a harsh game! You could give each player 1000 points to start with, just to avoid the ugly minus sign. But yeah, to keep play from getting weird and reckless, I agree some deterrent is necessary.

LiteS wrote:

In this contest the cabal was more skilled than the average contest player, as shown by the fact that more coins were won by the cabal than lost. If similarly skilled cabal members are selected for the launch contest, they'll likely shy away from the newbie penny stakes, since they will likely be able to use their skills at the higher tables to try to net some cash for themselves.

Hmm, that is an issue. The point of the contest is to allow anyone a chance to win-- or at least have a good chance of participating in the contest. If cabal play only happens at $1+ stakes... it might freeze out some new players.

LiteS wrote:

One thing we might not have considered is a new form of the dreaded collusion. The test contest wasn't long enough for this to be a problem, but imagine this scenario: I've lost to a cabal member and now have a score of -100. Regarding the contest, I now have a very strong incentive to create a new account than to continue playing. On top of that, if I didn't create a new account I'd have to win against two cabal members to break even. Perhaps allowing for negative scores is a bad idea after all.

I'm not sure this is a huge problem, since the player will still have to win games against a non-colluding account in order to climb the leaderboards. This isn't a "I won with one simple trick!" exploit. But yeah, the incentive for multiple account play is there. But then again, in any contest (except for one with an entry fee, which Jason wants to avoid), multiple accounts will always provide an advantage. The key is mitigating that advantage-- which this contest seems to do.

Congratulations again to all the winners!

joshwithguitar · 2015-04-14 15:13:52

I think an easy solution to stop people going into the negative while discouraging reckless play is to not allow scores to go below 0. I guess this might mean players might play recklessly until they get some points but once they have some then they should just play as normal.

Ghost Amount · 2015-04-14 15:31:38

I can tell you that when I saw that I took 99 points from a Cabal member - I almost stopped playing and started watching the leaderboard, I did not want to have the potential for losing any points. Once I saw people climbing up the ladder I started playing again but was very careful and even left one game when i was up ~40 because I figured 34 coins would be a nice add if it was a cabal game and it was!

Wreck-less playing was the complete opposite of what i was doing.

Cobblestone · 2015-04-15 01:58:55

The contest points carry the same risk/reward as the coins themselves, so I think letting them go negative isn't such a bad thing.

The only thing I can think of is maybe using the last ante of a game as some kind of modifier to reward more points if you win, or soften the blow if you lose, as you play longer.

Last edited by Cobblestone (2015-04-15 19:54:45)

.. · 2015-04-15 12:10:20

I would definitely hope that the contest structure punishes players who play recklessly, because it's annoying when the contest causes people to play differently. But for a launch contest that runs while most players are still learning the game, large numbers of players feeling that they've been knocked out of the contest isn't desirable. So I think Josh's suggestion is good, but hope for better.

I think the amulet contest definitely had a great idea: what mattered was the length of your winning streaks, which is a good proxy for your overall win rate but added excitement. (Of course if winning is instead defined by coins won/lost, it'll encourage people to leave early.) Additionally you had a separate chance to win every amulet, which means you had a very real chance to win whatever amulet you've managed to rack up a few wins on.

Unfortunately it seems impractical to have a separate scoreboard per cabal member (some wouldn't play many games; high risk of them colluding; etc), but that "you always have a chance on another amulet" seems like a good solution to the negative points problem. For example, you could hand out two amulets per day of the contest rather than all 36 at the end.

Last edited by .. (2015-04-15 12:13:58)

Dan_Dan84 · 2015-04-15 13:11:43

.. wrote:

Additionally you had a separate chance to win every amulet, which means you had a very real chance to win whatever amulet you've managed to rack up a few wins on.
Unfortunately it seems impractical to have a separate scoreboard per cabal member (some wouldn't play many games; high risk of them colluding; etc), but that "you always have a chance on another amulet" seems like a good solution to the negative points problem. For example, you could hand out two amulets per day of the contest rather than all 36 at the end.

Some good ideas here, I think. It's important that everyone feels they have a chance, so we don't have the "flurry of activity at the beginning, quickly dying out" phenomenon. Ideally, of course, people would find the game so enticing they'd continue playing anyway!

jere · 2015-04-15 13:25:16

The problem with flooring points at 0 is that a sequence of win-loss leaves you with 0 points and a sequence of loss-win leaves you with 99 points. Bleh! I do see the temptation to create another account after that first loss though.

Unfortunately it seems impractical to have a separate scoreboard per cabal member (some wouldn't play many games; high risk of them colluding; etc), but that "you always have a chance on another amulet" seems like a good solution to the negative points problem. For example, you could hand out two amulets per day of the contest rather than all 36 at the end.

I've always liked the idea of prizes being handed out more often, so you have an incentive to keep playing even if your chances at the big prize seem unlikely.

claspa · 2015-04-15 15:14:57

Cobblestone wrote:

The only thing I can think of is maybe using the last ante of a game as some kind of point modifier to reward longer games if you win, or soften the blow if you lose, as you play longer.

I do understand what Cobblestone is saying. We played the first game of the contest against each other. It took us almost an hour to get a decisive win, which was 30% of the whole contest time.
I would like to see a some sort of modifier/compensation, but I cannot think of something good.

CaravanDisturber wrote:

I'd be interested to know which cabal member scored the most coins off people in non-cabalVcabal battles.

During the final hour I had trouble to get into any game.
Only Caravan Disturber started playing in the last 15 minutes again and to have an impact on the player board, both played aggressively. I would not clap my back too hard for getting coins from me at the end of the tournament.

Cobblestone · 2015-04-15 20:40:02

claspa wrote:

I do understand what Cobblestone is saying. We played the first game of the contest against each other. It took us almost an hour to get a decisive win, which was 30% of the whole contest time.
I would like to see a some sort of modifier/compensation, but I cannot think of something good.

Exactly, ya. There were moments in our game where either of us could've left early and still come out ahead, but we played to the end and I don't think it's a bad idea to reward that.

What I was thinking is something along the lines of an ante of 20 would double a positive score or halve a negative one. So winning a longer game nets you a better score, or losing a longer game leaves you with less of a hole to climb out of. I'm not sure what the formula is exactly, but it would end up looking like:

Ante = Max Positive Score
1 = 99
10 = 149
20 = 198

Ante = Max Negative Score
1 = -99
10 = -75
20 = -50

LiteS · 2015-04-15 21:54:02

Cobblestone wrote:

claspa wrote:
I do understand what Cobblestone is saying. We played the first game of the contest against each other. It took us almost an hour to get a decisive win, which was 30% of the whole contest time.
I would like to see a some sort of modifier/compensation, but I cannot think of something good.
Exactly, ya. There were moments in our game where either of us could've left early and still come out ahead, but we played to the end and I don't think it's a bad idea to reward that.
What I was thinking is something along the lines of an ante of 20 would double a positive score or halve a negative one. So winning a longer game nets you a better score, or losing a longer game leaves you with less of a hole to climb out of. I'm not sure what the formula is exactly, but it would end up looking like:
Ante = Max Positive Score
1 = 99
10 = 149
20 = 198
Ante = Max Negative Score
1 = -99
10 = -75
20 = -50

I think what y'all are getting at is the tribute system as a whole "discourages" longer games and is in itself a different topic. The tribute system currently doesn't take a flat 10% from your winnings, instead it chips away money as the ante gets higher.

Cobblestone · 2015-04-15 23:29:24

Whoops, I forgot to take the tribute into account with my numbers. Obviously after 20 rounds, the max possible score couldn't be 99*2 since the tribute would've raked more than one coin. Maybe the winner should get points equal to coins + tribute?

.. · 2015-04-16 07:12:23

If you want to also penalise the loser less for losing after a long game (which I think is fair, because if you've lasted 20 rounds then you can't be playing much worse than your opponent), then the loser could also receive coins+tribute points, but then you have a situation where both players could receive positive points.

This is a problem that every CM tournament has encountered (with the exception of the amulet contest, due to its winning streak thing): they reward maximising your profit rate per minute, not profit rate per game (i.e. playing as well as possible). A structured tournament like a round-robin, Swiss, or double elimination, (or, I suppose, ranking by profit ratio instead of profit) seems necessary to reward those long games. I certainly hope we have those types of tournaments. Adding artificial points systems to change the incentives is sketchy.

LiteS · 2015-04-16 16:06:33

Dan_Dan84 wrote:

One piece of input I will give: while clearly it is better to have a "cabal" than just "The Creator" to create special matches, as more people will encounter these matches, the concept is not quite as exciting. The idea of "This might be a game against Jason Rohrer, and if I win, I score special points" is very powerful. Playing against a faceless cabal... not quite as interesting.

This is true, and a little unfortunate. Marketing could help fix this. "Jason Rohrer and his Cabal" or something to that tune might sound better.

.. wrote:

For example, you could hand out two amulets per day of the contest rather than all 36 at the end.

Resetting everyone's score daily would encourage reckless play, but rewarding daily standings in a week long competition should not be too bad.

If the contest structure stayed the same except floored scores at 0, I think it's safe to say we have successfully eliminated all collusion problems associated with the previous contests. Regarding the "Long games win less coins" thing, while true, if we consider this a problem I think it is solved by increasing contest duration. You still only have to score a minimum of 50 coins twice to defeat a more reckless player (I'm assuming we're associating a score of 99 with reckless play).

If I were to make a prediction for the leaderboards of the contest, I'd say a few players will rush to the top, then stay there, sitting on their position unless someone manages to climb above them. Following Ghost Amount's telling of how the contest went for him, he was nervous about risking his earned coins, and was more concerned with holding them for a possible second place than even trying to catch Arbiter Expression's lead. Loss Aversion is real, and both GA and AE were rewarded for risking their position.

CORDIAL MINUET ENSEMBLE

#1 2015-04-08 18:29:25

Fourth test contest with MONEY prizes

#2 2015-04-09 13:21:27

Re: Fourth test contest with MONEY prizes

#3 2015-04-09 14:04:46

Re: Fourth test contest with MONEY prizes

#4 2015-04-09 16:56:17

Re: Fourth test contest with MONEY prizes

#5 2015-04-09 19:09:37

Re: Fourth test contest with MONEY prizes

#6 2015-04-09 20:15:19

Re: Fourth test contest with MONEY prizes

#7 2015-04-09 20:33:05

Re: Fourth test contest with MONEY prizes

#8 2015-04-12 18:44:28

Re: Fourth test contest with MONEY prizes

#9 2015-04-13 07:44:48

Re: Fourth test contest with MONEY prizes

#10 2015-04-13 16:22:16

Re: Fourth test contest with MONEY prizes

#11 2015-04-13 17:32:11

Re: Fourth test contest with MONEY prizes

#12 2015-04-13 23:59:21

Re: Fourth test contest with MONEY prizes

#13 2015-04-14 04:56:20

Re: Fourth test contest with MONEY prizes

#14 2015-04-14 15:13:52

Re: Fourth test contest with MONEY prizes

#15 2015-04-14 15:31:38

Re: Fourth test contest with MONEY prizes

#16 2015-04-15 01:58:55

Re: Fourth test contest with MONEY prizes

#17 2015-04-15 12:10:20

Re: Fourth test contest with MONEY prizes

#18 2015-04-15 13:11:43

Re: Fourth test contest with MONEY prizes

#19 2015-04-15 13:25:16

Re: Fourth test contest with MONEY prizes

#20 2015-04-15 15:14:57

Re: Fourth test contest with MONEY prizes

#21 2015-04-15 20:40:02

Re: Fourth test contest with MONEY prizes

#22 2015-04-15 21:54:02

Re: Fourth test contest with MONEY prizes

#23 2015-04-15 23:29:24

Re: Fourth test contest with MONEY prizes

#24 2015-04-16 07:12:23

Re: Fourth test contest with MONEY prizes

#25 2015-04-16 16:06:33

Re: Fourth test contest with MONEY prizes

Board footer