I'm backburning the name "Odd Ducks" at this point, the original idea that odd value cards were special for stocks doesn't make sense right now, so the working title has become EvenOdd.
Development Journey!
After writing my last post, I spent hours tweaking and thinking about this game and went on a couple of development journeys before arriving at the game I tested today.
BIDDING
The same night I wrote the last post, I later realized that I could do away with the "shed-out bonus" cards and instead replaced them with a bidding phase after cards were dealt. The deck is made up of cards from 1-7, so I gave players 14 card hands, and we simultaneously picked one card to bid into the shed-out bonus pool. Just like I mentioned before, the player who bid the lowest card led first, and the player who bid the highest card would gain 2x those points if they shed out first, but instead lost that number of points if they didn't shed out first.
BREAKING TIES
The previous point led me to a problem. With cards valued 1-7, there WILL be tie situations. I decided to add tiebreaker numbers to each card so that they are all unique. This was originally done with dots under the cards like this:
The idea was that if two cards of the same value are bid, then the dots will break the tie to know which is higher or lower. I loved how these cards looked but wasn't sold on how easy they would be to parse.
BOMB BOONDOGGLE
From this point, I fell down a deep "bomb search" rabbit hole. I mentioned in the last post that I was quite interested in finding some way to add bombs to the game to add uncertainty and agency. The idea of a bomb is a meld that can be played outside of the led meld and one that is strictly stronger. I did come up with a great idea, in theory, and it involved finding essentially "straight flushes" inside the deck. If a set of cards had consecutively ascending dots, this would be a bomb with 3 or more, and a run of cards that had identical dots would also be a bomb. I was certainly worried it may be hard to parse these ideas, but figured if the mechanics worked well, then I'd figure out how to show it correctly later. I reached out to a good friend of mine who is far better at math than me, and he graciously built some simulations to determine the odds of finding these types of bombs at various player counts and bomb minimum lengths (3+ or 4+ cards for example). This is a visualization of the bomb types:
The results of my friend's math were quite interesting, but unfortunately, they didn't provide me with any kind of consistency. I was hoping to find a situation where the odds of a bomb with a 3p deck could be similar to that in a 4p deck, and thought I could likely get there by changing the minimum bomb length. In reality, based on my friend's simulation, the closest gap I could get between the 3p and 5p decks was around 20% which is way too big. I don't think the game will feel right if you have a 30% chance to get a bomb with 3p and only a 10% chance with 5p. I was disheartened by the results but also really happy to see them before I invested more time into this idea.
Feeling dejected, I decided I needed to come back to the WHY of this bomb search. This was based on key feedback from playtesters who wanted more agency in the middle of a hand. If someone played 1 & 1, then I wanted there to be the possibility that it wasn't strictly going to win. If there is a chance of bombs, then you may sweat a turn playing 1&1 because a bomb could come in and steal a "guaranteed lead" from you. I thought about this more and realized that bombs aren't the only way to do this...
BREAKING TIES AGAIN
Going off of the previous example of a 1&1 being unbeatable, well, what if they could be beaten by a stronger 1&1 from another player? I already put unique tie-breaker dots on the cards, so why couldn't one 1 card be stronger than another? This felt like a eureka moment, as well as a really head-slapper. Within a minute of this occurring to me, I had it mapped out, I'd use the up/down strength mechanic that is already in the game, and now a 1(2 dots) 1 (5 dots) would beat a 1(3dots) 1(4 dots). Two cards mean lower is stronger, so the 2 dots are stronger than the 3 dots. Sure this means that there will still be strictly "unbeatable" cards in the deck, but these will make non-decision plays far less often than before when you couldn't beat a pair with the same pair.
I realized that the "dots" on the cards wouldn't lend themselves very well to quickly parsing the strength of one card vs another, so I changed the cards to look like this instead. The deck was identical, it just showed differently.
Playtest! 11-1-22
With these changes in mind, I was itching to actually try them out and see if they would work.
I was able to get an impromptu playtest tonight with 3 players, and we played through 3 hands. The overall feedback was positive, both of my opponents not only enjoyed the even/odd low/high mechanic, but they also liked the tie-breaking with the second numbers on cards. I was worried this would feel clunky, but surprisingly it felt natural and simply worked well.
The main points of the discussion centered on the bidding part of the game. On the first hand one player bid a 4 and didn't expect that it would be the highest number. This meant they were essentially surprised with a bet, and when they subsequently didn't go out first, they had a strong "feels bad" moment as they lost 4 points. To them, it felt like the game randomly decided they had bet, and they would much prefer to actually make that decision themselves.
The other tester was also not sold on the entire bidding package of the game. I couldn't pull out any specific pain points, so I think it was more of an overall feeling, but it did foster a great conversation and brainstorming session between the three of us. Here are some of the key points discussed as well as how they affect my plans for the next test:
Slow it down
The hands finished surprisingly fast. I liked the quick pace of the game, we easily finished 3 hands in 15 minutes, but it almost felt too fast too me overall. Happening into a 1-7 card straight isn't going to be all that rare with a deck like this, and being able to dump over half your hand in that case, as well as more likely but slightly smaller cases, makes it easier to shed-out. I hate to say it, but I think I'm going to test with a 4 card ceiling next time to try and slow the hand down just a bit. Also, to force players into leading singles more often since they'll hypothetically have more remainders. That last part may not be true since low runs are still possible, but I'm curious to see how the dynamics shift. A 4 card ceiling means there are still 2 odd and 2 even options which feels like it still offers enough good options. Most hands were 4 cards or less anyway. I hate putting in arbitrary feeling limits like this, so I'll likely think on this part quite a bit after more testing.
Nothing for the loser!
It felt a little odd that the last player with cards still got points. With the score setting phase, there are cards equal to the players, so you may get lots of points and never actually shed-out. This led me to explore the idea of removing one of the score cards so that there is no card for the last player, which means they score nothing for that round. They are intended to be shed-out bonuses, so why give it to a player who didn't shed out? This design space got me considering what to do with that last, lowest, score card and this led me directly into...
High revealer bonus instead of bet
The player who revealed the highest card for scoring takes the lowest score card and puts it in front of them as a potential bonus if they go out first. If they do go out first, then they gain that number of points, however, if they don't go out first, then the card is simply discarded with no further effect. This means they won't accidentally find themselves having put their points on the line, but it also means there is an incentive to bid high because it gives the player the potential to gain these bonus points.
Bring back betting!
The previous idea feels like a relatively tame carrot for one player if they go out first, as well as a tame stick for their opponents to try and deny those bonus points. I don't mind these, but I do love a bet when it comes to shedding games and I want to bring it back. Again, a key piece of feedback was not being forced to bet by the game, so my thought is why not make it optional again. I see two ways of going about this right now, I'm not sure which I prefer:
- Once all score cards are revealed, and the high player takes the lowest score card as a potential bonus, that high player may choose to "double or nothing" that bonus score card. They do this before the low player leads the first hand. If the bonus card is a low value then that player may feel like they may as well bet, but of course they won't get much benefit either by doubling a small number. If it happens to be a higher value bonus card, then that incentive to try and double it could be interesting. I'm not sure if "double or nothing" is the right call, or if I should make it meaner and go with "double or lose points equal to the bonus". I imagine I'll test the tamer version first.
- Have all players be able to do a "double or nothing" bet before they play their first card. There could be betting tokens they could take to make it obvious, or they could just call it out like in Tichu. If they do this bet and go out first, then they gain double the points of that score card. If they do not go out first, then they still take the highest score card, but they flip it face down to show they gain no points for it due to busting. This may feel bad, since why work hard for inevitable 0 points, but it does deny those points from opponents, so hopefully they'd still be motivated to go out ASAP. The idea here is that multiple players could bet, which is always a fun time when it happens in other games like Tichu. This could be especially interesting to give multiple players "hail mary" options if they are far behind and want a way to potentially catch up even if they didn't reveal the highest score card for that round.
