I'm a long-time lurker and this seemed like a topic worth making an account for.
I should start by acknowledging that I'm not an expert compared to a number of the folks on this board and that, at this point, I spend a lot more time theory crafting and designing cubes than actually testing and playing them. That being said, over the past few years, I've been working on designing a synergy-based environment (many thanks to the posters here who I've learned from). It's been a lot of work, but I finally sleeved everything up this weekend. Here's the list: https://cubecobra.com/cube/list/10at
In terms of design philosophy, I have a lot of the same sentiments as other posters, but there are some other things that I want to highlight:
- Avoid over-supporting archetypes - The first iteration of this cube had far fewer archetypes and each of them was heavily supported. This led to the classic "drafting on wheels" scenario. It was so bad that sometimes two drafters sitting beside each other would be in the same archetype (e.g., +1/+1 counters), but would still feel like they were getting everything they wanted.
In my current iteration of my list, I've tried to cut back on supporting full blown decks / archetypes, and instead tried to only include a couple payoffs for each. A side effect of this is that there are a lot of different archetypes to be explored and I'm hoping they'll be merged together in creative ways. I'm hoping that rather than having people say "check out my sweet artifact deck" they'll end up saying "I never expected to end up in an artifact, dragon, madness deck".
- Have cards be universally playable - Rather than supporting narrow cards that only go in one deck, but are busted there, I've tried to include cards that go into a bunch of decks, but really shine when brought together. A card like Rampage of Valkyries is a decent stand-alone card in my environment, but when you start to add some other angels to your deck, it really gets going. At the same time, there are a bunch of different reasons a drafter may prioritize Rampage of Valkyries (e.g., if they already have an Archon of Sun's Grace and they want the constellation trigger, they have a Tergrid so they want to make their opponent sac creatures, they have an Esika's Chariot and want to populate more tokens, they have a Kor Skyfisher and they want bounce targets, etc.)
There are also a couple of ways to remove the risk of having synergy pieces be dead cards. One is having payoffs that combo with themselves. Something like Basri's Lieutenant is great because it both places a +1/+1 counter and benefits from having +1/+1 counters lying around. Playing a +1/+1 counter payoff like Lieutenant instead of one like Hardened Scales removes a lot of risk, because it's always going to do something. A second approach is looking for synergies that don't require very much special attention to benefit from. An example of this would be the draw trigger on Nadir Kraken. Players are going to draw a card every turn no matter what, but the card gets way better when you throw cantrips in the mix. A similar example is something like the "play your second card this turn" archetype from Kaldheim. In theory, every deck should have low drops that allow them to play multiple cards a turn, but the ability is way stronger than built around.
- Single card archetypes are great - I love cards like Approach of the Second Sun, Aetherf*ck Reservoir, Experimental Frenzy and Demonic Pact, because they're super compact to include in your list (just one card) and they can lead drafters to approach their draft in a totally different way. I'm particularly fond of something like Aetherflux Reservoir, since it's so open ended. Do I have any idea how people are going to storm off? No. Do I even know if it will ever happen? Also no. Do I know that my drafters and I will try to make it happen? Definitely yes!
- One thing I'm worried about in my list - My list is hyper-curated. I've spent wayyyyyy too much time thinking about how cards interact, to try to strike the perfect balance. My fear is that I've ended up with a list that limits the creativity of my drafters, because I've already pre-planned everything that I want to enable. This would be too bad since some of the best moments in my original cube (a Legacy list more in the MTGS vein) came when players built decks in completely unexpected ways. For example, the most unexpected deck someone's built was "mill wheel", where the drafter combined Wheel of Fortune and Memory Jar with Regrowth effects to mill the opponent out. The deck was terrible, but it was a blast for the player to draft and pilot. I hope my new list still has moments like this.
Okay, this is enough of an essay for one day. I'm not sure how often I'll post (job + baby is a lot), but I'll definitely keep lurking!