Can all prediction markets be numerical?

I like researching because the best ideas tend to come from immersion. When I come across a new topic, what I'm really trying to do is get across the Dunning-Kruger curve as quickly as possible. Not to become a fake expert overnight, but to figure out how much I don't know rapidly. Once I have a better sense of the depth of a topic, I can make a call on whether it's worth pursuing now, later, or not at all.

The way I do that is fairly simple. I read about the thing, use the thing, read what other people think about the thing, listen to people who know more than I do, and generally try to surround myself with it for a period of time. I'm not trying to reach a conclusion quickly. I'm trying to build enough context that the real problems become obvious.

Immersion allows your sub-conscience to do the heavy lifting of problem solving.

A recent example of this came from an interesting problem with functionSPACE. We've been building sports as an application layer, culminating in app.propspace.fun. The NBA finals were on, and numerical markets with our mech made for a very novel and fun trading experience.

We received feedback from a user in our competition. They liked the In/Out (range) functionality for the markets we had, but they also wanted the ability to just trade a simple win-or-loss outcome. That's the default mental model for most sports bettors, so it was a reasonable challenge.

Initially, I was thinking about it from the perspective of creating separate markets as we had in propSPACE, i.e. Total Knicks or Total Spurs. But given we've been spending enough time immersed in sports, with the world cup starting and us building something cool for that event as well, a quick walk around the block meant the answer turned out to be sitting in what we'd already built.

Take Spurs vs Knicks. The outcome space can be represented as a range from -50 to +50, where the value is simply the point differential of the Spurs over the Knicks. That's the underlying market.

On the front end, though, users don't need to see any of that complexity. You can simply ask, "Who is going to win?" and present odds in a familiar money line format. If a user wants to go a step further, they can express a view on the margin of victory. If they want to go further still, they can express a full probability curve across all possible outcomes, and so forth. This is cool because it really brings all trader types together on a level playing field.

The interesting thing is that these aren't actually different markets. They're all different interfaces to the same underlying market.

The casual user gets a win/loss experience.

The sophisticated trader gets spreads and margin-of-victory trading.

The more sophisticated trader gets the ability to express even more specificity, or full probability curves if they want.

functionSPACE already supports all of this natively. The unlock is simply a layer of UI/UX abstraction above the underlying mech.

Once the idea clicked, it felt obvious. Before this, I was thinking in terms of separate win/loss markets and total points markets for each team, which was layering on complexity the underlying market didn't need.

That's one of the reasons I like immersion so much. The answer didn't come from a brainstorming session or a framework. It came from spending enough time with the problem that eventually the problem revealed what it actually was. Once that happened, the solution was mostly self-evident.

Outside of sports, think about what other markets this opens up in functionSPACE:

Bitcoin v Gold.

Apple v Nvidia market cap.

Drake v Bad Bunny.

Lets see where we can take this.