Ask most sports platforms who you care about and they give you two buttons: follow or unfollow. That is the extent of their vocabulary for human attachment to a team. It is absurd. You do not "follow" your hometown team. You would run through a wall for them. You do not "unfollow" the division rival. You actively enjoy when bad things happen to their offensive line.
Binary preferences, broken product
Every platform's recommendation engine suffers from this same flatness. If the only signals you can give the product are "follow" and "unfollow," the product cannot distinguish between your 15 favorite players and your 15 least-favorite. So it treats them all identically, which means everyone ends up with a generic feed full of content they kind of tolerate.
Real sports preferences are not binary. They are graduated, and they swing. A player you loved last year may be a different story this year after a bad contract extension. A team you used to hate might have picked up a coach you respect and now you are grudgingly rooting for them. A receiver you fade hard for three weeks might flip into a trust-fall after a breakout game.
The five-state spectrum
RPO's Fire/Ice system gives you five states instead of two:
- Fire — ride-or-die. You actively want picks, news, and opportunities involving this player/team surfaced aggressively.
- Warm — positive lean. Interested, not obsessed. Surface when the opportunity is genuinely good.
- Neutral — no strong feeling. Treat normally; no bias either way.
- Cool — soft fade. Mild preference to avoid picks here; surface the alternative first when one exists.
- Ice — hard fade. Do not recommend picks for this player/team. I am never taking the bet.
Five states, not two. Intensity matters as much as direction. And because the scale is dynamic, you can toggle it any time — before a big game, after an injury, mid-season. Your feed stays honest to how you actually feel this week.
You do not "follow" your hometown team. You would run through a wall for them. Binary preferences do not match how fans actually feel.
What Fire/Ice changes in practice
When you mark a player Ice, RPO literally does not show you picks for them. When you mark a team Fire, their games bubble up earlier and their players get more real estate in your pick card carousel.
More subtly, the AI reads your Fire/Ice map when it generates rationale. "Here's a solid prop on Wilson — note you have him marked Ice, so I want to show my work on why this one is still worth your attention." That kind of personalized reasoning is only possible when the system knows what you care about at a granular level.
Identity, not filter
Fire/Ice is not a filter. It is an identity layer. Two bettors can look at the same slate and get completely different recommendations because one has the home team on Fire and the other has them on Ice. The platform stops pretending everyone wants the same picks and starts delivering the version of the day's slate that fits your worldview.
This is the part that feels most obviously right once you use it: your sports app should know who you ride for and who you fade. The current industry standard — a generic feed for everyone — is absurd once you have experienced the alternative.
Fire/Ice is not a filter. It is an identity layer. The platform stops pretending everyone wants the same picks.
Set it up in 90 seconds
The Fire/Ice map takes about 90 seconds to populate on first setup. Mark five Fires, five Ices, and a handful of neutrals. The rest you can fill in over the season as you notice a player you want to fade or a team you want to ride. The system gets sharper every time you adjust it.
Download RPO and set your Fire/Ice map during onboarding. Your feed will finally feel like yours.