Our Philosophy
From Vibes to PnL
Most prediction-market trading starts with a vibe: a belief about an outcome, a narrative, strong opinion, or a mental if/then model.
PredictBack is built to move you past that—toward rule-based, testable, repeatable strategies that can be evaluated on actual PnL, not intuition.
Instead of asking “Do I think this resolves Yes?”, PredictBack lets you ask and quantify:
What rules trigger entry and exit?
How does this logic behave across many markets?
What happens when liquidity disappears?
Where does execution destroy the edge?
What Makes Prediction Markets Different
Prediction markets look simple. They are not.
Binary outcomes hide path-dependent PnL Interim price movement, timing, and exits matter far more than resolution alone.
Liquidity is episodic and event-driven Volume clusters around news, narratives, and deadlines—not continuously.
Execution assumptions dominate returns Slippage, partial fills, and queue position often matter more than signal quality.
Strategies are regime-fragile Many approaches work briefly, then decay or invert.
Naive backtests that assume perfect fills or static liquidity produce misleading results. PredictBack is designed to surface where and why strategies break, not just where they look good.
What You Can Do With PredictBack
PredictBack lets you:
Test systematic, rule-based trading logic instead of one-off opinions
Evaluate strategies across many markets, not cherry-picked examples
Measure execution drag, capital usage, and drawdowns explicitly
Compare strategies on a like-for-like basis
Iterate quickly from hypothesis → evidence → refinement
The goal is not to tell you what to trade. It’s to help you understand what actually works, under realistic assumptions.
Our Philosophy
PredictBack is opinionated by design:
Realism over optimism
Research over narratives
Failure discovery over curve-fitting
If a strategy only works under perfect execution, infinite liquidity, or hindsight—it doesn’t work.
PredictBack exists to make that obvious before capital is at risk.
Last updated