You pick the setup. We show what's driving price right now, and who's exposed if it turns.
No buy signals. No price targets.
Every number carries the evidence behind it — and we publish the ones that die.
Leverage looks normal. No crowded side and no forced flow.
Where days like today landed after 24h · ~45 independent days · through 25 Jul 2026
Pick a side and the readings reorder by relevance. Nothing is added, removed or reworded — the market state is the same whichever side you are on.
You already chose the trade. The missing question is whether conditions give it room to work.
Built from individual trades rather than candles, separated by participant size, across spot and futures together.
Is new participation still arriving, or is the move running on what's already in?
Where leveraged positions entered, how much is offside, and how far the nearest forced exits sit.
Whether heavy buying or selling actually moves price — or gets quietly absorbed by someone taking the other side.
The full range of what happened after comparable conditions — not a single number pretending to be a forecast.
Leverage builds, something triggers it, it cascades, then it burns out. The panel names which phase the market is in and shows the components behind the call.
No crowded side and no forced flow. Positions are spread across both directions and nothing is being pushed out of the market.
Composed from funding percentile, liquidation bursts against their own 30-day distribution, and price displacement. Each component is measured; the stage label they compose is on trial.
From comparable states, BTC moved 1.2% against longs within 24 hours in 64% of cases.
We don't tell you to widen it, shrink the position, or skip the trade. You know your method; you just didn't have this number before you placed it.
Our best idea died in public.
The hypothesis that looked strongest failed its first out-of-sample test. We published the result and pulled it from the product — because a court date you only honour when you win isn't a court date.
See the full recordEach hypothesis is registered before the test period arrives, with its thresholds frozen and a verdict date fixed in advance. Results are published whether they survive or not.
Arithmetic on current market data. A fact you could recompute yourself from the same feed.
A pattern that repeats in our own history and has not been validated out of sample. We say so on the number itself.
Registered before the data arrived, thresholds frozen, judged on a date set in advance. It may well fail.
From the maker of TradingView tools used by 30,000+ traders. Plainflow is a separate, independent product.
Private beta. Cancel anytime.
No. You choose the trade. The product shows whether current conditions support it, work against it, or say nothing useful either way. There is no entry signal in the product today, and the panel says so plainly when you open it.
Participation, leverage and liquidity — measured from individual trades rather than candles. Who is buying and in what size, where leveraged positions entered, how much is offside, and whether pressure is being absorbed or moving price.
No. Liquidation exposure is one input. The product also measures whether leverage is still building, whether forced flow has actually started, and whether the market is absorbing it — which a heatmap cannot tell you.
It varies by claim, which is why every number is labelled. Some are direct measurements. Some are unvalidated historical patterns. Some are registered hypotheses awaiting a verdict date. Nothing is currently claimed as validated, and the research record shows the failures alongside the survivors.
Yes, but it assumes you already understand entries, stops and position sizing. It describes market conditions — it does not teach you to trade.
Plainflow is in private beta. Request an invite and we'll send your access when a seat opens — nothing else.