Crypto terms, explained plainly
124 short definitions and a handful of guides. Every term links to the coins and charts where you will meet it, and the ones the model actually uses say so.
Start here
B
- Base caseOur model
The median of a forecast distribution: half of the model's outcomes sit above it.
- Brier scoreOur model
Mean squared error of probability forecasts; 0.25 is the score of a coin flip.
C
- CalibrationOur model
How closely stated confidence matches the real hit rate of calls made at that confidence.
- ConfidenceOur model
The probability of the call's direction, as a number the site publishes next to every forecast.
D
Which way the model leans; neutral means it has no edge over a coin flip at that horizon.
E
- Expected returnOur model
The mean outcome a distribution implies; the base case sits near it when the direction model is on.
F
- Forecast range (low–base–high)Our model
A probabilistic band: the 10th percentile low, the median base and the 90th percentile high.
G
- GradingOur model
Automatically scoring a forecast at expiry against the recorded price, wins and losses alike.
H
- Hit rateOur model
The share of graded calls that were correct, by horizon and by model version.
M
- Model gateOur model
The rule that hides a forecast when the model could not beat a coin flip on it.
N
- No setupOur model
The hourly board's 'no trade' state: confidence did not clear the threshold, so nothing is called.
O
- OverfittingOur model
Fitting noise instead of signal, so a model looks brilliant on old data and fails live.
P
- Probability upOur model
The model's raw probability that the price is higher at expiry, before calibration.
R
- Risk scoreOur model
A 0–100 blend of volatility, liquidity, size, drawdown, exchange count and age.
T
- Track recordOur model
The public log of every graded call, including the losses, with the calibration behind the confidence.
W
- Walk-forward backtestOur model
Testing the model only on data after each training window, the way it would have been used.