How We Score

Every token analysis on CryptoGuide carries a Crypto Score from 0 to 100. This page explains exactly how it is built, so you can decide how much weight to give it — or ignore it and read the analysis instead, which is the better use of your time.

Five dimensions, equally weighted

The composite is a straight average of five sub-scores, each worth 20%:

  • On-chain health (20%) — network activity, active addresses, whale behaviour, exchange inflows and outflows, holder distribution.
  • Tokenomics quality (20%) — supply mechanics, unlock and vesting schedule, inflation, staking economics, distribution fairness, real token utility.
  • Sentiment & momentum (20%) — social volume and tone, news coverage, community and developer engagement, the Fear & Greed backdrop.
  • Technical setup (20%) — trend, momentum, volume confirmation, market structure, key support and resistance.
  • Fundamental strength (20%) — protocol revenue and adoption, team and treasury, competitive moat, regulatory exposure.

Grades and signals

ScoreGradeWhat it means
85–100A+Favourable across every dimension we measure
70–84AFavourable overall, with manageable identified risks
55–69BMixed — genuine strengths alongside genuine problems
40–54CNo clear edge in the factors we measure
25–39DSignificant headwinds or red flags
0–24FMajor red flags across multiple dimensions

What the score deliberately does not do

It does not forecast price, and it does not express a probability. Two tokens with the same score can have completely different risk profiles — a 70 built on strong fundamentals and weak technicals is a different asset from a 70 built the other way round. That is why the sub-scores are always shown, and why the bull and bear case matter more than the number.

The score is also a snapshot. It reflects the data available at the retrieval timestamp printed on the analysis. We do not silently re-score old articles; a new reading gets a new dated analysis.

Where the data comes from

Market data from public market-data APIs, protocol metrics from DefiLlama and protocol dashboards, on-chain data from block explorers and public analytics. Each analysis lists its own sources with links. Where a metric is self-reported by a protocol and cannot be independently verified, the analysis says so.

Limits worth knowing

  • Sentiment is the noisiest input and the easiest to manipulate. Treat it as context, not signal.
  • Technicals on a 24/7 market with thin weekend liquidity behave differently from equities. We do not pretend otherwise.
  • Small-cap tokens have shallower, less reliable data than majors. Their scores deserve wider error bars.

All of this is research, not advice — see the disclaimer.