1.2 Points Decided an A From a B
On our research record of November 2, 2026, the top three positions in the CryptoGrade ranking all carry an A.
Bitcoin (BTC), A / Prime, score 88.6, rank 1
Ethereum (ETH), A / Prime, score 85.2, rank 2
Aave (AAVE), A / Prime, score 80.4, rank 3
Compound (COMP), B / Sound, score 79.2, rank 4
All four were scored on the same record, so they are directly comparable. Look at the bottom two lines for a moment. Aave and Compound are 1.2 points apart and they are on opposite sides of a letter.
Our A band starts at 80. Aave is at 80.4. Compound is at 79.2. That is the whole of it.
If you stopped reading there, you would take away something true but thin: a letter grade is a threshold, and two assets can sit a rounding error apart with different letters on them. That is worth knowing. It is not worth an email.
Here is the part that is.
Where those 1.2 points actually came from
Our score is six weighted dimensions. Each one is scored from 0 to 5, each has a fixed weight, and the composite is the sum of weight times score divided by 5. The weights are the same for every asset we rate, which is what makes two assets comparable in the first place.
Aave, on the November 2, 2026 record:
Tokenomics and Value Accrual, weight 28%, score 4 of 5, worth 22.4 points
On-chain Usage and Economics, weight 22%, score 5 of 5, worth 22.0 points
Security, Decentralization and Durability, weight 18%, score 4 of 5, worth 14.4 points
Ecosystem and Moat, weight 12%, score 5 of 5, worth 12.0 points
Relative Valuation, weight 12%, score 2 of 5, worth 4.8 points
Market Structure, weight 8%, score 3 of 5, worth 4.8 points
Total: 80.4.
Compound, same record, same weights:
Tokenomics and Value Accrual, weight 28%, score 5 of 5, worth 28.0 points
On-chain Usage and Economics, weight 22%, score 4 of 5, worth 17.6 points
Security, Decentralization and Durability, weight 18%, score 4 of 5, worth 14.4 points
Ecosystem and Moat, weight 12%, score 4 of 5, worth 9.6 points
Relative Valuation, weight 12%, score 2 of 5, worth 4.8 points
Market Structure, weight 8%, score 3 of 5, worth 4.8 points
Total: 79.2.
Now the comparison is not a letter any more. It is three numbers:
Compound is 5.6 points ahead of Aave on Tokenomics and Value Accrual, the heaviest dimension in our framework at 28%.
Aave is 4.4 points ahead on On-chain Usage and Economics.
Aave is 2.4 points ahead on Ecosystem and Moat.
Add those up: plus 5.6, minus 4.4, minus 2.4. Net minus 1.2. The two assets are identical on the other three dimensions, to the point.
So the asset with the B beats the asset with the A on the single heaviest thing our model measures, and loses the letter on two lighter ones. Our model's standing assessment of Compound reads that its "supply is fully circulating, live protocol usage remains substantial, and the governance/security posture appears durable", and that its main limitation is incomplete fresh evidence on liquidity, volatility and insider distribution. Our standing assessment of Aave reads that the protocol "still shows large TVL, strong fee generation, and a durable lending moat", with valuation as the main restraint.
Neither of those is a reason to ignore the letter. The letter is our model's summary, and the summary is the point of a rating. But a reader who only ever sees the letter does not get to know that this particular A and this particular B disagree about which asset is better on 28% of the scale.
That is the whole reason we publish the dimension scores.
A number that reads the same on all four, for two different reasons
Look again at the four assets above and you will find one score they share.
Relative Valuation reads 2 of 5 on Bitcoin, on Ethereum, on Aave and on Compound, on the November 2, 2026 record. That is 4.8 points out of a possible 12, on all four, including the asset at rank 1.
Read our model's own reasoning and the 2s are not the same 2.
On Bitcoin, our model's rationale says that market cap equals fully diluted value and circulating supply is near the cap, but that no peer-relative set is provided, so "the evidence is insufficient to support a favorable relative-valuation tier". On Ethereum, it says the record "does not include peer-relative ratios", so only a cautious low-end view is supportable. Those are statements about what our model was given.
On Aave, it is a judgement: AAVE "screens expensive versus peers" on a fee-to-market-cap style comparison. On Compound, it is the middle of the road: the evidence does not establish a cheapest-third profile, "but it also does not show an extreme premium".
Same number, four different readings underneath it. If you only saw "2 of 5" you would conclude all four are richly priced by us. Two of them are not being called expensive at all. They are being told, in effect, that our model will not call them anything without the comparison set to do it with.
We would rather publish the restrained number and show you why it is restrained than publish a confident one we cannot stand behind.
The letter is not only the score
One more thing about where a letter comes from, because the 1.2 points above can make it look purely arithmetic.
Our bands are A at 80 and above, B from 60 to just under 80, C from 40 to just under 60, and F below 40. On top of the weighted score our framework runs nine named pass or fail screens, and a failed screen places an asset in F at any score, including a score in the A band. So a high score is not by itself a grade, and an F is not a statement that an asset scored badly.
That is a design choice, and it is why we always publish the letter our record carries rather than one worked out from a number.
What our model says about size, at the grade level
Our record carries a position limit for each grade. On November 2, 2026, our model's limit for an A reads "60% or less (BTC-class) / 30% or less for other A-grade assets", and its limit for a B reads "10% or less".
That is a statement about grades, not about you. We do not know what you own, what you are trying to do with it, or what you can afford to be wrong about, and we are not in the business of telling any individual person what to hold. What we can tell you is that our model treats the A row and the B row very differently, which is another way of saying that the 1.2 points between Aave and Compound are not cosmetic in our own framework.
If you hold any of these
The lookup takes seconds, and every one of these pages carries the grade, the score and the reasoning our record holds for that asset.
The ask is Alerts. It is free with a free CryptoGrade account, and it does one specific thing: it tells you when our grade on an asset you follow changes. The reason to want that is sitting in this issue. Compound is 0.8 of a point below our A band and Aave is 0.4 above it. Letters cross lines like that without anyone watching the number.
Create a free account, then turn on Alerts for what you hold.
The longer read. Erik's research deck, The Low-Effort Guide to Staking in 2026 and What the Yield Really Is, is the foundational piece we keep pointing people to. If you have ever accepted a yield number at face value, start there. And yesterday we published what a staking yield actually is, and the four things it does not tell you.
All grades, tier names, scores, positions, dimension scores and position limits above are CryptoGrade's own published readings from our research record of November 2, 2026, and they change.
CryptoGrade publishes investment research, asset grades, scores and ratings. Nothing here is individualized investment advice, and no one at CryptoGrade is a registered investment adviser or broker-dealer. Invest wisely with AI-powered crypto scores and guidance.
