Binance (wBETH): staking platforms rating breakdown
It is here as the exhibit, not as an option — Binance.com does not serve US persons. On 10 October 2025, wBETH printed as low as about $430 against ETH above $3,800. An 88% discount. On Binance’s own order book. On a fully-backed, 1:1 wrapped staking token worth $7 billion.
It is here as the exhibit, not as an option — Binance.com does not serve US persons. On 10 October 2025, wBETH printed as low as about $430 against ETH above $3,800. An 88% discount. On Binance’s own order book. On a fully-backed, 1:1 wrapped staking token worth $7 billion.
How the score is built
Each criterion is scored 0–10 and weighted. The median column is the middle score across every entry in this ranking, so a row reads as a position rather than a number.
| Criterion | What we measured | Weight | Score | Category median | Rank | Weighted gap |
|---|---|---|---|---|---|---|
| Custody and slashing | — | 35% | 2 | 3 | 4 of 4 | -0.35 |
| Do they tell you what they take? | — | 25% | 4 | 3.5 | 2 of 4 | +0.13 |
| Net yield against the real ceiling | — | 25% | 8 | 6 | 2 of 4 | +0.50 |
| Can you get out? | one — rounding to $0.0m on-chain pools for a $6.9bn token · sourced · 2026-07-15 · source | 15% | 1 | 2 | 3 of 3 | -0.15 |
Measured 15 July 2026 · weights and method · decided by net yield against the real ceiling, worth +0.50 points against the median
Custody and slashing: 2/10
Do you hold an asset or an IOU against a company balance sheet? Who signs, who can freeze, and what happens to your stake if the operator fails. When a validator is slashed, what actually covers it — a contractual claim you own, a fund of undisclosed size, or nothing at all. A solo validator scores here on the fact that no third party can freeze it, and is docked because 100% of the slashing risk lands on you.
Scored 2 of 10 against a category median of 3, which places it 4th of 4 among staking platforms on this criterion. At a 35% weight that is 0.35 points below the median contribution of the weighted total. The best score in the category is 6, the worst 2.
Do they tell you what they take?: 4/10
Not the fee percentage — whether the yield they SHOW you is the yield you GET. Kraken’s own support page states the in-app APYs "do not include Kraken’s commission". Coinbase states a retail rate and never states the take rate behind it, which we then have to derive by dividing against the network rate.
Scored 4 of 10 against a category median of 3.5, which places it 2nd of 4 among staking platforms on this criterion. At a 25% weight that is 0.13 points above the median contribution of the weighted total. The best score in the category is 10, the worst 2.
Net yield against the real ceiling: 8/10
Measured against the gross rate the network actually pays, which we derive rather than accept: the protocol APR before anyone’s cut. Every platform is scored on how much of that it gives back to you, not on how its headline compares to a competitor’s headline.
Scored 8 of 10 against a category median of 6, which places it 2nd of 4 among staking platforms on this criterion. At a 25% weight that is 0.50 points above the median contribution of the weighted total. The best score in the category is 10, the worst 3.
Can you get out?: 1/10
Unbonding and withdrawal mechanics as the platform documents them: the exit queue, any platform-imposed lock, and — where the platform issues a token instead of an unlock — the depth genuinely available to sell into. Marked not-applicable, with its weight redistributed, where there is no product between you and the protocol exit queue.
Scored 1 of 10 against a category median of 2, which places it 3rd of 3 among staking platforms on this criterion. At a 15% weight that is 0.15 points below the median contribution of the weighted total. The best score in the category is 7, the worst 1.
Other measurements
Incidents priced into this score
- 2025-10-10 — wBETH printed ~$430 against ETH above $3,800 — an 88% discount — on Binance’s own book, as thin-pair collateral pricing fed a liquidation cascade. Binance paid $283m in compensation and changed the pricing method. The underlying ETH was never impaired. [users made whole: confirmed] [source]
Questions about this score
What happened to wBETH in October 2025?
+
It traded at roughly $430 while ETH was above $3,800 — an 88% discount on a fully-backed token — because Binance priced the collateral off a thin internal pair rather than deeper external liquidity. The resulting liquidations fed more wBETH into the same book. Binance paid $283m in compensation and changed how the token is priced. The staked ETH behind it was never impaired.
Does full backing protect a staking token?
+
No, and wBETH is the most expensive demonstration of that on record. Four fully-backed tokens have now printed catastrophic discounts purely because there was no market to sell into: stETH in 2022, mSOL in 2023, ezETH in 2024 and wBETH in 2025. Backing tells you what the token is worth in theory; depth tells you what you can get for it today.