ETH Gains 1.85% This Week: 77 of 200 Weekly Samples Were Stronger
The week Ethereum's foundation shipped zkAPI on mainnet, ETHUSDT closed at 2737.11 USDT for a 1.85% weekly gain.
Binance is not available in your region. What follows is not a centralized exchange โ it is a different kind of product that you can use:
๐ Binance Web3
A self-custody wallet inside the Binance app; swaps route through on-chain DEX aggregation, not an exchange account
Bind via our link and wallet swap fees drop 30%. On-chain gas and slippage are not covered by the discount.
You must already be a verified Binance user to create a Binance Wallet (official FAQ Q13); wallet referrals and exchange referrals are two separate systems. Zero-fee pairs such as stablecoin swaps generate no rebatable fee.
The week Ethereum's foundation shipped zkAPI on mainnet, ETHUSDT moved from 2687.36 USDT to 2737.11 USDT, a 1.85% weekly gain. Over the same window, Binance's default maker fee on ETHUSDT perpetuals is 0.02% and the default taker fee is 0.05%; after a 20% rebate the effective rates are 0.016% and 0.04%. Narrative is one thing, the tape is another: the week closed green, but 77 of 200 weekly samples were stronger.
What the week actually looked like
Here are the daily candles, so the weekly number is not the only thing you remember.
| Date | open | close | change |
|---|---|---|---|
| 2026-09-25 | 2686.65 USDT | 2690.56 USDT | 0.15% |
| 2026-09-26 | 2690.56 USDT | 2694.97 USDT | 0.16% |
| 2026-09-27 | 2694.97 USDT | 2687.36 USDT | -0.28% |
| 2026-09-28 | 2687.36 USDT | 2687.55 USDT | 0.01% |
| 2026-09-29 | 2687.56 USDT | 2676.94 USDT | -0.4% |
| 2026-09-30 | 2676.94 USDT | 2684.68 USDT | 0.29% |
| 2026-10-01 | 2684.68 USDT | 2705.06 USDT | 0.76% |
| 2026-10-02 | 2705.05 USDT | 2737.11 USDT | 1.19% |
Three days tell most of the story. 09-29 was the ugliest, sliding from 2687.56 USDT to 2676.94 USDT for a 0.4% drop, the lowest close of the week; the zkAPI news did not lift the price in that moment. 10-01 was the turn, from 2684.68 USDT to 2705.06 USDT, up 0.76%. 10-02 was the strongest day, from 2705.05 USDT to 2737.11 USDT, up 1.19% and the highest close of the week.
Read the eight days together and the first four barely moved, all within 0.28%. The push came in the last two sessions, which delivered most of the weekly gain. The structure says the gap between news and money is measured in days, not minutes. The 1.85% weekly print was carried by the final two candles, not earned evenly across the week.
Where it ranks among 200 weekly samples
A 1.85% number alone invites้่ง, so it needs a reference. ETHUSDT has 200 weekly samples, and 77 of them were stronger. In other words, this week did not even crack the top third of the sample.
| Stronger week | open | close | change |
|---|---|---|---|
| 2025-05-05 | 1807.87 USDT | 2513.48 USDT | 39.03% |
| 2026-08-17 | 1875.21 USDT | 2462.58 USDT | 31.32% |
| 2024-11-04 | 2457.3 USDT | 3185.19 USDT | 29.62% |
| 2025-07-14 | 2970.37 USDT | 3755.67 USDT | 26.44% |
| 2024-05-20 | 3069.98 USDT | 3827.53 USDT | 24.68% |
The leaders are a different scale. The week of 2025-05-05 ran from 1807.87 USDT to 2513.48 USDT, up 39.03%; 2026-08-17 ran from 1875.21 USDT to 2462.58 USDT, up 31.32%; 2024-11-04 ran from 2457.3 USDT to 3185.19 USDT, up 29.62%. All three started lower than today's level and still covered far more ground in a single week.
Further down the list, 2025-07-14 gained 26.44%, 2024-05-20 gained 24.68%, 2025-08-04 gained 21.57%, and 2023-01-09 gained 20.46%. Those weeks each pushed price onto a new step. This week only moved 2687.36 USDT to 2737.11 USDT. The fair reading: ETH closed green on the zkAPI week, but within its own history this sits mid-to-lower on strength, not a breakout.
What the trending posts say
Trending posts on Square argue the mainnet zkAPI release matters for the privacy and scaling narrative rather than for near-term price (relayed, not this site's view). Two threads dominate. First, offering zkAPI at the mainnet level lowers the barrier to zero-knowledge calls, so developers do not have to build proof infrastructure from scratch. Second, if more applications adopt it, the way on-chain data is exposed changes over time, which is incremental for privacy-sensitive use cases.
Other posts caution that narrative and price often lag each other, and that shipping an interface on mainnet does not mean capital arrives immediately; the real test is how many projects actually call it. That caution matches the tape: 09-29 still fell 0.4%, and the gains only showed up on 10-01 and 10-02. Treat the posts as one interpretation, not as a realized catalyst.
Three things worth stating plainly
On gaps, do not imagine intraday trading from continuous daily closes. On 09-28 the open was 2687.36 USDT while the prior 09-27 close was 2687.36 USDT, which looks seamless; but 09-29 opened at 2687.56 USDT against the 09-28 close of 2687.55 USDT. The daily chart smooths that out; actual fills and momentary swings do not.
On leverage, being right on direction is not the same as staying in the trade. The first four days moved within 0.28%, and chop is the hardest part to sit through. High-leverage positions tend to get cleared on small swings, so by the time 10-01 and 10-02 delivered, the holder is often already out. A 1.85% weekly gain is a calculated result, not a lived experience.
On funding, perpetuals are not free to hold. ETHUSDT is a crypto USDT-margined perpetual, so any position held across a settlement point incurs funding, and direction plus sentiment decide whether you receive or pay. A long that only stares at the 1.85% weekly move and ignores carry cost will find the realized number is not the same as the printed one.
One more thing belongs in the open: the best week is always a conclusion in hindsight. The week of 2025-05-05 gained 39.03%, and nobody inside it knew they were standing in the strongest week; this week's 1.85% likewise only gets its rank once it is finished. Using a rear-view ranking to size a live position puts the cart before the horse.
The fee numbers
Binance's default maker fee on futures is 0.02% and the default taker fee is 0.05%. The rebate ratio is 20%, which brings the effective maker rate to 0.016% and the effective taker rate to 0.04%.
Put those against this week: a 1.85% weekly move gets eaten away by fees if you are posting and taking repeatedly, and the more often you trade the more visible it becomes. Remember 0.016% and 0.04% as the effective rates; they are closer to real cost than the headline numbers.
Source: https://fapi.binance.com/fapi/v1/exchangeInfo Data as of 2026-10-02; source: Binance official API. Not investment advice.
Related Articles
Binance is not available in your region. What follows is not a centralized exchange โ it is a different kind of product that you can use:
๐ Binance Web3
A self-custody wallet inside the Binance app; swaps route through on-chain DEX aggregation, not an exchange account
Bind via our link and wallet swap fees drop 30%. On-chain gas and slippage are not covered by the discount.
You must already be a verified Binance user to create a Binance Wallet (official FAQ Q13); wallet referrals and exchange referrals are two separate systems. Zero-fee pairs such as stablecoin swaps generate no rebatable fee.