MEXC fair price, in one formula

MEXC BTC_USDT figures below were pulled from the public MEXC contract market data at 3:55 PM ET on Thursday, Nov 5, 2026, a few hours after a sell-off in US tech and AI stocks dragged crypto lower.

Rule

MEXC defines it directly: Fair Price = Median (Funding Rate Premium, Mid-Price Basis Fair Price, Last Price). It is used to determine unrealized PNL and to avoid unnecessary liquidations caused by short-term volatility.

Formula

funding_rate_premium = index x (1 + funding_rate x hours_to_next_settlement / settlement_period_hours) mid_basis_fair = index + moving_average(best_bid/2 + best_ask/2 - index) fair = median(funding_rate_premium, mid_basis_fair, last)

Steps

1. Inputs at 3:55 PM ET: index 76,252.3, last 76,223.2, posted fair 76,227.1, funding 0.01% per 8h (0.0001), next settlement 7:00 PM ET, which was 3.0739 hours away.
2. Funding premium leg: 76,252.3 x (1 + 0.0001 x 3.0739 / 8) = 76,255.2.
3. Last leg: 76,223.2.
4. The posted fair, 76,227.1, sits between those two. For a median, that means it is the mid-basis leg: the moving-average midpoint of the book traded 25.2 under the index.
5. Last traded 29.1 under the index. After the flush, the perp traded at a small discount to spot, which tells you sellers were still leaning on the perp.

Why a median

A median ignores the most extreme of the three legs. In a cascade, one wild print on last cannot set the fair price unless the book's midpoint or the funding leg moves with it. That protects you from wicks. It does not protect you from a real trend: when the whole book reprices, two legs move together and the fair price follows within the moving-average window. Thursday's 10 AM hour was the second kind, not the first.

What changes it

• Fast candles. The 10:00 AM ET hourly candle went from 78,272.0 to 75,532.6, a 3.50% drop, and the 24h range ran from 80,927.5 down to 75,532.6 (6.67%). When last and the book's midpoint both sit at the low, the fair price follows, because two of the three legs agree.
• Time to settlement. As hours_to_next shrinks toward zero, the funding leg converges to the index.
• The funding cap. BTC_USDT funding is capped at 0.18% and floored at -0.18% per interval, which limits how far the funding leg can sit from the index.
• Fees. Trading fees on MEXC vary by region, so check the fee schedule for your region before you build fees into a liquidation buffer.

Mistakes

• Treating the index as your liquidation reference. The index feeds two legs, but the check runs on fair.
• Assuming a print on last at your liquidation price means you are gone. One leg alone cannot move a median.
• Assuming a bounce on last saves you. If the book's midpoint stays low, fair stays low.
• Reading a discount to index as free money. It is positioning data, not a signal.

FAQ

Why did my long show a gain the instant it filled? You filled at last and PnL is marked at fair. At 3:55 PM ET fair sat 3.9 above last, so a long filled at last showed a small unrealized gain at that moment, and a short a small loss.

Does the funding leg matter at 0.01%? Barely, three hours out: it put that leg only 2.9 above the index. It matters more when the rate runs near the cap with a full interval left.

Is the fair price the same on every pair? The formula is the same. The inputs differ: each pair has its own index, book and funding rate.

The formula restated: fair = median(index x (1 + rate x h/8), index + MA(mid - index), last). Know which leg is driving it before you put a stop near your line.