Buy Crypto
Markets
Spot
FuturesBTC
Earn
Event Center
Rewards Hub
More
100%+ PoR
Overview Per TokenInsight's July 2026 exchange liquidity report, a $1 million ETH futures sell order incurred measured slippage of 0.015% at MEXC, or $150, the lowest among the nine exchanges that repOverview Per TokenInsight's July 2026 exchange liquidity report, a $1 million ETH futures sell order incurred measured slippage of 0.015% at MEXC, or $150, the lowest among the nine exchanges that rep

Large Crypto Order Slippage: Which Exchanges Are More Predictable? MEXC Slipped 0.015% on $1M in ETH

Overview

 
Per TokenInsight's July 2026 exchange liquidity report, a $1 million ETH futures sell order incurred measured slippage of 0.015% at MEXC, or $150, the lowest among the nine exchanges that report sampled. What speaks more directly to the question of predictability is something else: the same measure at the same order size returned 0.015% in a separate, non-overlapping sampling window in that firm's May report, a difference of zero between the two readings.
 
A single low reading only shows that book depth was adequate during one sampling period, whereas the same measurement returning the same value across two non-overlapping windows shows the result is repeatable. For anyone placing large orders repeatedly at different times, the second is the useful information.
 
 

Key Takeaways

 
MEXC's combined near-touch depth for BTC and ETH futures in that firm's September 2026 report was $21.03 million measured at the 0.03% band, ranked first among nine venues, against $15.71 million at the same band in the July report, an increase of 33.9% between the two. Slippage in absolute dollars is small at this scale: per the July report's measured values, a $500,000 sell corresponds to 0.006% or $30 and a $1 million sell to 0.015% or $150, while per the May report's measured values a $5 million sell corresponds to 0.050% or $2,500. Cost grows faster than size, with a doubling from $500,000 to $1 million taking the rate to 2.50 times and the dollar amount to 5.0 times, implying roughly a 2.3 power relationship. Measuring surprise requires percentiles rather than averages: in that firm's August 2026 equity perpetuals report, MEXC's median and 90th-percentile slippage on $100,000 sells were 0.0112% and 0.0230% on SPCX, 0.0130% and 0.0278% on SOXL, and 0.0170% and 0.0241% on SK hynix, the lowest on three of the five contracts sampled. A final point on methodology: crypto markets have no mandated execution-quality disclosure equivalent to Rule 605 in US equities, so band definitions differ between reports, the band determines the ranking, and two liquidity tables reaching different conclusions can both be correct.
 

Why Large Orders Need Percentiles Rather Than Averages

 
Average slippage flattens out the worst fills, and the real risk on a large order sits precisely in those fills.
 
Percentiles address that problem. The 90th-percentile slippage means that across all sampled fills, 90% slipped less than that value and the remaining 10% slipped more. It answers not how much an order typically slips but how much it slips when conditions are poor. Investopedia's definition of slippage sets out the base concept, and applying percentiles to liquidity risk has a formalised framework in the academic literature: a slippage risk study published in March 2026 proposes measuring liquidation cost from the current state of the order book rather than from historical returns, and treats tail quantiles as a leading indicator, with its primary threshold set at the 95th percentile. That study's sample covers a single venue over roughly four weeks with validation largely in-sample, so its specific figures should not be extrapolated, but its methodological finding that the tail rather than the mean carries the warning is usable.
 
Contract
Median slippage
90th-percentile slippage
Ratio of P90 to median
Median cost on $100K
P90 cost
SPCX
0.0112%
0.0230%
2.054 times
$11.20
$23.00
SOXL
0.0130%
0.0278%
2.138 times
$13.00
$27.80
SK hynix
0.0170%
0.0241%
1.418 times
$17.00
$24.10
 
The figures above are MEXC's measured values in TokenInsight's August 2026 equity perpetuals market report, sampling five exchanges over weekdays from July 29 to August 6 at an order size of $100,000 sells, where MEXC was lowest on three of the five contracts sampled. Note this is the equity perpetuals market, not the same market as the crypto spot and futures reports discussed below.
 
One conclusion in that table is easy to skip past. The lowest median belongs to SPCX at 0.0112%, but the smallest ratio of 90th percentile to median belongs to SK hynix at 1.418 times against SPCX's 2.054 times. The contract with the best typical fill, in other words, is not the one with the least surprise. Reading only the median selects wrongly, reading only the 90th percentile selects wrongly, and the two values have to be read together.
 

Crypto Markets Have No Mandated Execution-Quality Disclosure

 
This is the premise behind every cross-exchange liquidity comparison, and without stating it first any ranking can be misread.
 
US equity markets require execution-quality disclosure. The Securities and Exchange Commission adopted amendments in 2024 expanding the disclosure of order execution information, and the final rule text specifies the statistics that must be reported including measures such as effective spread, with legal analysis of those definitions setting out the specifics. In Europe, the regime under the European Securities and Markets Authority likewise requires trading venues to publish execution-quality data periodically. Fidelity Institutional's research explains what such disclosure is worth to investors, while the implementation shortfall concept long used by the CFA Institute remains one of the standard frameworks for measuring execution cost.
 
Crypto markets carry no equivalent requirement. Cross-exchange slippage data available to readers therefore comes from third-party firms' own measurements, and those measurement methods are not uniform. Kaiko's best execution and exchange ranking products, research on liquidity measurement from the Bank for International Settlements, and the two firms cited below each use different bands, order sizes and sampling windows. That is not a data quality problem but an inevitable result where no common standard exists, and what readers need to do is establish what each table measures before reading its ranking. For a full compilation of MEXC's own order book depth and slippage figures alongside the third-party reports, how deep MEXC's liquidity runs itemises them.
 

The Same Measure Returned the Same Value in Two Separate Windows

 
Setting the citable measured values out by report, window and basis makes the repeatability visible.
 
Report and sampling window
Sample
Market and measure
Order size or band
MEXC value
Position in that sample
May 2026, April 1 to May 12
8 venues
ETH futures slippage
$1M sell
0.015%
First
May 2026, as above
8 venues
ETH futures slippage
$5M sell
0.050%
First
July 2026, June 15 to July 14
9 venues
ETH futures slippage
$500K sell
0.006%
Lowest
July 2026, as above
9 venues
ETH futures slippage
$1M sell
0.015%
Lowest
July 2026, as above
9 venues
BTC and ETH futures depth
0.03% band
$15.71M
First
September 2026
9 venues
BTC and ETH futures depth
0.03% band
$21.03M
First
September 2026
9 venues
BTC spot slippage
$500K sell
0.005%
Lowest
 
All values above come from TokenInsight's exchange liquidity report series, of which the September report is the most recent, while the July report release and the May report release set out each period's full basis and coverage of the September report provides a cross-reference.
 
The first and fourth rows are the core of this article. A $1 million ETH futures sell measured 0.015% in the window from April 1 to May 12, and the same 0.015% in a separate window from June 15 to July 14. The two windows do not overlap, the sample grew from eight venues to nine, and the difference between the results is zero. A single low reading can come from the chance composition of a sampling period, while two matching readings are harder to explain that way.
 
The depth side offers a comparison too. At the same 0.03% band, the $15.71 million in the July report became $21.03 million in the September report, an increase of 33.9%. Those two values come from the same firm on the same basis and can therefore be compared directly, whereas substituting another firm's band would invalidate the comparison, for the reason set out next.
 

The Band Determines the Ranking, So Two Tables Can Both Be Right

 
Where two liquidity reports from the same period give different rankings, in most cases neither is wrong and the two are simply measuring different things.
 
Source of the basis
Band definition
Converted at ETH near $1,825
What it actually measures
TokenInsight near-touch
0.01% either side of mid
About $0.18 wide
Immediate absorption at the touch
TokenInsight second band
0.03% either side of mid
About $0.55 wide
Depth available to mid-sized orders
CoinGecko spot study
$3 either side of price
0.164%
Resting orders further out
 
The fixed $3 band CoinGecko uses for ETH works out to roughly 0.164% at that price level, making it 16.4 times wider than the 0.01% near-touch band and 5.5 times wider than the 0.03% band. One is defined in dollars and the other in basis points, and the range of resting orders each captures differs by an order of magnitude, so differing rankings reflect method rather than contradiction. In practical terms, near-touch depth determines fill quality for ordinary order sizes while wider bands reflect the resting orders reached only in violent conditions.
 
In CoinGecko's 2026 spot study, sampling eight exchanges from July 6 to September 3, MEXC's SOL spot depth was about $934,000 and its DOGE spot depth over $443,000, both first in the sample. Setting those alongside the TokenInsight futures depth figures in the previous section requires noting that the markets and bands differ, so they can neither be added nor substituted. On what each band from 0.01% to 0.10% represents and why volume and liquidity are not the same thing, the difference between volume and order book depth works through them band by band.
 

Doubling the Order Does Not Double the Slippage Cost

 
This section answers one specific question: given the slippage on $100,000, what happens at $1 million.
 
The answer is not ten times. Per two measured points within the same July report window, doubling an order from $500,000 to $1 million took the slippage rate from 0.006% to 0.015%, a factor of 2.50, while the dollar cost went from $30 to $150, a factor of 5.0, implying roughly a 2.3 power relationship to size. Per two measured points within the same May report window, increasing from $1 million to $5 million, a factor of five, took the rate from 0.015% to 0.050%, a factor of 3.33, and the dollar cost from $150 to $2,500, a factor of 16.67, implying roughly a 1.75 power relationship.
 
The two exponents differ and come from different sampling windows, so they cannot be joined into a single curve and each holds only within its own window and size range. The direction is consistent, however: cost accelerates relative to size, because once an order consumes the near-touch band it must reach resting orders further out at worse prices. That is also why the near-touch depth figure carries the most weight in actual execution. On why $100,000 and $10,000 receive entirely different fill quality on the same order book, how large orders degrade execution sets out the order book mechanics.
 

Three Checks to Run Before Placing the Order

 
Third-party reports provide point-in-time snapshots while the actual fill happens at the moment of the order, so three checks are needed beyond the reports.
 
The first is the band. On seeing any depth figure, establish which band produced it, because 0.01% and 0.10% differ by an order of magnitude and report headlines usually omit this. The second is converting slippage into dollars and adding the fee. Both slippage and fees are charged on notional, and only their sum is the real cost: per the measured values cited here, a $1 million ETH futures sell carries $150 of slippage, while the fee is calculated separately and at a rate of six basis points comes to $600 on one side, four times the slippage. At this scale, in other words, the fee side carries more weight than the slippage side, and optimising only for slippage misreads the priority. On what a zero fee rate actually means and the full method for total cost, how to calculate your total trading cost itemises the components.
 
The third is estimating from your own size rather than the report's. Reports measure round points such as $100,000, $500,000 and $1 million, and the power relationship in the previous section means intermediate sizes cannot be interpolated linearly. A workable approach is to take the two nearest measured points, extrapolate using the exponent between them, and leave a margin. For the platform's composition and the range of markets covered, the platform overview sets it out.
 

Exclusive View from the MEXC Editorial Team

 
The easiest way to execute this topic is as a ranking table, and a ranking table happens to be the least informative presentation available. The reason is that crypto markets have no mandated execution-quality disclosure equivalent to Rule 605 in US equities, so every cross-exchange slippage dataset comes from a third-party firm's own measurement, and each uses different bands, order sizes and sampling windows. CoinGecko uses $3 either side of price for ETH, which converts to roughly 0.164% at $1,825, making it 16.4 times wider than TokenInsight's 0.01% near-touch band. Those two tables producing different rankings is not a matter of one being wrong but of measuring different things. Confirming the band before reading any liquidity comparison is considerably more useful than memorising a ranking.
 
On the question of predictability itself, a single low reading is insufficient evidence. The actual evidence is repeatability: a $1 million ETH futures sell measured 0.015% in TokenInsight's May report window of April 1 to May 12, and 0.015% again in the July report window of June 15 to July 14, two non-overlapping windows across which the sample grew from eight venues to nine and the readings differ by zero. A single low value can be attributed to the chance composition of a sampling period, while two matching readings resist that explanation. Combined near-touch depth for BTC and ETH futures at the same 0.03% band rose from $15.71 million in July to $21.03 million in September, an increase of 33.9%, a comparison that holds because both the basis and the firm are the same.
 
The percentile data illustrates something else that is routinely overlooked. In the August equity perpetuals report, MEXC's lowest median slippage was SPCX at 0.0112%, but the smallest ratio of 90th percentile to median was SK hynix at 1.418 times against SPCX's 2.054 times. The contract with the best typical fill is not the one with the least surprise, which means reading a single value necessarily selects wrongly. For anyone placing large orders repeatedly, the median determines long-run average cost while the 90th percentile determines the capital committed in the worst single case, and the two require separate accounting.
 
One further relationship of magnitude deserves isolating, because it changes the order of priorities. At the $1 million scale, the measured slippage cited here is $150 while a one-sided fee on the same notional at six basis points is $600, four times the slippage. Slippage is no longer the principal component of cost at this scale; fee rates and the number of fills are. Directing all attention at slippage while ignoring fee structure produces a wrong trade-off on total cost. Equally worth noting is that cost accelerates relative to size, with the $500,000 to $1 million range implying roughly a 2.3 power relationship and the $1 million to $5 million range roughly 1.75, two exponents from different windows that cannot be joined into one curve but that both indicate intermediate sizes cannot be interpolated linearly.
 

FAQ

 

Which measure should large orders be judged on

 
Both the median and the 90th percentile. The median answers how much an order typically slips while the 90th percentile answers how much it slips when conditions are poor, meaning 90% of sampled fills slipped less than that value. An average flattens out the worst fills, and the risk on a large order sits precisely in those fills, so averages are unsuited to this judgment.
 

What was MEXC's measured slippage in the third-party reports

 
Per TokenInsight's July 2026 report, sampling nine exchanges from June 15 to July 14, a $500,000 ETH futures sell measured 0.006% and a $1 million sell 0.015%, both the lowest in that sample. Per the May report, a $5 million sell measured 0.050%, ranked first. Per the September report, a $500,000 BTC spot sell measured 0.005%, the lowest in that sample.
 

Why is this result described as repeatable

 
Because the same basis returned the same value in two non-overlapping windows. A $1 million ETH futures sell measured 0.015% in the May report window of April 1 to May 12 and 0.015% again in the July report window of June 15 to July 14, during which the sample grew from eight venues to nine, leaving a difference of zero between the readings.
 

Different firms' liquidity reports disagree, so which should be trusted

 
Establish each one's band first, then decide which applies to your own order size. CoinGecko's spot study uses $3 either side of price for ETH, which at roughly $1,825 converts to 0.164%, making it 16.4 times wider than TokenInsight's 0.01% near-touch band. Near-touch depth determines fill quality for ordinary order sizes while wider bands reflect orders reached only in violent conditions, so the two do not contradict each other.
 

If the order size grows tenfold does slippage grow tenfold

 
No, cost grows faster than size. Per measured points within the same July report window, doubling from $500,000 to $1 million took the rate to 2.50 times and the dollar cost to 5.0 times. Per measured points within the same May report window, increasing from $1 million to $5 million took the rate to 3.33 times and the cost to 16.67 times. The two exponents come from different windows and cannot be joined into one curve.
 

Which contributes more to total cost, slippage or fees

 
At the million-dollar scale it is usually fees. A $1 million ETH futures sell carries $150 of slippage per the measured values cited here, while a one-sided fee on the same notional at six basis points comes to $600, four times the slippage. Both are charged on notional, and only their sum is the real cost.
 

Why do crypto exchanges not disclose execution quality the way US equities do

 
Because no equivalent regulatory requirement currently exists. The Securities and Exchange Commission adopted amendments in 2024 expanding the disclosure of order execution information and specifying the statistics to be reported, and Europe likewise has a regime requiring trading venues to publish execution-quality data periodically, while crypto markets lack an equivalent mandate, leaving readers dependent on third-party firms' own measurements, which is the root cause of the methodological differences between them.
 

What should be checked before placing an order

 
Three things. Which band produced any depth figure, since 0.01% and 0.10% differ by an order of magnitude; the slippage converted into dollars and added to the fee for a total cost; and an estimate built from your own size rather than the report's round figures, taking the two nearest measured points, extrapolating with the exponent between them, and leaving a margin.
 

Disclaimer

 
This article is for general information only and does not constitute investment, financial, legal or tax advice, nor any trading recommendation. Figures are current as of November 18, 2026, and the third-party firms' measured slippage and depth values are the results of their own calculations within specific sampling windows, with the firms differing in band definition, order size and sample composition, so rankings hold only within their respective samples and windows and do not represent actual execution results at any point in time, while the dollar conversions, percentile ratios and size exponents here are calculated from published data. Crypto assets and leveraged products are highly volatile and can produce losses exceeding the initial outlay, and actual execution prices and costs depend on market conditions at the time of the order. Readers should verify the original reports and judge independently, consulting a qualified professional where appropriate. MEXC operates in compliance with applicable laws and regulations and does not offer services to residents of restricted jurisdictions.
 

About the Author

 
MEXC Editorial Team provides the latest platform updates, product launches, trading feature guides, and industry insights. With a focus on developments in the global crypto market, it delivers transparent information, professional data, and forward-looking analysis. Both beginners and experienced traders can find firsthand updates, in-depth observations, and official announcements related to the MEXC ecosystem.
 

Research References

 
 
Want the fastest access to MEXC's latest updates? Join our official Telegram group now!
Join MEXC Community: X (Twitter) | Telegram | Discord
Account Verification: Understand KYC | How to Complete KYC
External Content Platforms: Substack | Medium | Paragraph | LinkedIn | X(News)
Market Opportunity
Orderly Network Logo
Orderly Network Price(ORDER)
$0.0283
$0.0283$0.0283
USD

The articles shared on this page are sourced from public platforms and are provided for reference only. They do not represent the position or views of MEXC. All rights belong to MEXC Editorial Team. If you believe any content infringes upon the rights of a third party, please contact [email protected] for prompt removal. MEXC does not guarantee the accuracy, completeness, or timeliness of any content and is not responsible for any actions taken based on the information provided. The content does not constitute financial, legal, or other professional advice, nor should it be interpreted as a recommendation or endorsement by MEXC. For expert insights and in-depth analysis, visit MEXC Learn.