Introduction

• The Merge was executed on September 15, 2022. This completed Ethereum's transition to proof-of-stake consensus, officially deprecating proof-of-work.
• The Merge was exciting, however, it also created uncertainties, which affected users behavior, especially around ETH token.
• ==In this dashboard, let’s examine the flow of ETH transferred to CEX to see how the merge effect on ETH transfers to CEX.==

Note

• Analysis time: Jul 1 - Sep 14, 2022 with focus on 3 days right before the merge from Sep 12 to Sep 14, 2022.

Approach

I look at:

• Destination and source of ETH transferred to CEX.
• Break down to new/return transferor.
• Examine return transferor by:
  • Return day bucket,
  • Transfer frequency,
  • Age of ETH.

→ all to evaluate the merge effect on these transfers.

Structure

• Overview
  • Jul 1 - Sep 14
  • Sep 12 - Sep 14
• Unknown transferor from Sep 12 to Sep 14, 2022
  • Where was ETH transferred to?
  • Source of ETH transferred to CEX?
  • Who were these transferors?
• Return transferor
  • Return to the period after how long?
  • Transfer frequency before the period?
  • Age of ETH?
• Key highlights
• Conclusion

1. Overview

Jul 1 - Sep 14

• ==27% in total transfer volume was to CEX.==
• ==3 days before the merge from Sep 12 to Sep 14, transfer volume to CEX surged 3x, 7x, & 13x vs Sep 11 → the merge really affected the ETH transfer to CEX.==

What drive the surge in these 3 days? Let’s find out next.

Sep 12 - Sep 14

• ==From Sep 12 to Sep 14, 46% in total transfer volume to CEX was internal or between CEXs. 54% was from unknown transferor.==

Next, let’s examine unknown transferors who transfer ETH to CEX from Sep 12 to Sep 14.

2. Unknown transferor from Sep 12 to Sep 14, 2022

From here to the end of this dashboard, if not specified:

• Transferor means unknown transferor in 3 days from Sep 12 to Sep 14.
• The period means 3 days from Sep 12 to Sep 14.
• All transfers are ETH transfers to CEXs.

Where was ETH transferred to?

• ==200K transferors transferred 3M ETH in 218K transfers.==
• ==Top 3 CEXs account for 80% of total volume: Binance, Bitfinex, & FTX.==

Who were these transferors?

New is transferor who transferred the 1st time in the period, otherwise is return transferor.

• ==37% of these transferors transferred the 1st time to CEX → highly likely triggered by the merge.==
• 74K new transferors transferred 871K ETH in 77K transfers.
• 125K return transferors transferred 2.2M ETH in 141K transfers.

Transfer frequency before the period?

0 day means transferor only transfer once before the period.

• ==41% of return transferors return after their 1st transfer → these transferors were very likely triggered by the merge.==
• 24% of return transferors transfer in every period from 8 to 30 days.
• 15% of return transferors transfer in every period from 30 to 90 days.
• 14% of return transferors transfer in every period from 2 to 7 days.
• 4% of return transferors transfer in every period within a day.
• 2% of return transferors transfer in every period > 90 days.

3. Return transferor

Return to the period after how long?

• 5K/125K returned after 1 day.
• 15K/125K returned after 2-7 days.
• 15K/125K returned after 8-30 days.
• 15K/125K returned after 30-90 days.
• ==76K/125K returned after > 90 days. → these transferors were very likely triggered by the merge.==

Age of ETH?

Age of ETH is number of day ETH stayed in a transferor address, which calculated by the difference between Sep 12, 2022 and the day of newest ETH inflow which have cumulative ETH from then till the last ETH inflow is greater than total ETH transferred to CEX by the transferor. Assuming the last ETH in is the first ETH out, & not yet account for ETH out before the period.

• 1.1M/2.2M ETH have age > 90 days.
• 247K/2.2M ETH have age from 30-90 days.
• 331K/2.2M ETH have age from 8-30 days.
• ==242K/2.2M ETH have age from 2-7 days. → likely late ETH transfers to CEX after withdrawing from other sources.==
• ==9K/2.2M ETH have age within 1 day. → likely late ETH transfers to CEX after withdrawing from other sources.==

• ==The largest age of ETH bucket by transferor is== > 90 days ==bucket.==

Source of ETH transferred to CEX?

These were source of the newest ETH inflow which have cumulative ETH from then till the last ETH inflow is greater than total ETH transferred to CEX by the transferors. Assuming the last ETH in is the first ETH out, & not yet account for ETH out before the period.

• 77% of ETH from unknown source.
• 18% of ETH from CEX.
• 3% of ETH from Defi protocols.

…

Key highlights

Overview

• 27% in total transfer volume was to CEX.

• 3 days before the merge from Sep 12 to Sep 14, transfer volume to CEX surged 3x, 7x, & 13x vs Sep 11 → the merge really affected the ETH transfer to CEX.

• From Sep 12 to Sep 14, 46% in total transfer volume to CEX was internal or between CEXs. 54% was from unknown transferor.

Unknown transferor from Sep 12 to Sep 14, 2022

• 200K transferors transferred 3M ETH in 218K transfers.

• Top 3 CEXs account for 80% of total volume: Binance, Bitfinex, & FTX.

• 37% of these transferors transferred the 1st time to CEX → highly likely triggered by the merge.

Return transferor

• 76K/125K returned after > 90 days. → these transferors were very likely triggered by the merge.

• 41% of return transferors return after their 1st transfer → these transferors were very likely triggered by the merge.

• 242K/2.2M ETH have age from 2-7 days. → likely late ETH transfers to CEX after withdrawing from other sources.

• 9K/2.2M ETH have age within 1 day. → likely late ETH transfers to CEX after withdrawing from other sources.

• The largest age of ETH bucket by transferor is > 90 days bucket.

Conclusion

The merge did affect how Ethereum users use their ETH, especially with transferring ETH to CEX. In this dashboard, we see that 3 days from Sep 12 to Sep 14, right before the merge, 200K users transferred 3M ETH in 218K transfers, each day did 3x, 7x, & 13x vs Sep 11 in volume. Most of ETH transferred to Binance, Bitfinex, & FTX. And we see evidence of late transfers from late withdrawing, & idle ETH stayed in wallet, or user return to transferring to CEX after quite a long time. All of these were likely triggered by the merge.

Thanks for reading!