Andrew Kang Profile picture
Sep 28, 2020 11 tweets 3 min read Read on X
1/ There seems to be a common misconception that AMMs won't work because LPs can be constantly arbitraged

Let's clear that up.
2/ First of all, yes LPs constantly get taken by arbitragers. There was rough analysis pre-DeFi explosion, that 70%+ of Uniswap volume came from arbitrage.

There is a lot more soft flow from directional traders these days, but significant volume still comes from arbitragers
3/ The is one false logical assessment that the misconception threads back to:

- Arbitragers make profit from taking from Uniswap LPs
- Uniswap LPs suffer IL from trades
- Trade volume predominantly come from arbitrageurs

-> Therefore, IL is value from arbitrage profit
4/ To understand why this is not true, consider a situation where the Uniswap ETH-USDT pool consists of only 1 LP.

T0 = $350
T1 = $400
T2= $350

Did arbitragers make a profit on the way up and down? Yes

Did ETH-USDT LPs suffer IL?
No IL + Fee revenue

It's really that simple.
5/ So where does arbitrage profit come from?

It comes from market participants (any traders, including MMs) that execute trades without utilizing Uniswap liquidity

E.g. Bid/Ask on CEXs move from 349.95/350.05 to 351.95/352.05. Uni @ $350

Arber takes from both Uni and CEX
6/ In this scenario, the MM got "arbitraged" but does that mean he lost money? No. In fact most MMs are "arbitraged" all the time from bids/asks on exchanges they aren't plugged into

You can't really even differentiate between these arbitrage trades and other types of flow
7/ The most efficient MMs are plugged into as many exchanges as possible, but there are many profitable MMs that only trade on a subset of exchanges as well

As long as volumes are high with takers crossing the bid/ask spread, MMs can make a profit. The same applies for AMMs.
8/ So where does IL come from? It comes from "changing deltas". Essentially your net exposure to each asset changing as price moves on each trade. LPs sell winners on the way up, and buy losers on the way down.

Just read this to understand the math:
medium.com/@pintail?sourc…
9/ The most important graph is shown here. The point is that as long as you get hit with enough volume, you can still absorb IL and still be in profit.
10/ Some people might pushback and say well in real-world settings AMMs don't really work and they are only profitable because of liquidity mining rewards

See below the return profile (IL adjusted) of ETH-USDT over a 90 day period where ETH appreciated 60%+
11/ Some of this was basic AMM 101, but it's important to understand where IL and arbitrage profits come from as it's not immediately clear.

Some AMM teams have even built flawed AMM designs because of the misconception stated in tweet 3/ .

Hope this is helpful

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Andrew Kang

Andrew Kang Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @Rewkang

Dec 18, 2024
1/ Ethereum has a fragmentation issue. Liquidity spread across dozens of chains/rollups not composable with each other

Justin Drake proposed an advanced scaling solution in 2023 called Based Roll Ups

@puffer_unifi @puffer_finance are the first to execute on this vision
2/ “A rollup is said to be based, or L1-sequenced, when its sequencing is driven by the base L1. More concretely, a based rollup is one where the next L1 proposer may, in collaboration with L1 searchers and builders, permissionlessly include the next rollup block as part of the next L1 block.” – Justin Drake
3/ Based Sequencing allows for

- liveness and decentralization of the Ethereum network, ensuring reliability without relying on a single point of failure.
- no need to operate a dedicated sequencer.
- 100ms execution due to preconfirmations
- Economic alignment with the L1, creating new revenue opportunities for existing validators through non-extractive MEVImage
Read 11 tweets
Dec 9, 2024
$ARB has had a great rally but is still fundamentally undervalued

Trades at a fraction of Sui, Avax, Tron, etc but has them beat on volume and TVL by multiples (even excluding HL activity)

There is a large segment of allocators (institutions) that don't invest in memes and pay attention to these metrics

Stylus which allows devs to build using Rust/C++, gaming deployments/Animechain and ongoing interop research look to be the biggest drivers of growth going forwardImage
Also, Robinhood has partnered with Arbitrum for swaps so I don’t think a listing would be unlikely

coindesk.com/tech/2024/02/2…
The DeFi Llama data above doesn’t include HL stats

L2 beat has Arbitrum TVL at $20b and HL at 10% of that
Read 4 tweets
Nov 21, 2024
Upping my odds of an eventual $DOGE ETF to 90%

Moon soon Image
Ngl I wouldn’t be surprised if a Doge ETF got more net inflows during a similar one year time period than the ETH ETF

Have actually heard of substantial institutional interest in DOGE

In the bull market, it’s like Bitcoin, but with more upside
Larry Fink’s face when he thinks about all the fees he’s about to collect from his Doge ETF Image
Read 5 tweets
Oct 27, 2024
It’s counterintuitive, but the best tokenomic design for a project (and retail) is to not have investor lock ups and have as much tokens to be as circulating as possible on Day 1 (except team, treasury)

One year cliff and 3-4 year vests are a poor standard that came about from a misunderstanding of capital markets and lazy copy pasting from prior projects

In reality, long vests have little impact on investor contribution post TGE. Good investors will be supportive whether tokens are vesting or not. Opposite for passive investors

The standard needs to changeImage
I wrote about why low float high FDV was bad in 2021. Back then projects started to copy the Serum model of 1% circulating - I pushed that projects should have at very minimum 15-20% circulating on TGE. Now I believe even that is too low. The standard should be 65-75%+

We've given a lot of this advice to new founders, but its tough because you are fighting against bad tokenomics advice from lawyers that misinterpret securities law and other VCs that try to push the status quo

But talk to any past founder and most will tell you that vesting + low float designs are a mistake and result in major headaches down the lineImage
Read 4 tweets
Aug 11, 2024
No knowledge of anything actually happening but combination of the below leading me to bet that there’s some interesting developments upcoming for $SUI

1. Raoul pal shill thread while he sits on advisory board
2. Large OTC bids
3. Relatively strong holdership through big unlocks
4. Aggressive price action with no pullback
5. Big recent performance upgrade with Mysiceti potentially allowing for interesting new apps
Many people commenting that they are giving grants to people to shill. If true, this is bullish
Potential speculation into Korea blockchain week announcement
Read 5 tweets
Mar 13, 2024
2020-2021 bull market coin - Doge
2024-2025 bull market coin - Pepe

Elon more hesitant to post Doge now but he has no issue consistently posting Pepe

As long as the market reacts he doesn’t even need to directly mention the token

The pattern self reinforces

Say his name Image
$PEPE is extremely bullish from a Holderbase perspective since they went through a handful of serious FUD events to leave only Diamond hands.

Insider sniping FUD, Multisig dump, Pepe fork. The combined efforts of these events was the definition of bullish selling. Anyone left is just not really selling
Wow Image
Read 4 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(