Overall Thesis

This is where stablecoins come into play. In short, if you have reason to believe that a stablecoin will depeg, then the odds are more in your favor compared to shorting other assets. If you’re wrong, the upside risk is very limited, as stablecoins rarely depeg above $1 and when they do, they usually don’t go very high above $1. If you’re right, you could theoretically end up catching a full depeg to ~$0. In this article I’ll be explaining this concept with a stablecoin called HUSD that recently depegged.

HUSD

HUSD is a stablecoin on Huobi's HECO chain that has remained relatively stable since it’s inception in November 2021. About 15 days ago Alameda sold 70k of the token which caused it to depeg quite a bit. It's recovered from it's $0.89 low, but is currently sitting around $0.97. This large sale came after Huobi made an announcement saying that they would be unlisting HUSD in the near future.

When I saw this tweet and checked the price of HUSD I understood that there may be an opportunity here. The price was no longer at $1 and was sitting around $0.97, which was still a great price to short at. As stated before, the upside risk is limited here. If I’m wrong and it repegs back to $1.00 I’m only losing 3% (assuming it stops at $1.00), and if it fully depegs, it could theoretically drop all the way to $0.

Setting up the Short Position

Once I had decided that this situation was within my risk/reward appetite, the next step was to setup a short position. This process varies from token to token, and is not always possible with more exotic tokens. Fortunately there was a money market called Channels Finance on HECO chain that allowed me to borrow HUSD. Once I found somewhere I could borrow HUSD the steps to setup a short were fairly simple:

1. Deposit some collateral (preferably another stablecoin to reduce liquidation risk)

2. Borrow HUSD against that collateral

3. Immediately sell the borrowed HUSD back to the collateral token

Being that this was my first time executing on this strategy I went with a relatively small amount of funds. To start, I deposited 360 USDT into Channels Finance and borrowed 250.447 HUSD against it.

Then I immediately sold that 250.447 HUSD for 241.52 USDT. With this, my short position was complete, and now I just had to wait for the price of HUSD to either decrease or increase. Broadly speaking, the way this will ideally work is that once the price depegs, I will be able to re-purchase HUSD at a cheaper price than what I borrowed it at. When I initially setup my short position, it would cost me ~241 USDT to re-purchase my borrowed (and swapped) 250 HUSD, but in the future hopefully the amount of USDT required will be lower.

The Depeggening

Fortunately for me the price of HUSD did in fact depeg quite a bit, and is currently sitting at $0.32. This is a ~67% decrease in price from where I initially opened my short position. While I didn’t enter this position with a large amount of funds, 67% is incredibly hard to come by in crypto markets these days so this is an amazing outcome, and serves as proof of the viability of this strategy.

With the price of HUSD now at ~$0.32, in order to repurchase my borrowed 250.89 HUSD I only need to pay 77.93 USDT. That’s a 164 USDT difference that I get to pocket. Right off the bat we can see how this can be easily scaled with larger amounts of funds. Had I done this with 1000 USDT, I could be walking away with an extra $670 in profit. That being said this doesn’t scale infinitely and there are risks that need to be considered.

Risks and Considerations

In the event of a complete depeg event most people holding HUSD will be selling it as soon as they realize what’s happening. This results in available liquidity on DEXs drying up pretty quickly. If I had done this with a lot more funds I could end up in a position where my short is technically profitable, but because there isn’t enough liquidity to sell my USDT back to HUSD, I may be stuck holding my USDT. That’s not the end of the world, but I need to repurchase HUSD to retrieve my collateral on Channels Finance.

Something else to keep in mind is that on the majority of money markets borrowers have to pay a variable rate while borrowing. Depending on the token you’re borrowing, these rates can get quite expensive. When I first setup this short position the rate that I was paying for borrowing HUSD was 2.37%. This isn’t crazy, but when I repay my loan to retrieve my deposited collateral, I will have to essentially pay a fee on top of the initial 250.447 HUSD that I borrowed. In my case if I were to close my position right now, I would have to repay 250.88 HUSD to retrieve my collateral. That’s only a 0.153% fee, but it’s something to keep in mind as it can eat into your profits.

Closing the Short Position

While I’m going to leave this short open a bit longer to see how it plays out, I’ll briefly explain how I’d exit this position and take my profit. It can be done in three steps:

1. Swap USDT back to HUSD. Since the HUSD price dropped, I should be left with some extra USDT

2. Repay HUSD debt on Channels Finance

3. Withdraw USDT collateral from Channels Finance

Once I’ve done that, if the price is still lower than what I opened my short at, I will be left with additional USDT.

Takeaways

Overall this was a successful experiment, but I do wish that I attempted this with more funds. Oh well, now I’m certain that this is a viable strategy, and I’ll be more aggressive the next time I come across a similar opportunity. Stablecoin depegs aren’t incredibly frequent, but it happens often enough.

As stated in the beginning, this is much better than traditional shorting since the upside risk is somewhat capped. Stablecoins rarely depeg above $1, and when they do, they oftentimes get arbitraged down immediately. This means that I can enter these kinds of positions with more confidence than usual.

One last thing I want to highlight is the power of decentralized money markets with strategies like this. Simply by giving you the ability to borrow tokens, you can use this functionality to short tokens. Decentralized money markets are much more willing to list more exotic and low market-cap tokens compared to centralized exchanged, and by extension short these tokens. With the markets how they are right now, this could result in some very profitable strategies.

If you’ve made it this far, thanks for your time and I hope I was able to share some new knowledge with you. I’ve really started liking the idea of shorting tokens on these markets so I will likely be on the look out for similar opportunities. If this article was interesting to you consider subscribing as I will writing more content like this in the future. If you have any questions or comments feel free to reach out on twitter @lorem___ (three underscores)

Here we can see the ForgottenRunesWizardsCult NFT collection on Sudoswap. It has a listed floor price of 1.75 ETH, a sell price of 1.520 ETH and an offer TVL of 7.83 ETH. The overall strategy for this arbitrage is to find NFTs on other marketplaces that can be bought for less than what the sell price is listed as on Sudoswap, and immediately sell the NFT into a Sudoswap pool.

From a broad point of view, this is how this kind of arbitrage would play out using the State of Mind NFT collection that is on Sudoswap as well as OpenSea:

Step 1: Buy State of Mind NFT for 0.0135 ETH on OpenSea

Step 2: Sell State of Mind NFT for 0.024 ETH on Sudoswap

This resulted in 0.01091 (~12.13 USD) ETH in profit. After transaction fees this wasn't profitable, but it's a proof of concept.

To find these opportunities I was manually sifting through all collections on Sudoswap and then comparing the floor/sell price against Opensea and Looksrare. This was working when Sudoswap first launched, but as more arbitragoooors got involved, it wasn't viable. Eventually I came across the NFT aggregator Genie.xyz which allowed me to view NFT prices across multiple different exchanges (OpenSea, LooksRare, X2Y2, etc.).

Here is an example of another arb I found which was worth ~$40 after fees:

Step 1: Buy PP NFT from Genie for 0.04 ETH

Step 2: Sell PP NFT to Sudoswap for 0.087 ETH

Automation

After manually searching for these and getting small profits, I decided that it was time to automate as much of this as I could. The first step was to create an off-chain bot that would check for arbitrage opportunities. I did this by using the unofficial Genie API. I got the endpoints by looking through the network calls on the site. On Genie.xyz, when you select a collection they make a request to their assets endpoint:

This endpoint returns a bunch of data about each NFT in the collection. This data includes the sell price in USD, owner, marketplace, and more. With this I could check would the cheapest NFT was. Once I had the cheapest NFT in a collection, I could then check the sell price on SudoSwap using a similar technique. At first I was using the unofficial Sudoswap API to get price data, but they have a convenient subgraph URL that you can query to get data on a specific collection. With all of this put together, I could let the bot run and it would constantly check prices between Sudoswap and Genie against multiple NFT collections.

With this out of the way I could start working on a smart contract that would allow me to execute these arbitrages atomically. This would eliminate the risk of me being caught holding the NFT while manually executing the arbitrage. This also opens the door for using flashloans, which would allow me to take advantage of arbitrages that I don't have enough capital to take advantage of.

Overall Strategy

1. Use Genie API + Sudoswap API to find NFTs that are on sale across various marketplaces, and filter the ones that can be sold on Sudoswap for more

2. Call smart contract (w/ the necessary parameters to buy/sell) that will:

3. Buy the NFT for X amount of ETH that is for sale on any of the markets provided by Genie. The funds are either provided by the caller or they are flashloaned from Euler finance

4. Approve NFT to be transferred into the Sudoswap pool

5. Sell NFT into Sudoswap pool for X+Y amount of ETH

6. Repay flashloan if funds were borrowed. (Euler has no borrowing fee)

7. Transfer profit to caller

Once I had this setup I attempted to execute on an arbitrage with less than $5 in profit. As soon as I submitted the transaction I was immediately front run by an MEV bot. The bot saw my transaction taking advantage of the arbitrage. They then used the execution data for my call to execute the arbitrage before me, causing my transaction to fail. I suspected that this may happen but I figured if the profit was low the miners wouldn't bother, but I quickly learned that they are willing to take any profit if possible. To get around this I had to make use of flashbots which would allow me to submit private transactions. By using flashbots RPC relayer I could send my transactions directly to miners instead of having them sit in the public mempool.

After running for about ~48 hours I had made about ~0.06 ETH ($93.45) in profit. Right off the bat I realized that this could turn into something much bigger. A few days later I managed to execute an arbitrage that netted me 2.70 ETH (~$3490) in profit. This was by far the most profit I've ever made in a single arbitrage. The NFT that I did this with was a ArtBlock NFT that cost 4.97 at the time.

Once my bot detected the arbitrage, it sent the necessary data to my smart-contact. First it borrowed 4.97 ETH from Euler (not shown in the screenshot) then purchased the NFT from OpenSea (Seaport 1.1). After the NFT was held by my contract, it immediately sold the NFT into a Sudoswap pool for 7.97 ETH. The 4.97 flashloan still needed to be repaid, but after that was repaid the contract was still left with 3 ETH. It then sent 0.3 ETH as a bribe to the miner to get my transaction to get submitted on chain quicker, and the remaining 2.70 was sent to my wallet. Seeing this happen was genuinely surreal.

After this first big win I managed to get a few more similar sized wins within the next few days. Below you can see where I was periodically recording my wallet balance as my bot continued to execute more arbitrages. Most of the profit came from 2 - 3 large arbitrages.

Conclusion

In total I made 7.38 ETH from this in just over a month. Unfortunately at this point the competition was starting to get very tough, and my bot was getting fewer and fewer arbitrages. There were a few bots that were getting every single one my bot attempted, and they managed to get their transactions submitted before mine every time. This meant that there was a lot of room for improvement, but I wasn't entirely sure what I needed to do. I tried making a few changes and optimizations to see if I could put myself back in the game but I had no luck. Eventually I decided that I was more than satisfied with the profit I'd made, so I called it quits. While I've been interested and searching for arbitrages for over a year now this was by far the most profitable strategy I've come up with, and quite frankly the strategy itself isn't that complex. Putting it together was a little tricky, but in total it took about two weeks to get up and running.

This is my first time writing a long-form article like this on a topic like this so I apologize if it's a bit messy and chaotic, but I hope that you enjoyed reading through this. I tend to experiment with arbitrages and MEV-esque strategies quite a bit, so I will continue to record my findings and post more articles here. If that’s something you’d be interested in reading about, feel free to subscribe by clicking the button below.

Thanks for reading :) If you learned something and would like to offer some amount of monetary support, I would appreciate if you collected this entry. To be honest I’m not 100% sure how this works yet, but I set the price to be ~$5, so it’s relatively cheap. No pressure though <3