DeFi Fund: Up, up and away – Colin Platt
Just a reminder that this is performance art and that nothing contained here should be construed as investment advice, nor an inducement to trade, lend, borrow any cryptocurrency or finance product. Transacting with these assets is risky and you are likely to get rekt. Do your own research and consult a financial advisor before transacting any financial product.
It’s a rainy February here. My new year’s resolution to get back in shape is slightly… underperforming, however, the DeFi Fund has more than made up for that.
It’s been an extremely interesting experiment so far, not only on the technical aspects but on the social side too. A number of people decided around the same time as me to do their own DeFi experiments, putting stablecoins into lending protocols to see how they fared. In addition, I received a number of solicitations to co invest into this DeFi Fund, I told them that they all missed the boat, suckas! A friend even called me a scammer (jokingly… well I hope), to call it a fund and not let outside investment in. Sorry, not sorry.
Anywho, now that I’ve danced around that, how are things going? Well, in fact extremely well. I closed up January just shy of $200… so close that I decided to wait until over the week to tally things up. And am currently sitting around $204.23. Not bad for an initial $130.87 investment just one month ago. Doing the math for you, that makes it about 56% return to date.
Right, but Colin, ETH is up too. Quit dancing around for making money when ETH made money.
Well funny you should say that, I too, am up against ETH. With my balance worth approximately 1.07 ETH after month one. I’ve also managed to outperform BTC by approximately 19.35%. Not bad…
Because we’re still BSing with stats, a couple of interesting ones.
Between 1 January, and 2 February, Ether has shown an annualised volatility of approximately 64.5%. My little DeFi Fund has been slightly lower, at roughly 53.1%. Why is this important? It shows that the returns which I managed, were not only greater (56% to ETH’s 46% return), but that my daily returns were more consistent. If I were a fund manager getting paid by risk adjusted returns, I’d be pleased with myself (not that I need an excuse)
In addition, I’ve made a lot of transactions, 360 since 1 January to be precise. Of course each transaction has a cost, and so far I’ve paid 0.116637 ETH, or about $21.78, which comes to about 10.66% of the latest value of the DeFi Fund. So the ETH miners too have benefited.
Just to mention again, all of this is completely transparent, so if you don’t believe me, go check the Ethereum blockchain.
In my last post, I showed the trades that I setup on Day 1, which were:
- Long SAI, lent @ 4.82%
- Borrowed DAI @ 2.91%, lent @ 4.57%, collateralised by long ETH
- Long WBTC, lent @ 6.6978%
- Short (borrowed) DAI @ 2.91%, Long (lent) SAI @ 4.82%, collateralised by long ETH
With the following holdings (on 2 January):
- ETH 0.606
- WBTC 0.00363
- DAI -25.128
- SAI 48.095
As of 2 February, my trades are:
- Long SAI, lent @ 7.6433%
- Long DAI, lent @ 8.04%
- Long WBTC, currently not lent
- Long ETH, lent @ 0.04%
- A tiny holding of USDC and DAI floating around in dYdX
With the following holdings (on 2 February):
- ETH 0.1573
- WBTC 0.00384
- DAI 37.7816
- SAI 100.2002
- USDC 0.0139
We get it Colin, you’re good. Absolutely, but not just that. Let’s discuss how I managed this.
So what did I do?
When I first set things up in early January, I noted that all the DeFi rates were variable. What I didn’t account for was just how variable. I first started to feel this in my DAI interest rate arbitrage (borrowing from dYdX, lending on Fulcrum). At first things were ok as the rate on dYdX dropped, while Fulcrum moved up slightly.
2019–01–03 @ 7:22 UTC
Later that day I noticed an interesting thing, an arbitrage opportunity on the price of the Fulcrum SAI interest rate token, and SAI. At first I thought that this might have been due to the mislabelling of the token in Uniswap. Naturally I traded it, netting a buck or two.
Yippie! It worked!
A few days later I noticed that it had happened again, so I did the same trade on Uniswap, this time netting $0.36
I also decided that it would be a good idea to play around with dYdX a bit more, and learn how their margin trading worked. The first thing that I was met with was an interesting disclaimer, about not carrying margin trades for more than 28 days. I always laugh to myself when DeFi platforms put these warnings on, especially when they don’t do KYC, so I could just be a refrigerator running margin trade on dYdX.
After I accepted that I wouldn’t do anything wrong, I then had to sign some kind of transaction. I’d not bothered to read much about what I was doing, so I just accepted and the order was out.
I instantly figured out that margin trades on dYdX only do market orders, you cannot do limit orders, so I was already in the hole a bit. Oh well.
A few hours later the rally continued, and I walked around with a few cents. Yay!
But those few cents cost dearly, with a $0.67 transaction fee associated with it on the way in.
And another $0.68 on the way out.
Ouch. This was not going to be sustainable on my little gas budget.
I reached out to the dYdX and learned that it was they, not I that footed the bill for the gas. Suckas! So I was back in the game.
I checked back in with my Uniswap trade and again took a few cents on my iSAI/SAI arbitrage!
And decided to do a limit long 5x on ETH on dYdX, because YOLO.
I forgot about it until the next day, when ETH had done a small run, and I had pocketed a cool 0.0224 ETH, about 8.33%
One day while I was doing actual work, I got an interesting notification, a strong shill in fact from my new best friend Marc Zeller
Aave had just come out, and the rates were all over the shop, but it was worth a go.
The Aave platform looked a bit different from Fulcrum, but looked like it was pretty similar. Like Fulcrum, Aave credits you with a tokenised balance that accrues interest, and can be traded like any other ERC20 token. Their tokens however stream the value of your balance in real time.
The promising high interest rates were there, so I closed my loan on Fulcrum and gave Aave a go with my 25.09 DAI.
Over the course of the week, Maker had increased the Dai Saving Rate, which first was reflected in Fulcrum and Aave, raising the rates on DAI. dYdX was slow to follow, but eventually did. I decided to swap my short DAI funding position out for USDC instead, as the rate was lower.
That Friday (10 January), and I had things to do over the weekend and didn’t want to leave a bunch of ETH floating around, or watch lending rates. So I decided to buy back some USDC and sell some ETH.
Upon returning on Monday, the rates over at Aave had dropped significantly, so I decided to move back to Fulcrum. I was quite surprised to see that the amount of gas needed to open and close Aave really ate into the interest I earned.
Over the next few days, I continued doing my iSAI/SAI arbitrage. It became particularly effective during periods where ETHUSD moved a lot, of which late January had a number of them. This brought my initial 48 SAI up to nearly 92
I also explored whether the same thing could work in Aave’s interest token (aDAI). Success!
By now I had burned through a lot of ETH in gas fees, and decided to sell off some of that DAI that I had earned to replenish my ETH bags in order to pay miners.
This left me in a pretty healthy place
I called it a week, and left things alone.
The following Monday I decided to try the aDAI/DAI game again, as it was looking promising
The trade went through Uniswap just fine, but oh no, it didn’t show up in Aave! Had I been scammed? Did someone list a fake Aave aDAI token?
I headed over to Aave’s Telegram group to see if I could find out more.
I checked Etherscan and everything looked correct, but still nothing. I decided to see if I could use Etherscan to try to write a transaction to burn the aDAI back into regular DAI.
And was met with a very unhelpful error.
While I was looking into exactly what I needed to do to force this thing through on a wallet, the Aave team came through and sorted out my issue. Apparently it was something on their frontend. But that was enough hassle that I decided to move it back to Fulcrum for the time being.
Several more ups and downs and I managed to work my iSAI/SAI arbitrage to build my SAI stack from 48 to more than 100. Happy times. And a similar strategy on aDAI/DAI brought in another ~15.5 DAI. Not horrible.
Bottom line!
The first month has gone well, extremely well by most measures. I found most of these platforms fairly easy to use, but they are hardly ready for mass adoption at this point. It’s not just the visible aspects of these platforms, which on the whole were mostly good, some of the somewhat hidden aspects had several gotchas. As an example, of the 360 transactions that I made, 11 failed or were reverted. Some of these were for “legitimate” reasons, such as price movements in Uniswap trades, but others ran out of gas because the platforms had not defaulted to offer enough padding. I confess that I am not an expert on Ethereum Gas, though have a better understanding than a great number of people that are being marketed to for using these platforms, but perhaps a simple solution is to just default to a slightly higher amount.
There is also a lot of debate in crypto about how efficient markets are. Clearly here I didn’t find that the case. A lot of money is left on the table. Did I just get lucky? Maybe. Will these opportunities disappear? To a certain extent that has started, with trades that I was making in early January with zero competition starting to get some competition. Maybe people reading this will copy what I was doing and beat me to it in the future. I don’t know!
This experiment is likely to be more time consuming, but also more interesting than I had initially envisaged.
While I’m still far from convinced that any of these products are ready for your average investor, and am concerned about some of the more esoteric risks related to stablecoins and DeFi, so far it’s still a fascinating experiment.
Until later!
Published at Mon, 03 Feb 2020 09:45:01 +0000
{flickr|100|campaign}
