The end of an (te)era

Satria Pamudji
10 min readJun 17, 2022


We’ll rebuild I guess, maybe. Maybe not. But here’s a dedication to the crazy ones who thought we could change the world.


What was once a thriving community — full of builders, developers, and users who were bullish on the narrative of decentralized money — has now been reduced to ruins ever since UST’s “great depegging”, causing LUNA to start minting exponentially in hopes of restoring UST to peg.

LUNA’s historical data from

In a short period of just one week, LUNA’s trading value dropped from an opening of $77.56 to $0.00011553, which is essentially $0 & a loss of about 21B in market cap. Now, how many times have you seen that?

Personally, it crushed me because I saw a drawdown of 90% in my portfolio happen right in front of my eyes, which I could not save despite my best efforts.

But of course, I’m not looking for sympathy as I have since dealt with it healthily — I started going for therapy (If you’ve watched Billions and are in the crypto space, you need a Wendy Rhoades) and got myself a personal trainer to keep me in check (xxx, call me maybe?).

This article merely serves as a fragment of my memory of the past year and what I have since learned from it: What really happened to LUNA, to revisit my thesis, and some lessons to keep in mind moving forward.

The Context and Thesis for UST

To fully grasp what really happened, we need to re-learn the mechanisms of UST (Terra’s USD) and the co-relation between UST & LUNA.


First, let’s touch on this topic of stablecoins. In crypto, there’s a category of tokens called “Stablecoins”… and as you can derive from the name itself, it’s a stable token that’s used for transacting.

Although Bitcoin in itself was created for peer-to-peer payments with low time and monetary costs, the problem with that is Bitcoin — when compared against fiat currencies — is volatile.

So how do we capitalize on the advantages of the blockchain, like sending/receiving value cross-border with quick finality and cheap fees, while being able to denominate and pay for everything with something stable?

Well here’s your solution: Stablecoins. If you’ve ever heard of USDT, USDC, DAI, MIM or even UST, they are all categorized as stablecoins with various ways of maintaining stability.

We talk about this more below. If you’re familiar with the concept of stablecoins, you can skip this part and move down to the thesis.

Fiat-pegged stablecoins: Fiat-pegged stablecoins, such as USDT and USDC are stablecoins that are backed 1:1 with fiat assets/currencies like USD, EUR, or GBP that are held by a counterparty. So in the case of Tether, if you have 10 USDT, you have to be able to redeem it anytime for $10 USD through exchanges or via the counterparty itself.

However, there are some problems that exist with this type of stablecoins:

  1. The counterparties have to be willing to hold billions of dollars with not much identifiable upside (ie. They don’t get any transaction fees for transactions that happens on the blockchain itself). Losing out on a % of fees for the counterparties in the billions might be manageable, but it gets harder to scale with demand.
  2. You can’t really trust that the counterparty really holds the full amount of collateral they say they do. Sure, there are audits, but after all, money makes the world go round, right?
  3. Since it’s all centralized, everyone is at the mercy of governments as well. I mean, if you go to the club today and piss off a high-ranking official, they can literally freeze your funds. You get what I mean.

Over-collateralized Crypto-Backed Stables: The next baseline for stablecoins such as DAI, works by collateralizing crypto assets to maintain the stable peg to USD.

Because cryptocurrencies are so volatile, the collateralization ratio has to be higher than 1 (ie. For 1 DAI in circulation, there needs to be $2 worth of crypto assets), thus the name over-collateralized crypto-backed stables.

There are a few caveats with this even though this tips the scale towards decentralization (it’s not fully decentralized since it is partly backed by USDC)

  1. It’s not capital efficient because of the over-collateralization.
  2. In terms of scalability, it gets harder to scale as more capital gets locked up and drained in order to mint DAI.

And this brings us to our next iteration, we now know that we know fiat-backed stablecoins are centralized, while over-collateralized stablecoins are hard to scale and not capital efficient.

Algorithmic Stablecoins: Algorithmic stablecoins (ie. UST or the famous IRON), does not depend on hard collaterals to back the stable coin. It’s actually a complete confidence game for algorithmic stablecoins, with the main safety mechanism being human behavior.

Why is it a confidence game? Let me explain.

LUNA is the governance token for Terra, and UST is the stablecoin that is pegged to $1, with the asset that softly backs UST being LUNA.

Think of this relationship between LUNA and UST as balancing a see-saw on both sides.

In the case that UST goes to $1.01, the protocol allows anyone to swap LUNA to UST at a price of $1. LUNA is then burnt out of circulation and UST minted into circulation, thus diluting the UST supply and bringing it back to peg.

In the case that UST goes to $0.99, the protocol allows anyone to swap UST to LUNA at a price of $1. LUNA is minted into circulation and UST burnt out of circulation, thus decreasing the UST supply and bringing it back to peg.

Thus the caveat for algorithmic stablecoins is this:

  1. There must be trust in the system for people to want to arbitrage and bring the stablecoin (In this case, UST) back to peg every single time it goes off-peg. This means that you need to increase demand to a point that the stablecoin is being used for day-to-day purposes, thus making it “too big to fail” (*coughs* U *coughs* S *coughs* D *coughs*).
  2. External safety mechanisms need to be fleshed out in full to account for the fact that humans and the markets can remain irrational for a long period of time. Relying on arbitrage by humans is by far the worst mistake any algorithmic stablecoin founder/creator can do.

The Thesis

To me, my thesis was pretty simple: Decentralized Finance (DeFi) needs Decentralized Money.

  1. It seemed pointless that people were using stablecoins such as USDT or USDC for DeFi when your funds can be frozen and controlled by a third-party entity at any single point. It doesn’t sound very decentralized, does it?
  2. To add to my thesis, Terra had a focus on creating demand for people to not only hold UST for the yields but to spend it too. This was apparent in the growth in numbers of dApps with multiple use cases for UST and developers that were motivated to grow the demand for UST and the ecosystem as a whole.

The last point held more weight for me, as previous algorithmic stablecoins such as IRON (and more) were only focused on creating demand through yields and the only safety mechanism was the behavior of humans to maintain the peg.

Thus, while capital flowed in quickly, this also meant that the moment the yield dried up, capital flowed out as quickly. And because there were no incentives to hold IRON apart from the yield, no one wanted to arb IRON back to peg.

But of course, as we know it, before the scale of demand for UST was fleshed out in full, Terra collapsed.

The great depegging of UST: What went wrong?

To be frank, there are a plethora of articles out there that explain what went wrong with Terra. So I’ll keep this section short while giving context and links for you to read up more.

TLDR; There were a lot of factors that came into play, but essentially Terra had: (1) Way too many attack vectors (by going cross-chain way too fast, making too many enemies along the way, neglecting the sustainability of its core dApps such as Anchor… and much more) with (2) Defense mechanisms that could barely hold off the attack on the peg (LFG only having 3/10B of BTC to soften the impact on depeg & not having the swap mechanism up, White Whale only being able to do on-chain arbitrage).

Below I will link some interesting threads/articles for you to read up on if you’re a more technical person/want to explore deeper. If not, you can head on to the learning points below.

@0xavarek on why Curve’s 3pool is the main factor behind Terra’s Collapse: Twitter Thread
Jump’s Report on the observations made during UST’s Depeg: Jump’s Report
Following the On-Chain trail by Nansen: Nansen’s take on UST’s depeg

Learning Points

Now after a month of processing and taking time off, I wondered what I can do better in the next run. This section will cover everything I’ve learned so far, in hopes that you as a reader won’t make the same mistake because it’s (1) costly and (2) really painful to process.

Have a thesis, but also have an invalidation thesis that’s quantified

As an investor, my timeframe and thesis are always created looking forward. Essentially, my timeframe for buying/selling/trading is never dictated in terms of months, but rather in years (ie. 3 to 5 years).

Why? It’s very simple. Web3 is still a very nascent space and the tech stack isn’t fully out yet. Experimentation and exploration are all part of innovation. I expect failure, but I also want to help to lower the chances of failure, instead of worrying about the micro, which is something we cannot control.

However, where I went wrong this time was that I did not quantify my thesis and I did not have an invalidation thesis. I did not consider my exit strategy in the short term, because I was too focused on building and making connections in the ecosystem to prevent it from collapsing. And I definitely did not consider the metrics I needed to look out for that would trigger that exit strategy.

Basically, I was sailing a boat without considering what happens in the event of a shark attack.

I was unprepared.

Leverage is your friend, but it can also be your worst enemy

A few months before Terra collapsed, I had a really brilliant idea, if it worked out. Honestly, a little bit too degenerate and it was fun while it lasted.

The end goal: To fully optimize all of my liquid assets, and get myself perpetual passive income, and this was how it worked.

  1. Convert all of my liquid assets to staked liquid assets (ie. LUNA to bLUNA, AVAX to wasAVAX).
  2. Start building my own personal treasury using all of the said assets by depositing them into Anchor.
  3. Use treasury to take out a loan from Anchor. For easier reference, let’s consider that I have 100k worth of bLUNA and wasAVAX, and I borrowed 30k worth of UST (to account for asset drawdowns).
  4. Use borrowed UST to purchase HUNT NFTs, which would automatically buy Crabada NFTs and auto farm TUS for you.
  5. Get perpetual income. 30k worth of UST would take about 120 days to recoup, given a stable TUS price.

This play caused me to lose extra assets apart from just LUNA itself as I faced 2 issues, (1) I could not repay back my loan as I had about 70 more days before I could fully repay back my loan. If the loan were to have been repaid, I would still have my wasAVAX and bATOM. (2) TUS inflated faster than the sinks could capture demand, and because of that TUS had more than a 10x drawdown.

Meaning for every $100 I invested, the $1 I was supposed to get back a day soon became $0.1, and it goes lower every day. Double whammy by leveraging into an unsustainable P2E game.

Given a choice, if I could choose to play this again, given the past market conditions, would I?

The answer is that I would, but I would not have taken as much loan as I did. Remember, leverage is your friend until it is not.

You’re not special

At the end of the day, being in crypto doesn’t mean that you’re special. From a top-level view, you are still affected by the macro environment, which I paid close to zero attention to.

A few months ago, I wouldn’t have understood what the difference between US10Y-US2Y was supposed to represent. VIXY? What the f*ck is that, I only care about Bitcorns and LUNA right?

And to put off criticisms which most LUNAtics did, that was valid but dismissed as FUD?

I felt that had a part to play in it. But I blame nobody but myself, for finding my way into an echo chamber without looking at things from a holistic point of view.

I’ve learned, and I am looking at risk-off assets differently now.

Next Steps

After taking some time off, I realized these were the main learning points for me. The major ones at least, with much more nuances that I’m not going to point out in writing because I could end up writing a book.

I will be taking the next few months to recoup my losses and larp on Twitter to find the next 100x. In tandem with that, I will still be writing for Flagship and deep-diving into Cosmos so you can expect many more articles in the next few weeks and months.

But hey, the blockchain space is still very nascent like I said, and there’s much more to be done and created. I still believe that one day there might be a capital-efficient stablecoin that works and is decentralized end-to-end. But that’s for the future.

For now, we pick the pieces up, and don’t let your mistakes get into your head. We press the green buttons again so that we can make it all back.

See you in my next few articles, it’ll be fun!

“Accept what is, let go of what was, and have faith in what will be.”



Satria Pamudji

I help you understand technical concepts by writing easy-to-digest articles