—

A death cross occurs when the 50d moving average intersects withe the 200d moving average. This is a significant technical analysis signal and indicates a bearish trend has begun.

If we look at the bitcoin chart, its pretty clear that we are going to see a death-cross in the occur next week or so. If we estimate, the death cross should occur at 87k (where the moving averages intersect).

In the history of bitcoin’s price action, there have been 10 different death crosses from 2013 onwards. This will be a short analysis on how bitcoin price action performed pre/post death cross. Hopefully we can use this as a basis for making some profitable decisions.

You can see that as we approach the death cross, there is a grind down in bitcoin price - we’ve seen the same thing in this instance. Bitcoin drew down from > 100k+ to the low 80k area.

What’s more interesting is how it performs after. From the chart and on average:

• bitcoin tends to bottom about 7 days after at -4.82% from the death cross closing price

• it makes a double bottom / higher low, and eventually goes up in a straight line starting around day 30

• it peaks at +12.74% on day 51 (within the +- 60 day time frame)

What’s also interesting is the relationship between the moving averages. We can we see that around day 30, bitcoin reclaims the 200 day moving average, and it peaks right below the 50 day moving average as well.

If we take a closer look at each of the instances:

• only one of the resulted in a negative return relative to the closing price when the death cross formed

• on average, the peak return ended up being +30% since the death cross, with over half of them being > 34%

• Additionally, it took 36 days (on average) for the peak post death cross to play out

• 3 of the instances had the peaks be within the first day after the death cross — this likely means price went down in a ~straight~ line afterwards

The numbers actually lines up well with how bitcoin price action performs after a -25% decline (see chart below)

You can read more about the bitcoin price action post 25% decline here:

—

NFA, DYOR

We’ve most recently had a > 25% decline in bitcoin price from peak to trough. This inspired me to take a closer look at how prior 25% declines have resulted before.

We’ve had 50 unique 25% (or greater) declines since 2013 on Bitcoin. On average, this happens about 4x a year, although it has only occurred once in 2024 and once in 2025.

On average, it takes about 25 days for a 25% (or greater) decline to play out. However historically, we’ve had these violent declines play out within 1 day.

Now, the real question is - after a 25% decline, how has price behaved afterwards? Does price tend to go up or down?

On average, price tends to rebound. Across all of our data points, price tends to bounce +26.7% over 24 days after a > 25% drop.

From our most recent > 25% drop, we peaked at +10% on Mar 26, 16 days after the bottom. We are now just hovering just above flat in terms of price.

Now let’s take a look at where bitcoin price peaks post 25% decline. In other words, this is the highest point that bitcoin bounces after a violent 25% (or higher) decline. We’ll vary the post decline window (60, 90, 120 days) and see how this affects the outcome.

Within 60 days, the median is +18.31%. This means more than half of the data points (n=40 here) result in bounces that are greater or equal to +18.31%. Only 5 of the data points were negative.

Within 90 days, the median is +26.26%. This means more than half of the data points (n=39 here) result in bounces that are greater or equal to +26.26%. Only 4 (10.3%) of the data points were negative.

Within 120 days, the median is +28.27%. This means more than half of the data points (n=38 here) result in bounces that are greater or equal to 28.27%. Only 4 (10.5%) of the data points were negative.

Note: the sample size slightly diminishes as we extend the window since this creates overlaps. E.g. you can have a > 25% decline end on 1/01/20 and another end on 03/01/20, but if they both peak on 06/01/20, we only include one data point. As the window gets larger (e.g. 60 days → 120 days) there are more of these overlaps.

—

There are a million factors to consider, but if you remove everything outside of historical data, we are due for a larger bounce.

NFA DYOR

Will assume reader has familiarity with MicroStrategy (check out the prior write-ups if not!)

Since Oct 11, 2024 (when I published [1]), MSTR had gone up nearly 3x ($188 to $540) before coming down and settling in the low $300 range.

Saylor has done a lot since, changing the company name to Strategy, acquiring over 506,000 BTC, and issuing out two other preferred shares, STRK and STRF.

I’ve seen a lot of inaccurate takes with respect to how Strategy works, and wanted to boil down Saylor’s approach so folks can make a more accurate judgment on whether his financial alchemy is (or isn’t sustainable).

Very simply - Saylor’s approach is two fold:

• Sell an option to purchase/convert MSTR shares where the strike (price to convert) ranges from +30% to +300% of the current price. Structure the sale to receive all the proceeds up front

  • This is done via convertible bonds and the STRK/STRF offering

• Print more shares of MSTR when opportune/appropriate

  • This is done via the ATM offering

What isn’t explicitly stated is that Saylor has no intention of every paying back the principle / proceeds. His intention is to pay the interest expense for as long as it takes for MSTR stock to appreciate up until the relevant strike.

The beauty of this strategy is that Saylor effectively has created a ton of optionality that allows him to raise capital opportunistically:

• Implied volatility high? Issue convertible bonds

• Premium on Nav high? Sell MSTR shares via ATM offering

• Neither? Issue out more preferred stock (STRK/STRF)

But what if Saylor blows up? Isn’t he taking on a lot of leverage?

If we just look at the facts:

• MSTR market cap is 85B and holds 506k BTC worth 44B

• It has 8.2B in debt with a .421% annual coupon (34M in interest expenses)

• 58.4M for STRK and 85M for STRF in annual dividend payouts

If you play with the ratios - looking at debt / pref (pref is referring to the total marketcap of the preferred shares STRK/STRF) / interest expense relative to MSTR market cap and its net asset values, its very clear that Saylor has not taken on much leverage at all.

Another lens to view MSTR’s bitcoin treasury strategy is to draw a comparison to the traditional banks. Banks effectively:

• Take deposits

• Buy “safe” government backed debt (mortgage backed securities, t-bills, etc)

• Pay you less than they get paid from the debt they purchase

To further juice returns, banks will lever up (fractional reserves). Occasionally, they fail.

The bitcoin treasury approach effectively:

• Raise capital (take deposits)

• Buy Bitcoin (instead of government debt)

• Pay you in some form of MSTR stock and a routine coupon/dividend

However, the treasury company won’t (shouldn’t!) take as much leverage given the government won’t step in in the case where things go wrong.

In both cases, depositors make $ - banks pass through some fraction of the debt interest, the BTC treasury strategy passes through volatility (check this out on how this trade works exactly and why it is profitable for the convertible bond buyer).

The banks make $ from the net interest margin (difference between what the purchased debt pays and what they pass through) while the bitcoin treasury company makes $ from capital appreciation (or devaluation of the dollar relative to bitcoin).

The sustainability of this strategy really comes down to whether bitcoin continues to appreciate relative to fiat currencies.

At this point, Saylor is one of the biggest advocates and has been able to convert a number of different companies into taking a similar approach.

The price appreciation on MSTR and BTC have been the greatest selling point, and we’ve crossed the chasm to where the US government has adopted some form of a strategy reserve as well.

At this point, I think a very small percent of folks can understand and appreciate the approach that Saylor and Strategy has taken. As BTC and MSTR continues to do its thing, you will get more folks trying to understand the nuances around the btc treasury approach which just adds to the reflexivity and the sustainability of this.

Until next time

[1]

[2]

So what’s with the underperformance of crypto equities?

I’m going to illustrate the general dynamics on how the options market effects the underlying stock price

A big factor is the monthly options expiry (mopex) coming up this Friday on Nov 15 -- the dynamic is specifically:

• the expectations on price movement were too high

• market makers unwinding their hedges

I’m gonna use Coinbase as a concrete example here, but these factors pretty much apply with most of the crypto equities.

I’ve mentioned before that you need to look at implied volatility before aping calls precisely post-Trump win given this dynamic I am going to describe.

These are the implied vol values on the Nov 15, 22, 29 expiries for Coinbase.

The 132% means that the 2 standard deviation (95% probability) expected move is + or - 16.62% per day in the underlying. These expectations are very very high

The implied vol percentile over 30 days is 80%. This means that this implied vol (or expectation of the upside on the Coinbase over the next 30 days) is higher than 80% of the time. This is ridiculously high considering this doesn’t overlap with a quarterly earnings call.

So what happens when the expectations are so high? Also what are the market makers doing here?

When MMs sell you out the money calls, they will hedge their short deltas by buying shares.

As you get closer to expiries, the delta value for those out the money calls will decrease and as a result, these MMs will sell their shares to ensure they have no directional exposure

As we approach Nov 15, the OTM calls lose deltas faster and faster (all things equal)

If we look at options OI, it skews heavily towards out of the money calls (barely any puts visible in the chart).

As we get closer to the expiry date, there is an acceleration in the directional move due to the reflexivity of this MM hedging dynamic. As a result, the MMs are selling shares more and more aggressively (all things equal or until an entity comes in and bids Coinbase more aggressively).

This is perpetuated as folks take profit as well (calls closed, MM dumps shares they bought to hedge). It’s one big reflexive loop

How did we end up in this spot?

Last week options positioning was quite bearish on Coinbase (look at historical put call ratio, and the 25 delta skews).

We had a Trump win that caused a large positioning unwind (resulting in precisely the opposite of what i described above) with MMs buying back shares as their short puts unwind.

The pendulum swung hard the other way, implied vol (price!) of these options spike higher - as this happens, you have higher delta values (and hence more aggressive hedging and the inevitable unwind) for OTM calls.

The expectation for price movement moved way too much in the positive direction as you had a mix of shorts covering, MMs unwinding their hedges, and general expectations of the “supercycle” playing out. This is visible in the implied volatility across all the crypto stocks

This is also why I opted to sell a lot of vol here - these expectations are rarely met and it is a great way to hedge spot .

There’s a lot of nervousness in markets leading up to this. We can see this with elevated VIX values, and bearish options positioning.

But historically, how have markets fared post election? Is the fear justified at all?

Here is the 1, 3, 7, 14, 21, and 30 day post election performance in SPY for every presidential election since 1980.

Including all years, the SP500 tend to chop around and ultimately resolving slightly higher 30 days post election. If you include 2000 (dot com bubble pop) and 2008 (great financial crisis), we end up consistently grinding higher post election. Also note, in the 2000 election was contested. In 2008, Lehman Brother’s filed for bankruptcy in Sept 15, 2008.

Let’s look at implied volatility measures — roughly, this is a gauge on how much the market expects the underlying to move.

Specifically:

• The VIX is a measure on the volatility over the next 30 days on SP500

• The MOVE index is effectively the bond market equivalent of the VIX

Barring 2000 and 2008, the VIX was lower 30 days post election across all election years. The market hates uncertainty and this makes it evidently clear — the desire to hedge positions diminishes regardless of who the winner is.

The relationship with the MOVE index is less clear.

So are those equity hedges justified?

We can answer this by looking at the implied volatility vs. realized volatility post election. A general rule of thumb is - if implied volatility is greater than realized volatility, it is better so sell (or not buy) insurance and vice versa.

Again, barring 2000 and 2008, the spread is always positive across 7, 14, and 30 days post election. What does this mean? It means hedges are generally overpriced - specifically, the implied volatility is higher than reality and the market moves less than expected.

And how much exactly are these hedges overpriced?

We can interpret this as:

• Across all election years, the price of hedges overestimates the expected move on SP500 by ~1.50%

• Removing 2000 and 2008, the price of hedges overestimates the expected move on SP500 by ~2.25%

And what about a contested election?

Maybe this time is different. Maybe we get an outsized move and hedges were the right move.

But the data really leans towards: Nothing ever happens

Read more

Read more

Read more

Since then, there’s been a lot of growth in the restaking ecosystem — other restaking infrastructure have come into the scene, as well as a number of liquid restaking providers.

Here are some updated thoughts on restaking / li…

Read more

Arthur reflects on his bearish risk assets call (bank term funding program). He is bearish again for the rest of April and goes through his reasoning

What happened to the bearish BTFP call?

• Even though BTFP was cancelled, the Fed was still printing $. How?

• The Fed allows banks to borrow against certain securities, namely treasuries and mortgage backed securities

• This mechanism is known as the “discount window” which allows banks to meet depositor withdrawals while holding these securities

• This basically replaced BTFP, but the issue is that it doesn’t account for losses

  • If banks bought a bond in 2021 for $100, but due to rates moving that bond is now worth $80, the $20 loss is still there

Bank Capital Requirements

• Banks effectively finance government debt at yields are less than the nominal GDP

• Why do this? Because regs are in place so that banks can purchase gov bonds with little to no collateral

• In the case where rates move adversely, the gov can implement a mechanism like BTFP or the discount window

• However, after 2008 GFC, regulators forced banks to hold more capital (rule called Basel III)

• This created a problem so banks can’t just hold unlimited amounts of US treasuries

• This rule went away during Mar 2020, came back after the covid dump but the subsequent set of events eventually led to the regional bank failures of Mar 2023

• As a result, stricter banking capital requirements were favored moving forward

• However, the US gov debt addiction hasn’t gone away – who is going to service their debt if the high capital requirements are still in place?

• It effectively came down to “if you, Bad Gurl Yellen, want us to buy dogshit government bonds, then we can only do it profitably with infinite leverage.”

• Hayes thinks relaxing capital reqs is inevitable for US treasuries (covert QE)

Bad Gurl Yellen

• Yellen was previously draining the reverse repo (RRP), effectively yeeting that $ into risk assets

• RRP is down to 400B which begs the question - where will the next bit of liquidity come from?

• Additionally, Apr 15 is when taxes are due - taxes remove liquidity for obvious reasons. The expectation is that 2023 taxes are on the higher side

• After May 1, the fed is expected to reduce the pace of quantitative tightening by 30B / month - this results in more liquidity within the system

• When the treasury general account (TGA) increases its balance, it removes liquidity, and when it decreases it adds liquidity to risk assets

• When tax payments are received, the TGA increases - TGA should blow up to 750B as tax payments are processed on Apr 15

• It is also an election year, so the base case is that risk assets should rip

• So the tricky period is Apr 15 to May 1 - when tax payments are removed, QT comes back, and before the TGA can be drained

Bitcoin Halving

• Scheduled for Apr 20 - very bullish in the med term

• Hayes is scared this is already well priced in and could dump in the short term “when most market participants agree the opposite usually happens”

• So he thinks, in the short term, prices slump after halving. He has taken profit on MEW, SOL, and NMT

• Halving also occurs during a time when dollar liquidity is tighter than usual (taxes)

• Hence, Hayes is bearish from now until May

We are in a bull market. Retail is trickling back, BTC is floating right above / around 2021 ATHs and there are so many bullish catalysts ahead of us. As someone that was around for all of 2023, it’s crazy to see a 7 in front of the Bitcoin price.

Read more

• Deposit incentives (regular acct and 401K)

• A new credit card

• A new UI / UX

Read more

I had a discussion about Grass that prompted me to dig a little deeper into the app. Here’s a TL;DR and some reflections on the overall concept:

Read more

Read more

Arthur thinks that Ethena will flip Tether as the largest stablecoin eventually.

He walks through why Tether has the best (but also the worst) business in crypto

Then provides a brief overview of Ethena, why he thinks it’ll flip Tether, its risks and a valuation model.

Here’s my summary:

If you want to try out Ethena and get those juicy yields, here's a ref link (helps me out too!)

Tether - the best and worst business in crypto

• Tether is a stablecoin (USD denominated)

  • Customers deposit USD, and received USDT. It also includes the redemption process (burn USDT, receive USD)

• The crux of Tether’s business is that it needs a Fed approved bank account

  • In order for Tether to move dollars around, especially across jurisdictions, it needs to get approval from the Fed

  • So if Tether doesn’t have a banking partner, it can’t function

• Tether acts as a fully reserved bank (meaning the dollar deposits it gets are not loaned out). What does Tether then do with these dollars then?

  • Tether shoves them into money market funds and treasury bills

  • At this point it has nearly 81B invested into MMF and treasuries

• Additionally, Tether pays 0% of the interest to customers

  • They have ABSOLUTELY printed over the last 18 months

Tether - too big to fail?

• It is now a liability in the system (too big to fail?) – if it was a forced seller of its US debt, it could wreck the treasury market

• So how would you off board Tether without knocking down the dominoes across the system?

  • Need to slowly sell off Tether’s US debt

  • Freeze Tether’s assets until the market is more capable of absorbing its assets

  

• Tether clears 62M per employee, order of magnitudes larger than any of the existing US banks

• US banks are certainly eyeing Tether’s business and also watching Circle’s IPO

• These stablecoin businesses don’t have a strong moat - especially since they have a dependency on the US banking system

  • What is stopping JP morgan from trying to issue out their own stablecoin?

• One other issue Tether has is an incentive/alignment one

  • Unclear whether banking partners have equity

  • They don’t have a gov token, so customers/users can’t share upside

  • Customers/users also don’t get to share in their interest income

Ethena - Overview, Luna?, Risks, Insurance Fund, Valuation

Overview

• Ethena is a protocol that generates a synthetically backed fiat crypto dollar

  • It combines a spot staked ETH position + short ETH perp in equal size

• Tether proved out product market fit for stablecoins and tee’d up the perfect opportunity for Ethena

• Ethena addresses the 2 major problems Tether has

  • Holders of USDe can stake directly and earn the vast majority of interest

  • Incentives/alignment are set up perfectly

    • Major exchanges all invested in Ethena during early rounds (in aggregate they cover 90% of ETH OI)

    • A governance token will also exist so users can gain upside to Ethena

• If USDe has such high yield why wouldn’t it trade at a premium to other stables?

  • There is an arb here - if USDe’s premium is high enough, you can mint USDe by depositing staked ETH into Ethena, than swap it out for stables

    • This puts downward pressure on USDe

Comparisons to Luna

• UST was backed by Luna + some BTC

• USDe is backed by staked ETH + short perps on various exchanges (no relationship to Ethena’s gov token)

• Additionally ETH is deflationary (ultrasound money!) and USD is inflationary

  • So in the long run, funding on ETH future derivatives will generally be positive (prem to spot)

• UST and USDe are very very different

Risks and Limitations

• Exchange/counterparty risk

  • Ethena isn’t decentralized and it isn’t trying to be

  • If exchanges run into problems, there will be an issue

  • Ethena tries to mitigate against exchange risk by placing funds on 3rd party custodians such as copper

  • Compare this to Tether - their CP risk is with TradFi banks vs. Ethena’s CP risks with CEXs and custodians

  • Relative to Tether, Ethena, incentives are more aligned. Many of the largest CEXs are investors (CEX derivatives trading is also a highly profitable business, why would they want to endanger that?)

• ETH OI as a bottleneck (to flip Tether)

  • Total ETH OI is 8.5B on the exchanges that Ethena operates on (total ETH OI is 12B across all exchanges)

  • There is an additional 31B of OI on BTC contracts if/when they decide to onboard BTC

  • Arthur also hints at the potential of using staked SOL + short SOL perps

  • In aggregate, this is still smaller than Tether’s marketcap today. But if the total crypto marketcap goes to 10T…

• LSD smart contract risk

  • This is if Lido’s smart contract has an issue

• Prolonged net negative funding (staking yield + perp funding)

  

• Smart contract risk

  • Ethena’s deposit vault is fairly simple and only approved participants can interact with it

  • Most of the complexity comes from a system that lives offchain

Insurance Fund

• Ethena does have an insurance fund (currently 16M) that is there to bid USDe in case some of the risks actually play out

  • The fund is capitalized from Ethena’s fundraising + portion of yield generated by USDe that hasn’t been staked yet

  • In the long run, the insurance funds will be capitalized by a long term take rate on USDe yield

Valuing Ethena

• Arthur assumes, long run:

  • 20% of yield goes to Ethena treasury

  • 4% yield on ETH staking

  • 20% funding rate on perps

• If these are the assumptions for revenue, what is the right multiple for an FDV calc? Arthur compares this against other protocols

• Ethena’s $820 million of assets generated a 67% yield this week. Extrapolating that out for a year given a 50% sUSDe to USDe supply ratio, Ethena’s annualised revenue is ~$300 million. Using an Ondo like valuation leads to an FDV of $189 billion. Does this mean Ethena’s FDV will approach $200 billion at launch? No. But it does mean that the market will pay up big for future Ethena revenues. 

The opportunity that HOOD presents feels similar to COIN in Oct ‘23 and CLSK in Nov ‘23. The combination of good fundamentals and technicals makes me incredibly bullish.

The asymmetric upside in buying HOOD comes from a confluence of:

• “Hated coins pump the hardest”

• Wall St is oblivious

• Charts look great

Hated coins pump the hardest

HOOD is a stock that has been completely beat-up:

From early ‘22 up until recently, this stock has hit rock-bottom from a sentiment PoV — and its pretty clear why.

But you know what they say “hated stocks tend to pump the hardest”. Why? After forced selling and time based capitulation occur, any attention is good attention (some of that attention leads to buying!)

And to add, Robinhood has been hustling - they’ve been busy adding features that differentiate them from other brokerages (but who’s paying attention?):

Wall St is oblivious

Bitcoin just hit ATHs, dogcoins are pumping equally hard. HOOD has been hustling, setting integrations up so that they too can provide shovels during this upcoming gold rush.

But Wall St doesn’t see that at all.

Wall St is accounting for 5% user growth in Q1 of 2024 and a 15% increase in crypto related transaction revenue.

HOOD went from 493 to 43 in the app store from Jan 1 to Mar 5. Look at Coinbase metrics as a comp and account for the explosion in downloads – it is so clear that Wall St’s estimates are way out of touch. HOOD is doing everything they can to absorb the crypto audience:

Charts look great

I’ve wrote a lot of words and painted a narrative around HOOD - but the charts tell the entire story.

Shortly after the initial IPO pop, HOOD has been down only. A few hundred days of down only followed by 700 days ranging in the $8 - $13 range.

Translation: people detested the stock for a few hundred days, and then they didn’t care for the next 2 years.

But now we’ve broken out of the range now.

Combine this with the fact that we are just entering the up only stages of global liquidity (aka the gold rush where risk assets are up only)...

I’m long HOOD (have been adding since $8-9)..

Not financial advice, do your own research