I'm building a local AI Chief of Staff on my Mac M3 Ultra.

No cloud. No API calls.
100% private. 100% under my control.
Here’s what it will do. 🧵👇 📩 Email Management

Auto-respond to inbound emails
Carry on threaded conversations
Summarize inbox daily
Draft follow-ups and schedule meetings

This is the absolute best analysis of Bitcoin, Money Printing, Power Law, Bitcoin versus Gold ever put together.

@1basemoney is criminally under-followed. I will put together my TLDR in the comments.

https://twitter.com/1basemoney/status/1912065202419540090

First of all, Mathew looks at population growth versus money supply growth over the last 50 years. Money Supply is growing at 13% while population growth is growing at 1-2%.

The impact: since 1980, Money Supply is up 25x, while population is up 2x.

The way I see it, Gold has essentially failed as hard money, and it's being held by central banks as a vague defense mechanism to prop up their fiat currencies in the (likely) event of an eventual "run on the bank".

Bitcoin could play a much different "reserve role".

A 🧵 1/N First, let's be clear: Gold has not been used as actual "money" for "transactions" in over 300 years.

- Certainly in the time of the Medici's (starting 1252) the gold florin was actually the main unit of exchange for large transactions.
- this got copied and pasted into the Ducat (1254), the British Noble (1344), the French Ecu (1377), The Spanish Escudo (1535), The French Louis d'Or (1640), The British Guinea (1663), the US Gold Eagle (1795) and finally the French Napolean (1803).

But starting in the min 1800s, banknotes replaced actual Gold coins.

These banknotes were initially Bank of England Pounds, US State Chartered Dollar bills, US Greenbacks starting in 1860, and then Federal Reserved backed dollars starting in 1913.

For almost 200 years we have used "gold backed paper" and not actual gold as money. And it has failed repeatedly

- Greenbacks were issued in 1860 as legal tender with no official backing
- The US officialy went on the Gold Standard in 1900, but off it in 1933 with EO 6102. The International redeemabilityy ended in 1971.
- Britain when on it in 1816, off of it during the Napoleanic wars (1797-1821), off during WW1, and finally off in 1831.

Similar patterns exist for France, Germany etc. The problem is the idea of "paper backed by gold" sounds good, but the reality is, it always fails. 2/N

Here's the real hard truth of where we are in America: 1/N

1. Over the last 24 years, we've increased Goverment Spending as a percentage of GDP from 17% to 24%. We're getting fatter, more bureaucratic and useless. @elon is right We spend basically 7 hours a day now wasting time of the Internet. We have become vegetables. 2/N

The arb desk perspective on MSTR. Math, analysis and limitations of cloning.

A thread 1/N

On September 24, 2024 MSTR issued 875 Million .625% convertible senior notes, maturing on September 15, 2028.

At the time of issuance, the conversion price of $183.19 was approximately 19% above the estimated spot price of $154. For simplicity, let’s use 20% as the number.

MSTR’s actual volatility is around 80%.

A four year call option 20% out of the money is worth 58% at 80% vol.

Let’s assume for simplicity that all of the bond is purchased by an arb desk.

The delta is about 79%. So the arb desk will short 79% of the issue to stay delta neutral. If the arb desk does 875 Million of the issue, they will need to short 691 Million.

They can expect to make 58% of the issue price or $400 million. There will be friction, and volatility could drop, so let’s assume 350 million to be safe.

With slippage, and including the .625% coupon, they get to 42%

They of course have credit risk. In an extreme scenario, they might not be “money good”. But let’s just assume that they are OK with the credit.

They do need to come up with 875 Million. Their funding rate on that might be closer to 8%.

Over 4 years, that is 36% of the principal, compounded.

The difference of 42% to 36% is the arb desks net profit. Something like 6 points, or 52 million. So what is needed to pull this off? Can others implement the MSTR strategy? 2/N

For the arb desk to be interested, they need three things:

1. volatility
2. a liquid share to short.
3. enough BTC in the vehicle to comfortably carry the debt.

At the beginning of 2024, MSTR has 154K BTC, valued at 5.8 Billion. BTC has gone up 50% since then, and MSTR has tapped the convert market for 2 Billion in debt.

Regarding point (1), all clones will have it. This is inherent to the Bitcoin backed strategy.

Point (2) really requires size. These convert arb guys wont be interested under 1 Billion market cap. Not deep enough.

Roughly speaking, per point (3), a rate of 5x BTC to new convert debt per year is probably doable.

So if a new clone company was created with 1 Billion in fresh BTC capital, it might be able to tap the market with 250MM in convertible debt in 2025.

Fresh look at the Power Law Data
Date: Oct 6, 2024
Monthly Data.

Observation 1: We're on track to hit BTC 1 Million (trend) in 10 years (1.4 in Log since Genesis) Observation 2: Vol is estimated at 0.3. So 1 standard deviation is 100% from trend. 2 is 200% of trend.

So, depending on timing we could hit 1 million earlier if we are say at trend 250K and we are at 2 standard dev of this.

This could happen in June 2028 (4 years from now)

Money Printing
The Last 100 Years
A short presentation

1/N Central Banking In Place WorldWide by 1914

- Bank of England created in 1694
- Bank de France 1800
- State Bank of Russian Empire 1860
- German Reichsbank created in 1876
- Federal Reserve created in 1913

All major powers now have central banks, and the ability to print money and finance the first world war! 2/N

A little mathematical diversion on Poisson Densities,
as mentioned in the WhitePaper (p7)

@moneyordebt If you have independent, identically distributed events, such as the number of phone calls you get in an hour, or, as Satoshi was modeling, the number of blocks you can mine in 10 min * z, where z is the amount of confirmations a receiver has been given in advance of a planned double spend.

If you let lambda = z * q / p

Then the probability of getting exactly k events (blocks) is

Bitcoin 69K in November 2021 is very different from Bitcoin 69K in April 2024.

In November 2021, the Power Law "fair value" was 24.5K. We were trading at 2.8X that. In log terms, thats about 1.6 standard deviations! That's VERY expensive.

Today the power law model "fair value" is 64K. We are at 8% over fair value. Thats just 0.1 standard deviation. Basically, we are "on trend".

The biggest deviation overshoot was in March 2021 when we hit 58K. At that point the trendline was 18.2K, 3.2X. That's about 2 standard deviations!

Most people don't appreciate this. It's one of the big benefits of the power law tool of analysis! I would argue that with the ETF, we are now VERY cheap in fundamental terms.

Here's the best intuition behind power law; it's slowing exponential growth. A 🧵1/N

10 years ago (2014) Bitcoin price was growing at 200% a year.

5 years ago (2019) that growth rate slowed to 79%

Today it's 47%

In 5 years (2029), it will be a "mere" 34%

And by 2034 we will be down to 19%

I'll explain all these numbers, as well as intuition for number of addresses and hash rates next A power law in time simply states that

y(t) = a * t ^ b for some constants a and b

y could be bitcoin price, hash rate or number of addresses. @Giovann35084111 has shown that all of these are power laws 2/N

The power law is an extremely simple model that was replicated here by 20+ people.

It has a correlation of 96% and has held up over 6 orders of magnitude of price data for 15 years.

All known data fits within the 2 standard deviation band and 90% of the data fits within 1 standard deviation.

It may not work forever, but statistically, it's an exact fit.

And people are ignoring it because it is a "guestimate". Meanwhile, the entire concept of "Stock To Flow" does not work at all for Gold prices.

Over the last 20 years, we have moved up from 400 to 1800, went down in a 10 year bear market, and are now at all time high.

The stock to flow was more or less unchanged.

The emperor has no clothes.

Episode 45 Bitcoin Power Laws with Giovani.

This is very long deep dive into what I tend to think is a very strong predictive model for Bitcoin.

The uncut version is presented here. In the comments, I am going to dumb it down and chop it up piece by piece.

There is math involved, and I have a google sheets that you can follow. It's worth it. 1/N 2/N

The idea of power laws is to look at relationships where
y = a * x ^ n

You have all heard of parabolas -- which is n=2, or y=x^2

Giovani and I will make the case that Bitcoin is a roughly a power low in time, where n = 5.8 roughly

In other words, the price of bitcoin is roughly the time elapsed since genesis block to the power 6.