"Why the Stock-to-Flow Bitcoin Valuation Model is Wrong"

Brendan Burgess

Brendan Burgess

Founder
Messages
53,618
www.coindesk.com

Why the Stock-to-Flow Bitcoin Valuation Model Is Wrong

The popular stock-to-flow bitcoin valuation model has the air of academic rigor. Unfortunately, it's just math-laden marketing.
www.coindesk.com www.coindesk.com

The stock-to-flow model (SF), popularized by a pseudonymous Dutch institutional investor who operates under the Twitter account “PlanB,” has been widely praised and is the leading valuation model for bitcoin proponents. SF has achieved viral popularity and inspired rags-to-riches dreams for those gambling it all on the future of bitcoin. However, we believe the model’s accuracy will likely be about as successful at forecasting bitcoin’s future price as the astrological models of the past were at predicting financial outcomes.

Over to you tecate and Duke...
 
'All Models are Wrong but Some are Useful'

Brendan Burgess said:
Over to you tecate and Duke...
Click to expand...
Thank you Brendan. :D

So I referred to it briefly in recent threads and back then I said that it was a supply side model. So it accounts for half the picture. Furthermore, be aware that stock-to-flow modelling has not come from the bitcoin sector. Before bitcoin ever existed, quants were (and still are) applying it to try to determine the price of gold.

All PlanB did (a quantitative analyst himself) was adjust the model and apply it to another scarce asset - bitcoin.

There are mixed views on how far it goes in determining value and there has been much debate about exactly that within crypto circles. I think that it is a tool and an indicator of sorts - and that's as far as it goes. Having said that, what financial model can be relied upon? PlanB has acknowledged that with regard to modelling generally. I believe the phrase he has used is that 'all models are wrong but some are useful'. It has also to be considered that we're in the early stages of the development of a new asset. There isn't that much historical data to work with for quants to develop models on. However, PlanB is to be credited with trying to apply a model. There is no panacea when it comes to modelling but it would appear that his model still has some utility.

For those that want to look it over in more detail, PlanB has published a brief outline of the model here ->

Bitcoin Stock-to-Flow Cross Asset Model

Historical data aside, it would seem that those in the bitcoin/crypto sector have the same difficulty as those in the $9 trillion gold industry in terms of financial modelling.
 
Last edited:
Brendan Burgess said:
Over to you tecate and Duke...
Click to expand...
Boss that is wicked mischief. I will make only one posting on this thread. I have more self respect than to get sucked into this rabbit hole.

I have never seen such utter pompous drivel in all my puff. This convinces me that bitcoin, indeed all crypto, is a total nonsense. It is one massive con job. I can't tell whether it is a cult style group self con or whether the likes of PlanB are having a good laugh at the mugs, probably a mix of both.

The appropriately named SF model predicts, as your link points out, that in the final tranche bitcoins will be valued at $235bn each:eek: This was based on a Regression analysis no less. Of course that is more than 100 years away and not quite so persuasive as a promotional strap line. So in 2019 PlanB restricted his forecasts to the next halvening which we know was in May this year. Surprise surprise, based on his SF model, a halvening greatly does boost the price. PlanB went so far as to predict a $55k price as a result of this May's halvening and asked that he be judged by the accuracy of his predictions. (Gosh, does he actually believe this stuff?).

But he did give himself a wee bit of wriggle room - he was targeting 2020-24. Amazingly his latest enhanced "cross" SF model has made him double down on this prediction, he now plumps for $288k per bitcoin in 2020-24. The new insight is that bitcoin is going through phases like water from ice to liquid to vapour to ionized (? a new one on me.) Jayz I am nearly throwing up on my keyboard.

Boss I don't know whether you ever tried to value your poetry on the SF model. But I can help you on your toe nails. If you are constructed in that department along the same lines as myself, the flow should be zero (correction, I suppose if you IPO'd your current set, a new set would grow in its place). That makes your toe nails priceless:cool:
 
Last edited:
It seems the double standards are contagious. The only mention of a stock-to-flow based model previously in this series of recent discussions was in relation to its use over many years by those who trade gold. Not in relation to bitcoin.

PlanB himself points out that 'all models are wrong but some are useful' yet that doesn't seem to be good enough for the Duke.

You're scathing with regard to stock to flow as the model is applied to bitcoin yet no mention of it as it has been used for years in the gold industry to determine (or be an indicator of) gold price.

This is why consideration of how a fair price of gold is arrived at is very much pertinent to the consideration of same with regard to bitcoin.

Your 'analysis' is that the price of gold is determined by a consideration of supply and demand whilst Brendan refuses to answer. As I write this, the gold price is $1,788/oz. Neither of you can present a neat calculation as to whether this is a fair price or not. You're accepting of that yet when it comes to bitcoin, you don't have the same accomodation strangely.
 
Last edited:
@tecate Why am I hearing a sucking noise? I don't expect you ever to admit the sheer nonsense of what PlanB is on about here. Makes Kelleher look rational.
For other AAM contributors, I have read the various links so far and this is my summary of what PlanB has done. Make up your own mind.
His SF model posits that the ratio of existing supply (S) to new annual supply (F) is the main driver of overall market cap and therefore bitcoin price. In fact his regression analysis shows that the price is 99.7% determined by this statistic.
So what is "regression analysis"? He calculates the SF of Gold to be 58.3 and that gives a market cap of $10tr. Silver has an SF of 33 and a market cap of $0.6tr. Bitcoin in its current phase has a SF of 56 and so if PlanB's thesis is right bitcoin should have a market cap close to that of Gold. Problem is that two sightings (Gold and Silver) isn't really much of a basis for a regression analysis. So he invents 4 more data points - the 4 phases of bitcoin, just as H2O has solid, liquid, gas and ionised(?) phases. Now he has 6 data points and, to cut a long story short, his model predicts bitcoin to have a market cap of $5.5tr and a price therefore of $288k with 99.7% confidence. But we are now in that phase where the SF of bitcoin is 56 and yet its price languishes at $9k, so much for the model having 99.7% prediction capability.
What sort of a spoof is this? :eek: :eek: :eek:
 
Last edited:
Duke of Marmalade said:
What sort of a spoof is this? :eek: :eek: :eek:
Click to expand...
So is that the purpose here? Brendan drops mention of the model in here and we all take turns kicking the daylights out of it?
Nobody has presented here claiming a bullish case for bitcoin based on stock to flow. PlanB has acknowledged repeatedly that 'all models are wrong but some are useful'.
There's little historical data for quants and statisticians to work with in order to establish an appropriate model. PlanB has tried to do so with stock-to-flow. You can take it or leave it. The efficacy of a bullish view on bitcoin doesn't in any way depend on the stock to flow model.

I still maintain that the interesting take away from this discussion is that you reserve your wrath for stock to flow being applied to bitcoin but not to gold. Still not a word from either of you as to how the fair price of gold is arrived at. Are you both telling us that there is no neat little formula to determine price in the $9 trillion gold business?
 
tecate said:
So is that the purpose here? Brendan drops mention of the model in here and we all take turns kicking the daylights out of it?
Nobody has presented here claiming a bullish case for bitcoin based on stock to flow. PlanB has acknowledged repeatedly that 'all models are wrong but some are useful'.
There's little historical data for quants and statisticians to work with in order to establish an appropriate model. PlanB has tried to do so with stock-to-flow. You can take it or leave it. The efficacy of a bullish view on bitcoin doesn't in any way depend on the stock to flow model.

I still maintain that the interesting take away from this discussion is that you reserve your wrath for stock to flow being applied to bitcoin but not to gold. Still not a word from either of you as to how the fair price of gold is arrived at. Are you both telling us that there is no neat little formula to determine price in the $9 trillion gold business?
Click to expand...
I will address your questions re gold if you will admit that PlanB's model is an absolute nonsense, otherwise you are just a troll.
PlanB bemoans the fact that he has only 6 data points and suggests there should be further research on his thesis. Can I suggest that he includes the 2,000 other cryptos in his regression analysis and then we will see if market cap is truly determined by SF.
Absolute, total, complete, utter etc. nonsense. OP posted a link highlighting the implications of this nonsense (infinite price when F becomes 0), but that link actually is counterproductive as it suggests that the model is worthy of some intelligent critique. You are being duped my dear tecate.
 
Duke of Marmalade said:
I will address your questions re gold if you will admit that PlanB's model is an absolute nonsense, otherwise you are just a troll.
Click to expand...
Ergo, your answer to this would be yes then? ->
tecate said:
So is that the purpose here? Brendan drops mention of the model in here and we all take turns kicking the daylights out of it?
Click to expand...
It's no better nor worse than the use of stock to flow as it has been used in attempts to price gold over the years.
Furthermore, it's the first application of such a model to this new asset. Others will attempt to develop other models. It's an iterative process but of course they're still just models and all models are limited.
Duke of Marmalade said:
PlanB bemoans the fact that he has only 6 data points and suggests there should be further research on his thesis.
Click to expand...
He's referring to a lack of historical data. I mean, if you want to suggest that bitcoin has the same amount of historical data available as gold or commodities where stock to flow analysis might be used, then do yourself that disservice.

Duke of Marmalade said:
Absolute, total, complete, utter etc. nonsense.
Click to expand...
Yet you don't reserve the same ire and venom for the use of stock to flow in the assessment of gold? Remember, stock to flow analysis is not his original thesis. It's borrowed from the gold industry by him and adjusted for use with bitcoin. I've no problem with you expressing the opinion that stock to flow holds little value for you, but it would be good to bring some objectivity back into these discussions. Can you confirm that stock to flow is neither useful for gold or bitcoin?
Duke of Marmalade said:
OP posted a link highlighting the implications of this nonsense (infinite price when F becomes 0), but that link actually is counterproductive as it suggests that the model is worthy of some intelligent critique.
Click to expand...
See above. This form of analysis has been around for years but it's only today that you're venomous about it (as in this instance, it is applied to bitcoin).

Duke of Marmalade said:
You are being duped my dear tecate.
Click to expand...
How am I being duped your dukeness when I stated above that the efficacy of a bullish view on bitcoin doesn't in any way depend on the stock to flow model?
 
Last edited:
tecate said:
Ergo, your answer to this would be yes then? ->
It's no better nor worse than the use of stock to flow as it has been used in attempts to price gold over the years.
Furthermore, it's the first application of such a model to this new asset. Others will attempt to develop other models. It's an iterative process but of course they're still just models and all models are limited.

He's referring to a lack of historical data. I mean, if you want to suggest that bitcoin has the same amount of historical data available as gold or commodities where stock to flow analysis might be used, then do yourself that disservice.


Yet you don't reserve the same ire and venom for the use of stock to flow in the assessment of gold? Remember, stock to flow analysis is not his original thesis. It's borrowed from the gold industry by him and adjusted for use with bitcoin. I've no problem with you expressing the opinion that stock to flow holds little value for you, but it would be good to bring some objectivity back into these discussions. Can you confirm that stock to flow is neither useful for gold or bitcoin?

See above. This form of analysis has been around for years but it's only today that you're venomous about it (as in this instance, it is applied to bitcoin).


How am I being duped your dukeness when I stated above that the efficacy of a bullish view on bitcoin doesn't in any way depend on the stock to flow model?
Click to expand...
Bye
 
Here are a few articles that ponder the direction of the gold price - referencing stock-to-flow in the process.

Gold's Outlook for 2020 : "Annual mine production adds 3,000-3,500 tonnes, giving a stock to flow ratio of over 50 times."
Bloomberg from Dec. 2019 - Commodities Roundup: The Top Trades of 2020 : At around 6:50 precious metals specialist Everett Millman tells Bloomberg that "gold has a high stock to flow ratio" .

Here's an examination of the price of silver - with a number of references to stock to flow:
Zerohedge: Silver Above-Ground Stocks a Mere Fraction of Global Assets

Here's the title of a whole section of a report on gold by a research group belonging to a major European bank:

"The Stock to Flow ratio as the most important significant reason for gold's monetary importance."

People can make what they want out of it (I've never mentioned it on AAM in the context of bitcoin) but the model has been in existence long before bitcoin ever appeared on the scene.
 
Last edited:
Brendan Burgess said:
Twitter account “PlanB,” has been widely praised and is the leading valuation model for bitcoin proponents.
Click to expand...

I think the contention is the use of the words 'is the' above.
Had the author used 'is a' then perhaps the article wouldn't attract as much attention?

I've never heard of Plan B, and it makes little sense to me anyway.
 
WolfeTone said:
I think the contention is the use of the words 'is the' above.
Had the author used 'is a' then perhaps the article wouldn't attract as much attention?

I've never heard of Plan B, and it makes little sense to me anyway.
Click to expand...
The issue is that it's a new asset at an early stage in its development. His is the first attempt at a model that might provide some indication of price. However, it's the same stock to flow analysis as applied to gold. The approach is equally flawed in its application to gold or to bitcoin. That's not something that's getting recognition from the crypto naysayers amongst us.

Other than that, there doesn't seem to be any significantly better methodology for figuring out the price of gold despite that market being in existence for donkeys years. Yet it hasn't stopped participants in that market from trading gold. It hasn't stopped them from arriving at buy/sell price every trading day of the week in a $9 trillion market.
 
Last edited:
Here's an interview with Jesse Proudman of Strix Leviathan, the company that provided its analysis on Stock to Flow which forms the basis of the article that Brendan linked to in the initial post. Over the course of the discussion which was uploaded yesterday, Proudman confirms...

- That the company disagrees with the stock to flow model as applied to bitcoin as it doesn't work for gold either.

- That despite the model he's bullish on bitcoin as a market that will see further growth:
"We believe that this asset class is valuable.We believe that bitcoin will rise in price."

- When asked about models and price analysis generally he states:
"There is no one magical model that predicts price."
 
Last edited:
Fidelity has made other news today ( See Gus1970's thread on its new Bitcoin Fund ). However, it's interesting that the asset manager (with $8.3 trillion under management) also finds it relevant to consider the stock to flow model.
considering the merits of bitcoin investment published last month, the head of research at Fidelity Digital Assets considers Plan B's stock to flow model worthy of discussion. See pages 8-9 of the report.
 
Last edited:
Hilariously:

Stock-to-Flow Multiple (463d) 2021-04-29, 23:59 UTC
Actual price: $53,567.08
Model price: $53,391.54

s2f.jpeg
(source )
 
You must log in or register to reply here.
Back
Top