One interesting thing with the Kelly Criterion is that it sort of necessarily shows you the importance of diversification.

Even with insanely good odds, you never bet a huge amount of your bankroll
(I did a short explanation of Kelly here if you're unfamiliar)

Here's an insane example:

You have a bet which pays 100 to 1 and you are 80% sure that you are correct, how much should you bet?

(Try to pick a % of your assets/bankroll in your head)
The "Full Kelly" answer is 79.8%.

However, going full Kelly in real life is insane.

Half Kelly is very aggressive.

Most practitioners I know of use quarter kelly.

(This has to do wit the marginal risk/profit as you move out the Kelly Curve as @NickYoder86 shows here)
At half Kelly, the suggested allocation is ~40%

At quarter Kelly, it's ~20%.

I picked this example (80% confidence of 100 to 1 odds) because it's the most extreme example I could think of - most people will never see this and, at most, you would see it 1-3 times in a lifetime.
I think it's a reasonable rule of thumb that you should basically never allocate more than ~25% to any single bet if your goal is to maximize long-term wealth.
Necessarily, this means you need to spread your bets out more than intuitively seems right even when you have amazing odds and high confidence!

(And, yes, I am subtweeting a bunch of crypto people)

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Taylor Pearson

Taylor Pearson Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @TaylorPearsonMe

3 Jun
One of the interesting elements of crypto/digital currency that doesn't get talked about enough is the auditability of having everything being digital.
Part of the 2008 GFC story that isn't as widely talked about was that a lot of the problems were not just that a lot of bad mortgages had been handed out (they had), but that it was all buried in giant paper contracts so no one know how bad (or not bad) it was.
A lot of the traders that made the most money in 2008/9 were actually buying mortgage-backed securities (MBS) that were trading too low because people were afraid things were even worse than they were.
Read 9 tweets
1 Jun
I think understanding basic game theory concepts like the prisoner's dilemma is really useful, but it leaves at least two important concepts out:

1. Reputation

2. Context Dependence.
The gist of the prisoner's dilemma that [[Robert Axelrod]] showed was that by getting people to engage in iterated prisoner's dilemmas instead of one off, you promote cooperative
(If you're unfamiliar with the example, I wrote a short explanation) behavior.taylorpearson.me/interestingtim…
Read 14 tweets
19 May
I've never found a way to articulate it cleanly, but this is a really important point from Taleb's work (and others) that goes underappreciated.

If an environment is becoming more fat tailed, we would actually expect less variance in the short-term.
It's got a Minsky-esque quality to it that more stability can actually suggest greater future instability.
If you remove all the stressors from an environment by delaying risk, you not only make the eventual collapse worse, you make people less prepared for it.

Read 4 tweets
14 May
Just posted a summary and my notes from John Galls' wonderful book Systemantics, a wonderful (and funny) book on how systems work

Some of my favorite lines....

taylorpearson.me/bookreview/sys…
SYSTEMS IN GENERAL WORK POORLY OR NOT AT ALL More technically stated: COMPLICATED SYSTEMS SELDOM EXCEED FIVE PERCENT EFFICIENCY
SYSTEMS TEND TO MALFUNCTION CONSPICUOUSLY JUST AFTER THEIR GREATEST TRIUMPH

Toynbee explains this effect by pointing out the strong tendency to apply a previously-successful strategy to the new challenge....
Read 28 tweets
26 Apr
The rapid sell-off in Bitcoin last week is a good example of how exogenous market factors can trigger endogenous market structure factors leading to a cascading sell-off.

This phenomenon is an important part of markets and (IMO) underappreciated.
In the case of Bitcoin, Phase 1 of the sell of was that there was a large hashrate drop which triggered a wave of selling.

However, that also forced a lot of overlevered players to cover their levered long positions (or they got liquidated), causing a second leg down.
I think this is important because the common understanding of price movements is that they are reflective of investors saying "I have updated my beliefs about the future value of this asset and am buying/selling based on that."
Read 12 tweets
23 Apr
I’ve been (rental) house hunting for the last couple of months. I don’t really know anything about real estate investing, but I’ve been trying to read up (John T. Reed’s Best Practices for the Intelligent Real Estate Investor is my favorite so far).
It’s been interesting seeing the market and how homes are priced.

Factors which the market seems to price really efficiently include:
-Square footage
-Neighborhood
-Amenities/Finishes
-View
However, there are a lot of factors that (in my experience) have very high quality of life implications and basically don’t seem priced in at all.
Read 19 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!

:(