10 minute coin flip experiment

[Walkeia]

Following an article in the economist a weekend or two ago. I attach a link to a coin-flip game / experiment conducted by Elm Wealth in the USA. I thought the results from the experiment with finance graduates and young professionals were fascinating. Have a read of the instructions and a short think before playing. It last 10 minutes - the pay-off remains 50/50 per coin flip despite the coin being weighted 60/40 to heads.

https://elmwealth.com/coin-flip/

Linked below is the results of Elm Wealth's experiment and the lessons for investing they take from it. If you'd like to play then click the link above and stop reading my post here.

https://elmwealth.com/lessons-from-bett ... ary-tales/

Their take away / argument is that many investors pay too much attention to 'what you invest in' as opposed to 'how much to invest in a given opportunity'. This is something I wholeheartedly agree with hence me posting this in 'Investment Strategies'. Finally a hat tip to those participants in their original experiment which went all in on Tails... Vegas exists for a reason I guess

[dealtn]

50 spins 25 heads 25 tails including a run of 7 tails to end.

[Bubblesofearth]

An interesting experiment but not sure how it informs wrt the stock market, i.e. what approach it would best advise. I found myself betting approx. 20% of my balance on heads each time so fortuitously quite close to the Kelly criteria. But how would you invest in shares in this way, especially if, like me, you prefer to remain fully invested with some 80% of investable wealth.

Any thoughts?

BoE

[vand]

I got to $300 after 5 mins but it was boring

As a once professional gambler I understand well the concept of betting in proportion to how much edge you have while still keeping it small enough to be able to survive a string of statistically likely random losses.

Kelly Criterion etc

The problem in most real life trades is that you don't even know for sure if you have any edge, and even if you do you certainly can't be sure how large it is

[moorfield]

Took me 9 flips to double my money, at which point I called it quits.

[1nvest]

A main stock index fund isn't likely to sequentially/repeatedly halve (tails) without at some point doubling - heads (albeit from relatively low levels). Accordingly the failure of a Martingale in a Casino isn't evident in stocks (unless your major market stock index totally fails) and as such is a reasonable 'bet'. 1920's US stocks doubled, doubled again, doubled yet again (Roaring 20's), before halving, halving again (Wall Street Crash), If you started at the peak, that was followed by a -75% type decline (halved 100 to 50, halved 50 to 25) then with a Martingale betting approach you were into profit again once the value had rebounded from 25 to 50 (that was still half of the start date level).

How to play something like that in practice? 67/33 stock/gold perhaps. Whatever might drive stocks halving, 67 stock value dropping to 33, may be inclined to see gold prices double, 33 value rises to 67. At that 33/67 stock/gold weightings you've not lost any money and rebalancing has you back to holding 67/33 stock/gold again, the only difference is you hold twice as many stock shares as before - after prices had halved, and half as much ounces of gold as before (after prices had doubled).

[dealtn]

An interesting experiment but not sure how it informs wrt the stock market, i.e. what approach it would best advise. I found myself betting approx. 20% of my balance on heads each time so fortuitously quite close to the Kelly criteria. But how would you invest in shares in this way, especially if, like me, you prefer to remain fully invested with some 80% of investable wealth.

Any thoughts?

BoE

You don't need to do anything, you are already running that as a real time experiment.

In the real world your companies are investing (gambling) everyday with each business investment decision the majority of which have a positive return, with a minority that has an economic loss. Every £ invested on average has a £+r return as an average. Those unseen "coin tosses" at the corporate level are creating a gambling machine the benefits of which you are accruing within the companies finances.

Separately to this you have a stockmarket that is (somewhat inefficiently) calculating the worth of that corporate "gambling game". You are invested long-term and benefiting from that compounding. Furthermore you are (probably) invested in multiple versions of the game by choosing numerous companies to diversify the risk of an individual game having a poor run of "tails".

[Sternumator]

I don't believe Kelly is optimal for this game. Does Kelly assume the game goes on forever?

If someone has a 6 figure net worth and $300 left on the game, it's got to be optimal to go all in on the last flip. They will probably win or lose more than that on the stock market that day for nothing like 60/40 odds.

You don't want to bust out at the start either so optimum stategy should be function of time left in the game as well as account balance.

In terms of investing strategies, this kind of thinking draws me towards leveraged ETFs. Some people say daily rebalancing is a bad thing but, costs aside, the more rebalancing the better. It's playing a positive EV game and adjusting the stake size up and down according to pot size.

[vand]

This very experimental is discussed here

https://ofdollarsanddata.com/losing-wit ... ning-hand/

[vand]

And, of course, CGT is well known for placing risk management at the forefront of its strategy:

https://www.youtube.com/watch?v=_o_7REFkp8E

[Bubblesofearth]

In terms of investing strategies, this kind of thinking draws me towards leveraged ETFs. Some people say daily rebalancing is a bad thing but, costs aside, the more rebalancing the better. It's playing a positive EV game and adjusting the stake size up and down according to pot size.

2X FTSE100 has risen 63% over the last 10 years;

https://fundcentres.lgim.com/en/uk/inst ... FTSE%20100®,(the%20%22Index%22).

FTSE100 TR has risen about 100% over the same time period;

https://uk.investing.com/indices/ftse-100-total-return

So in an overall rising market you would have done worse in the leveraged ETF than the underlying. To be fair you would probably not have quite matched the underlying by investing in a FTSE100 ETF but still better than the leveraged version.

BoE

[vand]

In terms of investing strategies, this kind of thinking draws me towards leveraged ETFs. Some people say daily rebalancing is a bad thing but, costs aside, the more rebalancing the better. It's playing a positive EV game and adjusting the stake size up and down according to pot size.

2X FTSE100 has risen 63% over the last 10 years;

https://fundcentres.lgim.com/en/uk/inst ... FTSE%20100®,(the%20%22Index%22).

FTSE100 TR has risen about 100% over the same time period;

https://uk.investing.com/indices/ftse-100-total-return

So in an overall rising market you would have done worse in the leveraged ETF than the underlying. To be fair you would probably not have quite matched the underlying by investing in a FTSE100 ETF but still better than the leveraged version.

BoE

Look at TQQQ - great if you get a stretch like Apr 2020 - Dec 22... but investing trading is rarely that easy, and in downways, sideway, or even upwards markets that are whipsawing around a lot you will lose out.

QQQ has done 30% in last 3yrs despite everything - actually a reasonable return for the risk you are taking.
TQQQ has done 0% in the same time.

We know that stocks can reasonably only return 8-9% nominally over the long term (assuming inflation 2%), so the last 3 years should be seen as the "norm", not the exception.

[minnow]

Very interesting link ! This experiment was also covered on a recent Rational Reminder episode. As the article mentions, a worryingly large proportion of players (many of whom are financially literate) manage to go bankrupt eventually. They do irrational things like betting on tails because it hasn't come up recently, etc. Anyway, I made $1000 after ten minutes. Anyone know where I collect my winnings ?

[CliffEdge]

This another example where probability fails in practice. Like the blind bend near where I live which always has something coming round it just when I get there and nothing after it for 20 miles, or at least that's the impression I get.

[dealtn]

This another example where probability fails in practice. Like the blind bend near where I live which always has something coming round it just when I get there and nothing after it for 20 miles, or at least that's the impression I get.

No it isn't. It is the opposite. It is a practical demonstration of probability. If you, or anyone else, think that then it just shows you don't understand the mathematics of probability.

[GoSeigen]

This another example where probability fails in practice. Like the blind bend near where I live which always has something coming round it just when I get there and nothing after it for 20 miles, or at least that's the impression I get.

No it isn't. It is the opposite. It is a practical demonstration of probability. If you, or anyone else, think that then it just shows you don't understand the mathematics of probability.

I think he was being tongue in cheek. Hard to tell with text -- don't mind giving a heads up when I notice.

GS

[CliffEdge]

This another example where probability fails in practice. Like the blind bend near where I live which always has something coming round it just when I get there and nothing after it for 20 miles, or at least that's the impression I get.

No it isn't. It is the opposite. It is a practical demonstration of probability. If you, or anyone else, think that then it just shows you don't understand the mathematics of probability.

Interesting point of view. Is this borne out by experience?

[Urbandreamer]

Humm.

Can I add my opinion. I have been "investing" on the stock market for some 30 years and I didn't start by investing at all!

I even spent some time spread betting.

OK, so I didn't do badly in all this speculation, however it has only a passing relationship to investment.

As per the link, there ARE actually methods to either loose less or make money on games of chance. What most of us do investing is not that.
YES I have done that, but we are talking something different from investing!

The point about both the engineered example and what I and others try to do is that it's not a zero sum game.

The example was engineered to give, yes and limit the amount, money to those who took part.
It's not so clear with the stock market, but usually profits are provided by actually providing goods or services to people who value the fact.

The original link is hugely interesting in understanding ourselves and others (as I said I joined a less than zero sum game).
Understand yourself and to yourself be true!
Now given that, learn to live with yourself.

BTW one of my posts was best of boards on TMF. It was the one talking about this and how I started spread betting to allow me to invest in boring stuff.

[CliffEdge]

Humm.

Can I add my opinion. I have been "investing" on the stock market for some 30 years and I didn't start by investing at all!

I even spent some time spread betting.

OK, so I didn't do badly in all this speculation, however it has only a passing relationship to investment.

As per the link, there ARE actually methods to either loose less or make money on games of chance. What most of us do investing is not that.
YES I have done that, but we are talking something different from investing!

The point about both the engineered example and what I and others try to do is that it's not a zero sum game.

The example was engineered to give, yes and limit the amount, money to those who took part.
It's not so clear with the stock market, but usually profits are provided by actually providing goods or services to people who value the fact.

The original link is hugely interesting in understanding ourselves and others (as I said I joined a less than zero sum game).
Understand yourself and to yourself be true!
Now given that, learn to live with yourself.

BTW one of my posts was best of boards on TMF. It was the one talking about this and how I started spread betting to allow me to invest in boring stuff.

Exactly, that's what I have found. There's more to this.

[dealtn]

No it isn't. It is the opposite. It is a practical demonstration of probability. If you, or anyone else, think that then it just shows you don't understand the mathematics of probability.

Interesting point of view. Is this borne out by experience?

Yes. And quoted if you read all the original articles. You can set it up and do Monte Carlo analysis too to prove it to yourself if you don't trust other's results.