2 year investment journey so far, trying not to feel disheartened.

SneakySpectator

masonic said:

But you have to accept that there is a range of possible outcomes when using equities, and if you are unlucky you may not materially beat cash returns, even drip feeding over long time horizons.

I do accept the is a range of possible outcomes, but I only accept realistic outcomes, meaning outcomes that have high probabilities of actually happening. 

There's no point for example in factoring in the possibility of my average annual return being ~1% over a 25 year period because the probability (in my mind) is just so extraordinarily low and the only reason I say the probability is so low is because it's never happened so far in over 100 years...

masonic

SneakySpectator said:

masonic said:

SneakySpectator said:

masonic said:

SneakySpectator said:

masonic said:

What happens at the start of the 25 years will be largely immaterial. It is what happens at the end that is likely to have the greatest effect.

If you were to continue to invest about £20k per year for the next 25 years, your ending portfolio could be anywhere between an inflation adjusted £800k and £3.8m based on historic returns, which would have cost you £540k in today's money, so anywhere from 1% - 8% real return (based on the 5th-95th percentile), whereas the average would have been about 5% real. So the pound cost averaging and time in the market reduced the range, but you can still end up not performing materially better than cash/bonds, even over large time horizons. Backtesting can only take you so far though, as it won't consider sequences of returns not seen before.

I'm sorry did you just say that I may not perform materially better than cash or bonds?! I'm assuming you meant a cash isa and not actual cash, considering cash loses value each year. 

But even now in times of good interest rate I'm only getting 4.5% return, which when you deduct inflation off that it's only like 1.7%, and bonds are not much better only recently picking up again. 

So in what universe does saving in a cash isa or government bonds work out the same as a 5% real return in a global index fund??

I'm not usually a suspicious person but when people say things like this, the only conclusion I can come to is that they don't want you to invest, they don't want you to make money. Either through jealousy or because they just don't like the stock market or something 

No need to apologise. The range of possible returns from backtesting is 1% - 8% real (based on the 5th-95th percentile). If you achieve the lower bound outcome of 1% real, then that is about what you'd achieve from the long term returns of a rolling 1 year fixed term account, or bond investment. That is the best you could expect in 5% of universes. Only 50% of universes will give you at least a 5% real return, and you might not be living in any of them.

For transparency, 70% of my net worth is in equities and until recently it's been more like 90%, but I am much further into my investment journey than you. I'd be 100% equities if I were only 2 years in, but that does not change the fact that even drip feeding over long time horizons, you cannot rely on achieving a historical average return.

Oh I remember this discussion from a few months ago where you said a similar thing.

And I responded that never in the history of the stock market has a 25 year period resulted in a 1% averaged return and you said that wasn't the point, it's a possibility etc etc. 

I'll take my chances with the stock market and an average expected real return of 5% than a cash isa / bond guaranteed 1.7% real return. 

You seem to have invented that exchange in your own mind. The numbers I have mentioned are based on actual historical periods, such as the period around the 1929 and 1970s depressions. Other such periods will come along in the future. Nobody knows when.

It is simply preposterous to suggest that an average return or mathematical expectation is what will actually happen. It demonstrates that you have a lot to learn and an unwillingness to accept that fact. It doesn't matter how hard you believe you are certain to achieve a 5% real return from your equity investments, it won't make it so.

Definitely not in my head.

So basically what you're saying is there is a probability (1.26%) of getting the average, which I agree, getting smack bang on the average is very unlikely, but getting somewhere close to the average is very high.

It appears 6 people thanked my post correcting your misconceptions about flipping coins. It is sometimes nice to receive confirmation you are not going mad when facing a barrage of specious arguments counter to reality. But I don't see anywhere where you said "never in the history of the stock market has a 25 year period resulted in a 1% averaged return" and I said "that wasn't the point, it's a possibility etc etc"

SneakySpectator said:

So tell me again about this 1%?

Here is a crude cFireSim simulation for a 100% global equities portfolio with starting value $40,000 and contributions of $20,000 per year all inflation adjusted ($540k contributed in total):

You'll have to excuse it being valued in dollars, but that's life. The range of possible outcomes is $577k to $3.78m and you can see the distribution of outcomes from the chart. There are quite a few periods where the outcome was considerably lower than a 5% real return, and 5-10% of periods where returns ended up being cash-like (the 1% you are asking about). Like I said before, it is what happens at the end of the journey that makes the most impact. A crash at the start is generally good for returns, whereas a crash at the end is very damaging.

So with that, I am done with this discussion as I know when someone is so entrenched in a position that they will not listen to facts.

SneakySpectator

masonic said:

SneakySpectator said:

masonic said:

SneakySpectator said:

masonic said:

SneakySpectator said:

masonic said:

What happens at the start of the 25 years will be largely immaterial. It is what happens at the end that is likely to have the greatest effect.

If you were to continue to invest about £20k per year for the next 25 years, your ending portfolio could be anywhere between an inflation adjusted £800k and £3.8m based on historic returns, which would have cost you £540k in today's money, so anywhere from 1% - 8% real return (based on the 5th-95th percentile), whereas the average would have been about 5% real. So the pound cost averaging and time in the market reduced the range, but you can still end up not performing materially better than cash/bonds, even over large time horizons. Backtesting can only take you so far though, as it won't consider sequences of returns not seen before.

I'm sorry did you just say that I may not perform materially better than cash or bonds?! I'm assuming you meant a cash isa and not actual cash, considering cash loses value each year. 

But even now in times of good interest rate I'm only getting 4.5% return, which when you deduct inflation off that it's only like 1.7%, and bonds are not much better only recently picking up again. 

So in what universe does saving in a cash isa or government bonds work out the same as a 5% real return in a global index fund??

I'm not usually a suspicious person but when people say things like this, the only conclusion I can come to is that they don't want you to invest, they don't want you to make money. Either through jealousy or because they just don't like the stock market or something 

No need to apologise. The range of possible returns from backtesting is 1% - 8% real (based on the 5th-95th percentile). If you achieve the lower bound outcome of 1% real, then that is about what you'd achieve from the long term returns of a rolling 1 year fixed term account, or bond investment. That is the best you could expect in 5% of universes. Only 50% of universes will give you at least a 5% real return, and you might not be living in any of them.

For transparency, 70% of my net worth is in equities and until recently it's been more like 90%, but I am much further into my investment journey than you. I'd be 100% equities if I were only 2 years in, but that does not change the fact that even drip feeding over long time horizons, you cannot rely on achieving a historical average return.

Oh I remember this discussion from a few months ago where you said a similar thing.

And I responded that never in the history of the stock market has a 25 year period resulted in a 1% averaged return and you said that wasn't the point, it's a possibility etc etc. 

I'll take my chances with the stock market and an average expected real return of 5% than a cash isa / bond guaranteed 1.7% real return. 

You seem to have invented that exchange in your own mind. The numbers I have mentioned are based on actual historical periods, such as the period around the 1929 and 1970s depressions. Other such periods will come along in the future. Nobody knows when.

It is simply preposterous to suggest that an average return or mathematical expectation is what will actually happen. It demonstrates that you have a lot to learn and an unwillingness to accept that fact. It doesn't matter how hard you believe you are certain to achieve a 5% real return from your equity investments, it won't make it so.

Definitely not in my head.

So basically what you're saying is there is a probability (1.26%) of getting the average, which I agree, getting smack bang on the average is very unlikely, but getting somewhere close to the average is very high.

It appears 6 people thanked my post correcting your misconceptions about flipping coins. It is sometimes nice to receive confirmation you are not going mad when facing a barrage of specious arguments counter to reality. But I don't see anywhere where you said "never in the history of the stock market has a 25 year period resulted in a 1% averaged return" and I said "that wasn't the point, it's a possibility etc etc"

SneakySpectator said:

So tell me again about this 1%?

Here is a crude cFireSim simulation for a 100% global equities portfolio with starting value $40,000 and contributions of $20,000 per year all inflation adjusted ($540k contributed in total):

You'll have to excuse it being valued in dollars, but that's life. The range of possible outcomes is $577k to $3.78m and you can see the distribution of outcomes from the chart. There are quite a few periods where the outcome was considerably lower than a 5% real return, and 5-10% of periods where returns ended up being cash-like (the 1% you are asking about).

So with that, I am done with this discussion as I know when someone is so entrenched in a position that they will not listen to facts.

I do understand what you're saying and I know you won't respond any further but it's hard to tell from that chart because there's so many lines. But which 25 year resulted in your 1% average return? 

First you said 1929 and 1970 but those 25 year periods had a solid return. 

If you can point to a specific 25 year time frame that would help me to visualise it. 

I mean I know there's a chance I could end up with a negative average annualised return if the global market crashes 90% I mean it could right? But we don't tend to include silly probabilities when we're trying to calculate what is likely to happen.

masonic

I was referring to end dates. You should be able to see from the chart that 25 year periods ending between 1978-1981 and 1918-1922 were particularly poor.

There is no point considering the worst possible outcome, but outcomes with a frequency of 5-10% are worth bearing in mind as not vanishingly unlikely. That's 1 in every 10-20 periods within the historical data.

SneakySpectator

masonic said:

I was referring to end dates. You should be able to see from the chart that 25 year periods ending between 1978-1981 and 1918-1922 were particularly poor.

There is no point considering the worst possible outcome, but outcomes with a frequency of 5-10% are worth bearing in mind as not vanishingly unlikely.

1978 - 1981 is only 3 years and 1918 - 1922 is only 4 years though?

I'm not investing for such a short period of time, I'm investing for 25 years... 

This is what I expect to happen long term for the global stock market



This is coin flips but it applies to my stock market idea. So if you take any short time frame in this chart you may get a wildly different outcome instead of the expected 50-50 average. but cumulatively over time it teeters out to the expected average. 

That's why I was up 26% in just 2 years, next 2 years I might be down 30% but over a long period of time it should approach somewhere around the average. This doesn't mean it absolutely will, but there is a high probability that it will. 

I hope you understand what I'm trying to explain.

masonic

I am referring to 25 year periods ending in those years. All of those lines are 25 year periods spaced a month apart and you can see how varied the value of the portfolio is depending on the end date.

Ergo, investing in the stock market is nothing like flipping a coin.

LHW99

The issue with comparing to a coin flip, is that it is the probability over many trials (of 4000 flips) that equates to 50% heads. The distribution of results is approximately Gaussian / normal, so the graph is bell-shaped.

Our problem as investors is that we could be anywhere on that distribution, and we only get a single (trial) result.That could be anywhere on the curve, where there are a few very low results (<10 heads out of 4000) and a few very high ones (>3990 heads out of 4000). However, the distribution is based on not just doing a single trial of 4000 flips, but on doing a lot of them, where each result will (generally) be different.

We only get one trial - we can't say 25 years down the line - I don't like the result of this trial, I'll start again!

All we can do is try to make sure that we can at least cover ourselves if we do end up with one of the poorer results (if they get one of the miraculously good ones, I doubt anyone would argue )

SneakySpectator

LHW99 said:

The issue with comparing to a coin flip, is that it is the probability over many trials (of 4000 flips) that equates to 50% heads. The distribution of results is approximately Gaussian / normal, so the graph is bell-shaped.

Our problem as investors is that we could be anywhere on that distribution, and we only get a single (trial) result.That could be anywhere on the curve, where there are a few very low results (<10 heads out of 4000) and a few very high ones (>3990 heads out of 4000). However, the distribution is based on not just doing a single trial of 4000 flips, but on doing a lot of them, where each result will (generally) be different.

We only get one trial - we can't say 25 years down the line - I don't like the result of this trial, I'll start again!

All we can do is try to make sure that we can at least cover ourselves if we do end up with one of the poorer results (if they get one of the miraculously good ones, I doubt anyone would argue )

I agree but like the 400 coin flips we could say 400 months. 

Yes on any given month it could be anything from -30% to +30% and you might even get consecutive negatives (or positives) but over many months (coin flips) you would expect to start approaching somewhere towards the average. 

So sure we only get 1 trial, but 1 trial is not 1 year, it's 25 years.

Just like with our coin flip analogy there is a chance that the end average would be significantly different to the expected average. Maybe instead of 49.7% - 50.3%, we get 40% - 60%, that is of course possible but it's on the lower end of possible. 

So with the global stock market we're expecting somewhere around 6% - 8% but we could get 2% - 3%, however that doesn't seem like a relatively likely probability.

kimwp

SneakySpectator said:

LHW99 said:

The issue with comparing to a coin flip, is that it is the probability over many trials (of 4000 flips) that equates to 50% heads. The distribution of results is approximately Gaussian / normal, so the graph is bell-shaped.

Our problem as investors is that we could be anywhere on that distribution, and we only get a single (trial) result.That could be anywhere on the curve, where there are a few very low results (<10 heads out of 4000) and a few very high ones (>3990 heads out of 4000). However, the distribution is based on not just doing a single trial of 4000 flips, but on doing a lot of them, where each result will (generally) be different.

We only get one trial - we can't say 25 years down the line - I don't like the result of this trial, I'll start again!

All we can do is try to make sure that we can at least cover ourselves if we do end up with one of the poorer results (if they get one of the miraculously good ones, I doubt anyone would argue )

I agree but like the 400 coin flips we could say 400 months. 

Yes on any given month it could be anything from -30% to +30% and you might even get consecutive negatives (or positives) but over many months (coin flips) you would expect to start approaching somewhere towards the average. 

So sure we only get 1 trial, but 1 trial is not 1 year, it's 25 years.

Just like with our coin flip analogy there is a chance that the end average would be significantly different to the expected average. Maybe instead of 49.7% - 50.3%, we get 40% - 60%, that is of course possible but it's on the lower end of possible. 

So with the global stock market we're expecting somewhere around 6% - 8% but we could get 2% - 3%, however that doesn't seem like a relatively likely probability.

But you're not investing all your money for a month in all these months simultaneously, so you can't average out the outcomes and produce anything sensible for you as an individual.

SneakySpectator

kimwp said:

SneakySpectator said:

LHW99 said:

The issue with comparing to a coin flip, is that it is the probability over many trials (of 4000 flips) that equates to 50% heads. The distribution of results is approximately Gaussian / normal, so the graph is bell-shaped.

Our problem as investors is that we could be anywhere on that distribution, and we only get a single (trial) result.That could be anywhere on the curve, where there are a few very low results (<10 heads out of 4000) and a few very high ones (>3990 heads out of 4000). However, the distribution is based on not just doing a single trial of 4000 flips, but on doing a lot of them, where each result will (generally) be different.

We only get one trial - we can't say 25 years down the line - I don't like the result of this trial, I'll start again!

All we can do is try to make sure that we can at least cover ourselves if we do end up with one of the poorer results (if they get one of the miraculously good ones, I doubt anyone would argue )

I agree but like the 400 coin flips we could say 400 months. 

Yes on any given month it could be anything from -30% to +30% and you might even get consecutive negatives (or positives) but over many months (coin flips) you would expect to start approaching somewhere towards the average. 

So sure we only get 1 trial, but 1 trial is not 1 year, it's 25 years.

Just like with our coin flip analogy there is a chance that the end average would be significantly different to the expected average. Maybe instead of 49.7% - 50.3%, we get 40% - 60%, that is of course possible but it's on the lower end of possible. 

So with the global stock market we're expecting somewhere around 6% - 8% but we could get 2% - 3%, however that doesn't seem like a relatively likely probability.

But you're not investing all your money for a month in all these months simultaneously, so you can't average out the outcomes and produce anything sensible for you as an individual.

Sure you can.

The money I invest today will compound far greater than the money I invest in the later years.

Let's say I invest £100 on month 1 and the price of the stock is £100. Then on year 24 I invest £100 and the price is £300. The first £100 has increased by 200% and the last £100 has not gone up at all.

Finally at the end of the year the price is £312. Well from year 24 - 25 the market only went up 4% on my last £100, but the first £100 went up from 200% by 12%

Isn't that the entire binding principle of "the earlier you start the better compounding works for you" mentality?