Even the chart from JP Morgan is a bit fishy. Why analyze only 19 years when there are over 100 years of S&P 500 data? Why start in Jan 2003? I wonder what the chart would look like with the full data?
What would be more interesting is a chart that shows attempted market timing where you time perfectly the buy and sell (this one already exists) and then charts for each day you are wrong on the buy and the sell.
That way you can see how accurate you need to be for this to pay off and also whether it is better to be too early or too late on the buy or sell.
1 Like
It’s actually 20 years of data and I think they picked this time frame because they had to pick something which is familiar to most people in current times.
No one in real world invests for 100 years , so it doesn’t represent investment horizon of a real person.
I am sure such a chart will also play out similarly if we take 50 years of data because the reality is on average market moved 9-10% per annum so if you miss the days which moved 10-15% in a day, it would definitely make a dent in your returns.
You don’t mean this here, right? It’s only about buying, I’d be curious to see perfect buying and selling.
Perfect buying and selling timing would result in astronomical gains. Assuming you mean that investor always sold at tops and then rebought at bottoms and did this for every move up and every move down
This is what traders dream off and mostly never achieve
I am not so sure of “astronomical” gains unless we include massive leverage/derivatives. Maybe 2-3% more CAGR at best without them? This is a number I pulled out of my backside and would compound a ton over time, but still not astronomical.
I will try to do some math later . But conceptually perfect timing buy-sell means following
- On days market is moving up you were invested
- On negative days you were in Money market fund
This means you only experience positive change in your portfolio and no negative changes. If you just look at 2025 daily changes , you will start to get a picture of how this would play out.
I would settle on a chart showing the returns of missing the 10 best days and an arbitrary buffer of days (say 5) :
- before the best day;
- after the best day;
- both before and after the best day.
That would give me more fodder for thoughts about the great days happening right after terrible days.
For the sake of completedness, the same chart but around the worst days may help me understand how precise market timing needs to be to be efficient.
I know what it means, I am just challenging “astronomical”. I mean, if we’re going full hog “world’s best insider trader” it’d be fun to see what could be done assuming perfect use of leverage both upwards and downwards.
ChatGPT says this, probably can be much improved with a better worded question: ChatGPT - VWRL Positive Day Strategy
See here. It seems more than 40% of the trading days have negative returns.
Imagine if you were able to magically eliminate them from your investment portfolio exposure. I don’t think CAGR difference would be 2-3%. I think it would be much much higher.
Why do both of you think? Go on Yahoo finance, download the data, and just calculate it.
3 Likes
Probably because Abs_max doesn’t have the time, and definitely because I don’t have the skill 
Yeah. Don’t have access to laptop right now.
Will calculate 2024 data when I do have access
Why not ask ChatGPT? I gave it the last 12 months of the SP500.
- Actual total return: 9.51%
- Hypothetical total return (avoiding all negative days): 133.15%
@Mirager
It’s a huge difference, like 14x better return.
It is, but given loads of people around the world have gotten the same longing lucky or dead obvious stock picks like nVidia in 2022 I don’t see it as astronomical. I’d see 100-500x as astronomical.
I think that’s how RenTech is actually doing it with the Medallion Fund. Being right 55% of the time and using a lot of leverage.
But weren’t we talking about 20 year period ?
Do this for 20 years and I think numbers would be staggering due to compounding
Well yes. The difference would grow astronomically after only a couple of years, for example 10 years: +148% vs. 474’530% (3’200x better return).
Of course, or hold Berkshire from when Warren Buffett was an unknown fund manager in the least likely place to find a fund manager 
I think finding Berkshire would have been nothing compared to finding the magic mirror which tells you which days are negative for S&P 500 in advance.
Your table shows 10% additional gains per year. The magic mirror will means maybe 100-200% additional gains per year.