HN2new | past | comments | ask | show | jobs | submitlogin

So if we look at companies that didn't do as well, we find that they didn't do as well as those that did really well?


I guess the story here is more like

If you take the Top 500 companies as an indicator how well your economy is doing, but everything is carried on the shoulders of only 1.4% of those 500 companies then what is the point of looking at the 500 companies in the first place.

Make an S&P10 then instead. or in addition to the S&P500

That would probably be a good way to look at it anyway. Looking at the trend of the S&P10 vs. S&P500 and if they agree then thumbs up, but if they disagree then things might not be as rosy as everyone thinks


The problem with that approach is you miss out on the gains of tracking a company from when its market cap is small to when it is large, and only capture the opposite as it leaves the top 10.

This is one reason people are so concerned with companies going public later. If they just appear in their fully formed embodiments then you can only capture their growth after they are large and their death.


> If you take the Top 500 companies as an indicator how well your economy is doing, but everything is carried on the shoulders of only 1.4% of those 500 companies then what is the point of looking at the 500 companies in the first place.

It's valid to argue that success of companies in the S&P 500 is not evenly distributed, but if you really want to understand the impact of that distribution on the economy you have to look at the value of each company rather than treating them equally.

The top 7 companies might only be 1.4% of the companies in the S&P 500, but they represent roughly 35% of the market cap of companies in the index. Because of that, they have an impact on the broader economy much larger than the raw number of companies would suggest. "Just" Nvidia doing well is going to have a much larger economic impact than just Newell Brands, which I have never heard of but is apparently one of the smallest on the index. In fact Nvidia's market cap is roughly 1000 that of Newell Brands with presumably similarly disproportionate economic impact.


> Make an S&P10 then instead

How do you pick these 10? After the fact, necessarily. So then meaningless for tracking the economy.


Them not agreeing isn't a bug. I bet it's pretty historically normal.


How about S&PMedian? Wonder how this would do over the time.


Probably not well. You'll be consistently betting against the market with very little reason to believe you're really adding information.

Also, cap weighting reduces your need to trade and thus your overhead and tax costs.


Interesting idea, even average would probably already work, it would give high performers some impact but not as much as now.

But it kinda misses what the S&P is in my pov. An index tracking how the value of those companies developed over the years measures by what you would have now if you would have invested in each of them


The problem is that the set that "did really well" consists of only seven companies.


And those companies still did extremely well. It just so happens that the top 7 of them did insanely well.


No, it's about relative value. Obviously it's a tautology to announce that you've discovered that stocks that did really well outperformed ones that didn't.

The interesting thing is that the aggregate market gains are almost entirely concentrated in those stocks. It's not a statement about sorting, it's a statement about distribution.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: