I've been thinking about this a little more lately, and am curious of this sub's thoughts.
Right now, ***everyone compares their performance to the S&P 500***. I don't think it's wrong, necessarily, and I want to make sure that I'm not naive in picking a low hurdle just to beat it - "yeah, I underperformed the market, but *excuses*..." sounds like something a loser says. But there are some pitfalls to using it as your lone benchmark.
Part of the problem is hindsight bias. If I buy a basket of stocks and I have an expected return for the group of 12% for example, and they all hit that mark, I should be pretty happy. But then you pull up the sp500 and see that it's done 20% and all of a sudden your returns become inferior. ***But, you didn't know how the sp500 was going to perform in advance so why benchmark to an unknown, moving target.***
I think this can be dangerous because now we all have this silent pull towards the index. Top comments in other investing subs are often "*just own the S&P 500, you dunce*". But doing that increases concentration and reduces diversification benefits because eventually it may matter, and is ultimately performance chasing anyway.
So I think it's worth considering something else to gauge peformance against.
# Alternative Benchmarks
Scratching out the S&P 500 leaves us with plenty of other options. One might be the ***equal weight index***. I think this is actually a great first step, but the law of averages will still make the winners overly representative (if 499 were flat and 1 went up 1,000%, you're going to lag unless you owned the 1 that carried the index - this a caricature, obviously, but it highlights that a small set of stocks can still carry significant influence).
You can ***compare against peers***. I.e., how did you do compared to professional active managers? But I think that can also be misleading because active funds often end up being closet indexers, so we can end up right back where we started.
Many funds will ***track against a target index***. I.e., if you're a small cap manager, you track your performance against the universe of small cap stocks. I think this is fine, but I often see portfolio drift. Like, I've seen RIAs talk about how they beat their 60/40 benchmark, but you look under the hood and it's because they allocated 80/20. Well of course you're going to beat a benchmark that has a bigger portion allocated to bonds... So unless you're exclusively going to buy from within an index (*small caps or industrials or whatever*), it doesn't make sense to narrow down your benchmark. And you still have to deal with the law of averages - small cap performance is likely still driven by a handful of small cap stocks.
I think the best benchmark is to use the ***median stock performance***. This answers "if you picked a stock at random, how likely are you to have outperformed that stock?" and it solves the equal weight index issue above - one stock can't decide the benchmark performance. At the very least, this could act as a Mendoza line, and answers whether you have any business picking stocks at all.
Or better yet, ***what decile do you rank in among the universe of stocks***?
I think ranking against the universe (*whether it's excess returns against the median or a decile rank*) is an objectively better way to track performance. We may want to downselect that universe - *filter out micro caps or only include profitable companies if that's a steadfast criteria for you* \- so more to be fleshed out there.
I think the big thing is to move away from the mega cap-weighted benchmark if you don't explicitly and exclusively own only mega cap stocks. Owning the S&P 500 is a terrific idea, and has it's own advantages (Nasim Taleb's 'winners win' observation), but if you want to feel comfortable holding diversifiers, you don't benchmark against the thing you're trying to diversify away from.
Some pitfalls for using the ranking method above is that we'd be comparing a portfolio of stocks againt a single stock. Now your portfolio has the advantage where one or two holdings can outperform the median in your universe. Maybe the answer to that is to rank your individual holdings. Or run some sort of monte carlo where the benchmark is the average performance for a random group of stocks (maybe 10-20 stocks)...not sure how materially different that would be from simply looking at the median. You also then have the question of holding period. The median pick turns over every year, but you might hold the same positions through many years.
Anyways, just kind of fleshing out some ideas. What do you guys think?