A follow-up to some of the points made by other posters above.
1. Re: the market is a discounting machine - this is probably the lesson that I have learned the most over time. To make exceptional returns, you need to contrarian (and right). Part of the reason people who call themselves value investors have had such poor returns over the last few years is that a lot of people are looking at the exact same stocks and thinking the exact same thing. Consequently, all that information is priced in. My performance has really improved after figuring out what processes other investors were using to come to the stocks they ended up choosing (the short answer is virtually everyone does the same thing).
2. Having a sizable cash position - this is a very challenging problem. On the one hand, all historical data suggests that the long run returns on equities are substantial, so having a substantial cash position is a major opportunity cost. On the other hand, having been through March 2020, can clearly see that the mispricing is far greater in times of extreme pessimism than at other times. So the question really is how do you co-manage these two phenomena. John Hempton has one solution by running a diversified short portfolio that he re-loads into longs at these panic moments (which appears to be fundamentally a very intelligent strategy), but this is very very hard to do and also the short positions are a drag a lot of the time. Any long term active investor needs to have a framework by which to deal with this specific situation.
3. Diversification This is really a bigger question about risk management. Everyone needs to have appopriate risk management in place. IMO it is far better to diversify by having multiple sources of income that are unrelated to each other (even for instance, having a job unrelated to the capital markets such as in the public service) than worrying too much about the specifics of diversification. The truth is that correlations are pretty spurious at the best of times! I currently have only 7 stocks, of which 3 are >80% of my portfolio. Concentration focuses the mind. My simple rule for buying a new stock is- am I certain that this position offers better risk adjusted reward than adding to any current position. I also have in my head my weakest stock, which is by default my newest addition which has underperformed the most. Ie if I need to sell to buy a new position, I will sell this stock unless there are good reasons not to do so. I also try to sleep on these decisions.
4. Other investors The best investors I am aware of are not very public about it, for good reason. They also tend to be pretty independent, have a dose of humility and some other unique factors. I am unconvinced by many public figures, particularly in the context of the recent crisis, which almost noone called correctly, and the few that did appeared to do so for incorrect reasons. I can't say I have the same respect for some of the other public investors mentioned in this thread on those grounds. I will note that the other posters on this thread are almost certainly some of the best investors around, so I would certainly learn from them (and I do have a lot of respect for- even if I don't agree with them on a few things).
*** For what it's worth I use very simple first-principles approach to investing: 1. The signal for further investigation must start first on price - that is the clue to investigate further, and if the investigations find undue pessimism, you may be on to a winner. 2. A lot of market truisms are in fact largely true, but misapplied and don't apply to every specific situation: - eg mean reversion does apply (except when it doesn't) - the market is generally right (except when it isnt) - value beats growth (except when value is hidden and it's not a value trap) - compounding works (but only if you are right and you didn't overpay and you hold long enough) 3. Ideas are not generated out of an ether- figure out how you came to an idea, and if it is pretty easy for someone to come up with it with similar exposure, and that exposure is fairly easy to come by, the idea is most definitely NOT unique. 4. You need a systematic approach to both risk management and portfolio allocation which go hand in hand, that has to reflect the way the world is, rather than the optimal situation. This needs to take into account things that are real, such as tax, making a mistake, liquidity + your position size in the context of the volume, tail risks, risk of fraud and bad behaviour. - my approach is using Kelly principles based on actual risk versus reward calculations, where my at risk position is based upon the actual price that I think I could get being a relatively nimble player in an adequate liquidity stock. The reward is based on fundamental valuation (I use PEG as rule of thumb, which anchors very well to 2 stage DCFs for growth stocks) 5. Modelling error is bar the most fruitful way of finding stocks that are statistically likely to outperform. Specifically- is is when the majority of market participants do something in a certain way because it is the cookie cutter way of doing things, and a bespoke model is actually correct. If there is a very large discrepancy between the bespoke model and the standard model, extraordinary profits can be made. 6. Meta-cognition is critical- I need to step outside myself and objectively analyse what I am doing, to be certain that I am on the right path. There are ways to do this.
