It will be very useful for me if you can tell me the conflicts that you see. I maintain a very detailed account of the kind of bet I am making (shown below). My corpus size allow me to be flexible with my investing decisions i.e. I don’t need to own only high quality businesses. I am happy with lower quality if risk-reward is in my favor.
I haven’t come across any research that shows owning a large basket of stocks leads to market performance. What you are probably referring to is the number of stocks an investor needs to have adequate diversification, which is very different from tracking error. You can have a 1000 stock portfolio and have a large tracking error from the underlying index. Past academic research have identified a few factors such as quality, company size, valuation, etc. that can help explain excess returns. In order to harness one factor, we need a large basket of stocks (atleast 100), that’s the fundamental attribute of long-short strategies that try to harness a given factor.
Coming back to reality, my thumb rule is to keep 20-30 stocks with largely diversified cashflow (geographic, industry, etc.). This provides adequate diverification and I can track them reasonably well. Plus, I will be glad if you can help me identify unidentified risks 