Once I actually started collecting and choosing model games from targeted players in openings I wanted to learn, I immediately started to see problems with my approach.

First of all, I didn’t have Anatoly Karpov around (or even the equivalent of Mr. Shorman) to explain the ideas behind his moves. I could immerse myself in the patterns, but without knowing why they worked I couldn’t reproduce them well in my own games. Karpov did annotate some of his games, but frankly I thought he wasn’t a very good annotator - many of the moves I least understood had very little annotation, and what annotation there was seemed cursory.

Second, often Karpov’s opponents played sub-optimally. They would make a move that I felt I could easily understand how to beat, but I had no idea what to do if they had played something else. Again, annotation helped a little there, but it really depended on the skill and interest level of the annotator and Karpov (and others) weren’t always the best.

Third, it was really hard to build an opening repertoire this way. The openings Karpov played weren’t being played much today, and if I just used Karpov games there were huge holes in the repertoire. I would have common situations that I had no idea how to handle because Karpov had never been faced with this thing - at least not his public, annotated games.

Fourth, and most importantly, I learned that I didn’t always like the decisions made by the player I was trying to emulate. Although Karpov was an amazing player, sometimes his decisions weren’t always the best. Even when they were objectively good, I learned that I had clear preferences when it came to chess, and they didn’t always match Karpov’s. Some of this was because Karpov was way better than I was, but some of this was because Karpov was different than I was - and I would be better served to honor my own personal preferences that were independent of strong chess play than to try to become something else.

This last point was the one that really iced the initial model games plan I had created, and also left me confused about how best to move forward. I didn’t have a flexible range of players to emulate - there were only so many strong players available, and none of them were me. How could I find a player who I would always want to emulate, and who could explain and annotate all of their moves, and who could map out modern openings?

In retrospect, the answer was obvious. The answer was Stockfish. Or, rather, the answer was me with access to Stockfish.

I would never be able to fully emulate Stockfish, because it’s inhuman and plays that way. It sees way more tactics than a human would ever be able to see, and doesn’t clearly identify long-term plans the way a human needs to. But “me with access to Stockfish” can make long-term plans. That player can explain what’s going on in a given position and can annotate thoroughly. That player will look at whatever opening lines I want him to look at. That player will always make move choices that I like. As an aside, that player may not be quite as good as Stockfish alone but is way, way stronger than Karpov.