Nitpicking the Author

Nitpicking the Author

| 2

It's difficult for a chess author nowadays. Someone will email "Fritz says your analysis on p.11, move 16, line 42 is wrong. Black's move should be 23...Rxf7! -0.34"

Just like the GMs whose games are followed online, and criticized by average players whose engines find every mistake, the author now has to watch each line of analysis because everyone can use engine "microscopes".

Even if the author and editor check every move by computer (we tried, using Houdini 2, to do this for in my new book The World's Most Instructive Amateur Game Book this fall but it's not easy...), they still have to draw a line on how much computer analysis is meaningful. And meaningful is not the only issue. Providing too many tedious lines may make the analysis more correct and complete, but it may also make it much more difficult to read and possibly even less instructive.

If a player chooses a move that is only 0.08 less than Houdini's, that is usually not too meaninful and the author might ignore or write "Possibly better is X" (Note: Houdini's evaluation may not be that accurate; for the big picture in evaluation refer to Steinitz, Zermelo, and Elkies at

However, if the player makes a 0.08 error it five moves in a row then, assuming the opponent played perfectly, the evaluation may go from 0 to -0.4, which could become very meaningful. Then the author may have to write "Without making an outright mistake, White has chosen a series of second best moves which have accumulated into a noticeable disadvantage." Smile