The A-factor

Absolute playing strength is not an easy term to define. But it's mainly a question about the objective strength of moves played, and can to some extent be measured by today's strongest chess players - the PC chess engines. It would have been very interesting to magically put together chess players from different eras to fight it out over the chess board. That will never happen and that's exactly why it's such a great discussion subject.

My preliminary nomination for the top ten players of all times, based on actual peak playing strength is as follows:

1 Kasparov
2 Anand
3 Topalov
4 Kramnik
5 Fischer
6 Karpov
7 Shirov
8 Leko
9 Capablanca
10 Spassky

No doubt many players of the older generations would quickly adopt to the new situation and raise their level of play. It would not take them too long to catch up on the opening theory in their repertoire and - possibly more important, simply to realize that chess is a professional activity, requiring their full attention. That could even happen during the relatively short duration of a tournament or a match. Still I believe today's professionals, battle hardened from countless tournament games since their early youth and grown up with huge databases and strong analysis engines would come out on top.

The R-Factor

In a previous post (January 20th, 2007) I discussed the theme of "The Greatest Player", and claimed that one of the major factors that must be taken into consideration is a player's relative strength compared to his contemporaries - the R-factor. I believe this should be one of the easier lists to agree on. A very relevant source is the Chessmetrics site. Among other statistics, they provide lists of the best "peak years". After having set up my own preliminary list, I compared it with Chessmetric's list of "Peak Average Ratings: 3 year peak range", added a couple of names that obviously belonged in the top and did some other rather minor adjustments.

In the list below I have tried to take into account peak strength as well as the time span at the very top. I admit my methods have been far from scientific, but for now this is my suggested top-10 list of "Relative strength compared to contemporary competitors":

1 Garry Kasparov
2 Emanuel Lasker
3 Bobby Fischer
4 José Capablanca
5 Anatoly Karpov
6 Paul Morphy
7 Alexander Alekhine
8 Mikhail Botvinnik
9 Wilhelm Steinitz
10 Mikhail Tal

These just missed my top 10 - mainly because it's not clear at what time they would have a claim to being the clearly strongest player on earth:

11 Tigran Petrosian
12 Viswanathan Anand
13 Vladimir Kramnik
14 Vassily Smyslov
15 Akiba Rubinstein

Compared to Chessmetrics, the most obvious difference is that Morphy made it to a 6th place in my list, while ending on a modest 66th in the "3-year peak range". I must admit there is some doubt about his world dominance at his time. But I still cannot see who of his contemporaries that could have given him a good match when he was at his peak around 1860.

My next entry on this subject will probably be on the "A-factor" - a list of the objectively strongest players based solely on the quality of their moves. I will not go into details yet, but suspect that most of the players on that list will have been active during the last decade or so.

The Greatest?

A favorite pastime for chess fans (as for most other fans of competitive activities, I guess) is discussing who was "the greatest player of all time". I plan to return to that subject as soon as I have worked out my own lists.

For the moment I only want to share some thoughts on method:

I believe that "greatness" consists mainly of three components:

1. Relative strength compared to one's own generation of competitors. I will call this factor 'R'.
   
2. Absolute strength (the quality of moves when the complexity of positions is taken into account). I will call this factor 'A'.
   
3. Potential strength (estimated strength if given the same conditions as players from other epochs or areas). It should be noted that this is not exactly the same as talent, as work ethic is an important ingredient in potential strength. I will call this factor 'P'.

At least for the two first of these components it should be possible to work out a list based on fairly objective criteria. When it comes to potential strength, there obviously is more room for discussion, but that's after all part of the charm.

I suppose the main question is how these factors should be weighted and combined into one final ranking. For starters you could just give all three the same weight, simply to see whether the result looks reasonable. Then I suppose you must decide whether to add the numbers or multiply them. Probably the results for the top contenders will not be too different however you choose to do it, but instinctively I prefer multiplication (if for no other reason, because I called them 'factors').