Illustration 99 in Chess: East and West, Past and Present (New York, 1968)
Matt Goddard (Hiram, GA, USA) mentions a peculiar list of piece values on page 155 of Hoyle’s Complete and Authoritative Book of Games (the 1952 edition from Garden City Books, Garden City, New York, originally published in 1907 by the McClure Company):
Our correspondent writes:
‘Are there other examples where these were given as the common values, and where specifically did these values originate?’
In no other edition of Hoyle that we have consulted is the queen valued at more than ten pawns, and the evaluations given in the illustration above are certainly unauthoritative. Below we briefly list some further examples (pre-Second World War) of how this subject has been treated.
The ‘final relative values’ (page 136) were:
Pierce then explained the conclusions in the Biddle paper:
‘The author lays down the law that the entire value of a piece is the product of its separate values for attack and defence. He shows that the value of a piece for defensive purposes varies inversely as its entire value and directly as its mean scope. Hence the square of the entire value is equal to the product of its attacking value and mean scope.
Thus we get ultimately the entire values Q=10, R=5.84, B=3.53, Kt=2.89, P=.5.’
Page 83 of the March 1916 BCM gave a tepid reception to Vogler’s ‘curious’ article, although on page 125 of the April 1916 issue a reader, S.H. Gould, opined that the BCM had dismissed it too lightly. Vogler’s article was reproduced on pages 65-70 of La Stratégie, March 1916.
The above listings are, of course, by no means exhaustive but offer an overview of the old-timers’ attempts to allocate exact values to the chess pieces.
From Robert John McCrary (Columbia, SC, USA):
‘Page 57 of my publication The Hall-of-Fame History of US Chess (1998) has an article tracing the scale of powers that were given in Amusements in Chess and in Staunton to an earlier source: an 1817 edition of Philidor’s analysis, apparently done by Peter Pratt. In that edition, Pratt devoted 40 pages to mathematical analysis of various factors, leading to his concluding scale of Q=9.94, R=5.48, B=3.5, N=3.05, P=1. He suggested rounding off the scale, but Staunton and Steinitz printed it with full decimal places.
I have subsequently found that Pratt, as early as 1805 and probably earlier, gave two different scales, based on different lines of reasoning. The first scale had P=2, N=12, B=14, R=15, Q=28, K=9, whereas his second scale was P=2, N=9.25, B=9.75, R=15, Q=23.75, no K value. Apparently Pratt later developed the mathematical analysis printed in 1817. The same figures were used not only by Staunton in his Handbook but also by Steinitz (on page xxxiii of The Modern Chess Instructor), and ultimately became our modern scale.’
A question from Stuart Rachels (Tuscaloosa, AL, USA): who introduced the famous elementary system of values (pawn 1, knight 3, bishop 3, rook 5, queen 9)?
We can do no better than quote from page 59 of The Hall-of-Fame History of US Chess by Robert John McCrary (1998):
‘By the 1930s, the idea of a single numerical scale of values seemed to be losing favor. Instead, great players such as Capablanca and Tarrasch were generally avoiding giving a single scale of values in their books, preferring to discuss various kinds of exchanges on a case-by-case basis.
Then in 1942 Reuben Fine published his Chess The Easy Way. On page 23 is the following historically important section:
“Since there are six different kinds of pieces it is necessary to set up a table of equivalents in order to be able to know whether an exchange is favorable or not. Again such a table is based partly on the elementary mates (R can mate, B or Kt cannot) and partly on practise. If we take the pawn as 1 we may set up a table such as this:pawn = 1
bishop (or knight) = 3
Fine goes on to say that his table is satisfactory for rough calculation but not as accurate as a set of equivalents for certain exchanges he gives on the next page. Nevertheless, Fine apparently popularized our modern scale and indeed may have invented it. Does any reader know of an earlier publication of the modern scale? Perhaps Fine then should be considered the true “point-count” man of chess.’
Mark N. Taylor (Mt Berry, GA, USA) notes this passage on page 227 of A History of Chess by H.J.R. Murray (Oxford, 1913):
Robert John McCrary (Columbia, SC, USA) writes:
‘Lasker’s Chess Magazine published a series of copyrighted articles by Emanuel Lasker on basic chess instruction, and on pages 188-189 of the August 1905 issue he discussed the value of pieces. After describing the methods of Jaenisch and the mathematician Euler, he gave, on page 189, a peculiar scale:
“The values of the pieces calculated with sufficient accuracy for over the board play are as follows:
Pawn, 6; knight, 17; bishop, 17; rook, 24; queen, 47.
These values have not been found by executing the indicated calculation, but are those which come nearest to express my experience in hard fought games.”
Lasker then added some provisos and concluded by giving the king a value of 20.
I have yet to find any occurrence of these values elsewhere in Lasker’s writings.’
To the Chess Notes main page.
To the Archives for other feature articles.
Copyright: Edward Winter. All rights reserved.