|
I have a dilemma with this article.
The book under review is written in the form of a quiz, and each
of the 122 problems in the quiz deserves to be given your full
attention and careful consideration. Picking some example
positions and quoting from any of the solutions could spoil your
fun when you actually buy and read the book. Because, I can assure
you, this is a book which you will NEED to own if you are serious
about backgammon.
But I am getting ahead of myself. First, here’s a rundown of what
you get for your $45 (or, by the time it has got to you via Chris
Ternel’s BG shop, £40 or so). The book is a high quality softcover,
close in size to A4, and the same page size as Marty Storer’s
‘Backgammon Praxis’ books (actually US ‘Letter’ size, or 8.5 x 11
inches). However, at 303 pages, it is chunkier than either of the
two ‘Praxis’ volumes. The glossy and attractively designed cover
shows one of the positions (Problem 48) from the book, and this
particularly challenging problem never seems to go away. It is
repeated on the back cover, as well as on the first inside page,
daring the reader to stick his neck out and pick one of the six or
seven seemingly quite reasonable choices which the roll of 53
presents. Here is the position:
|
|
Without giving anything away about
how specific plays rank, I can tell you that the top five plays
span an equity range of about one fifth of a point! Choosing the
right game plan here makes a big difference.
Moving inside the book, the first 14 pages include a detailed
‘Methodology’ describing how positions were selected and rollouts
performed, and also explaining the meaning of all the figures in
the rollout tables for each problem. The eXtreme Gammon rollout
settings which were used throughout inspire confidence in the
results: for most play options this was 2592 trials, with variance
reduction, 4-ply for both checker play and cube actions, and huge
search space. The 95% confidence intervals are pleasingly tight.
The next 21 pages present the problem positions, with six per
page. They are all money play problems (no match play or cube
action), and there are no pip counts, themed headings or any other
clues to the solution. This enables the reader to try and solve
each problem without any bias, and is a great feature of the book.
I went through a page of six problems at a time before checking
the solutions. One thing I did not do, but which is recommended
(and which might have reduced my blunder-count!) was to set the
positions up on a board.
The bulk of the book comprises the solutions, with either two or
three pages devoted to each. For every solution the initial
position diagram is again shown, this time with pip counts. The
diagram is preceded by a themed heading and a sentence (or two)
which expands a little on the theme. For example:
Fight or Flight? All the safe plays leave you in some
danger. Should you strike first?
These introductions are a good reason for keeping the solutions
separate from the initial problems.
There follows the analysis, with discussions of the merits of
different types of game plan, and the choices of play associated
with them. This discussion is very detailed and extremely
thorough. If you liked Marty Storer’s Praxis books, you will
appreciate the effort put into this analysis. The authors home in
on the correct game plan by examining features of the position
including the race, timing, relative home board strengths,
relative priorities of your objectives, tactics versus strategy,
and the tried and tested safe versus bold criteria of Magriel.
Sometimes a problem will have only two reasonable (or even legal)
plays to choose from; sometimes six or seven. Frequently the
authors will examine how a change in checker placement, or cube
location (or imminent cube action), affect the correct play. These
insights are particularly informative.
The lack of typos or other errors was impressive and indicative of
the care which has gone into this work. I thought I had found an
incorrect count of shot numbers in one solution, but it was I who
had erred. Problem 92 failed to show the dice roll (44) at the
start of the solution. In one problem I thought that duplication
had not been mentioned when it perhaps should have been. Only
occasionally did I find that I was not completely convinced by a
solution, and that is perhaps more my problem than the authors’!
The last sentence or paragraph of the analysis summarises the
conclusion and confirms the best move, and this is followed by a
diagram showing the position after the correct move has been
played. Finally there are two tables of rollout data; the first
showing equities for each play and the second giving percentages
of plain games, gammons and backgammons. The difference in equity
between the top two plays is usually at least 0.060, and in many
cases it exceeds 0.100, so we are talking about second best plays
which are significant errors, and sometimes mega blunders.
Readers might think that I have been bribed if I don’t make any
adverse comments about the book, so I had better get a few
possible criticisms on the record:
1. Some of the analysis might seem too long and detailed for those
who prefer bite-sized bullet points. The format for every solution
consists of a themed heading, followed by an introduction,
followed by the analysis itself, and ending with a conclusion and
summary. This formula sometimes felt a little cumbersome.
2. To my mind, a couple of the positions were closer to ‘End Game’
than ‘Middle Game’ but who cares?
3. A few colloquialisms jarred a little (for me at least). For
example, “In the early game, if you have to choose between blah
and vicious, go with vicious!”
4. The typeface used in the main text is clear and easy to read;
however the rollout data is in a much lighter, less pleasing,
font.
To resolve the dilemma mentioned at the start of this review, I
will just present two more positions for you to think about
without quoting from the answers. Both are fairly
unremarkable-looking middle game positions of the type which crop
up frequently. In both cases the second best play is a blunder:
|
