the situation, may be briefly explained as follows: Before a game has
been won, the securing of a large bonus in the honor column places the
fortunate doubler in a most advantageous position, as he starts the
rubber insured against loss unless he suffer a similar penalty.

When the only game finished has been won by the adversaries, a large
bonus should be preferred to game. As the adversaries already have a
game, the next hand may give them the rubber, and should it do so, its
amount will be most materially affected by the action of the player who
has the chance either to score a bonus or win a game. If the first game
be of normal size, a large bonus will nullify the result of the rubber,
but if instead a game be taken in the adversaries will score an average
rubber.

When the player considering a double has a game and the adversaries
have not, he is in a most excellent position to double with the hope of
a big winning. To secure the enlarged rubber, it is only necessary for
him to obtain one game before the adversaries get two, and as the odds
are greatly in his favor it is a chance worth taking.

When, however, each side has a game and the question is whether to
obtain a bonus or score rubber, the bonus must be large and sure to
justify giving up a rubber practically won for merely an equal chance
of capturing a larger one. It has been elsewhere stated that when a
player who has an opportunity to win a rubber fails to avail himself of
it, and on the next hand the adversaries reach the goal, the loss may
be roughly estimated at 600 points. The player who doubles during the
third game knows that the next hand may see the adversaries score the
rubber. Even if he obtain 400 points by doubling, and this happens, the
adversaries gain to the extent of approximately 200 points by his
action. On the other hand, he has an equal chance for the game, and if
he win it, he will be the gainer by the amount secured by the double.
When he has a sure 400 in sight, or even a sure 300, with a reasonable
chance of more, the odds favor the double, but it is the height of
folly to take an even chance of losing 600 unless 300 be the minimum
return.

Advice as to whether to double or go game is useful only for players
who can with accuracy estimate the trick-taking value of their hands.
To refuse a double which would net several hundred for the sake of
going game and then fall a trick short of both the game and the
declaration is most e