Rubber bridge strategy differs from matchpoints or IMPs in ways worth knowing:
Overtricks barely matter. Safety plays to guarantee your contract are almost always right — there is no field to beat.
Part-scores are precious. With 60 below, 1NT or 2♣ makes a game; opponents will compete fiercely to stop you. Light protective bidding gains.
Vulnerability shapes sacrifice maths. Once vulnerable, going three down doubled (−800) is rarely a good save against a game (~600 with the rubber equity).
Honours change close decisions. Holding A K Q J x of a suit, playing in your own trump suit quietly adds 100 to the ledger.
The rubber bonus dominates. Bidding thin games — especially when the opponents are vulnerable and you are not — is the engine of winning rubber bridge.
The general shape of it: rubber bridge rewards not losing. There is
no reward for a brilliant top and no punishment for a dull average, so the
percentage play, the safe contract and the quiet part-score are all worth more
here than the field-beating alternative would be at a duplicate table.
A part-score is a loaded gun on the table, and both sides should bid as
though it is.
When you hold the part-score:
Count how much you need. With 60 on, any contract of 2♣ upwards completes
the game — so a bid of 1NT or 2♦ that would be a modest part-score at duplicate
is a game try here.
Do not sell out cheaply. The opponents competing to 2♠ over your 2♥ are not
just buying the contract, they are protecting themselves from your game.
Be readier to double them. Their competing is often desperate, and their
undertricks are worth more to you than a thin contract of your own.
When the opponents hold the part-score:
Compete harder than the raw points justify. Letting them make 1NT for the
game is worse than going one down yourself.
Balance in the pass-out seat on values you would ordinarily pass — three
points lighter is a normal adjustment.
Consider bidding a game of your own on a hand you would otherwise invite
with: winning the game wipes their part-score off the pad entirely.
The app's engine knows the part-score: the current below-the-line totals are
passed into every robot bidding decision, so the robots really do compete harder
when they need less for a game. You can see it happen — the same hand bid twice
at different part-scores can produce two different auctions.
Every competitive decision in rubber bridge is a comparison between two
numbers: what their contract will score, and what your alternative costs.
The four vulnerability states and how they feel:
Neither vulnerable — the cheapest time to compete and to
preempt. Saves cost 300 or 500 against games worth 400-plus with equity.
You vulnerable, them not — the most dangerous seat. Your
penalties double while their sacrifices stay cheap. Bid your own hand and
double theirs; do not save.
Them vulnerable, you not — the best seat in the game. Save
freely, preempt aggressively, and bid thin games.
Both vulnerable — everything is expensive. Sound bidding
pays; speculation does not.
The sacrifice arithmetic. A non-vulnerable save against a
vulnerable game: their game is worth roughly 500–600 once the rubber equity is
counted, so three down doubled (500) is roughly break-even and two down (300) is
a clear gain. Vulnerable against non-vulnerable, the same three down costs 800
against a game worth perhaps 400 — a losing trade twice over.
Doubling. The most profitable doubles are of part-score
contracts the opponents have competed into, not of games they have bid
confidently. If they have found a fit and bid to their level, they will usually
make it; if they have been pushed one level too high, they will not.
The rubber bonus makes a game worth roughly 300 more than the trick score
alone suggests when you are non-vulnerable, and roughly 500 when vulnerable and
one game up. That figure is what should govern a marginal decision, and it says:
bid more games than a cautious player would.
A game that makes half the time is a fine gamble when it is worth 400-plus
and the alternative part-score is worth 60. The traditional rule of thumb — bid
a vulnerable game at 50% and a non-vulnerable game at 45% — is sound rubber
bridge and is more aggressive than most players' instincts.
Three practical consequences:
Take the push. If the opponents' competing has forced the
decision on you at the three level with 24 points, bid the game rather than
settle.
Prefer a major-suit game to a minor. 4♥ needs ten tricks
for a game; 5♣ needs eleven for the same reward.
3NT is often the pragmatic game even with an eight-card
major fit, when your hands are flat and the ninth trick is easier than the
tenth.
The counterweight: do not bid a game that jeopardises the contract you
already have when you are one game up and vulnerable. At that point you are
close to 500 for the rubber, and the safe route is worth more than the
speculative one.
Try it against three robots
Rase plays a full rubber, scores it the traditional way, and grades every bid and card afterwards.