Somebody says "double," and suddenly the whole table is doing arithmetic. Most players know a doubled contract is worth more. Far fewer know exactly how much more, and that gap is where a lot of points quietly leak out of home games.

I'm talking about the penalty kind here, not the takeout double that asks partner to bid. Once a double is left in, the scoring changes shape. The penalties aren't linear. That one fact explains nearly every doubled disaster I've watched unfold over a coffee table.

The doubled undertrick table, in plain numbers

Start with the case that stings most often: you go down after being doubled. Duplicate scoring charges you like this.

  • Not vulnerable: 100 for the first undertrick, 200 each for the second and third, then 300 for every one after that. Down one is 100, down two is 300, down three is 500, down four is 800.
  • Vulnerable: 200 for the first undertrick, 300 for each one after. Down one is 200, down two is 500, down three is 800.

Compare that with the undoubled penalties of 50 or 100 a trick that the basic scoring chart covers. Down three not vulnerable costs 150 undoubled. Doubled, it costs 500. Same cards, same play, more than three times the damage.

Redoubled? Take the doubled figure and double it again. Down two vulnerable and redoubled is 1000. You'll rarely see it. When you do, somebody at the table remembers it for years.

When the doubled contract makes

This is the half people forget, and it's the half that turns a quiet partscore into a swing. If a doubled contract makes, the trick points are doubled. Sounds modest. Then you notice what it does to the game line.

Take 2♥ doubled, making exactly. Undoubled, two hearts is 60 trick points, a partscore worth 110. Doubled, those 60 become 120, which crosses 100. You've just been doubled into game. Not vulnerable, that's 120 plus a 300 game bonus plus 50 for the "insult" (the bonus for making a doubled contract): 470. Vulnerable, it's 670.

From 110 to 670 on one card. That's why experienced players are so reluctant to double a low-level partscore unless they're sure it's going down, and why good doublers think about the downside before the upside.

Overtricks swell too: 100 each not vulnerable, 200 each vulnerable, twice that if redoubled. One doubled overtrick is often the difference between a fine score and a top.

The numbers that actually decide boards

Memorizing the table is fine. Knowing which numbers matter is better.

200 is the painful one. Most competitive partscores are worth 110 or 140. Step in over the opponents' 2♠, go down one doubled while vulnerable, and you've written minus 200 on a hand where they could only have scored 110 or 140. At matchpoints that's close to a zero, even though you were "only" one off.

300 is the defender's target. Say the opponents compete to 3♣ not vulnerable, and your side could have made 2♥ for 110. Doubling them only pays if they go down two (300). Down one doubled gives you 100, which is less than you'd have scored by declaring yourself. So a double that beats the contract can still lose the board. Odd, but true.

500 against 620 is sacrifice arithmetic. If the opponents are vulnerable and will make 4♠ for 620, going down three doubled at favorable vulnerability costs you 500, a profitable save. Down four costs 800, and suddenly it isn't. The post on sacrifices gets into the judgment side; the scoring side is just this subtraction, and it's worth doing before you bid rather than after.

Why it goes wrong at home

In plenty of home games, doubled contracts get scored slightly wrong. Nothing dramatic. Someone forgets that the second undertrick costs 200, not 100, or nobody remembers that 2♥ doubled and made counts as a game. A number goes on the pad, everyone nods, the cards get shuffled.

Then there's the bigger problem. Even with perfect arithmetic, one table can't tell you whether minus 500 was a disaster or a bargain. If every other table would have let the opponents score 620, your 500 was a win. If everyone else defended 3♠ and collected plus 140, it was a catastrophe. The raw number is silent on which.

That's the gap a tool like Bridge@Home closes for a home table: the board is scored properly (doubled, redoubled, vulnerable, all of it), and you see how your result compares with other players who held exactly the same cards. "We went for 500" stops being a vague regret and becomes a question with an actual answer.

A five-second habit

Before you double, or before you bid on over a double, run one quick sum. What's the score if I'm right? What's the score if I'm wrong? Now compare both to what you'd get by simply passing.

Most of the time the answer is obvious. Sometimes it isn't, and those are exactly the hands where the double card deserves a second look (partner will thank you, eventually).