What a Simulation Answers
A simulation answers a question your table judgment cannot: given what I know from the auction, which contract (or which opening lead) actually scores best? BridgeDummy generates up to 1000 random deals that match your constraints, plays every one of them double dummy, and aggregates the results into rankings. A 4-part demo video series accompanies this guide on the custom simulations page.
Each run costs 15 diamonds, charged only on success: if the run fails, the diamonds are refunded. Premium subscribers simulate without limits. You will find the simulator at bridgedummy.com/simulations after logging in.
The Two Modes
Bidding mode compares up to 4 contracts at once. For each contract you get average score, average IMPs, making percentage and average matchpoints, so you can see whether 4 spades or 3NT is the percentage action with your actual hand.
Lead mode flips the question: pick one contract played by your right-hand opponent and find which of your 13 cards is the best opening lead. You are always the opening leader, declarer is your RHO, and the engine tries every card in your hand on every generated deal.
Entering Your Hand
Click cards in the suit-by-suit selector until the counter reads 13/13. Set the vulnerability to match the real deal, since it changes the scoring of every contract. You can optionally enter some or all of partner's exact cards; for unknown hands, use constraints instead. The Clear All button resets everything.
Constraints: Describing the Unseen Hands
For each unseen player (Partner, LHO, RHO) you translate the auction into ranges: a High Card Points slider (0 to 37) and a min-max length slider per suit. Partner opened 1NT? Set HCP 15 to 17 and Balanced. RHO overcalled 1 spade? Set spades 5 to 13.
The rule of thumb: constrain only what the auction actually told you. Over-tight constraints leave too few matching deals, and you will see a low sample warning if fewer than 100 deals were generated.
Advanced Constraints
The Advanced button on each seat opens more options. Balanced requires a 4-3-3-3, 4-4-3-2 or 5-3-3-2 shape; Unbalanced excludes all three. Good Suit requires at least 3 top honours (A K Q J T) in every suit you set to minimum length 5 or more.
Key Cards counts all four aces plus the king of a suit you select (0 to 5), built for Blackwood decisions. Control in a suit means ace, king, singleton or void; Stopper means A, or Kx or longer, or Qxx or longer. Each of these has a negative version too, so you can require that a hand does NOT hold a stopper or control.
Sometimes one set of ranges cannot describe a bid, like a 1NT rebid showing 12-14 or 18-19. Use Add Alternative to create up to 4 options per seat. A hand matches if it fits any included option and no excluded one.
Conditional Contracts
A real auction is not a single guess: you bid 4 spades when partner shows support, otherwise you settle in 3NT. Conditional contracts simulate exactly that. In the bid editor, Add Conditions opens a rule builder where each rule tests an unseen hand (HCP, suit length, key cards, balanced, control or stopper, each with a NOT toggle) and names the contract to play when the rule matches.
Conditions inside one rule combine with AND; rules are checked top to bottom and the first match wins; an Else contract covers deals where no rule matches. On every generated deal the engine plays whichever contract your rules would have reached. In the results the contract shows with an asterisk, and hovering reveals how often each variant was chosen.
Reading Bidding Results
Contracts are ranked with four metrics. Average score is the vulnerability-aware bridge score across all deals. Average IMPs comes from per-deal pairwise comparison against the other contracts you selected. Making percentage is how often the contract makes. Average matchpoints scores each deal as a pairwise win, tie or loss.
Practical reading: at IMPs follow average IMPs and making percentage, since safety comes first. At matchpoints follow the matchpoint percentage, since frequency beats size.
Reading Lead Results
Lead simulations produce two rankings that measure different things and can disagree. Best Leads for MP ranks by average declarer tricks, where lower is better, because at matchpoints every overtrick matters. Best Leads for IMP ranks by defeat rate, where higher is better, because at IMPs beating the contract is almost all that matters.
Leads with identical numbers are grouped on one row, like the queen and ten from the same holding. The View Sample Hands buttons open 5 full deals showing the score of every lead, so you can see why the winner wins.
Templates and Saved Simulations
The Templates button opens a library of pre-built scenarios: each template loads a bidding sequence, contracts and constraints for every seat in one click, with search and type filters. After a successful run, the Save button stores the whole simulation (hand, constraints, contracts and results) under an auto-generated name; My Saved lists up to 30 of them for instant reload.
How the Engine Works
Deals are produced by constrained random generation across parallel workers, and every deal is solved double dummy: both sides see all cards and play perfectly. Real defenders are worse, so making percentages are slightly pessimistic, but the comparison between contracts is what matters. A run typically takes 20 to 60 seconds.
The page validates your setup before any diamonds are spent: impossible card counts, HCP totals above 40 across the four hands, contradictory shape flags and similar errors block the Simulate button with an explanation.