Golf Solitaire Strategy Guide

Golf is a matter of fitting thirty-five column cards in between sixteen stock cards. This guide goes a step past "make long runs": what to count, which column to take from, and what to aim for on the rankings.

September 23, 2026 · 8-minute read

Start with the arithmetic: 35 cards, 17 starts

Golf on this site starts with one card already turned onto the waste pile and sixteen in the stock. Cards can only go to that one pile, so a game splits into runs that begin from at most seventeen starting points: the first card, plus one for every draw.

That gives you an average to beat. Even using the whole stock, 35 ÷ 17 ≈ 2.1, so you need a little over two cards from every start just to break even. To finish with eight draws you have nine starts and need about four cards from each (35 ÷ 9 ≈ 3.9). If you keep drawing and taking a single card, the arithmetic is already turning against you.

Your move count comes from the same sum. A win always takes 35 cards, so moves = 35 + draws. Finishing in 43 moves means you drew eight times.

A 2 for every Ace, a Queen for every King

Because Ace and King don't connect, an Ace in the columns can only be taken when a 2 is on top of the waste pile, and a King only when a Queen is. Once an Ace is on the pile, only a 2 can follow it, so one 2 can never carry off two Aces. In the end every Ace left in the columns needs its own 2, and every King its own Queen.

Everything you need for the count is visible: the column cards and the first waste card are face up, so you can work out exactly which cards are in the stock. Only their order is hidden.

Example: you can see three Aces and one 2 in the columns, and the first waste card is a 7. The other three 2s must be in the stock. Three Aces need three 2s, and four are available, so you have one spare. Take the column 2 onto a 3 without an Ace following, and the spare is gone: from then on, every time a 2 comes out of the stock, an Ace has to be sitting at the front of a column.

Every deal served here is solvable, so the count always works at the start. What breaks it is spending a 2 or a Queen for nothing. Before taking a 2, check that an Ace can follow it or that you have a 2 to spare.

Runs turn back at Aces and Kings

Only a 2 can follow an Ace, and only a Queen can follow a King. A run that passes through an Ace has to turn back, 2 → A → 2, and without a second 2 the run ends at the Ace. Keller's example 3-2-A-2-A-2 is what taking two Aces in one run looks like. Kings work the same way: Q → K → Q.

At a fork, the count decides. Say the waste card is a 2, an Ace and a 3 are both at the front of columns, and no other 2 is available. Take the Ace and the run is over. Go up with the 3 and the run continues, but this 2 is buried without carrying off an Ace. If you have no spare 2, take the Ace; if you have one, the longer run up from the 3 is usually better.

Kings are less scary here than the old books say

Morehead and Mott-Smith (1949) call Kings the most dangerous cards and advise saving Queens to take them off. Under their rules nothing may go on a King, so taking one ends the run. On this site a Queen can go on a King, so a run carries on if a Queen is showing; Q-K-Q-K takes two Kings in one go. The count itself still holds: one Queen per King.

Which column to take from

When the same rank is at the front of two columns, choose by these rules.

  1. Take from the longer column. Empty columns are never refilled. Taking from a column with cards behind uncovers a new card, so you keep the same number of playable spots; taking a column's last card loses one. Long columns also take longest to dig out, and Wolter's solver prefers them too. An empty column may sound satisfying, but it gains you nothing.
  2. Look at what you will uncover. If the card behind continues your current run, take that one.
  3. Dig out Aces and Kings. If neither card continues the run, uncover an Ace or King, or a card with plenty of neighbours showing (a 5 when several 4s and 6s are out). A buried Ace can't be taken even when a 2 turns up.

Read the whole run before you move

Every card is face up and you can change direction on any play (6-7-6-5-4), so plan the whole run before taking the first card. The first run you spot is not necessarily the best.

The waste card is a 5. Reading each column from the front (assume no other cards connect): column A 6·7·8, column B 6·5·4, column C 9·3.

(1) The obvious run: 5 → 6(A) → 7 → 8 → 9(C). Four cards.
(2) One change of direction: 5 → 6(A) → 7 → 6(B) → 5 → 4. Five cards.
(3) Down first, then up: 5 → 6(B) → 5 → 6(A) → 7 → 8 → 9(C). Six cards.

Line (3) uses column B's 6 and 5 as a stepping stone to climb all of column A. The 4 (B) and 3 (C) it leaves behind are connected too, so any draw from 2 to 5 takes two more cards.

As Morehead and Mott-Smith say, the longest run is usually the right one, though other things can matter more, especially early on. When runs are about the same length, prefer the one that leaves a connected group for the next draw: with an 8 and a 6 showing, uncover a 7 rather than an unconnected 2.

Handling the stock

Common mistakes

Practising with deal numbers

Play a nine-hole round

Borrow the scoring from the old books. Open nine deal numbers in a row (say #101 to #109). If you get stuck, score the cards left on the table (the "left" count at the top of the screen). If you clear it, score minus the cards left in the stock, which is 16 minus the draws shown on the results screen. Add up the nine. Morehead and Mott-Smith's par of 4 a hole (36 for nine) assumed nothing on Kings and deals that might be unwinnable. Every deal here can be won, so aim higher: zero or below is what Wikipedia calls a perfect round.

Replay the same number

The same number deals the stock in the same order too. Play it blind first, then open it again, this time knowing the order. Part of the difference in draws comes from not knowing the order; the rest is reading you missed. Finding the moves you could have avoided even without knowing the order, such as a wasted 2 or a short column cleared too early, is the fastest way to improve.

Play the same nine with a friend

In the 1949 Tournament Golf each player used their own deck, so luck of the deal crept in. Pick the same nine numbers and you both play identical deals. You can keep that book's scoring: one point for winning each hole, three for the lower nine-hole total.

The rankings: fewest moves and fastest time

Fewest moves means fewer draws

Since moves = 35 + draws, the moves ranking rewards saving draws; the results screen shows your draw count separately. Everything above helps: longer runs, saving bridge cards, taking from long columns.

Fastest time means less hesitation

The clock starts when the deal opens. The game ignores input while cards are moving, so setting the card speed on the start screen to Fast or Instant saves time. On a keyboard, Space draws from the stock. The biggest gains come from your head, not your hands: read a whole run, play it in one go, and stop to think only right after each draw.

Don't chase both at once

Saving draws takes long reading; playing fast means reading less. Time and moves are ranked separately, so it is realistic to play a number twice, once for each. Under Rankings, "By deal" shows one number's rankings, and "All deals" shows the best records across every number, with a "Try it" button beside each to jump straight into that deal. You need a name to submit records.

Play Golf now

Sources

  1. The Complete Book of Solitaire and Patience Games — Morehead, Mott-Smith, 1949
  2. Golf Solitaire Strategy Guide — Dan Fletcher, Solitaire Central
  3. Rules for Golf — Semicolon Software
  4. Golf Solitaire — Michael Keller
  5. Experimental Analysis of Golf Solitaire — Jan Wolter, 2013
  6. Golf (patience) — Wikipedia

Keep reading: How to Play Golf Solitaire · Golf Solitaire: History, Records and Competitions · TriPeaks Strategy Guide · Solitaire Glossary