Klondike 2011

I still play Klondike Solitaire, but you’ll note there hasn’t been a peep out of me on this topic  in almost two years.

In my last 2009 post I noted that I’d achieved 50% win ratios but had then sunk back down to around 48%. That pattern has repeated itself several times since the 2009 post. I hit 50.4% today and had to check to confirm that’s a new personal record, though I’ve hit 50.3% at least twice before. It won’t last.

I’d also written that I had a new article on Klondike [strategy] in the works, but that I was putting it on hold. Now I’m just going to put it on ice.

The image below shows I’ve played over 3,000 games in the current “session” (and there have been others). So, if I’d been playing for a dollar a hand, I’d be about twelve bucks ahead. Not what you’d call a bell-ringer at the casinos.

Based on my own long experience, it’s my personal belief that Klondike Solitaire odds are close to exactly 50% if you do everything perfectly. I’ve been playng Klondike for the better part of 10 years, and I started saving off results in 2007.  It’s always the same old pattern. There is no mathematically possible way my current 50.4% win ratio has anything to do with a good luck streak or a bad luck streak someplace back in the past. If you start recording a new session, and win the first hand, that’s a 100% “win ratio” – but watch how fast it sinks back down to the same old 50% range!

I read a short story in The New Yorker the other week where the author wrote that he plays marathon Solitaire – a lot of it, like me. To him, it’s not about winning an individual game. It’s about the thrill of  getting on a roll and winning a streak of ten or twelve games in a row.

It is all much like surfing the Sargasso Sea. Once in a while you’ll get to ride a few gentle swells that roll in from a thousand miles away. And then it is over and you are in the doldrums again.

So it’s not about beating the house after all. It’s about relaxing, and maybe it’s about all the things you can think about while “wasting time,” like that book you’re trying to write.

168 total views, no views today

I Still Play Klondike

… and I have a major update wrapping up the wit and wisdom of my previous posts on Klondike Strategy. But that update has been gathering dust for months. I’m feeling rather humble about the whole thing.

You see, I climbed to over 50% win ratio on that same long streak of games I’ve been accumulating for years. So, I reasoned, I must be doing way better than 50% in order to raise the long-term average that much. And, I started a brand-new game under a new user name.

I haven’t been able to hit better than 48.3% since.

So, my “new strategy” makes a lot of sense but I haven’t seen any improvement in my score. I can’t even beat my old record. Maybe I was just riding the 100-year Tsunami.  I play more Sudoku as a consolation. Perhaps I’ll publish that new Klondike article one day. When, I just don’t know …

148 total views, no views today

Klondike Wins 49%

Klondike-49-0.jpg

Klondike-49-0.jpg

I don’t know what the official record is for this game, if it exists, but 49% is a record for me.

If you do the math, this isn’t really 49%, it’s rounded up from 0.4896675… or 48.966%. I wrote the SolSuite support folks about this. Their answer:

Our programmers have explained to me that it is possible to have only one decimal after the point, so the percetages [sic] have to rounded. Anyway, they know about your suggestion. Thank you!

I’ve worked with programmers most of my working life, and this answer is of course balderdash, though (as I just explained) not entirely unexpected from the profession. If the higher score is “better”, and you only support one decimal place, the answer should be reported as 48.9% (truncate excess digits; don’t round up).

137 total views, no views today

Fastest Klondike win

Klondike-48-4.jpg

Klondike-48-4.jpg

This Klondike win, with its new jump to 48.4% wins, was a long time in coming, but it was my fastest ever: 1.33 minutes. I slogged up through the wins by tenths from 47.7 to here, but it took over a month. As in the October Klondike post, I busted through a record and then fell back at every step. It takes patience – too much of it. My main strategy change is even more aggressive shuffling of cards from top to bottom, and vice-versa, to give myself every possible chance to promote just one extra card to the top foundation piles – while the iron is hot.

“Aggressive” means:  if you have a two of hearts burying its own ace at end of game, and if you can disassemble a couple of large stacks on the top foundation piles all the way down to the three of clubs, and play that deuce, you have another win. The alternative is saying you’re stuck and willing to lose the game.

179 total views, no views today

More Klondike

The thinner the ice, the more anxious is everyone to see whether it will bear. — Josh Billings

The mighty falleth, and riseth anew, so that we may see them fall once again. I’ve busted through this 47.7% barrier three times. Klondike isn’t just a great way to pass the time away. It’s a tough Solitaire variant requiring more concentration than you might suppose. If you play it long enough, you’ll see that even one oversight or mistake can make the difference between win or loss.

To get this measly one percent gain in the stats took a lot of work. I’m doing at least one thing different, by which I can account for the gain. I’m spending a lot more time interrupting myself, so to speak, to work the “tableaus” (7 piles on the bottom) – the cards I’ve already played. For example, say I need a five of clubs on the foundations (4 piles on top). I have it, but it’s buried by a red 4 of hearts and a black 3 of spades. I do have a five of spades I can get to, on the bottom of another foundation.  What to do? Simple, move the 4 of hearts (and the 3 on top of it) over on top of the five of spades.

It’s worth the extra effort.

Now I can move the five of clubs to the clubs foundation pile. It’s only one card. But sometimes it pays to get “greedy” while you can — if I waited, the five of clubs might get buried too, and then it may be too late.

UPDATE 10/28/2008

Doing this is also a double whammy. Not only do you build your foundations, you get to turn over the next card buried in the tableaus and play that too. At a minimum, you’ve parlayed this into two good plays. With a little luck near the end of the game, this can cascade into a win that otherwise would have been a loss.

And we knew there was a specific reason for mentioning this little strategy. See the image below.

This game isn’t even finished, but you can see it’s a guaranteed win: the tableaus are now all played out. All we have to do is move those cards up top to complete the foundations (or click “Game/Autoplay” from the menu, which takes the repetitious grunt work out of it.

This looked like a loss, but what I did is remove my six of diamonds (yes, the image is hard to read) back off the diamonds foundation, and put it down over my seven of spades. This allowed me to play my five of spades from one of the stagnant tableaus. Before I saw this move, I thought I was stuck with a loss. But, true to my earlier real-time prediction, this little stunt cascaded into a win situation.

Was it a win? Of course it was a win. When the big picture seems too cut-and-dried, it’s time to start sweating the small stuff.

127 total views, no views today

Klondike Wins

I thought I better post this now because I’m on another losing streak. I’ve been stuck in the 45.0-46.5% “wins” plateau for months, but briefly set a new record.

4/17/2008: It took me all this time to get back to the old record. You can see it took a number of additinal games, too.

6/7/08: Incremental gains. A new record, hard-won by a very small margin!

6/11/08: The audacity of hope. Chance of busting 47%: slim. Chance of backsliding: do the numbers.

6/14/08: The audacity of hope. I did it.

9/6/2008: like the last butterfly of Spring, you’d best get a last look.

9/11/2008: more ephemera.

115 total views, no views today

Klondike (Solitaire)

Late last year in My Notes, I posted a somewhat tongue-in-cheek tip sheet on playing Solitaire. Solitaire is a pretty simple game. I’ve been playing it in one form or another since I was a pretty simple kid. I play a variant called “Klondike” on a PC game application by Solsuite. Solsuite bundles a collection of over 400 card games into one slick, compact and inexpensive game package. I only play one of them, “Klondike”. At over 1,400 games, I’m forced to admit I take my Klondike pretty seriously.

This image also appeared in my glib My Notes post. Examining it for a minute, this game was lost with very little on the board.

For a person who spends so much time on one game, it took me quite a while to realize I had never really figured out logical strategic guidelines. I’d like to lay out excerpts from the publisher’s Help menu. Then, I’d like to set some high-level priorities, and finally, accumulatived tips I’ve been writing down for a couple of months now.

The basic difference between Klondike and the regular 3-at-a-time Solitaire I learned as a kid is that you get to deal from the waste (deck) one at a time, and you can recycle the pile until you either win the game or run out of plays, in which case you lose. When it’s this simple, you’d think there would be a tremendous advantage. Not so. Games that are won almost always developed a lot of action in the first couple of rounds through the pile.

RULES

Klondike type; 1 deck (52 cards); unlimited redeals
Chance of winning: Medium (about 1 in 5)
Object of the game: To move all the cards to the foundations.

definitions

Foundations (4 piles: complete these piles to win the game)

Build up in suit from Ace to King (for example, a 2 can be played on an Ace ).

Tableau (7 columns)

Build down in alternating colors (for example, a 10 can be played on a Jack ).
The top card of each pile is available for play to another tableau pile or to the foundations.
A packed sequence, or any portion of it, may be moved to another tableau pile.
Spaces may be filled only with a King or a King-sequence.

Stock

Turn up one card at a time from the stock to the waste by clicking.
Unlimited redeals are permitted until the game is blocked or won.

Waste (1 pile)

The top card is available for play to the foundations or to the tableau.

SolSuite Help, Copyright © 1998 – 2001 TreeCardGames.com. All right reserved. No portion of this help file may be reproduced in any form without the express written permission of TreeCardGames.com

PRIORITIES

  • I think your absolute top priority has to be to un-bury every possible card from the tableaus (7 piles) as quickly as possible.
  • The entire purpose of doing so is to move cards from the foundations to the top four piles (foundations), so that’s your second priority. If you “bury” a pile (tableau) with a card that can’t be played (say, the three of hearts, but you don’t have your ace or deuce of hearts in the foundation pile yet), you’ve locked up all the other cards in the tableau.
  • Keeping your tableaus “clean” is your third priority – nay, imperative – at all times.

Rule # 3 was the hardest for me. I trained myself to build elaborate ordered tableaus with cards wherever I could get them (from stock). Un-learning this has been the most difficult for me in actual practice. Don’t place a card on the tableau unless you are willing to bet you know what you’ll be able to do with it next. Betting that the ace of hearts will arrive in time to save you turns out to be a losing strategy.

OBSERVATIONS

Avoid the urge to play a card just because you can. Housekeeping and orderliness can freeze your tableaus so you can’t continue and lose the game.

1. Don’t freeze the tableau: Think ahead carefully before you “bury” a Tableau with a new card off Stock that’s in a suit for which you haven’t started a Foundation. If you play the deuce of clubs on the Tableau and haven’t an ace upstairs to start the required Foundation, you’ve frozen the whole Tableau. This is Rule #3 previously stated. I am coming around to the opinion that even medium and high-numbered cards (six, ten) need not be played from stock unless you see a good opportunity in doing so.

I think of it like the Japanese “just in time” inventory philosophy: if you have enough left-front fenders in stock to last to June 23, don’t have 4,000 of them on order to arrive April 15th.

Stock cards will always come back around the next time. If you can’t actually play off that card once you put it down, there’s no good reason for playing it right now.

2. Low cards are dangerous: The lowest numbered cards like the deuce and three, therefore, are particularly dangerous unless you have a specific reason to believe they’ll play back up to the Foundations quickly. Don’t play them until you really need them; they’ll still be there in Stock the next time around. You can build on fours and fives and sixes, so you’ll probably just play them, but you can’t build very far below a deuce or trey. I will play a three to move a deuce (and free up a tableau), but offhand I can’t think of any other reason why I would.

This perspective on Rule # 3 just restates earlier observations. Restating what I noted before, I am getting better results being stricter in this gambit.

3. Play it later if you don’t have to play it now. Leave all such “playable” dead-end cards in the stock (or waste) until you need them. Getting it on the tableau serves no purpose if you can’t do anything with it when it gets there! It’ll come around again. If you can play it the next time around, now harm done. And, if you still can’t play it when it next comes around, you made the right decision, didn’t you?

4. The fat lady really does sing. Sometimes the system will just hand you a win, and all you have to do is play it. Mostly, it will bring you to a point where you are sorely tempted to just quit and deal a new hand. Many of my games look hopeless but are won by fighting “tooth and nail” out of stagnant situations.

Extract the very last possibility of progress out of the game, even if you don’t see a breakout path exactly yet.

5. Un-do: You’ll make mistakes no matter how fresh your eyes are. Most games will allow “undo” (Control-Z) if you catch it right away.

6. “Reverse” is a gear too: Don’t be afraid to take a card back off the Foundation if you can use it to build a better situation on your Tableau.

7. Lopsided builds: Generally, your first priority is to build Foundations, even if you can also use the card to build a Tableau. But, an over-representation of a single suite in the four Foudations may also cause problems building and moving the Tableaus. For example, if you have Tableaus in all four suites Club-(4 each), Spade-(4), Heart-(4) and Diamond-(4), that’s as evenly proportioned as you can get. But, if you have Club-(2 each), Spade-(1), Heart-(0) and Diamond-(9), you’ve got a shortage of medium-numbered red cards to build your Tableaus. You might get away with it, but you might regret it. Don’t play the Foundations TOO far ahead of the Tableaus!

You can sometimes un-build a Foundation in order to play a Tableau card (usually to get at a new face-down card), but you gain a slight tactical advantage in building your tableuas “just in time” – unless you bury good foundation cards with unplayable ones!

8. Consolidating piles: Even in moving cards around on the Tableaus, don’t consolidate two piles into one until you need to. Doing so reduces the number of Tableaus in play. Again, tidiness for its own sake can ruin your day. If you can move a “packed sequence” onto another tableau pile, freeing a blank space for a King when you get one, resist the urge to tidy housekeeping. Just wait until you get that King. You may find a better opportunity for those two playable tableaus in the meantime.

9. Blank tableaus: suppose you have a blank (empty) tableau. The rules say this can only be populated with a king. Supposing also you have a red quenn high on another tableau. A red king comes up on the stock pile. Should you play it? I would wait and see if you have a black king, and then play that if you do.

10. Musical Chairs: Be aggressive about shuffling parts of the tableaus around to get at cards you can use to build the foundations.

11. If your game has an autoplay, turn it off. You should make the decision where each card is most needed and when it should be moved. Cards on the foundations, or eligible to go there, can still sometimes be just what you need to break up a logjam in the tableaus). You can use autoplay to quickly wrap up the game after all the face-down cards in the tableaus have been used up in play. Congratulations!

12. Accept your losses when you run out of plays: Some deals will be just awful and your game will get stuck before you really thought you’d begun. There’s probably nothing you could have done about those games. I find I can only predict a loss about half the time, so don’t toss in the towel early.

13. When you’re ready to toss in the towel, you have about an even chance of busting. So pay attention. You will win the hand the other half of the time – and the very “last chance card” is where I find at least half of all my wins. (3-4-2008)

14. Plug-Ugly Corollary: Very often I find that the really ugly game, the one you just can’t wait to bust on so you can start a new game with a decent deal … will string you along until all of a sudden your luck changes and the ugly game is a WIN. So just be patient and watch for openings like a hawk. (4-22-08)

IMAGES

Solitaire_Breakthrough.jpg – Here I broke a big losing streak with a bit of luck. The one remaining buried card, under the Ten of Hearts, has to be the Five of Diamonds. After cycling through Stock a couple of times, I was ready to give up, but found one last move in the tableaus which triggered a cascade of moves up top to the foundations. Here, I need to play the Nine and Ten of Hearts, and I’m home free. Once all cards in the tableaus are unburied, a win is guaranteed. You can now allow Autoplay to finish, saving you time. (Though the game is timed, those times aren’t recorded and don’t improve your points.)

Solitaire_Promising.jpg – This moribund game looked like it was going nowhere fast. Suddenly the tableaus (piles) opened up. I had all the covered cards un-buried in a few plays. Note the under-developed Foundation piles at top. It’s rare to be guaranteed a win with such a poor initial showing upstairs.

Alex Forbes © July 16, 2007

653 total views, no views today