Win stats can be faked? Log Out | Topics | Search
Moderators | Register | Edit Profile

Goodsol Forum » Pretty Good Solitaire Questions » Win stats can be faked? « Previous Next »

  Thread Last Poster Posts Pages Last Post
Is this fixed?Ken Millar 15 1-08-10  5:39 pm
  ClosedClosed: New threads not accepted on this page        

Author Message
Jeff Dunn (Jdunnpm)
New Solitaire Player
Username: Jdunnpm

Post Number: 1
Registered: 5-2007
Posted on Friday, June 01, 2007 - 7:52 pm:   

Am I right in stating that you can play the same game over and over and build statistics, i.e., there's no difference between playing 100 different games of the same type and playing one numbered game 100 times?

This is the only way I can explain people posting 100% stats for How They Run. At least 10% of those games cannot be won, as far as I can see.
(This must have been covered before, but I don't see it looking at past discussions.)

Jeralyn Taylor (Annika)
Advanced Solitaire Player
Username: Annika

Post Number: 95
Registered: 4-2004
Posted on Friday, June 01, 2007 - 9:18 pm:   

Unfortunately, yes. I cannot understand why someone would want to do this, but it is surely possible. There are some games, which I believe are 100% winnable, such as Maze; and others, for which other people post 100% scores, and I cannot see that it could be done at all, other than than to repeat a winning game or games enough times to post.

Later you may want to try Action Sol, in which repeating a game is not possible, and one plays timed rounds to post high a cumulative score.

I had not played How They Run, and I like it. I played 4 games before I came to one I have not been able to win. Yet.....
Richard Mechen (Richardscotland)
Master Solitaire Player
Username: Richardscotland

Post Number: 772
Registered: 9-2006
Posted on Saturday, June 02, 2007 - 3:48 am:   

The simplest and laziest way to fake your stats is to find a winning game and then save it. Next clear your stats for that particular game. Now open the saved game and it will record as a win in seconds. Repeat as often as your moronic mind allows.
Yes, as you've both observed, some stats are works of pure fiction, but on the other hand some 100% ones are entirely genuine. The problem is you can't tell which is which. This devalues the efforts of people who have possibly struggled for days, perhaps weeks, to win a problem game to maintain their 100% record.
Just ignore the stats. YOU know if you've not cheated, and that's all that matters.
Jeff Dunn (Jdunnpm)
New Solitaire Player
Username: Jdunnpm

Post Number: 2
Registered: 5-2007
Posted on Saturday, June 02, 2007 - 3:53 pm:   

Well then, I would charge that the two top scores for How They Run are faked, and should be removed.

In How They Run, you get deals 5-15% of the time where cards can't be moved off the free cells, and the three right tableau rows get blocked. An example is game #1723319423, which I'm sure cannot be won.
Richard Mechen (Richardscotland)
Master Solitaire Player
Username: Richardscotland

Post Number: 776
Registered: 9-2006
Posted on Saturday, June 02, 2007 - 5:05 pm:   

Hi Jeff, Whilst I sympathise with this position we don't know which games the people in question have played, and without that knowledge it can't be proved that they have encountered any unwinnable games. If you were to take sufficient time to examine each game before you made any moves, you would probably see a great number that appeared to be unwinnable. You could then move on to the next game, and the next and the next until you got one you were happy with. You could then amass a very high percentage without doing anything unethical. You have actually made a libellous statement which I'd advise you to retract. Give them the benefit of any doubt. It's only a game.
I agree that #1723319423 is probably unwinnable.
Jeff Dunn (Jdunnpm)
New Solitaire Player
Username: Jdunnpm

Post Number: 3
Registered: 5-2007
Posted on Saturday, June 02, 2007 - 9:32 pm:   

I'm not out to point fingers, just to get familiar with the statistical rules and options players have. Other online sites, e.g., freecell.org, won't count repeated games in the stats.

Now in the case of just the game in question, How They Run, I would submit that it IS unethical to shop for an easier game. The reason is that you can tell immediately whether or not the two cards in the free cells can be moved off. In almost all cases, as far as I know, if you can move one card off, you can win; if not, you have a high probability of losing.

Another method of score faking that occurs to me is that a person could have two IDs, two copies of Goodsol, and only play games under one ID that they had won under the other.

Anyway, I agree it's only a game. Perhaps I myself play unethically because I use the Undo button all the time. But my goal is to win any game that's possible to win and reduce the effects of chance as much as possible. However, I do accept the initial deal, because that's the only way you get an idea of the statistical winnability of a game. It seems to me that the chances of winning quoted in the rules are too low, and perhaps don't take Undoing into account.
Richard Mechen (Richardscotland)
Master Solitaire Player
Username: Richardscotland

Post Number: 777
Registered: 9-2006
Posted on Sunday, June 03, 2007 - 4:06 am:   

Hi Jeff, There are many, many ways to fake your stats in PGS. I know of several that you've probably not even thought of. The less said about that the better.
Some players would agree that shopping round for soft games is unethical. Purists would argue the case that the Undo button should be removed. If this was the case you would have players wtiting down all their moves and then replaying them differently. Whatever your point of view, there is nothing in the rules to say that this is illegal, and the referee couldn't whistle for a foul.
In the tiny sample of these games that you've played, it is possible to get a very inaccurate win percentage, and it's therefore possible to win 100 or more consecutive games.
At the time of my last post to you, I hadn't bothered to consult the stats to see who you were complaining about. Now that I have, I contend that it's not unreasonable for the board leaders to have achieved this completely ethically. Furthermore I contend that YOU could do it too using your own high standards. It may take a LONG time, but by clearing your stats if you hit an unwinnable game, you will eventually hit a long winning run.
The stats used to bug me too. I cannot abide injustice. I have now formed the opinion that all we are trying to say in the stats is "I'm better than you", and by so doing we are making some decent people feel inferior. We're making them unhappy. I don't think that's what any of us want.
If you really want something to infuriate you, try a sample of 50 games of Robert. If you win any of them I will personally send you $200. PROMISE. After you have failed miserably, which you will, have a look at the online stats. I guarantee that you will NEVER get anywhere close to the board leader, supposing you outlive Methusalah.
Nice hearing your opinions.
Richard Mechen (Richardscotland)
Master Solitaire Player
Username: Richardscotland

Post Number: 779
Registered: 9-2006
Posted on Sunday, June 03, 2007 - 3:44 pm:   

Hello again Jeff, I decided to see if I could reach 50 without a loss using your ethics. I got to 14 before I came unstuck. Try 1864276735 and tell me what you make of it. I'm thinking of wiping my stats again and seeing if I can get at least 15, and so on. It's an enjoyable game and it forces you to think your moves through.
Jeff Dunn (Jdunnpm)
Junior Solitaire Player
Username: Jdunnpm

Post Number: 4
Registered: 5-2007
Posted on Monday, June 04, 2007 - 12:46 am:   

I took a look at the game, but didn't play it because I didn't want to affect my stats. I don't see how it's winnable: the lack of the 6 of spades, 2 of spades and 7 of clubs in the layout makes it impossible to remove any of the key cards and free up hidden ones.

My guess is you'd have to reset your stats at least 50 times before you could play 50 games without a loss. I'm not a statistician, but this could be estimated by some standard formula assuming two outcomes, one of which is 20% failure, and determining the probability of 50 non-failures in a row.

You're right, Richard, the game forces you to think your moves through. That's why I like it. Opening up a free cell is the highest priority. Even if you've exposed all the cards, you have to be careful how you sequence them so you won't get blocked.
Richard Mechen (Richardscotland)
Master Solitaire Player
Username: Richardscotland

Post Number: 781
Registered: 9-2006
Posted on Monday, June 04, 2007 - 3:16 am:   

Hi Jeff, Create another player and try the game under that name. That way your stats won't be affected. Your analysis of the game is way out however. I scored 50 with it, but at no time could I free up both cells in order to sort out the low clubs. Uncovering the hidden cards wasn't a huge problem.
You could do me another favour though. I love messing with other people's losing games. Could you possibly post a list of the games you lost?
Your online stats show 12 losers. It would be nice if we could establish if all of them were definitely unwinnable.
Thomas Warfield (Support)
Moderator
Username: Support

Post Number: 1144
Registered: 12-2002
Posted on Monday, June 04, 2007 - 12:25 pm:   

There certainly are ways that people can cheat the statistics. I've tried to make it as difficult as possible, but there are ways that you simply cannot stop programmically.

The playing of the same game over and over again is an example. The only way to stop that would be to not allow the same game to be in the statistics more than once, but there are serious running time issues with that when you get up to thousands of games played. It would simply slow the game down too much to do that.

Another way of cheating that some people have found is to simply turn off the computer every time you have a losing game. Simple and nothing the game can possibly do about it. I wouldn't recommend this, it would hardly be worth the wear and tear on the computer.

Nevertheless, the vast majority of people are honest and most of the scores, even the ones that look incredible, are legitimate. Some people have come up with remarkably good strategies for certain games. When I see a score that is clearly not legitimate, I delete it.

As a shareware author, my philosophy is simply not to worry about the dishonest people and just deal with the honest people. This is easy because in shareware, the dishonest people simply aren't customers to begin with.
Jeff Dunn (Jdunnpm)
Junior Solitaire Player
Username: Jdunnpm

Post Number: 5
Registered: 5-2007
Posted on Monday, June 04, 2007 - 1:14 pm:   

Hi Richard,

Sorry, I don't know how to create another player. I'm using the client version. Were you playing an online version?
Jeff Dunn (Jdunnpm)
Junior Solitaire Player
Username: Jdunnpm

Post Number: 6
Registered: 5-2007
Posted on Monday, June 04, 2007 - 1:17 pm:   

Oh, I found it in the Options--sorry again.
Jeff Dunn (Jdunnpm)
Junior Solitaire Player
Username: Jdunnpm

Post Number: 7
Registered: 5-2007
Posted on Monday, June 04, 2007 - 2:10 pm:   

OK--I played 1864276735. My first analysis was completely wrong--it applies to a different game. I had entered the wrong game number from my note pad (1723319423-a horrible game I could only score 9 on). Your game 1864276735 is an interesting game in that you can free up one of the free cells, but never both, and it takes both cells to break up the initial club configuration of 4-2-3. I too could get no more than 50 points.

The other games I've lost are (with scores in parens): 1010235392(42), 1848783359(45), 1298580992(46), 2142861567(37), 1092242560(31), 2113591295(25), 1966436479(37), 330104193(43), 1297409024(25), 214250753(15), 1141217024(23). Now I'll replay them under another player as you suggest.
Richard Mechen (Richardscotland)
Master Solitaire Player
Username: Richardscotland

Post Number: 783
Registered: 9-2006
Posted on Monday, June 04, 2007 - 3:14 pm:   

Thanks Jeff. Opened a new thread under "Is It Winnable?"
Mike Bailin (Mikeb)
Master Solitaire Player
Username: Mikeb

Post Number: 102
Registered: 3-2003
Posted on Tuesday, June 05, 2007 - 10:30 am:   

Slightly (or more than slightly) off-topic, this relates to the expected-percentage thing in the rules.

In many games there's options to make the game easier, most commonly turning the King-Only requirement (for empty columns) off. We've discussed elsewhere that the change in difficulty really makes them totally new games which in fairness shouldn't have their stats compared with the default settings. But Thomas has pointed out that it would be an outrageously complex task (and more importantly would require much larger databases) to keep separate sets of stats for these.

Back to where I started. Is it a reasonable guess that the expected-percentage in the rules is based on the default settings?

Jeralyn Taylor (Annika)
Advanced Solitaire Player
Username: Annika

Post Number: 97
Registered: 4-2004
Posted on Saturday, June 16, 2007 - 3:42 pm:   

I have been trying to see whether I could get 50 wins in a row at How They Run, and today I have 49 wins, and #50 is not winnable! I cannot get even one card off the 2 cells. Want to try it? Number 1784881663.

I have been playing every game as it comes up, skipping none, and while some are easy, some are the very devil to find the winning moves.

The game is a lot of fun. It was new to me when this discussion started.

Jerri
Richard Mechen (Richardscotland)
Master Solitaire Player
Username: Richardscotland

Post Number: 862
Registered: 9-2006
Posted on Saturday, June 16, 2007 - 4:12 pm:   

Hi Jerri,
Looks like a duck of the deceased variety.
Kathy Quade (Kathyquade)
Intermediate Solitaire Player
Username: Kathyquade

Post Number: 47
Registered: 2-2006
Posted on Sunday, January 10, 2010 - 8:34 pm:   

I learned today of different ways players cheat, i.e., using different names and playing almost the same games giving them consecutive top score wins, etc. After this, I'm ignoring those ridiculous top scores (unless they look legitimate), but still in the spirit of competition, I'll base my goals on topping the list of the legitimate players. That way I can be honest and still compete for high scores, and have fun at the same time.
Kathy Quade (Kathyquade)
Intermediate Solitaire Player
Username: Kathyquade

Post Number: 49
Registered: 2-2006
Posted on Sunday, January 10, 2010 - 11:46 pm:   

How can you have more than one score on the same game using the same name and location. I thought whatever your last statistics posted overwrote the other score. I don't want to do it. I'm just curious as I just saw some with three top scores on the same game for three different dates. Also, just for a laugh, I found today that someone played a game over 50 million times (maybe more). Is that possible? This is my day to just browse around after playing, and I've found some strange things.
Richard Mechen (Richardscotland)
Master Solitaire Player
Username: Richardscotland

Post Number: 3683
Registered: 9-2006
Posted on Monday, January 11, 2010 - 5:43 am:   

Look more closely and you should see subtle differences, usually in punctuation, when a player has been listed more than once.
I'm glad that you've come to terms with the less credible scores in the stats. Just competing with the ones you believe is a good policy.
Ken Millar (Tpa_ken)
Master Solitaire Player
Username: Tpa_ken

Post Number: 1285
Registered: 2-2006
Posted on Monday, January 11, 2010 - 6:24 am:   

Also, I believe Thomas once said that if one submits stats from another computer even if the name and location are exactly the same, it creates another posting.
Gregg Seelhoff (Seelhoff)
Master Solitaire Player
Username: Seelhoff

Post Number: 416
Registered: 1-2003
Posted on Thursday, January 14, 2010 - 11:06 am:   

>How can you have more than one score on the same game using the same name and location.

One can submit from different games (say, PGS and GS101), or one can submit from different player names. I, for instance, have a 'Gregg' player that has serious scores, and also a 'test' player for testing, verifying bugs, etc. I also (of course) have copies of all of the major games installed. The submission information can be identical for all of these, but the different player/registration information prevents the server from overwriting the scores. (You may have noticed that the climb mode pages display the game and version from which the scores are submitted.)

[Note: I do try to put "(test)" at the end of my testing submissions, and I always plan to delete these, but there are still some games where my test entries are the bulk of the high score table. :-)]

Topics | Last Day | Last Week | Tree View | Search | Help/Instructions | Program Credits Administration