Subject: I'm sorry to have to bring this up . . .
Date: Fri Mar 31 15:49:27 2017
User: ConcernedFreeceller
Message:But I feel I have to. I have been playing Freecell on this site for a long time and I have recently noticed something that doesn't seem quite right. In the 8x4 where I play, a large percentage of the games I am dealt have solve rates in the 80-100% range, which I know is normal, but I also am dealt a fairly large number of games with solve rates that are much lower than 80%. In fact, I routinely am dealt games with solve rates in the 20%, 30% 40% and 50% range, ie, tough games. Once in fact, in the span of 30 games, I was dealt a game with a 12% solve rate and then another with a 7% solve rate. I suppose this is all normal too but here is the source of my concern.
The leader in the 8x4 is Ki. For a long time I have been monitoring the games this player is dealt and what I found is very different from what I described above. The vast majority of the games Ki plays have solve rates above 90% with most of the rest being in the 80-90% range and, occasionally, a game below 80%. Even the games that are below 80% are usually in the 60-70% range. I have looked at hundreds of games that Ki has played and rarely have I seen a game with a solve rate below 60%. I don't understand how this can be as I regularly am dealt games with solve rates below 60% as noted above. Earlier today I was dealt a game with a 33% solve rate and then just a few games later, one with a 26% solve rate. If the games are dealt randomly to every player, then over time every player should see the same basic mix of easy games and hard games, but that does not appear to be happening. I feel that I am playing a much tougher mix of games than the current leader is playing. And to reiterate, I am basing this on analysis of hundreds of games played, both by me and by Ki. Over hundreds of games the solve rates should even out, but they aren't. This pattern has held for a very long time.
I really hesitated to post this and in fact, I waited a long time to do so. I know this is provocative but I feel it needs to be said. I have posted this under an anonymous name because it is so provocative but will gladly share my player name and info with Freecell management if requested.
Thanks,
A Concerned Freeceller
Date: Fri Mar 31 19:52:43 2017
User: joeygray
Message:I just checked. Please be aware that out of all of the 32767 8x4 level 10 games, which are all the ones Ki plays in his streak, only 1450 have win rates below 70%. This is less than 4.5% of them. I also just checked Ki's last 100 games as of ten minutes ago, and you are right that he doesn't have any of these 1450 games in his list. I'm pretty sure that you would see this 'most' of the time you looked, though.
Denny doesn't save history past the last 100 games you played, so I can't look back any further and neither can Denny without going to his backups.
I could do a similar check for you if I knew who you were. It would be a bit harder because I bet you play various levels between 5 and ten since you probably lose games like most of us. Let me know if you want me to.
But also be aware, that level 5 games have very misleading win/loss records w/r to actual difficulty. Many newbies have played here and if you are not good (shall we say, "not yet good") you will be playing a lot of very winnable level 5 games, and losing them.
Date: Fri Mar 31 22:08:39 2017
User: joeygray
Message:Yeah, so the level 5 newbie bias is not, on the surface, as bad as I made it out to be: 4435 of those are less than 70%, or about 13.5% of them. That's three times as many though.
Date: Fri Mar 31 23:51:44 2017
User: jamesblackburn-lynch
Message:I was going to pooh pooh this with some probability but it wasn't as straightforward as I thought. The level 10 games are mostly 90% success rates and if you play those for the most part (like Ki or anyone at the top of the leaderboard does, almost exclusively) it is rare you run into any 20% or 30% games like are all over the place in the earlier levels.
But in a run of 100 games to see nothing less than 70? Not unheard of but after verifying that that is the case for Ki, I checked the number two guy, rgk16, and he had four games in the 50's in his last 100. Then I looked at a few more up at the top and others had 50's and I even saw a 30. I was starting to really buy into it until I looked at Astronz and his/her numbers were much like Ki's.
Now, OP says this has been going on for some time. If so..that is weird. But what is the suggestion? That Ki has hacked the code to only give him/her occasional 70's? Why not only 90's?
I am curious and will probably check again in a couple of days but I still think it is highly unlikely. We could calculate some expected values and see what the variance is.
James
Date: Sat Apr 1 05:38:37 2017
User: Goosey_Goosey_Gander
Message:I can't explain CF's problem. I have played 11,345 games of 8x4 with a 99.19% win rate. I keep a record of games I solve with less than a 40% win rate, 19 in total or 0.17%. Of those the lowest win rate was 21.7%, I don't think I have ever seen an 8x4 with a win rate less than that. I might have missed recording a couple. What about the games I lost? I record most of those too. In the last year I have lost 8, with win percents ranging from a low of 33% to a high of 95%, average 63%. Yeah, carelessness is the problem, not super difficult games.
Date: Sat Apr 1 11:43:50 2017
User: ConcernedFreeceller
Message:Thanks for your replies. First, let me say that I am not suggesting that Ki has hacked the code or cheated in any way. I am just pointing out that he/she seems to get an inordinate number of easy games, at least compared to what I am seeing. Last evening I looked at the most recent 100 games Ki played and the most recent 100 games that I played. Here is what I found:
Ki played 81 games with a solve rate of 90% or higher and 15 games with a solve rate between 80 and 90%. That's 96 of the 100 games over 80%. The other four games were in the 70-80 range and none of the games was below 70%. Of my 100 games, 72 were over 90%, 14 were in the 80s, 10 in the 70s and 4 below 70: 1 with a 51% solve rate, 1 with a 41% solve rate, 1 with a 32% solve rate and 1 with a 26% solve rate. So I played 4 extremely tough games compared to 0 for Ki, plus 10 games in the 70% range compared to just 4 for Ki. And as I noted in my earlier posting, this has been the pattern for a very long time. I am seeing quite a few games with solve rates below 50% but hardly ever see them for Ki. The law of averages says this shouldn't be so.
Lastly, I am not a newbie. I have been playing Freecell on this site for over 15 years and I have some of the best all-time records in the 8x4. And I only play the 8x4.
Date: Sat Apr 1 12:00:32 2017
User: jamesblackburn-lynch
Message:So then..what is the "concern" if it isn't hacking or cheating. Just that Ki is extremely lucky?
I don't think your numbers for the last 100 are very shocking. rgk16 faced more tough ones than you. Ki less. But if this has been true for the last 1000, 5000 that does seem statistically significant. But we'd have to do the math to really make that claim.
James
Date: Sat Apr 1 12:19:00 2017
User: ConcernedFreeceller
Message:James, the last 100 games are not the issue. What I'm saying is that the pattern of the last 100 games has pretty much been the pattern of the past 1000, 2000 games or more. That can't be normal and this is my concern. Joeygray mentioned looking at the backups. If that is possible I would suggest it be done because the logs will confirm what I am saying.
Goosey said he/she has played 11,345 games of 8x4 and of those only 19 had a solve rate below 40%, and none below a solve rate of 21.7%. I am currently playing under the name CocoLoco and the figures I cited in my previous posts are for CocoLoco. I have played under other names in the past, including Michelangelo. As CocoLoco I played one game with a solve rate of 12% and not even 30 games later another one with a solve rate of 7%. These are extremely difficult games and to get two so closely together is odd, don't you think? I won both by the way. I am certain that in the 1,700+ games that I have played as CocoLoco (less than 20% of what Goosey has played) that I have seen a lot more than 19 games with solve rates below 40%.
Date: Sat Apr 1 12:41:58 2017
User: ginxie
Message: I think there are different tupes of people who play and it is nice to have variety in all categories some play to divert their attention. some are serious vompetetors some waver. it would be very booorine to play all of the same games. but discouraging if you are the type who works hard at it sometines and never win. I think the reason there are so many players is it is a game for all types of players. and that is a good thing for all
Date: Sat Apr 1 13:58:05 2017
User: Goosey_Goosey_Gander
Message:I agree that the odds of you getting that many difficult games is slim to vanishing.
Date: Sat Apr 1 18:19:26 2017
User: jamesblackburn-lynch
Message:Hah! I didn't know you were still playing Freecell. If I have my freecell history right, you are also Bucephalus.
But I still don't know what you are suggesting. What is your theory as to why you are getting tougher games than Ki is?
Looking at the game stats for win rate in 8x4 at level 10, a 7% is number three all time while 12% is number seven.
Out of 32000 possible deals, you got two of the top seven most difficult in a short time. My current theory is that you have been incredibly unlucky.
But I also have to guess that since you have been around for so long and have always played 8x4 you have probably played an enormous number of games. I only use the one nickname, so I can easily tell my total number of games. I realize now I haven't a clue how many you have played. Perhaps there were times you were as lucky as Ki has been the last 1000 or so games.
Or again...what is the alternative explanation?
James
Date: Sat Apr 1 18:42:21 2017
User: joeygray
Message:Whoa! Sorry about that, you are NOT one of those "of the rest of us" who lose games sometimes. My observations about level 5 win rates are irrelevant. .bow Michelangelo
As an aside, why did you find it necessary to create another username to post with. (Frankly that's why I pegged you for a newcomer to begin with.) I would have thought anybody who's been around as long as you would be beyond that.
Date: Sun Apr 2 13:02:36 2017
User: ConcernedFreeceller
Message:I stopped playing for a long time and when I recently returned to Freecell I thought it was time for a new player name. I'll tell you what Joey, if and when I lose as CocoLoco I will resuscitate Michelangelo (if I can remember my password!). I also played as Bucephalus (and a couple other names) as James noted.
As far as a theory as to why Ki has been so lucky - or why I've been so unlucky - I don't really have one. All I know is that I have played thousands of games of Freecell and I don't ever remember such a prolonged stretch where I've gotten so many of the most difficult games thrown my way. Not that I mind the challenge but something just didn't seem right, which is why I posted this after, and only after, I felt I could no longer chalk it up to a "tough stretch". I was hoping someone at Freecell would have a theory as to what was going on. If this is in fact just a tough stretch, it has been a real doozie.
Date: Sun Apr 2 13:25:32 2017
User: gumbybros
Message:Here's something I've noticed. If I get a tough one, (60 to 70 percent win range is tough for me) I can almost always count on another one following it before I go back to a bunch of games in the 80 to 90 percent range. They almost always seem to come in two's. In my last streak of 500+ I commented to my brother that I didn't think I had a single game below a 50 percent win rate.
Date: Sun Apr 2 21:56:23 2017
User: joeygray
Message:Denny always says "it's the tough luck you remember" when poo-pooing this kind of observation; meaning, of course, that we forget about those long runs of easy ones... *shrug*
Since I respect you so much, and since I've recently built up some automation freecell tools, I might add a gizmo to save all the games I play from now on, and then revisit the question statistically in six months or so. Soon as I finish the (exciting (sic)) current project, that is.
As for the nic thing, I wasn't commenting on the new player nickname (which is a different topic for some other time), I was just asking how come you didn't just post as CocoLoco to begin with. I would have seen how awesome you are!
Date: Mon Apr 3 11:48:18 2017
User: Ki
Message:Well if it helps,
I just had a 63.86 and a 54.17 in my last 5 games.
.....................
Date: Mon Apr 3 13:48:06 2017
User: CocoLoco
Message:I'm happy to hear that Ki (LOL). By the way, congratulations on your streak in the 8x4. I know how hard it is to keep a streak going and your current streak is amazing!
Date: Mon Apr 3 15:10:00 2017
User: joeygray
Message:The 54.17 game is down at 1.2 percent of games as measured by their solve rate.
The 294th hardest game is at 50%. Meaning of all the level 10 games, only 0.8% of them haven't been solved at least half the time they been played.
Date: Tue Apr 4 16:19:50 2017
User: Ki
Message:Thanks CocoLoco, I have enjoyed building this streak.
I find ConcernedFreeceller's comments and the stats on the last 100 games interesting.
I've played a lot of games under different nicks (PudongPete and QingpuKid being the most successful) and am often surprised at how streaks of mostly easy games come and go and are sometimes quite lengthy.
Being a backgammon player I am quite used to runs of seemingly impossible good luck (mostly favoring my opponents :-) ).
I have been recently enjoying a very long run of mostly easy games. The last 1,000 games I think have been very rich in 90%+ games.
However I do also remember a very nasty streak where I encountered what seemed like a impossibly high number of more difficult games.
This seemed to last for several hundred games with 1 stretch of 100 or so games with a very high number of low percentage games, 60's, 50's an occasional 40's, I thought maybe an 18.
My conclusion: Nothing more than statistics doing their strange and baffling thing!
Date: Tue Apr 4 16:41:13 2017
User: The_Inquisitor
Message:Who knows what evil lurks in the hearts of men?
The Denny knows.
Date: Tue Apr 4 22:04:27 2017
User: jamesblackburn-lynch
Message:Wait! Ki is PudongPete! Man, I wish the nicks were easier to spot. That means in the top 10 all time, two players have the top six scores and seven of the top ten. Unless one of them is also PhoebeDee or stringlug or Cophomanch. At least I know I'm me.
But...the PudongPete nick has a live streak...why not play it?
James
Date: Fri Apr 14 20:52:57 2017
User: SMIDDY
Message:If you are worried about game difficulty you might want to stay away from solitaire as it would seem there are about a 60 to 40 ratio of high percentage wins. By the way, hi Joey.
Date: Fri Apr 14 22:10:48 2017
User: SilvioManuel
Message: I have no theories on the topic but Michelangelo and QingpuKid are my heroes. I make prostrations to both of you. :)
Date: Sat Apr 15 02:46:44 2017
User: TNmountainman
Message:And now *I* really hate to bring *this* up (been biting my tongue), but........
Remember that "PudongPete/QingpuKid/Rolling Thunder" has admitted to using a card deck, at least in the past. (As has SilvioManuel, for that matter - but I believe in his case it was a whiteboard.)
See his (PudongPete's) long(ish) post @Mon May 24 15:56:23 2010 in the below link.
And note that back in 2010, I was still very equivocal about external aids being "cheating". Since that time, just to be clear, I, and most others (but far from unanimous) who have posted about it have come down on the side of it definitely being cheating. There's just no way around it.
And I have no idea if "Ki" is continuing that practice. I suppose he can answer for himself, if he so chooses. The average time/game for Ki is less than it was for QingpuKid and PudongPete, but still more than most.
And I re-post this not to slander, belittle, harangue, or insult anyone. But we do need to know if we're comparing apples to apples, or apples to oranges.
Link: "Is PudongPete for real?"
Date: Sat Apr 15 10:31:18 2017
User: jamesblackburn-lynch
Message:TN,
I don't agree that "we need to know" just because it's just not that big of a deal. But I'd say Ki's average time is such a vast improvement over Pete and Qingpu's time that now I want to know how. Ki's 6:48 average isn't that much more than most of the folks at the top of the streaker list. I've gotten mine down to 5:19 over endless years. There are a few significantly quicker folks and some significantly slower folks (one with an average time more than 11 minutes) but Ki has cut his average time by multiple minutes. Perhaps he remembers to save games rather than let them clock out after fifteen minutes.
I don't know but my main question remains...why not play with the Pudong Pete nick that is the all time best streaker and has an active streak. Saving it for when someone thinks they are going to get ahead?
J
Date: Sat Apr 15 13:30:39 2017
User: outskirts
Message:It is a web site for playing cards. No such thing as cheating. Do what you want and don't let anyone tell you that you are cheating.
Date: Sat Apr 15 14:16:51 2017
User: joeygray
Message:Disagree 100%. That is no different than "it's only A stupid board game, Candyland. So of course it's not cheating if you draw Mr Plumpee and just ignore it and stay where you are."
You may have some valid argument for why it's not cheating, but that one- that it's just a card game- is not one.
Date: Sat Apr 15 14:21:02 2017
User: joeygray
Message:Oh, and Hi Smiddy, nice to see ya!
Date: Sat Apr 15 17:52:30 2017
User: outskirts
Message:An ON LINE card game. No one can control people on the internet, so why bother trying? It's futile.
Date: Sat Apr 15 18:10:57 2017
User: joeygray
Message:That is a different argument. Not saying that it's not cheating, but giving reasons why its ok to do it whether it is or not. I respond to that with a shrug. Not interested. My interest is that if something is cheating then I'm not gonna do it, and maybe persuading others to do the same.
Date: Sun Apr 16 00:27:19 2017
User: outskirts
Message:The game is solitaire. I am not "arguing" anything. You can try to make it that way. Good luck with that. You are wasting your precious time. Plenty I could say but will not bother.
Date: Sun Apr 16 17:09:02 2017
User: BuzzClik
Message:Blowing through a stop sign because there is no traffic cop does not make my action any less illegal. The absence of someone to monitor the rules is not the bar for "not cheating."
joey has it on this one.
This is an old debate with a lot of facets, but I don't recall previously hearing this particular rationalization before. It goes on the bottom of my Convincing Arguments list.
Date: Sun Apr 16 17:48:02 2017
User: The_Interpreter
Message:In the end, it doesn't matter. We all die.
Date: Sun Apr 16 17:49:26 2017
User: The_Inquisitor
Message:Jesus lives.
Date: Sun Apr 16 17:51:26 2017
User: The_Interpreter
Message:Ok, ok, but, technically he died.
And, do we really know if he uses cards, or not, to solve games?
Date: Sun Apr 16 19:31:59 2017
User: BuzzClik
Message:If Jesus cheats at Freecell, that would be important news.
Date: Sun Apr 16 21:52:04 2017
User: joeygray
Message:Come on, outskirts, that's not fair. I mean, I probably don't have as open a mind as I want to have but it's as open as I can make it. If you got more to say that will convince me it's not cheating, I invite, or even beg, you to bother to do so. Lord knows something about the topic interested you in the first place cause you posted. Make a case! By the way, that's all I meant by argument, its dictionary-approved meaning of a case toward a particular opinion. I never was implying pejoratively that you were arguing. Anyways I hate it when people say they got something else to say and then they won't say it.
Date: Sun Apr 16 23:01:39 2017
User: TNmountainman
Message:The case that use of external aids, in the discussions of 2015 and 2016, is "cheating", is profoundly strong.
While I can 'understand' the point outskirts is trying to make, it completely ignores the 'damage' to the millions-of-collective-hours, semi-sacred database. The value of that database is not being appreciated by outskirts, it appears (or by those who cheat).
The unreliability of that data lessens it's ability to add historical context to the play of thousands on this site, and thus enjoyment, especially to those of us who have engaged in the competitive aspects hereon. To reiterate all the weight of "argument" earlier brought against those practices isn't necessary, methinks, although it's possible there are those here now who missed said very deep and thorough discussion. They could be directed to the relevant passages if need be.
BuzzClik's analogy above is pretty accurate. Any excuses have been dealt with fairly clearly and definitively in the past. I *do* think those who have remained unconvinced are in that state due to either: a) being so inured to the practice that they can't see it as others do; and/or b) have never appreciated the importance of the millions-of-collective-hours database. There may be some other category that I can't think of at the moment. It's a form of collective wisdom earned over years, now polluted to some unknown degree. It's a bit like a generations-old saga of a people being lost, or at least severely muddied by an interloping storyteller who has come in and changed the narrative - thus changing the heritage of those to come.
And I realize that that's a very dogmatic (and that example being knowingly overdramatic) viewpoint; but it's been long- and finely-honed over the years in discussions on this board with some quite wise and erudite folks, capable of incisive insights. It's cheating because it's an unfair advantage over those who don't have that same edge - very importantly including those who played long ago, and some of those no longer with us.
So sayeth TNmountainman.
Date: Mon Apr 17 00:52:26 2017
User: joeygray
Message:I see the words: "The game is solitaire", and assume for the moment, outskirts can correct me if I'm wrong, that he (pardon me for using he here, if you are a woman please read it as she) is saying "How can you even call something that is solitaire a competition? How can you cheat against anybody in a thing called solitaire."
To address that admitted strawman, I would first opine that, while the game itself is solitaire, this site is mostly about competition. Back to the Paul Alfille days and the earliest version that Denny made available on the WWW in its relative infancy, it was always about who was fastest, who could get the most in a row, tournaments, competitions, etc. My feeling is that's what attracts and keeps most people who've stayed here longish times. But I can be dead wrong about that and still say this:
To those of you who ARE just using this site to play solitaire without regard to any of its competitive aspects, I am not addressing you here. Because the point is, who is using solvers, cards, or aids to solve games? It's competitors, that's who. There is no other reason to use them, rather than just playin and then, if it gets hopeless, just hitting quit and then replay if you feel you must solve that game or hit play if you want to get on to the next one. There really isn't. It's because you want to see your name move up in a streak list. It's because you are competing, and more: competing unfairly. Until the day that Denny grants unlimited replay to all usernames, I'll continue to consider it obvious on it's face that the intended meaning of a streak win is winning it on the first try, and that any scheme that allows you to replay a game repetitively until you win it or figure out how to win it is a scheme that merely tricks Denny's record keeping software into thinking you have won a game on the first try when you have not.
If you are 'just playing solitare', then I'm confident you're not doing this. Sure, actually.
Date: Mon Apr 17 01:46:32 2017
User: TNmountainman
Message:Very well stated, joey. That's nearly exactly what I almost wrote, but decided not to re-visit the already-clearly-laid-out logic from earlier threads. (And I likely would have re-stated the steroids analogy, too.)
And no, of course you're not wrong - it's the competition aspect that elevates this site above its would-be also-rans. And what attracted and kept quality players for years and years.
Date: Mon Apr 17 09:23:42 2017
User: The_Interpreter
Message:Under the subject "back button" Denny posted this"
Date: Fri Nov 22 03:30:06 2013
User: Denny
Message:
What competitive sport is it again that has undo?
Date: Sun Apr 23 02:55:02 2017
User: Mastermind
Message:Alas, I don't belong to the category that constantly gets 90% and above win games.
#32758 8x4 9 592/53 8.95%
Date: Sun Apr 23 06:18:20 2017
User: joeygray
Message:That sucks. At least when you get to level 10 games you can knuckle down in your mind, and prepare to eventually strike one of the 3 games (.01%) that are below a 10% solve rate. But to hit one before you even get to a fifty streak? Unfair!
Date: Sun Apr 23 06:50:22 2017
User: Mastermind
Message:Unfair,theoretically yes; but since the game did not derail my streak no complaint. Actually I only lose games which have 80% plus wins.
Date: Sun Apr 23 11:26:11 2017
User: TNmountainman
Message:There are also eight climate 9 games with <10% solve rate, and for climate 8, there are four.
Date: Sun Apr 23 13:22:25 2017
User: gumbybros
Message:I know Denny used to have all the games for each level with their respective win percentages posted, but I don't see them anymore. Are they still available somewhere for everybody? I see several posts that mention them.
Getting back to the original topic of this thread, CocoLoco, I feel your pain. Of the last 160 games I've played at level 10 I've had 28 games with a solve rate below 80% with 14 of those in the 70's, 10 in the 60's, and 4 in the 50's. My toughest was a 47% solve rate. That's some consolation.
Date: Sun Apr 23 13:34:20 2017
User: TNmountainman
Message:Yes. Go to "scores", then on the *right* side of that page, click on "game stats". Click on the variant you want (in your case it's apparently 8x4), and you can order the results by win % by clicking on "Win%" at the top of that column.
Post follow-up