Subject: ELO TOP TEN PLAYERS

There is a rumor out there, started by some brigand, that all of you use some sort of external aid to solve your games.  Here is your chance to refute this nefarious scoundrel and set the record straight.

Just post your umbrage below.

It would be unfortunate, but a lack of response from any of you would probably be construed  by other players as admission that this blackguard has assumed correctly.

OK, TN?

It was a mere wondering - not a "call to action".  But, ok.......   I can just hear all the admissions being conjured up.

Other than the vocal mis-achievers, I believe the rest is best left unknown. Some people just prefer to go about their daily ways without having to look over their shoulder. I honestly believe that there are smarter, luckier people than I. Those same don't owe me an explanation for their fortune. I'd hate to find out now that Dimaggio used steroids.

While we are waiting, I would like to ask the programmers on this site how hard is it to use a solver to work on winning the games that comes up in streak play? 

did you try googling "freecell solver"?

you don't have to be a programmer, but it helps when trying to get it to work for you

The premise in this discussion is that certain players spend thousands of hours loading games then use a solver to solve the games. I guess the next part of the argument is that they are doing this to impress random strangers that they are exceptionally talented freecell players.

Seems like there's probably a simpler explanation.

Facts have always been stranger than fiction. What motivated the guy who made a twine ball bigger than he was?

Maybe people need to accept the fact that there are better freecell players than themselves.  I can't see why anyone should find it unbelievable that 10 people could have ELO's over 2250 without cheating.

It's just that those Elo numbers actually mean something, rbw, and we get well into superhuman territory in the top 10 here. If this were like chess and players could compete head to head, a 2100-rated player would outsolve an average 1500-rated player 97% of the time, and a 2700-rated player would in turn outsolve that player 97% of the time. Hop would outsolve Uberman 98.7% of the time based on current ratings, and Uberman is very, very good.

It doesn't mean the top 10 all cheated to get there. They could be protecting ratings that were obtained playing games that were overrated at the time. The ratings certainly do look inflated though, and those players are probably kidding themselves. They'd find out really fast if they started playing more games with ratings close to their own where most of the games are winnable. They could try, for instance, 4x8-8, 4x9-12, 5x5-6, 5x6-12, 6x3-5, 6x4-12, or 7x2-6. Basically, if you're not playing game levels where you sometimes win and sometimes lose, the Elo system doesn't know what to do with you and your rating should just say "undefined."

I understand it's a statistical distribution. It also cannot accurately translate into a head to head comparison because there is no time limit on how those ELO points are accumulated. I think if Uberman focused strictly on raising his ELO it would be significantly higher. 9 sum and 10 sum don't reward very well and have the extra step of large numbers of unwinnables.

 It seems tournament play clusters in a similar manner.

Do you really believe Uberman would stand still long enough for Hop to take 98.7 jabs at him without striking back?

I'd probably be so starstruck at being in the same room as Hop, red, that I'd give him the last 1.3 jabs for free, too 🥊

My Elo score is highly inflated. The Elo system here had bugs at the beginning. So just in a few day I accumulated so many (undeserved) points. I myself also complained about this and suggested a reset.

I usually play easy variants and this Elo system still gives me points.

If you look at a person's recent play, you can see the time spent solving (or losing) each individual game listed. It seems to me that that factor must be somehow relevant to this discussion; but I'm not equipped to see precisely how. I doubt that any conclusions can be drawn without establishing a Freecell Police Force with right of entry to a player's home and computer (or device).

This seems to have branched off into two lines of thought, both of which may be true. Within the framework of how ELO is calculated it is possible that the top ELO's aren't a result of cheating. Also those ELO's may not accurately reflect the difference in skill level between players (based on standard deviations). 

The word cheating has not been used in this thread or the other.  Actually, Denny, our founder, doesn't consider the use of external aids to be cheating.  Others may feel differently.

Titanic Tony admits he uses a deck of cards.  He's being honest.  Therefore, it can be assumed that if he didn't use a deck of cards, his Elo score would be lower.  How much lower, we don't know.  But, from the multitude of games he posts on the multitude of other threads, it may be assumed that it would be significantly lower.

We don't know how good a player he is, and neither does he.

Even though hop didn't declare one way or the other on external aids, his suggestion of an Elo reset is a good idea.  

Well .... MoscowMitch didn't mince words, and "cheat" was unminced in his opening post. He limited his accusations to a single person. 

This has been a productive thread with some interesting comments.  

I stand (well, sit) corrected.

An Elo reset is off the table as it would introduce bad data all across the game ratings. It would be a disaster, frankly. Instead you could maybe add a rule that a player is still considered idle if they haven't played a game within x number of rating points of their own rating. Doesn't fix the external aids problem but neither would a reset.

As far as the word cheating goes it's been used elsewhere relating to using cards. I can't imagine anyone would disagree a solver would be cheating. I could have said external aid but the idea is the same. Hop and some others could not be using cards because the time signatures would be suspect. I don't agree with the idea that the best rated players owe a disclosure and my guess is most would feel it's offensive asking them. 

It’s an internet game. Don’t take it so seriously. There ain’t no World Federation of Freecell that I know of.

That said go read about online bridge and online chess and how they’ve dealt with things. Same problem space but they also have tournaments IRL.

Another factor is freecell is “aging out”. No next generation coming up to make it worthwhile to have tournaments IRL.

I haven’t seen proposals for Elo reset. No point unless something changes, right?

MrFixit aka Denny

I adopted the philosophy of not taking freecell seriously from the beginning. (It's easy to be philosophical when one could never be competitive.) 

However, this game is taken quite seriously by many, as illustrated by the array of tournaments, competitive metrics, detailed statistics, and recurring threads on the discussion board lauding the talents of the players. There's no money involved, just the desire to compete. 

Denny, I have dabbled on and off in playing bridge, and it's shocking the depths to which players will sink to find ways to cheat. Online tourneys have bigtime safeguards, and even the casual online apps take steps to avoid cheating. Again, no money is on the line, just rankings, but cheating is commonplace. Those who play even casually (and honestly) just want a level playing field. 

So, in response to the question, "Are they doing this to impress random strangers that they are exceptionally talented players?" Yep. That simple. 

I have noticed that some of the high scorers find obscure variants that are not really that difficult but seem to give a large ELO score for the games that have been played a very small number of times.  I tried some of the different combinations but found that my way of thinking needed to be modified and I didn't do as well, so I have focused on the standard 8 X 4.  I don't use any external aids, but I will admit the one thing I do to "cheat".  I have found that I can keep 20 to 25 moves in my head to try and weed out getting stuck in a box with no forward path.  If I study a particularly tough one long enough the game times out.  BUT, you can make 12 or 13 moves (if you do it fairly quickly) before the game lets you know that it has timed out.  When that happens I take a screenshot of where I am, reset the game, and then start my mental 20 moves from the screenshot (thereby giving me over 30 moves ahead).  This feels to me sometimes like I am cheating, but we are a community that loves this game and everything should be shared.  I have spent many (TOO many!  LOL) hours thoroughly enjoying this game and I love the challenge of no back button and appreciate the work done to keep track of all of the statistics.  Thank you so much Denny (or MrFixit (?) not sure who to thank) for keeping this site alive.

Hardly cheating, spugeddy2.  I find it impressive that you can think that many moves ahead.  I barely have the patience for maybe ten moves before I start getting confused.  Thank you for your response.

i wouldn't do an Elo reset, meaning set the numbers to 0 and start over

However a recalulation can be done based on the daily playlogs
In fact Hop has already done it See Hop Style Elo... read the info about how the calculations are done. You can see the Top 25 players by clicking Daily Top-25 tab

I said earlier: "Instead you could maybe add a rule that a player is still considered idle if they haven't played a game within x number of rating points of their own rating." Just building on this idea a little, currently to get listed on the Elo Ratings scores page you can't be idle more than 14 days. We'd now clarify you're still considered idle if you haven't played a single streak game that was close to you in ratings in 14 days.

The magic number for x is probably a difference of 800 rating points. At that point a player is expected to win 99% of the time, so we're not asking a lot. If you wanted to make it more stringent you could make it 675 points. The player would still be expected to win that game 98% of the time, but at least now they're challenging themselves a tiny bit and might even lose a game now and then.

If someone did miss the cutoff by not playing any challenging games we wouldn't even have to consider them idle, we could just italicize their rating and maybe show it in a lighter typeface to indicate that it's unverified. Then if players want to mentally exclude them they could, or we could add a toggle to "remove unverified."

Of the current top 10 players, most would still make it based on recent play:

Hop played and won a 2188-rated game recently that is now 2/103. (He admitted earlier he usually plays easy variants, but he also plays a lot of tough games.) Flocci would miss the 675 cutoff but make the 800 despite only playing a handful of games recently. Wasjun I assume is fake so don't care.

Stringlug would just miss 675 but make 800. Lindy makes it either way. More fake names, then spugeddy, LeftNuttt, and act make it either way. I think it's a good potential addition to the page, and would probably go with the 800-point cutoff so we don't exclude players like stringlug who only play 8x4.

1.  Help me figure out why taking a screenshot and then building from that isn't the same as TT's infamous whiteboard.  Just raising the question sort of as a thought experiment.  I actually know what I think about it (I think), but would like to hear others' opinions.

2a.  Denny, you may have forgotten.........that the original plan was to re-set things (elo) after a proof-of-concept (my characterization) period.  While Hop (and many others) took that opportunity to dramatically rachet up scores, there is at least one player that I know very well who took the opposite approach, and played a bunch of hard/impossible games just for fun, as they weren't going to count.  That elo score ended up being somewhere between 1100 and 1150, if I recall correctly.  Assurances were made by ElGuapo (and yourself, also if I remember correctly) that very quickly those scores would not only gradually lose meaningful mathematical relevance, but would "disappear" in short order.  Hop's statement above sort of argues against that - altho surely his level of play mitigates against *that*, if you follow my logic.

2b.  But no...........one ("we") can't re-set now - unless one takes the verrrry long view of things. 

2c.  However..............does ElGuapo's statement above take into consideration the idea of re-setting *only* the player elos, and leave the game elos as they have already been adjusted?  Surely the game elos are vastly, vastly more 'accurate' than they were at the beginning.  I'm *not* saying that it *should* be done, but is it even an idea with any merit at all??

My post immediately above was composed before knowledge of ElGuapo's last post, so I was referencing *his* post of 10:18:02 2025.

I guess I'm out of the loop. By "Fake" does that mean they are using cards or a solver? 

Also how many solvers are there? If there are only a couple or so and if there were a capture of move logs for a players games, specific games could be compared to known solvers. If there were repeat identical moves to known solvers a player could be deleted. A lot of ifs there.

rbw--3, the time signatures idea on what you all call "streak" play doesn't work because TT (among others), pauses the clock just for the purpose of making it appear to be a faster solve whilst laying out cards and manipulating them over and over as 'necessary'.

And apparently there are many solvers (I have no personal knowledge, however), and cellmate developed his own after 'breaking' Denny's algorithm. 

"rbw--3, the time signatures idea on what you all call "streak" play doesn't work because TT (among others), pauses the clock just for the purpose of making it appear to be a faster solve whilst laying out cards and manipulating them over and over as 'necessary'."

I was referring to end time - start time for a game should equal game time during a streak. I think you mentioned that you noticed with TT in the past.

Of course it's not perfect. Someone could leave the clock running also, but it takes a very long time to set and reset cards. Clearly (just as an example) you can tell Hops times could not be done with cards. You can't prove cards are being used with the times, but I think you can tell when they are not being used. 

@Kumquat-of-Conciliation

I assume your number 1 above was directed at my post.  I acknowledged that when I’d do that I kind of felt like it was cheating, and from the way you phrased things it seems like you do too.  Your TT whiteboard analogy got me to thinking and I have decided to stop doing the screenshot thingy.  My competitive nature got the better of me, but I think the integrity of the scores should be more important.  Even though I have a huge competitive gene, I also have a huge fairness gene that (in my logic) supersedes.  Thanks for your thoughts.

And if you weren’t talking about me, then….nevermind.  😊

Yes, I was referring to your 'admission', but was stopping short of being accusatory.  And.......I was somewhat surprised to see firenze's (apparent) acceptance of that strategy.  Some time ago there was a somewhat in-depth discussion of how many moves ahead each of those participating could plan.  Somebody, and I don't remember who, had what most considered an outlandish number.  I also don't remember what that number was, but I wanna say 40(???).  Most of us mere mortals said something like 15-20, as I recall.  And even that seems pretty impressive.  

Point being, obviously, that, at least to me..........being able to plan 30 moves ahead instead of 20 seems to be a clear advantage.  And the whole point(s) of the competitons here is stacking up one's brainbone to those of others.  When one uses something/*anything* outside of, or beyond, one's own noggin's capabilities subverts the comparisons. 

You're clearly a very good player - but again, it's hard to know *precisely* how good.  If you indeed hang up the screenshot technique........it would be interesting, down the road, to hear your feelings about it.  Thanks for your engagement on this. 😊

Point being, obviously, that, at least to me..........being able to plan 30 moves ahead instead of 20 seems to be a clear advantage.

That is the same conclusion that I came to after further thought.  I honestly don’t do it that often, but even doing it once is in conflict with your other statement (which I agree with and thought was succinctly put) When one uses something/*anything* outside of, or beyond, one's own noggin's capabilities subverts the comparisons. 

I am enjoying having a good conversation about this without politics or emotions (although the fact that we feel the same makes that easier).  😊

I really don't have a problem with someone staring at a screen/screenshot for hours, as long as you are only using your brain to solve the game.  My personal patience lasts about two minutes, then I have to make a move.  I blame mother.

This discussion makes me feel ethically naive. I'm rather sorry I ever became aware of ELO, as it has messed up the way I'd like to play.

As others have mentioned and suggested........ignore it - or at least don't worry about it.  One can't help but be aware of it since it flashes up after every game, but you don't have to chase it.  I certainly don't.  I wouldn't call it a "false god" like winnables are (🙄🤔😉), as it does have a noble purpose, but like streak lengths, it's a metric that can be - and is - gamed by those with outsized egos.  The last paragraph of Buzz's last post above is sadly all too accurate. 

Ohhh... spugeddy2,  you're on TN's naughty list forever now. you'll be reminded in some future message

LOL @cellmate

Hopefully I didn’t ascend to TN’s naughty list (want to avoid that at all costs!).  I would think that he would actually agree with my decision as it was made from wanting my ELO to be a true reflection of my abilities rather than potentially inflating it to ‘impress random strangers’.  With any luck I will not be emotionally destroyed if not doing the screenshot thingy causes me to fall out of the top 10 (although Hop’s Daily Top 25 already has me out of it!).  I do my best to keep my ego insized (is that the opposite of outsized?) so I am hopeful...

@firenze

My personal patience lasts about two minutes, then I have to make a move.

That is a really interesting statement because a few years ago there was a ceiling that I couldn’t seem to get past and I identified a common thread to games that I lost, and that was my impatience.  Now, as soon as I notice my brain saying “just make a move, it’ll work out” then I stop and go do something else until my patience refills. 

spungeddy2, the boy can't help it.  He got his father's brain, whoever he was.

LOL 

ElGuapo, in your above post you identify some players in the top ten as fake.  I am not sure what you mean by that term.  Could you clarify, please?

I was wondering the same. I asked above. I guess I'm naive if there's several of the top 10 that don't play unaided. If a player doesn't say they use some kind of aid, I'm not sure how anyone would know. 

If there are so many streakers using aids, can we trust tournament players and hot streak players. Since consensus seems to be they do it to impress others, it would take much less time to impress people winning a tournament. 

I don't mean they're bots, I imagine everyone here is an actual human playing the game in some form or fashion. I have no insight into how the top players are obtaining their streaks. I mean "fake" in the sense that they've either said or I've somehow concluded they aren't playing the same game I am in any meaningful sense. That could of course apply to many other top players, but without any firsthand information I tend to view names on these lists on a sort of continuum of credulity. Speaking up like spugeddy did here goes a long way in my book, so that's at least a starting point.

I'm not sure it works that way with regard to hot streaking - the play period expires 30 minutes after the start time, and pausing games does not circumvent that limitation. I don't play tournaments, so I can't say for sure, but I imagine their time limits are calculated the same way. As far as I know, game numbers are not available before a game is completed, so a 'solver' would require the competitor to input the deal.

I may have misinformed myself there, so this output may have come from an unreliable orifice...

"Continuum of Credulity" is just about the perfect descriptor - with an awesome catchphrase-worthy ring to it.  I would only point out that - imo - it's not a linear progression/scale.  It's not *exactly* like being "a little bit pregnant"............but let's say - again imo - that it's heavily front-loaded.

Thanks, El Guapo.  

This thread has pretty much played out.  Because of the title, I assume that most, if not all the top ten Elo players have followed it.  With the exception of spugeddy2 and TitanicTony (who has continually defended his use of externals), none of the other players has declared one way or the other on their use of these aids.  If I were any one of them, and I did not use these external aids, I would declare it loud and clear. If I did use, I probably would be too embarrassed to say so because it would openly nullify my 25 years on this site.

Therefore, I am left with only my suspicions.  And, they are strong.

Boy, you need a life.