Based on your move choices, our estimate of your Elo rating is 1624, with a 95% confidence interval of [1405...1842].

First try. I timed out on at least four puzzles... And a few more probably weren't cosidered because I made invalid moves by accident. I doubt my ability to play these puzzle moves like I play real games, and I really hate time pressure.

Not sure what to think about this estimate's accuracy, although it's relatively close to what I was expecting. Expected a little lower, though. Then again, there were a couple of puzzles where I blundered (valid moves, but I picked up on some of my mistakes/blunders immediately after making them) due to nothing but time pressure.

I feel like repeating the test again, but overall my experience with it actually mirrors my real game experience: very inconsistent between good and bad play. So I guess that overall this is a relatively accurate estimate and subsequent tries probably won't change it that much. Shall see.

Thanks for the site MarcoBR444 really enjoyed it!

mine was...

Based on your move choices, our estimate of your Elo rating is 1938, with a 95% confidence interval of [1753...2123].

The time was an issue but im happy!!!

Very nice site.

Based on your move choices, our estimate of your Elo rating is 2083, with a 95% confidence interval of[1903...2263].

A little high maybe, but pretty good.

Based on your move choices, our estimate of your Elo rating is 1940, with a 95% confidence interval of [1755...2125].

I'm not so good

I just tried again and did a little better. (still timed out on two puzzles)

Based on your move choices, our estimate of your Elo rating is 1767, with a 95% confidence interval of [1566...1967].

Don't think this is particularly realistic... Or maybe this simply tests raw skill like an IQ test, and the first attempt is the most accurate representation of one's raw skill/potential.

Retaking might be cheating then.

For those who can't retake the test: try to clear your browser's cookies or history.

EDIT: I take back that retaking might be cheating here. I don't think this parallel can be made between IQ tests and chess puzzles.

On the sort of IQ tests that assess fluid intelligence (the sort of skills that are independent from acquired knowledge - pattern recognition, logical puzzles, etc,) it is true that one can eventually learn the answers by repeating the same type of problems several times. Over time, this translates into an inaccurately inflated score.

Don't think this is the case with chess, though. It seems to me that one's chess skill is a lot more fixed and requires a significant amount of effort and time to improve upon. So repeating the same chess puzzles again and again over a relatively short period of time shouldn't result in noticeable improvement. If anything, the mean of multiple attempts should be a more accurate estimate of one's true skill than just one or two tries...

I doubt this test I got 1840...(pretty high for me)

Just did it. The site 'feels' scientific/professional - I was honestly expecting some clickbait/spam/ad site that was masquerading as an analysis test site, but so far so good.

I'm not so sure about the result though - 

My chess.com 5-min blitz rating as of this second: 1440

My chess.com bullet rating: 1233

(My chess.com standard rating probably isn't valid as I've only played <3 games)

Elometer rating: 1935 (1750-2120)

I've seen 1935 players play before, and honestly, I'm not one of them! Although to be fair/honest, despite not having played a tournament game in over 10 years now, in the few games I did play as a 1400ish UCSF rated player, my best win was against a true 1950-level player, so I do have my moments. I'm also a clearly better player now than I was back then as a 1400 UCSF player - I have no idea what my UCSF would be now, but I seriously doubt it would be 1935. 

I also feel like I missed well over 50% of the problems on that test - I probably confidently solved about 8 of them, felt 'ok' about my answer on about 10 of them, and was totally clueless on the other 50% in that 30-sec time window they give you.

Another moronic time pressure tactics test that aspires to pass  off as a valid  indicator of true playing strength in Classical chess.Pathetic...

1457

Seriously!

Lets call guessing the correct move a false positive.

To help avoid this, positions where the captures and checks score very poorly are needed. Other than simple positional puzzles, an easy way to marry this with a quick format is to have tactical positions with defensive solutions.

1873, and I have 1877 on chess.com online chess. Impressive (or lucky).

Goodness. I think 30 seconds it too fast for me on one problem, I got a rating of 1800.

<Robert0905> this is surprising coming from you (as we consider your bullet performances and your USCF blitz rating). 

Just went for the third attempt. Gives me 1865 with [1675 - 2056] confidence interval...

This value slightly exceeds the confidence interval of my first attempt [1405...1842]. It would bold well to this test's accuracy aspirations that I'd never exceed those limits... But the difference isn't much and I don't think I can do much better than this.

If there is after all an element of cumulative improvement in these tests and the methodology of this test already takes that factor into account, then the first result is probably accurate enough. In which case, none of my subsequent tries should significantly exceed the limits of the confidence interval deduced in my first attempt [1405...1842]... I think...

our estimate of your Elo rating is 1561, with a 95% confidence interval of[1334...1788]

Given my correspondence rating of 1430, I'm fairly pleased with that.

My blitz rating is only 1200 but I play 3 | 0 so you can't spend 30 seconds on a move, so I'm not surprised it's lower.

Based on your move choices, our estimate of your Elo rating is 1682, with a 95% confidence interval of [1471...1893].