The first time I took the test, it gave me a score of 2055. I checked out wafflemaster's answers, and retook it twice. First, I sped through it, getting a 2576. (I beat you waffle! ) The 3rd time, I went much more slowly, getting a 2470. Now, I realize that it's not a perfect test, but could you share the algorithm you used to determine time's factor on rating? I'm impatient, so I only took about 7-8 minutes longer, but that cuts off 100 points?
New Elo Estimator


Wow, thanks everybody for all the feedback, I'm definitely going to improve now with all of this advice. The first thing I'm going to do (which I can do very quickly, is make the board a bit smaller (some people can't see the bottom ranks) and change the squares to an easier colour! All of the other feedback has been noted, and I am going to improve upon this.

Really? I'm 2200 and play 'like a' 2300. Good job, anyways.
I am actually still quite happy with that accuracy! Some people have +/- 400

That makes little more sense. Saying chess isn't 99% tactics is not saying that tactics are not part of every game. Do you struggle to understand English or something?
Yet you still don't seem to understand how high the correlation is between tactics and chess strength. Do I have to draw you a picture so your fragile mind can comprehend?

Oh and I forgot to add:
Why can't you understand anything I've already written? Did you even read my comments?
(It's so fun to make daft remarks like a third grader.)

This is the last post I make here because you clearly don't have the intelligence required to understand anything being said. Let me try another example.
Fielding is an important part of 100% of cricket games yet players will virtually never be picked to play for a side because of their abilities in the field, as their batting and/or bowling skills are astronomically more important. So, fielding is a part of 100% of a player's games yet it plays about a 3% role (ball park) in whether or not they are picked. Tactics are a part of 100% of chess games yet they play about a 40% role (just a random guess, don't argue about this) in the outcome of the average game.
I cannot comprehend what part of this simple concept it is that you don't understand. Unless you can reply to this with something uncharacteristically intelligent, I CBA to reply. I am done spoon feeding you.

The problem on post #14 is a bit of a cook since Qxe6+ is also completely winning.
It's a mate in like 15 instead of a mate in 7.
I went through it incredibly quickly and got ~1700 or so. I was incredibly dissappointed when it didn't show any solutions. I took it again more carefully and got ~1900. I think it's fairly accurate.
Yeah, I played Qxe6 on that puzzle too. Completely winning. Really no way I would have found a fancy mate like Ne4 unless you're telling me "mate in 7" or whatever it was.

Estimated Elo 1587
Keep it up the good work. The elo rating matches my Chess Tempo Tactics rating to the number. I played fast on this Quiz

The problem on post #14 is a bit of a cook since Qxe6+ is also completely winning.
It's a mate in like 15 instead of a mate in 7.
I went through it incredibly quickly and got ~1700 or so. I was incredibly dissappointed when it didn't show any solutions. I took it again more carefully and got ~1900. I think it's fairly accurate.
Yeah, I played Qxe6 on that puzzle too. Completely winning. Really no way I would have found a fancy mate like Ne4 unless you're telling me "mate in 7" or whatever it was.
Hey, just to let you guys know, Qxe6 is a credited move on that position!

What I don't understand is, I can get around 2k on all these ELO predictors, and something similar at chesstempo but I'm painfully awful at chess and always lose to simple oversights 2 moves deep into a combination.

I've been over and underrated by them before.
One problem with a test like this is, you can choose the right move for the wrong reasons... or you can calculate a lot (but incorrectly) and seemingly invalidate the best move.
For example on the 6th puzzles (post 18) of course the first thing I looked at was Bxg5. But after Qxg5 Qxg5 hg5 Bd3 f5 I didn't see a good way to continue. e.g. takes en passant then Nxf6 and black is fine. For puzzles like this, IMO, the credit should be given for seeing there is no defense to doubling the rooks on the h file to win the pinned bishop.
In one puzzle book that offered tactical tests I remember one of the last puzzles I was running out of time, so I guessed a sacrificing move and I was awarded a bunch of points even though there's no way I would have followed up correctly (it was the finish to a great game by Anand).
But anyway Scottrf, maybe there's something you could try? Try to focus on calculating forcing moves more deeply in your own games (pretending a tactic exists?) At least when I had a break from chess for a while and wasn't playing too well I found this helped me catch more tactics that I felt I should have been seeing. I just wasn't sticking with the lines long enough during the game.

This is the last post I make here because you clearly don't have the intelligence required to understand anything being said. Let me try another example.
Fielding is an important part of 100% of cricket games yet players will virtually never be picked to play for a side because of their abilities in the field, as their batting and/or bowling skills are astronomically more important. So, fielding is a part of 100% of a player's games yet it plays about a 3% role (ball park) in whether or not they are picked. Tactics are a part of 100% of chess games yet they play about a 40% role (just a random guess, don't argue about this) in the outcome of the average game.
I cannot comprehend what part of this simple concept it is that you don't understand. Unless you can reply to this with something uncharacteristically intelligent, I CBA to reply. I am done spoon feeding you.
It's literally like I'm speaking with a child.
How about you reread what I already said and come back to speak with me once you're done because obviously you are far too dull to grasp my ideas in one pass-over.
Each post you make is less sensible than the last one. Maybe you will learn to properly formulate an argument by the next time you reply so that my time isn't wasted any further.

How does the ELO Estimator find a person's rating more accurately than say a computer at fixed ratings?
What I would like to see is an ELO Player which allows you to play against various GM personalities. Take games and use the most frequently made moves by them. For wildcard moves, throw in blunders or sacrificies (winning and losing). If a player's rating is higher, these events could happen less often.
More or less what the chessmaster program does.
In any case, computers can't simulate human play. Random blunders mixed with moves found by calculating millions of positions always feels artificial.
That makes little more sense. Saying chess isn't 99% tactics is not saying that tactics are not part of every game. Do you struggle to understand English or something?