Tactics puzzle rating system is ridiculous

How many point you get depends on the puzzle's rating, the solver's rating, and the time take to solve. If the rating of the puzzle is a lot lower than you'll get a lot fewer points. If. You talk too long over the average, you'll also get fewer points.

In addition, the site implemented a new Difficulty setting for puzzles. The default is Standard which are going to be generally easier and give fewer points. Change it to Hard or Extra Hard and you should get more puzzles that give higher amounts.

Was it a request from people? Why would you (chess.com) change a system that has been working for years? What about people such as myself who have been doing puzzles daily for years? Is there even a setting to revert things to the previous system? I don't want extra hard puzzles for an hour straight, and I certainly don't want +5 for easy puzzles for an hour straight either. I want variety. . Please consider implementing an option that would allow people to choose which setting to play from...

My guess is that something like that bhad been requested over the years but I don't know if the change was in direct relation to previous requests or something staff wanted to do.

Change the difficulty to Extra Hard and that should be very close to what the original level/algorithm was at.

I am curious why puzzles work different than regular chess matches

In chess matches players get opponents of their rating while in puzzles players keep getting puzzles of predefined level on and on no matter what puzzle rating these players have

Hi I am IKK

And?

Puzzles are chosen in a range around the member's puzzle rating. What that range is depends on the difficulty level. I believe there are some other considerations in the algorithm but those aren't the biggest factor.

Yes, I meant this as "predefined level"
A player can select the low-level puzzles and keep grinding the rating up to infinity
What is the value of this rating?

If I kept playing against, lets say, ~400 rated players all the time, my rating would increase up to ~1400 and from this moment I would stop getting rating points at all.

In puzzles I can keep solving low-rated puzzles and still get rating points (as I see is it +5 points minimum) regardless how high rating I have already got

As in +/- around your current rating. At some point, someone with Standard is likely to get to a new relatively stable rating. While you may be getting +5 on most puzzles, those you miss are likely going to drop 15+ points. If you're getting 75% correct, you'll eventually reach a point where you're gaining a similar amount to what you're losing, overall.

If that doesn't end up working on practice, staff will likely modify things.

I came here for this conversation and to also make a request about the "newest" puzzle scoring system and to ask why it's always changing.

I'm 75 years old and retired, and I'm using chess.com to keep my brain active. I have no interest in competitive chess and probably spend 1 to 2 hours every day just solving puzzles - and I really enjoy it - until very recently.

After about a year I've managed to drag myself up to a puzzle rating of ~2,000 - but chess.com is driving me crazy with the constant changes to the puzzle system.

During the last 12 months, both the interface and the points awarded and deducted algorithm have changed a few times, and my first question is - why?

It's bad enough that you're always messing about with the interface, but what's really grinding my gears is the constant tinkering with the core of puzzles - the scoring system!

A few weeks ago we were served a flashy new interface which lasted a few weeks then - poof... we're back to the old one again! 

During this time, we were rewarded by earning anything from a few points all the way up to 14 or 15 for a longer, more complex solve - or perhaps for those considered more difficult - and it all made sense!

Now we're back to the old interface again but in addition, we seem to have yet another "new" scoring algorithm - that doesn't make any sense!

It makes no sense that I completed 36 puzzles in a row yesterday earning a measly 5 points for each solve - even for for the more complex 4 and 5 move solutions or one deemed as !! Brilliant by the algorithm - only to lose 25 points on my next incorrect attempt and 20 points on the one after that and 15 in the next one!

So I gained 180 points for 36 consecutive correct attempts, and lose 60 points for getting the next three wrong!

How does that make any sense? As it stands I'm finding the current format frustrating and demotivating...

Please... either go back to a value algorithm that has some meaning like it did for most of the last year, and which clearly motivates people - or just have a simple +10 for a solve and -10 for a fail.

And please, once you've corrected it (as I'm sure you'll agree you have to 😉) - just stop messing about with it - because from an old man's POV you're sucking the joy out of what has become an important part of my daily life.

Cheers,
Michael 😊👍🏻

The overall puzzle scoring system hasn't changed. It's still based on the ratings of the puzzles, the player rating, and solve time versus the average solve time.

The scoring formula may be the same, but the inputs have changed. You say yourself in another thread that the puzzle ratings that are used in the formula are now some new "backend" rating that is different from what is presented to the user on the front end.

The scoring system is effectively changed by the algorithm now presenting puzzles with ratings that are well below the user rating, resulting in penalties that were never seen before. For example, users are now seeing penalties of -30, which were simply never seen before. This is because the algorithm used to not present puzzles with a rating so low that this was even possible. A very easy partial fix for this problem would be to cap the penalty at some level (say -10), just as they changed the scoring system a few years ago to always give a minimum of +5 for a correct solution.

Frankly, these changes to the Puzzles section seem both unnecessary and bungled. It's a bunch of new features/changes that are upsetting and confusing users. It's not clear to me that any users asked for these changes, and the changes seem to be inconsistent across platforms, suggesting to me that they were pushed out too quickly by the developers.