points in tactics


This is older.
https://support.chess.com/customer/en/portal/articles/1444811-how-does-the-time-and-ratings-work-
The method is similar but the mechanism has recently changed a bit.
https://www.chess.com/blog/News/chess-com-dev-update-april-28-2017

In a very stupid way.
I like most of the change. I think there is still a bit too much upward pressure on ratings giving some inflation but I have been able to solve almost 70% of the puzzles I've been given. I'm seeing the success drop so I think it is feeding me appropriate puzzles too.
Some of the inflation is likely due to scoring, which still may be modified some.

with all the changes my rating has been fluctuating quite a bit. Last week I lost 3600 points in a day, then gained about half of it back. A few days ago I gained an additional 400. Weird stuff.

It depends on the day, these days / weeks chess.com seems to change / tweak the settings.
Sometimes u get 20 points for an easy problem even with exceeding the target time, another day u get only 1 for a puzzle similar.
No stability at the moment and not to recommend the tactics, not at all. Not now!

It depends on the day, these days / weeks chess.com seems to change / tweak the settings.
Sometimes u get 20 points for an easy problem even with exceeding the target time, another day u get only 1 for a puzzle similar.
No stability at the moment and not to recommend the tactics, not at all. Not now!
I haven't noticed anything like that since the last obvious change. I even reset my tactics, which dropped me back to 400 and worked my way back to 2200 in a couple of hours and about 250 tactics. Tactic selection seemed good and no weirdness in scoring.
Went up another 130 point the next session and dropped a bit in my last attempt.

With the target time somehow being part of the resulting points now it can get ridiculous, target time 2sec a few times. How can you achieve more than 1 point there ...
"Tactic selection seemed good and no weirdness in scoring."
Yeah sure, countless threads with people complaining about the tactics gone weird and what for you is obvious mustn't be obvious for other people without the possibility to run tactics a couple of hours.

I have been posting in a bunch of the topics and am very aware of previous issues. Since the last change, earlier this week (end of last week?), there hasn't been the weirdness. It has been good, with a bit of inflation over the old scoring mechanism, pre-end-of April change.
Tactics with a very low Target Time are ones rated a lot lower than your current rating and the idea is that you should be able to solve it quickly. Or ones with a low Average Time maybe; haven't paid that much attention for those.

Nothing new you write here and I just write my experience nothing backed by dates of changes. Your last part of the post, surely that's the way it is. Nonetheless, if you take a closer look those things were handled in a quite imbalanced way. 2-4 sec target time and average time 43-45 sec. Get it in less than 10 sec (about a forth of the average) and u get 1 point. Hence checking against the history u see that most people are not able to achieve 100% on such easy tactics due to an imbalance.
The correlations between time solved / strength / points can't be linear or alike. Now it looks like the easiest tactics the least people get 100% of the possible points.
I don't know the way tactics algo / settings are checked against but

I think the site would have been better off to implement the Target Time but not display it. Explain how it is calculated in a help article but leave it at that.
I actually think I'm getting more 100% than I ever got in the old TT but that is a subjective feel since I can't back it up with numbers (history doesn't go back that far and I doubt I've posted about it enough to verify in old posts).
I've had almost 25% of my tactics solved in the Target Time in the last session, with one of those being a significantly lower rated puzzle requiring fast times. And target time is unique to the rating of the player, based on what I've read. If we had the same tactic, say a 1700, you would have a longer Target Time, in general, and I would only have a couple of seconds, assuming the Average Time wasn't too long.
Also, if the average time of a puzzle was 45 seconds, the target time was 4 seconds and you solved it in 10, you would get more than 1 point unless the rating of the puzzle was a lot lower than your current rating.
For example, you solved a 700 rated puzzle earlier. It is pretty close to what you describe. It was rated over 400 points lower than your current rating. The expectation is that you would solve it very quickly which you did, but even had you received 100%, you wouldn't get many points. A 400 point lower rated puzzle will not give as many points as one closer to your rating; and it shouldn't. It should also create a larger loss on a failure.
Likewise, a tactic 400 points higher than your current rating should decrease your rating less on a failure and give you more points on a full success.

https://www.chess.com/blog/News/chess-com-dev-update-may-26-2017
There are some additional details on the changes made to TT.