Iäm just wondering out of curiosity what would be consider as normal/good roi
at levels:
1.5
3.5
7
15
30$
???

I think 20% ROI would be excellent.

Depends on speed and table count. Better to think in hourly than in ROI as a default.

So many variables. a normal roi is -rake, a good roi depends on speed, game selection standards, number of tables, site etc.
You'd likely have amuch better $5 roi on party than stars,
a higher roi at 10 min levels than 3 min
a lower roi if you just reg every $15 sng than if you carefully table select from 7/15/30.
Also, if you picked the best $300+ buy in player, and had him play $5 games, he'd crush at a superhuman rate, but this is unrealistic because nobody would ever crush $5 games @20% for $15 p/h instead of winning 1% @$300 games for $100 p/h or w/e
Hope you can see what we're driving at

The best players are having 10% at 15-100$, i dont know about 7$ and under.

Since you are here asking this question i assume you are not a top player so dont expect these numbers...dont worry about roi and work on youre game.

Great advice!

It's kinda like groundhog day and catch 22 with this question.
It's gets asked over and over, and the answer is always totally irrelevant, because if you have to ask it, you can't achieve the result yourself.
And as we know the asker is always thinking "how much could I win at this stake?" the whole shabang becomes one great exercise in futility.

I think maybe we just tell people from now -
if you have to ask, you don't need to know

Eh, it's very useful to know where you stand compared to other good players at the same level.

I used to play $10 speed sit'n'gos (10 man) on party with ~20% ROI, not sure if it's more difficult nowadays.

Why/how is it usefull to compare youre self with the highest rois and you dont have much experiance/sample size?
Over what sample you had 20%?...cause that happens now too over 500 or 1k games regularly, actually, 30% over 500 games happened last year. If you had 20% over 5 K+games...than yes, it is much more difficult nowadays...in any case its much more difficult nowadays.

How many games you played with 20%.?and when was this?

Agree with Pop.
I don't want to get all flamey and the like, but if you aren't sure if it's more difficult today than back when you had a 20% roi on party, then whilst I'm confident you're a good guy and a smart/capable person, you probably don't have much to add on the topic of online poker.

As for knowing where you stand v other good players at your level -
if I have to ask on a public forum what good players at my level are winning then I am not a good player at that level and I should not be comparing myself to those that are.

How can he make that assumption I hear you ask...

For one thing, if I don't know the difference between an average reg, and the best reg in my games (which I don't if I don't know what anybody wins at), then I'm not able table select in a whatsoever meaningful way, that alone makes me a pretty poor player in today's poker economy.

Also, by logical deduction I'm not going to have spent the hundreds of ICm hours and postflop/population tendancy hours required to achieve anything like the top roi's at any reasonable stake, if I haven't even bothered to research who the good players in my games are and to table select around them.

Cliffs: One statement demonstrating ignorance (not being insulting, it's just the correct word to use in this context) on the topic of what it takes to achieve the best roi's invariably leads to a plethora of others, culminating in the inevitable conclusion that the person asking about roi's has no reasonable use of the answer.

Cliffs CLIFFS: If you have to ask how much it is, then you can't afford it.

I’d say if you’re starting at the micros you don’t have to be a good poker player but you have to be good at knowing what the players are doing … keep track of preflop jamming ranges , check jams , 5-10 x open sizes … see a lot of showdowns cheaply … see what hands are donking see how many tourneys each player is in … the information is gold and can cost you quite a few buyins to figure it all out but soon you will have a huge edge just buy paying attention and adjusting to the style of play your witnessing .. take tons of notes … most of the time it’s easy to identify leaks .. probably up to ten dollars sng you will see either no bluffs or way too many bluffs not protecting your checking range very unbalanced and after a few weeks you should be able to win at 10-15 percent roi … I would say 20-25 percent but you will run into variance along the way … at some point you will get your money in good every hand and lose a **** ton .. know this is going to happen … stay at these stakes and start investing in tools like icmizer and gto wizard these are valuable tools you will need as you move up in stakes … I would get the tools study them utilize the drills they offer … win your investment back and move up to win more money … although the higher sng you play your return percentage will be smaller the actual gains will be more / but you will also be at more of a risk of busting your roll so make sure to move down in stakes if you get down to around 35/40 buyins … either way it’s a great question and it’s one that should be answered truthfully and with empathy … no one started poker a great player it takes time… but if you work hard and play often in games you find you have an edge in it’s extremely profitable … sng is also the easiest/ fastest way to build your bankroll … just be careful for collision though because it’s very simple to join a table that have 2-3 players all sharing information