Thanks for the snowie graph.
I feel like my intuition was mostly good but def some suprises

I was referring to the fact that an open with A2s implies that all other higher EV Ax hands are opened first: AK, AQ, AJs, ATs, A5s, A4s, A3s, and possibly A9s are chosen as opens before A2s, and each new ace you add to your range saturates your range with Ax that much more. This saturation reduces the EV of your range on boards that aren't Axx, and makes it harder for you to get action with your big hands on the Axx flops.

Incuding A2s doesn't necessarily imply you are opening all A2s+ combos. This is especially true given the context of "board coverage".

eg; I used to play a lot of nl25 on bovada. I was trying to work out my UTG range for 6 max.
I asked two friends, both pros.
They had different approaches UTG. One opened A8s+ the other opened A2s+. My friend (the one who was only opening A8s+ asked his coach (a well known and respected coach who does training vids for IPokerVIP) the advice he received and that I adopted was to open A2s-A5s, A9s+, ATo+ but to open fold A6s, A7s, A8s

the idea was that the pairs you flop, weak pair top kicker or top pair no kicker, are virtually the same hands so you may as well play the ones that have straight potential. (The word on the street is that board coverage matters some too
Hope thars coherent. I jusr fell asleep abd dropped my galaxy.
guess its bed ti5

I've traditionally raised AJo UTG virtually 100% of the time in live frnl games. I think it takes a tough game for it to be a fold. I'd have guessed Snowie opens it more than 17%

Most people I know struggle to make money with AJo UTG in 6-max games, let alone full ring.

Live full ring Arty

SNOWIE 6 MAX NLH 200BB GTO OPENING RANGES

UTG: 17.8% | 44 (19%) – AA, ATo-AKo, A2s-AKs, KJo-KQo, K9s-KQs, QJo, QTs-QJs, J9s-JTs, T9s, 76s (74%), 65s

Arty how do these look?

^ That's more or less what Snowie suggests for 200bb deep 6-max at 5/10. It plays a bit tighter (fewer SCs) at 100NL/200NL (and lower) due to rake considerations. You can see the "standard" 6-max chart in the free Pre-Flop Advisor app.
As mentioned before, it prefers 65s/76s over 98s/87s and doesn't like baby pairs much at all. It seems like a solid range to me, but obviously you should tailor your range according to game conditions and your own abilities. I don't play live, but I doubt it's very profitable to open 65s UTG if you get called by several players that have position on you if they wont fold a pair post-flop. To that extent, AJo is almost certainly a more profitable open in a game full of calling stations.

Sorry,
I now realize you werent thinking A2s's inclusion necessarily means A2s+ but you were assuming all higher EV combos of Axs would implicitly be in a range containing A2s

new question for everyone, as this raises an interesting point;
Could it make sense to chose some combo X and exclude Y if the EV of X > Y, for the benefit of the entire range? Or is that ~ a loss leader (which are not a feature of GTO play)?

Ive played a good amoubt of 1-2 and 2-5 live.
I think AJo is break even at best UTG.
You only have two options at live low stakes regarding UTG opens. You can use sensible open sizings (something like 4xbb) and go to the flop against multiple opponens the vast majority of the time or you can use a monsterous open sizing (something like 7xbb) and occasionally take the blinds, but still usually get called and play a bloated pot OOP.

I don't think AJo is very good in either case.

If you want to use relatively sane open sizing for small stakes live cash (4xbb ballpark) you have to expect to get called most of the time and go 3 or 4 way to the flop an awful lot. For that reason, I will often draw the line at AJs/KQs and open to about $9 UTG at 1-2.
At 2-5 the games really vary a lot from one table to the next.

Derailed my own post;

Could it be correct to open hand X and food hand Y where Y is higher EV than X for the sake of our overall range performance? Or is that, more or less, a loss leader?

it would be a lossleader

but if you look it that way, KK is a lossleader too if the only hands that GII would be AA

No, it can't be correct. In GTO solutions though it can certainly be correct for hands to be a raise/fold mix. In this case the raise is 0EV at that frequency, but this is a misleading figure because that EV is reflected in the increased EV of stronger hands in your range that come from adding the marginal hand. In other words, people have to defend against you a little bit more aggressively to make the bottom of your opening range 0EV, and as a result they increase the EV of the top of your range a little bit.

Sounds like a (pointless) loss leader to me. The GTO solution is the one that maximises your total EV. You don't maximise EV by including -EV hands in the strategy. Every hand you play in a particular spot should at least break even. Hands with an EV of zero often utilize mixed strategies, and they are included because they increase the EV of your overall gameplan, but you wouldn't include hands that lose money if you had others that make money.

As a quick (extreme) example... suppose that the GTO solution to 6-max had J9s being raised UTG 65% of the time, and that hand would break precisely even in the long run. What benefit would you gain by folding that and instead raised J2s, if the EV of that hand is negative? Putting J2s into your range instead of J9s wouldn't magically make your whole range more profitable.

It can be higher EV to take slightly -EV plays in theory vs. certain opponents imho. Just because people over-adjust/adjust badly/psychology. Same kind of thing why live players 3b 72o and show it (obviously this is awful for numerous reasons, but the reasoning they have for doing it isn't just completely wrong).

But yeah, just play solid and you'll crush pretty good. So focus on that first.

I think this concept is important in live play. Things like table image and setting up players can be pretty big.

However, I don't think any of these things should be a part of our overall strategy in how we should construct ranges or betting frequencies. I think its something we should consider or do to create or avoid certain dynamics.

The best or simplest example I can think of is if I'm in a game where the majority of players are very sticky/call stations, I may expand my opening ranges if I haven't played a hand in a while, especially early in my session, so I don't get too tight of an image. Its possible that some of the hands I may open could be slight loss but as long as I'm not getting carried away, I think I should be fine. The objective of doing this would be to avoid having my opponents peg me as tight and have them play more correct against me.

noone has actually considered metagame as a gto-factor

we are humans, and even if not, we likely would have a mind that can be manipulated into poor decisions(gto vs a supercomputer would be to pull the plug)

gto imo HAS to consider doing things that are purely done to manipulate the others
you could have a showbluff/value range and it wouldnt show anything because we play gto

but it could change others into playing badly, thus we should do it from time to time

we can even consider things like prizepsychology into the hole thing

if you type in random numbers in a tournament after your original bet, it doesnt change your bet meaningfull (like blinds are 1/2million, adding a 3 digit number at the end doesnt change anything)
but maybe it makes one play spazz out for whatever reason(a abstract one would be arbitary code when bots play vs bots)

Yes, it sounds like you understand. I think effective stack sizes is important to consider as well. In NL, the deeper the effective stacks, the more important it becomes to have an uncapped range on every board, and if you never have hands like 45s in your range you can never have the nuts on the A23, 236, or 367 board, for example.

In regards to playing slightly -EV hands to create or avoid a particular image;
I don't buy it.
Firstly, I am not convinced there is such a thing as a "bad image" only an image that is bad for X but good for Y.
If youre having a tough time getting action from opponents then simply adjust your ranges to exploit their folding mistake propensity.
Of course our exploitive adjustments wont be the highest EV plays against an equilibrium playing opponent sort if by definition. Thats not the same as making an intentionally losing play so you can show it to cultivate a particular image.
Also, even if you desperately wanted to create (for example) a very wild and reckless image to get lighter call downs later on in the session you could do it without making a blatantly -EV play. The truth is, so many of our decisions at the felt (especially starting hands to open or not) are so nearly break even that playing every hand that breaks about even compared to mucking every hand that breaks about even could generate VERY different strategies that induce very different perceptions about our play without changing our actual EV in any meaningful way.

If youre not getting enough action you can start opening more marginal hands and err to the side of aggression when a spot is close. If you are unable to get respect from your opponents then you could cut out all the marginal ( close to break even) hands from your preflop ranges and pass on all the "break even" bluffs until your oppobents start to adjust.

I doubt it ever makes sense to just go totally off to create an image, it simply isnt necessary

I agree. I do definitely think it exists as an exploit vs. bad players in theory, but it is impossible to quantify and therefore difficult to prove.

For example, I 3b/4b a lot and then people end up 5b jamming AJ vs me and it doesn't work out to well for them.