Should I purchases snowie learn cash game for micro and low stack ??

Hi everyone,

I'm sorry if this has been answered before, but I have a PRO subscription on Pokersnowie and I was wondering if there was an easy way to have access to preflop ranges (a tool a bit like the free preflop ranges but updated with current solutions) or must I use the scenario tool and manually find the preflop ranges for each spot?

Thanks in advance

U can train against Snowie and u can check preflop advice in analysis. Also can import hands from Pokerstars and check preflop advice in analysis. For what u r asking for ,u need to create your own charts in excel or whatever after checking Snowie's advice for a spot.

Thank you

Yes

Follow the free preflop advisor headings to create the scenario. (eg 100nl)
Take the preflop advice screenshot

It doesn't take much time

I use a word document with the index linked to the images

caveat: snowie keeps updating itself - so keep updating the 3bet/4bet spots once in 3months or so

Hi,

Apologies if this has already been answered but sometimes Pokersnowie gives you a suggested pre flop bet size but when you look at other pre flop bet sizes they have a higher EV. Is this due to Pokersnowie trying to balance your range?

Also, there doesn't seem to be much in the way of a pattern to pokersnowies pre flop bet sizing, it all seems a bit random to me. Is there a pattern to it that I'm missing?

Thank you

Because it picks one size for its entire range (rather than "range-splitting" with multiple sizes), some hands are 'forced' into using sizes that might not be the most profitable in a vacuum. e.g. On the river with the nuts, you'd often love to overbet jam for max value, but if the range as a whole does best by betting small, Snowie would bet small with the nuts.

There isn't much of a pattern pre-flop at some stakes (rake is a big factor), but it's basically just the case that the artificial neural net tried various sizes and ranges, and had a slight preference for one or the other, and this varied from position to position.

Thanks Arty, that makes sense!

I was expecting to find some kind of pattern pre flop but couldn't work it out so I'm glad I asked as I could have been looking forever lol.

I do really like and value Pokersnowie as it has helped me fix some big leaks in my game that I didn't know that I had but in the back of my mind, I have trouble trusting some of the plays.

For example Snowie seems like a very passive player in comparison to my previous game, I struggle to get comfortable skipping so many c bets and bluffing opportunities as I feel at the micros I'm losing a lot of pots that I could have taken down.
Also, the bet sizing is hard to get the hang of. All the 1/4 pot bets seem really counter intuitive after years of betting 50%+

As an average player winning a little at 10NL micros, I feel that it makes sense to adjust my game to try to emulate Snowie, learn how to play with 1/4 pot size bets, skip a lot of C bets and become a more passive player like Snowie.

I guess what I'm hoping for is some confirmation that this is the sensible thing to do so that I can have the confidence to adjust to a Snowie style of play without a part of me holding me back as it feels wrong.

It makes sense that I should try to get out of old habits and learn how to emulate Snowie for the good of my game, doesn't it?

There are definitely elements of Snowie's play that will be useful even for 10NL. You just have to experiment in your real life games. I got into the habit of c-betting 33% of pot, as 25% just looks like a pathetic amount of money at 5NL/10NL (it's like a minbet) and has almost no fold equity at all.
Lowering your c-bet frequency (in order to reduce spews, and to protect your checking range, and/or enable you to pick off turn bluffs by aggro regs) is likely to be even more useful for when you move up. If your player pool is over-folding to half-pot c-bets, then by all means keep c-betting a higher frequency.

Yeah, I have been playing about with 25% and 33% bets. When you put it like that, realistically they're probably not over folding at these stakes to be fair are they lol. I'm sure that I do just need to reduce the frequency and size of my c-bets. I'm just having a look on pokersnowie for situations where I can cut out some of my cbets and how to play on the turn. Thank you!

Does anybody have any pokersnowie stats by any chance? I would be interested to see how mine compare.

Snowie uses a 1/2 pot open raise size from all positions except the button, where it uses a MIN open (with the exception of the 0.5/1 stakes, where it uses a pot sized open on the button).

Whenever Snowie bets a fraction of the pot, it then rounds its bet to the nearest 1. So if you're playing 0.5/1, all the half pot (2.25BB) opens get rounded down to 2, effectively becoming a min open. If you're playing 5/10, all the half pot opens get rounded from 22.5 up to 23. If you're playing 3/6, all the half pot opens get rounded up from 13.5 to 14. Etc.

This might explain some of the confusing preflop sizings. They are actually consistent across all stakes, just getting rounded.

Yes, but the range is almost the same for 2.25bb open or 2bb open.For example, at 0.1/0.2 Every position is opening 3.5bb(pot size bet) except HJ for which Snowie suggests to open 0.5 pot(which is 2.25bb) but if u train against it u will notice that it is choosing 2.5bb open from the HJ.

Play a training session for 100+ hands or so - statistics section will give you the breakdown incl stats for you and all the snowies

https://www.pokersnowie.com/help/statistics.html

I feel the same way - I have become too nitty pre and weak tight post - but guess what? -

what stakes you play? i used snowie mainly for training esp 3bet/4bet pots

otherwise i would advise you to tweak your game to the pool you are playing - be aggressive

I´m still using snowie a lot, and love to practice with the training function before playing for real, usually at the lowest limit (20NL).

I know we were told the preflop ranges were built considering the rake structure of the stake, but after that, does snowie adjust its play postflop? (like a solver will depending on the rake inputs we make?).

I remember, from watching a few vids and also from real, seemingly good opponents, that we´re not really cold calling from MP and CO for example, which really makes sense considering rake for example. Aggressive squeezers imho obv impact this a lot. And I´m sure I heard Peter Clarke mentioning this when he was working with monker. Since snowie still calls from those positions, is the 3b/fold strategy mostly exploitative, or is snowie actually wrong/outdated?

Thank you, any input will be hugely appreciated.

Rake impacts all of snowie's outputs pre and post. This is why you can observe some small changes in its strategy when you increase the stake.

Snowie's preflop ranges are pretty close monker ranges, at least for 100bb cash. I think the places you see differences are mostly due to the limited number of bet sizes snowie has and the way monker ranges are solved. The biggest of these is the choice of 3b/4b size snowie use vs what most monker solutions use.

The monker ranges I have also have CC range in all spots except the SB(removed for simplification purposes).

It seems to me that it's more about simplification and saving time and money than a strategy built around exploitation.

I'm far from an expert on the subject, but my understanding is that adding additional actions like more bet sizes and cold calling in the early positions and the SB into preflop solvers increases the size of the game tree. Increasing the size of the game tree requires more RAM to solve it, like hundreds or even over a thousand GBs of RAM to solve really big preflop game trees. Renting servers with that much RAM isn't cheap. Bigger game trees also take longer to converge to the desired accuracy. Using less bet sizes and removing less frequent actions to make the game tree much smaller requires less GBs of RAM, results in faster solves and doesn't affect the total achievable EV too significantly.

Without knowing how your opponents deviate and how to exploit them, the highest achievable EV would be attained from allowing the solver to use all possible actions and bet sizes. Even if this was possible with access to infinite RAM and time, it would be impossible for any human to come anywhere close to implementing the exact output of that solution at the tables. For this reason, many people making solutions attempt to simplify things and save time and money by using less bet sizes and eliminating less frequent actions while trying to maintain as much of the total achievable EV as possible. Because of that, you will come across several preflop solves that don't have cold calling from the early positions and the SB and possibly other less common scenarios removed from their game tree.

We have been told by the developers that Snowie trained against exploitative agents. I can only assume that excessive squeezing would have been one of the areas it trained its entire strategy for and found that the hands that it chooses to cold call with were still more +EV for its entire strategy to cold call with than to develop a strategy where it only 3bet or folded from those positions.

As mike1270 pointed out in the post above, given the option to cold call in the early positions and the SB, even preflop solvers will still provide a cold calling range in many spots. Altering your strategy from the suggested cold calling ranges to a 3bet or fold strategy because of excessive squeezing might not even be the most EV way to exploit excessive squeezing from your opponents. It certainly doesn't appear to be the most +EV way to play your entire range if both Snowie and the preflop solvers suggest having a cold calling range when given the option to do so.

@mike1270, @MCAChiTown Good explanation, understood.

Thank you very much!

Is anything being done to make Snowie stronger?

Hey folks,

So I want to check my real life hands with pokersnowie (I know it´s flawed, but I´m still convinced big mistakes - blunders are still blunders no matter what), and since I don´t really understand snowie points/session evaluation concept, I did my own spreadsheet.

Basically, I assume a cheater using snowie advice 100% of the time would achieve 10 bb/100 at stars full ring games (5nl and 10nl) and then subtracts the mistakes to estimate my true EV.

Here comes the questions:
- Are there any really big flaws in working that way?
- Is 10 bb/100 too much (or too little) to expect from pokersnowie at those games? Obviously we can´t really know snowie´s true winrate, but we can say if smth is too far off (above or below). 10 bb/100 seemed a conservative "guesstimation" to me, but what do you guys think?

Thanks

Know one knows how much snowie would win/loss at 10nl.

The good news is it doesn't matter. Just put in your HHs or run sims and snowie will give you feedback. There is no reason to overcomplicate things.

I'm brand new to Snowie, so I'm sure I'm doing something wrong or misinterpreting something, but could someone explain to me why Snowie says I should call here with 6 high and no draw?

We´re going to bluff a lot of rivers with this combo if villain x´s to you.

That’s the only rationale I could think of myself. But here’s the problem, if we are calling with that, what are we folding? And if Snowie’s strat in my position is to almost never fold, then the SB’s strat can’t include any bluffs, therefore I can’t call, etc. it’s not a balanced strat.

A friend of mine put it in a solver and it said it’s a fold.

Anything you look at in snowie that isn't going to be a high frequency situation should be taken with a grain of salt.

This means anywhere on turns/rivers, or spots like this where its a 25% psb, multiway spots, or where action is very convoluted (limp, min-raise, multiway etc.)