I see so many RIO videos where good players 3bet all kind of polorized range
strong hands like TT+ AQ+ but also a lot of semi garbage like:

A2-A5o A2-A5s 65s-98s K5s+ for blockers and so on.

Whenever I try to mimick this strategy and have polorized range from the blinds even when they fold a lot I end up losing money like -2bb with hands like that over 200 instances of those hands so basically I lost more than if I would have folded hands like this.

Also if we have TT+ AQ+ 3.5% range and then 6.5% polorized weak hands we become exploitable when we getting 4bet - and we will be OOP in a big pot with low SPR if we call.

Can someone confirm that hands like this A2-A5 / A2s-A5s 65s-98s or K5s+ can be profitable from blinds? / vs CO and BTN opens.

I know in theory it should work cause we have blockers - so if they fold 60% to 3bet whenever we have hand like this they should have strong hands with lower frequency and hence fold even more so if they fold 66%+ we end up with immediate profit.

But in reality my database saying otherwise.
Looking for different opinions and suggestions.

Thanks

they aint folding to a 3bet enough. adjust.

1) Just because some high stakes pros are doing something doesn't mean it'll win the most money in your games.

2) They don't 3b hands like A5s to autoprofit from folds. They 3b so their range isn't super face up and they're tougher to play against (and have a bit more board coverage) in 3b pots.

Imagine a player who only 3b AA + KK over a big sample -- any time they 3b, you could just fold...or take flops with some hands (to setmine, for example) and you'd know exactly what they had when you're trying to navigate postflop. But let's take an opponent who's 3b AA, KK, A5s, 97s, JTs...well they'll be much trickier to play against on a wide variety of flops, and they'll now sometimes be bluffing and sometimes vbetting post, which gives you trickier decisions than just knowing they only have AA/KK.

Is this sort of disguise important in your games? Who knows.

Also, if you only 3b a super value heavy 5%, they'll be able to open wider vs your bb because you don't put them in tough pre spots by 3b appropriately.

If someone is folding 60% to 3b, they're probably not folding 66%+ if you have A5. Yes, you have the A blocker, but you need to think about this more because you're missing something.

You're posting in beginner's questions, so one assumes you're playing at beginners stakes - why do you think that low limits play the same as high limits and that you should replicate what higher stakes players do against their opposition, rather than exploiting the hell out of your opponents?

I think your SB 3-betting range (at 100bb deep) should be more linear than polarized, since you're more likely to be called by a player in position, than cause him to 4-bet or fold. An optimal SB 3-bet range will look somewhat like an UTG opening range, since it's made from hands that play pretty well OOP.

If I run the filter "3-bet=true" for the SB position on my latest database, I get this heat map:


Note that not every hand on that chart is always a 3-bet. I'd only ever 3-bet ATo, KJo, KTo, QJo and Q9s/Q8s if the opener was on the button, and even then I'd only do it vs wide openers with high fold to 3b numbers.

FWIW, that chart is for 2NLz, where people don't fold to 3-bets particularly often. But provided you have half a clue about how to play bloated pots OOP, 3-betting is a very profitable strategy. A lot of regs 3-bet much more than me, and there's a fairly strong correlation between high 3-bet frequency and overall winrate.

Lots of good posts already ...

1) Who opened from where .. and how sticky are they or any callers?
2) Stack sizes?
3) Will this V pay me off?
4) What is my image/style right now compared to the others still in the hand?

Heed the comments about table dynamic ... every table, regardless of stakes, has it's own ebb and flow. If they think you are running the table over, you will get some wider backlash eventually. If you continue to get lots of folds then keep it up, but just be aware of pot control when you start seeing a lot of Turns with these same opponents. GL

+1 this is really good stuff, take notes op.

I can’t count how many times back in the day, a student asked me this question. “Chart XYZ tells me to 3bet those hands but my database shows I am losing money when doing that!”.

The answer to that is obviously complicated and multi-faceted and there’s lots of valuable information in the previous replies in this thread, but for the vast majority of micro stakes players the problem isn’t a too wide or to tight 3betting range. Sure, 3betting a little more or less from the blinds might be an improvement, but the reason why so many players lose money OOP in 3bet pots with hands like A5s is simply a lack of postflop skill.

If you 3bet from the SB and BU calls, you have to play postflop poker. If I am not able to do that well because I lack experience in 3bet pots, am scared money or have no idea what hands BU calls 3bets with, I’m in a world of pain. High stakes regs know how to play in 3bet pots, most micro stakes players don’t. Therefore they can play a wider range of hands in those spots. It’s just so much easier to play a 3% range than a 30% range.

Bottom line: it’s way cheaper to be exploited for folding too much preflop than for folding too little on the river in a 3bet pot. Fixing postflop leaks might be a good approach before 3betting more pre.

That might be survivor bias. 3-betting with a lot of hands increases variance - which should polarize the player base into groups that seem to win a lot and groups that bust out of poker for good. Obviously the latter no longer post any win stats.

For this to be true, increased 3-betting would have to cause a lower WR versus lower 3-betting. The reverse is true, imo, because increased 3-betting leads to higher WR which reduces variance.

As has been pointed out though, if post flop skills are lacking, then no balanced preflop strat will cure bad poker.

Well, let's take an (arguably contrived*) example.

Player base with
- 1024 players who play only coin-flip all-ins
- Several thousand "normal" players (who may be on the other side of these coin flips or not...doesn't matter)

After 10 all-ins per player in the first group there is (on average) only one left. All other's of that group have lost their bankroll and left poker for good. That one player then proceeds to play normal poker.
The 'high variance' play then looks to have a better win rate than the average play. (as the departed players no longer appear in any current statistic)

*contrived because this probably doesn't hold over large player bases and large hand samples in skewing the issue perceptibly.

I understand what survivor bias is. Coin flips are not balanced 3-bets in a well constructed poker strategy. Increasing WR is the best way to increase the odds of success, with the added bonus of less variance.

I'm not sure if it's survivorship bias or another named bias, but one reason why the biggest winners in my database (which still only has small sample sizes) tend to have higher 3-bet frequencies is that they probably got dealt better hands than other players did on average over the sample sizes I've got on them.
Hence, their 3-bet frequencies and their winrates are higher than average, because they were getting better hands on average.
With larger samples, however, it's pretty clear to me that someone playing 19/5/2 doesn't win as much as someone playing 21/17/7. To make more money with the top of your range, (and also to win more with bluffs) you need to build bigger pots.

Sorry, Arty, but I can't take this chart seriously when AKs is that dark red.

You are misapplying the term "variance." Variance is a term with a specific mathematical meaning which represents the average of the squared deviation from the mean (EV).

Folding has a variance of zero at that decision point, exactly equal to the EV of folding which is also zero. Any action which puts money in the pot almost necessarily increases variance relative to folding, which in turn increases the variance of the entire strategy. Starting to 3-bet hands that you are currently folding necessarily increases variance.

The data points are the completed hands of poker, the data is no more granular than that.

What do you mean you can't take it seriously? Everything on that chart that isn't grey is something that I've 3-bet in the SB. It's a small sample size, and AKs has lost money over that small sample (as has JJ, unfortunately), but it's still a fairly standard SB 3-betting range, and I'll continue to 3-bet with the bulk of the hands in that range (depending on who opened), since I believe them to be +EV on average.

Out of curiosity do you ever call out of the small blind or is it always a 3 bet if someone has opened before you?

I read an article that said you should only 3 bet out of the sb and almost never just call.

Is that a strategy you agree with?

I flat with a handful of hands if the opener was UTG or MP (stuff like 99/88, AQo), and some slightly weaker pairs (77-55) vs CO or BTN, but mostly it's a '3-bet or fold' position.
If the open is a minraise then you can flat with a couple of suited hands like K9s, A8s if you think calling with them will have a higher expectation than 3-betting, but you shouldn't let flatting become a bad habit.

Ok thanks mate.

ArtyMcFly makes the best posts.

OP,

Mimicking the pros only take you so far, in some cases it's counter to good thinking. Their decisions are based on....essentially thousands of pages from their personal "strategy book", and upwards of millions of hands to build a knowledge base. Their betting patterns are functions of precise sequencing and calculations - they aren't 'systems' (meaning 'shortcuts'), and they shouldn't be easy to replicate.

If someone is making a profit over your 3-bets by 4-betting, there is technically a mistake in your ranges. It is calculated by thinking about the preflop only (some likely have solved it all the way with solvers).

You might have to call with up to all of your good hands when you are polarizing and you get 4-bet. If your range is too strong, the opponent might exploit you by folding. Normally, you exploit their overall play, including their 4-bets. You can think about your usual opponent and exploit that, losing some back when you get exploited and adjust the strategy if you get exploited too often.

With smaller stacks, like in tourneys, you don't necessarily polarize about at all but when you use blockers (possibly other hands also) or just go all in with more or less weaker kicker high cards and small pairs. It is not exploitable there; they use showdown value charts to prove their ranges, though they are regularly not the best ranges to use because of the opponent's opening and calling/reraise ranges (and the players left to act), and other stuff like ICM and weaker players, multi-tabling.

Small blind in NLH likes to 3-bet liner because there is more value in getting the opponent to fold preflop and later in NLH (this is why the 3-bets can be more or less dominated when they get called in NLH), and there is some BB squeeze phobia. When not having a line on the players, it is the easier way.

The rake might also play some role here, e.g. if the rake cap disappears earlier when one 3-bets, it makes more sense to make the pots bigger (and play deeper). The pot being bigger has a fine line in the value of hands one is holding and in the size of the pots one is re-stealing. Not making the pot bigger when you have stuff like that doesn't look good to me, and you do prefer the opponent to fold most of the time, so you give him that option also.

I am not sure what you mean to say. If we estimate variance empirically we look at the results of individual hands. If we calculate it analytically we look at the result of every possible line in the decision tree of our strategy.

I probably just misunderstood the point of posting that chart. I thought you were trying to show what types of hands are reasonable to have in a SB 3-betting range, but the chart doesn't support you when the sample size is clearly too small.