Fold more often to 3-bets when OOP
Consider a pretty hand: A♠J♠8♦7♣. You open from middle position and face a 3-bet from the button, a player with a 7% 3-bet width1 in this situation. You want to call, right? Let’s look at a few reasons why this might not be the smartest idea.
This hand is in the top 16% of all starting hands, but so are most (almost all?) of your hands opening first in from middle position. A quick equity calculation against a 7% range gives us 41% pre-flop.. that’s not bad is it?2 We’ll see another way of formulating this question at the base of the article. For now, let us dig a little into how this hand will behave on the flop. This hand is essentially a weaker version of many other Axxx hands which you have in your pre-flop opening range, including such luminaries as {AKQJ, AQJ9, AJT8ds}. We certainly don’t need it to improve our range on Ace high flops. Does it help us on other flops? Consider the most obvious flop it helps us on: T9x. When we hold a wrap (as opposed to a made straight), the {QJT}-components of our opponent’s range have a 60-40 edge on us, as do all of the {overpair+FD} hands whenever the flop is duochrome. Even when we flop top two pair we only have a 58% edge against {overpair + gutter}.
Of course, the most common flop we will encounter that will entail some thought is the Ace high flop. The ‘good news’ is that we have a 53% edge on an A high flop. The not-so-good news is that whenever our opponent also has an Ace we are a 68% dog! This will inevitably lead us into the 3BP OOP no-man’s land of check-calling the flop and waiting for our opponent to pick between realizing his equity on the turn with a weak draw and firing the turn and river with a polarized range into our ‘known’ top pair.
Another favorite hand of the newly-blooded Omaha player is ‘good’ Queens, say QQ65ss. Why it’s a suited connector with set potential, what’s not to like? Indeed this hand is a 7th percentile hand and yet is an auto-muck against a 7% IP 3-bet. The problem is that your opponent will have a higher flush draw than you around 1/4 of the time, and so the 12% of time that you “fist-pump” with a flopped flush draw you aren’t in that great a shape. This problem persists when you turn a ‘back-door flush draw’. To compound the misery, even when you do flop a set, your opponent will have 30%+ equity against you around 27% of the time! Set-mining OOP in 3bet-pots is a losing proposition; a solid guideline with disconnected pairs is to only call a 3-bet if you would be happy shoving the flop over a C-bet with {Overpair+gutter}.
I’ll conclude this discussion with an introduction to examining flop equity with a more sophisticated metric. Consider a situation where you ‘know’ you are going to face pot-sized continuation bets on both the flop and the turn. How much equity do you need on the flop to call them both profitably (disregarding implied odds)? The pot will be 9 PSFBs (Pot-sized flop bets) by the river, of which you will have contributed 4 post-flop, meaning you need 45% equity against your opponent’s range to call on the flop and turn. AJ87ss has flopped hand vs range equity of at least 45% around 38% of the time against a top 7% range. QQ65ss has flopped hand vs range equity of at least 45% only 30% of the time against that same range! It should be immediately apparent to the reader that, against a range composed of many such hands, the 3-bettor can turn a profit by potting many flops with his entire range3 . I would encourage all of my readers to stop calling 3-bets of this 5-9% width from those 40-50BB stacks that proliferate on some well-known poker sites. You do not flop sufficient equity with sufficient frequency that your strategy of ‘making good flop push/fold decisions’ will render this line +EV.
Thanks for reading! As hinted by the “Step” title, for a few months this blog is going to trend towards releasing more articles intending to take a ‘new’ Omaha player from the basics of play to a competitive poker game. I am reserving the more dense and detailed articles for my subscriber section (sign-up in the side bar if you haven’t already). This shouldn’t discourage the more advanced reader, since many of the leaks I will be discussing remain as core features of your opponents’ strategies in the mid/high-stakes games.
This week I am launching a 5 seminar series on “Defending the Big Blind against a Button open”. The series will cover the following topics:
Constructing Calling Ranges and general flop principles in a single-raised pot
Constructing 3-betting Ranges and general flop principles as the 3-bettor
A Case Study in a single-raised pot
A Case Study in a 3-bet pot
Defending turn/river transitions in a single-raised pot
I have had sufficient reservations from present clients for the first session this Friday 3rd May. If you would like to join a second, parallel seminar stream starting next week, please get in touch. It is possible to attend individual sessions or reserve a spot on all 5 at a discount. Places are limited to 5 per seminar so get in early!
That’s enough promotion for this week, good luck at the tables,
Quad
Hi Quad!
I’m novice in poker math and I have question : you said that AsJs8d7c in the top 16% of all starting hands.
But when I have used
1) hand compare option in propokertools : AsJs8d7c vs 16% – I have seen difference in equity ~ 45% vs 55% => 16% range better than AsJs8d7c
2) Omaha Hi Hand Ranking in propokertools : AsJs8d7c has average mark 16 versus random hand
When I play poker and use HUD – for example I can see that opponent 3bet range = 16% => Am I need to fold AsJs8d7c? if no money from another players in the pot.
Please explain
Hi Evgenii,
Thanks for your comment. When a hand is “in the top 16% of starting hands” it can be anywhere from the 1st to 16th percentile inclusive. In this instance, AsJs8d7c is in the 16th percentile. As a consequence, when you run the HandvRange equity you would expect AJ87 to be a dog versus 16% because most of the hands in the top 16% dominate AJ87. In fact this effect gets more pronounced the tighter the range is due to the concentration of AA/AKK/KKds. The reason that we aren’t too concerned about equity with say a 50th percentile hand against a 50% 3-bet range is that those {AA…} hands form a tiny part of such a wide range. The reason this effect is very important against say a 7% range OOP is that the hands that dominate us constitute a great fraction of such a tight range.
In answer to your response to a 3-bet with AJ87 versus a 16% range, many other factors need to be considered. For example we should always continue having opened on the BTN at deep stacks because otherwise we fold so frequently that our opponent immediately profits by 3-betting virtually any 4 cards. In contrast when we open from UTG OOP we have such a tight range that our opponent is not exploiting us if we fold this hand. Whether to continue with this hand OOP would depend on exactly how our opponent constructs his 3-bet range (more ds trash makes this a better call, broadway concentrations make it worse) and how well he plays post flop. Since this hand has mediocre nut potential, we are looking for either a very passive player (who allows us to get to SD often) or a player who C-bets the flop with such a high frequency that we can check/shove or reverse float light.
That should get you started on understanding a much underestimated part of PLO- preflop strategy in 3bet pots.
Quad
Thank you for response.
My Odds Oracle trial expired and I haven’t bought it yet so I ran simulation on their website, I used “Count” function and got 5.25% answer (9%-base count,5.25%-optymized count). I used to think it is correct way to get answer on my question but I miss something. Seems I have some fundamental errors in my thinking.
Hi! I have a question about an influence blockers on a width of 3 betting range. Assume that we open KsKhJs7h and get 3bet from BU who has 9% range. Suppose for the sake of simplicity that he 3bets top 9% hands. When i put this situation to PPT I found out that 9% range drops to 5.25%! Does it mean that he 3bets me in only 5.25% times when i hold this hand? AA!AAA:xx composes 1.83% of starting hands => he’ll have aces in 1.83/5.25=35% times? Am I right?
Hi xazervl,
Thank you for your comment.
When I ran the simulation, “How often does Player 1 Match Hand Range 9%,” with those cards as blockers PPT returned a frequency of 7.3%. I’m curious as to what simulation you ran to get 5.25%, since 9-5.25 =3.75% >> all combos of KK. I would caution against the assumption that a BT 3-bettor drops most trip Aces from his range (I certainly don’t). If you accept my methodology, then the ‘blocker effect’ you wanted to calculate comes to 2.2%/7.3% = 30% of the time your opponent will have aces.
That should help,
Quad
Thanks for another good article. Please keep free info to us micro donks (perhaps slightly less so now) in addition to your paid services.