HU SNG strategy

I started a thread about goofy hands from HU SNGs in the brags forum, but you know, the intertubes will run out of room to post all of them.

I don't know why, but HU SNG's are by far the spewiest form of poker I've ever seen. Basic plays that 2NL fish are too dumb to make happen all the time. People shoving 90 bbs into a 4 BB pot drawing dead happens all the time.

I'm not going to say I'm an expert, but I've been playing a lot of low-BI games and I have some thoughts on them. Most of these will be basic, and involves basic poker concepts, but some cannot be stressed enough.

1. Play somewhat straightforward. Like, if you raise preflop with AK and then cbet an A-high board, many HU players I've run into will do unbelievably spewtastic things. They'll call you down with king high. They'll check-raise you with bottom. They'll 4-bet fold the flop.

2. Raise the button constantly. Some people could probably get away with raising like 100+% of buttons, I play more conservatively and raise like 80-90%.

3. Use a lot of minraises. Seriously, a lot of players will fold to a minraise constantly and hardly ever 3-bet. If they do, you can profitably minraise 100%.

Try to keep pots small without a made hand. Semibluffing and bluffing with air obviously have their place, however, you won't believe how light many players at the $1-$5 levels will call down. You can bet 1/2-2/3 pot on flop, turn and river with TPTK and you'll get value from A-high constantly.

4. Everyone floats wide OOP. I'm not saying I don't do it or that you shouldn't do it, but I can't see how doing it constantly can be profitable. It'd be nice to have a semblance or equity when you do obviously. Against some people who float too much, you need to cut down on how often you cbet with air and widen your value range and never slowplay.

5. Betsizing is hugely important and is the biggest problem I see among players at this level.

You have the type who will minbet 3 streets with the nuts and others who will mash the bet-pot button or massively overbet shove. Or they have serious bet-sizing tells: donk min-bet with air or a weak draw and then bet pot with a real hand.

I took down a 4-man HU shootout in 4 hands.

Tournament 1
Hand 1


No Limit Holdem Tournament
2 Players
Hand Conversion Powered by weaktight.com
Step 1 Heads Up Shootout

Stacks:
SB (1,500)
Hero (BB) (1,500)

Blinds: 15/30

Pre-Flop: (45, 2 players) Hero is BB A♣ A♦
SB raises to 60, Hero raises to 120, SB calls 60

Flop: 8♥ 6♠ 3♠ (240, 2 players)
Hero bets 180, SB calls 180

Turn: 8♣ (600, 2 players)
Hero bets 330, SB goes all-in 1,200, Hero goes all-in 870

River: 2♥ (3,000, 2 players, 2 all-in)

Final Pot: 3,000
Hero shows two pair, Aces and Eights
A♣ A♦
SB shows two pair, Eights and Sixes
6♦ 7♣

Hero wins 3,000 (net +1,500)

SB lost 1,500


Tournament 2
Hand 1

No Limit Holdem Tournament
1 Players
Hand Conversion Powered by weaktight.com
Step 1 Heads Up Shootout

Stacks:
SB (1,500)
Hero (BB) (1,500)

Blinds: 15/30

Pre-Flop: (45, 1 players) Hero is BB A♥ Q♣
SB folds

Final Pot: 45
Hero shows
A♥ Q♣

Hero wins 45 (net +15)

Hand 2

No Limit Holdem Tournament
2 Players
Hand Conversion Powered by weaktight.com
Step 1 Heads Up Shootout

Stacks:
Hero (SB) (1,515)
BB (1,485)

Blinds: 15/30

Pre-Flop: (45, 2 players) Hero is SB 5♥ 5♦
Hero raises to 60, BB calls 30

Flop: 3♠ 9♥ 10♦ (120, 2 players)
BB checks, Hero bets 60, BB calls 60

Turn: 3♥ (240, 2 players)
BB checks, Hero bets 90, BB calls 90

River: Q♠ (420, 2 players)
BB checks, Hero checks

Final Pot: 420
Hero shows two pair, Fives and Threes
5♥ 5♦
BB shows
K♠ 5♠

Hero wins 420 (net +210)

BB lost 210


No Limit Holdem Tournament
2 Players
Hand Conversion Powered by weaktight.com
Step 1 Heads Up Shootout

Stacks:
SB (1,275)
Hero (BB) (1,725)

Blinds: 15/30

Pre-Flop: (45, 2 players) Hero is BB 7♦ 2♠
SB raises to 90, Hero folds

Final Pot: 120

SB wins 120 (net +30)


No Limit Holdem Tournament
2 Players
Hand Conversion Powered by weaktight.com
Step 1 Heads Up Shootout

Stacks:
Hero (SB) (1,695)
BB (1,305)

Blinds: 15/30

Pre-Flop: (45, 2 players) Hero is SB J♠ Q♦
Hero raises to 60, BB calls 30

Flop: 8♦ J♥ 4♦ (120, 2 players)
BB checks, Hero bets 60, BB raises to 180, Hero calls 120

Turn: 10♣ (480, 2 players)
BB goes all-in 1,065, Hero calls 1,065

River: Q♠ (2,610, 2 players, 1 all-in)

Final Pot: 2,610
Hero shows two pair, Queens and Jacks
J♠ Q♦
BB shows a pair of Jacks
7♥ J♦

Hero wins 2,610 (net +1,305)

BB lost 1,305

Hm, color me intrigued.

So do you min-raise pre-flop most of the time? What advantages does that have, other than making your flop c-bets smaller?

Also how often are you calling from the BB?

Min-raising allows us to get our opponant to call and play lots of pots OOP. We're also giving him odds to call with just about anything, so if he folds he's making a mistake, if he calls he's probably making a mistake. But yeah, mainly the positional advantage (huge in HU) as Baudib mentions.

yeah, I usually minraise but if I notice someone is calling like 100% I'll start raising more.

I like the deep stacked (3,000 chips) slow blinds structure on FT. The turbos obviously force you to make a lot more moves and gamble more.

also, if you raise the btn constantly and villain starts to adjust by 3-betting a ton, limp once in a while. if you are always raise and then limp, everyone suspects u have a monster, so bet flop always and you'll take it down most of the time.

Yeah, the deep stacked HUSNGs are a godsend. I really like them. They may not be as quick to play as the normal ones, but I think they provide the least variance of all poker formats. And that's always good. Plus, you can have a huge edge over most weaker players.

I've just started playing them and I think they're great bankroll builders. I've read on a lot of places that you can go as low as 10 BI with the deep stacked normal speed ones, if you're confident enough in your ability (because of the very low variance).

I was considering playing PLO HUSNGs, but I don't really know if it's even worth it. There is a lot higher variance and with the tournament structure, you'd be forced to gamble. If there are a lot of fish I might give it a second thought, though.

Another note:
No one can fold a pair of Aces heads up. In fact, if someone makes a pair of Aces, they have the immortal nuts and believe you will call them with worse.


No Limit Holdem Tournament
2 Players
Hand Conversion Powered by weaktight.com
$5 + $0.25 Heads Up Sit & Go

Stacks:
Hero (SB) (3,950)
BB (2,050)

Blinds: 15/30

Pre-Flop: (45, 2 players) Hero is SB 3♥ 3♦
Hero raises to 60, BB raises to 90, Hero calls 30

Flop: K♥ A♥ 3♠ (180, 2 players)
BB checks, Hero bets 180, BB goes all-in 1,960, Hero calls 1,780

Turn: 8♣ (4,100, 2 players, 1 all-in)

River: Q♠ (4,100, 2 players, 1 all-in)

Final Pot: 4,100
BB shows a pair of Aces
A♦ 4♦
Hero shows three of a kind, Threes
3♥ 3♦

Hero wins 4,100 (net +2,050)

BB lost 2,050

Here's a situation I sometimes struggle with. In a vacuum, when you and villain have an aggressive dynamic going with a lot of 3-betting, I'd probably always stack off with TT+/AQ/AK and maybe AJ.

This spot is weird because this guy was so terribad and I felt I could own his soul forever. He folded his BB to minraises a ton and basically played his cards face-up postflop. He check-folded when he missed and check-called when he hit something. On occasion he would just c-r shove $2,000 into a $120 pot. As preflop raiser, sometimes he would cbet about 40%. If his draw came in, he'd bet pot. He made serious light calldowns as PFR if I donked 3 streets into him.

Pretty easy to play against.

3 example hands

No Limit Holdem Tournament
2 Players
Hand Conversion Powered by weaktight.com
$5 + $0.25 Heads Up Sit & Go
first hand
No Limit Holdem Tournament
2 Players
Hand Conversion Powered by weaktight.com
$5 + $0.25 Heads Up Sit & Go

Stacks:
SB (3,000)
Hero (BB) (3,000)

Blinds: 10/20

Pre-Flop: (30, 2 players) Hero is BB 9♣ K♣
SB raises to 50, Hero calls 30

Flop: 6♣ 6♥ 8♦ (100, 2 players)
Hero checks, SB bets 70, Hero calls 70

Turn: J♣ (240, 2 players)
Hero checks, SB bets 140, Hero raises to 380, SB folds

Final Pot: 760

Hero wins 760 (net +260)

SB lost 260

After this hand, he almost never cbet.

Hand 2
2nd pair is pretty much the nuts in this hand, but can't be too careful.

Stacks:
Hero (SB) (3,967)
BB (2,033)

Blinds: 15/30

Pre-Flop: (45, 2 players) Hero is SB 7♦ 7♥
Hero raises to 60, BB calls 30

Flop: J♦ 2♥ Q♠ (120, 2 players)
BB checks, Hero checks

Turn: 9♣ (120, 2 players)
BB checks, Hero checks

River: 5♦ (120, 2 players)
BB checks, Hero checks

Final Pot: 120
Hero shows
7♦ 7♥
BB shows a pair of Jacks
J♥ K♦

BB wins 120 (net +60)

Hero lost 60

No Limit Holdem Tournament
2 Players
Hand Conversion Powered by weaktight.com
$5 + $0.25 Heads Up Sit & Go

Stacks:
SB (2,588)
Hero (BB) (3,412)

Blinds: 15/30

Pre-Flop: (45, 2 players) Hero is BB Q♥ J♦
SB raises to 60, Hero calls 30

Flop: 3♣ 9♠ 10♠ (120, 2 players)
Hero bets 60, SB calls 60

Turn: Q♦ (240, 2 players)
Hero bets 120, SB calls 120

River: Q♠ (480, 2 players)
Hero bets 240, SB calls 240

Final Pot: 960
Hero shows three of a kind, Queens
Q♥ J♦
SB shows
A♥ K♠

Hero wins 960 (net +480)

SB lost 480


Then this


No Limit Holdem Tournament
2 Players
Hand Conversion Powered by weaktight.com
$5 + $0.25 Heads Up Sit & Go

Stacks:
Hero (SB) (3,937)
BB (2,063)

Blinds: 20/40

Pre-Flop: (60, 2 players) Hero is SB 10♦ 10♥
Hero raises to 80, BB goes all-in 2,063, Hero

Okay I've played ~25 of these now, and I feel like I'm not 3-betting/raising limpers enough - maybe that's just compared to how my opponents are playing, which is pretty poorly, obv. I'm 3-betting like ~16%, and raising a limp ~24%, with overall 26/16 stats from BB. My SB is 94/91 lol. Insignificant sample size obviously, but yeah.

That sounds about optimal, play nitty in small pots in BB and be a maniac from BTN.

Okay cool, that makes me feel a little better.

I think the main thing though is being able to adjust to your opponent, which is something I'm still learning. Like, there were a few villains I played who would always donk bet into me for a psb, and I didn't know really what to do. What adjustments would be necessary?

Or even if an opponent didn't fold to a c-bet ever, even if they had like J high or whatever. I guess you just stop bluffing against that type of opponent, but v-bet pretty thin? Do you keep raising pre?

I just played a guy who would donk like $650 into $120 pots. I just folded until I hit a hand hard enough (TPWK) to call him down. He bet another $650 on turn and we both checked river, he had 2nd pair.

On the ones who float every cbet, continue c-betting your good hands and strongest draws most of the time and weaker draws like gutshot+overs at least some of the time. yeah i keep raising pre because we still want to play bigger pots and if they're calling ridiculously light so much the better.

Yeah, last night I played a guy that donked every single flop. Every time he was in the BB he would donk. It didn't matter if I was the PFR or not. I really didn't know how to deal with this kind of guy. I decided to wait for a decent hand, but he just folded when I raised him or check/folded the turn when I called him on the flop, unless he had a decent hand too.

Just for lolz: Before that I played a guy that would call an all-in every time. Every single time, lol. I don't know how I found it out, but it was really funny beating him 5 games in a row, lol.

raise w/bottom pair and good draws, flat call with 2nd pair and top pair+ and A-high/K-high/backdoor draws. Bet all turns when checked to.

Also, fold button more often against this guy.

If hero plays fit/fold here, which a lot of low stakes players do, the donkbetting is very profitable, hence villain's strategy. From villain's perspective hero only hits the flop ~43% of the time, and you may even be folding bottom pair / 2nd pair no kicker until hero adjusts. That means around half the time you do hit, you still fold to a donk bet. Thus villain's donk bet works >75% of the time, which is clearly profitable.

Clearly one adjusts by playing back at villain more -- since he's missing just as often as you. Raise or call with some strong made hands / some good draws, but also as a semibluff occasionally. If you can make his donk bet work <50% of the time, his strategy becomes -EV. This means you need to call/raise (profitably) more than 50% of the time.

As a simplified example, let's say you respond by shoving.

About 20% of the time you have top pair or better and the decision is easy. What about the other 30% of the time you're bluffing/semibluffing?

I've been playing with some PQL queries (http://www.propokertools.com/pql):

Assuming hero raises 90% of buttons, and BB calls with the top 25% of his range.

If hero holds a pair or better, then his equity is 52% (see below), making a shove profitable. Remember, this is only if villain donks 100% of pots.

Since hero is playing 90% of his range, he has a pair or better on the flop just over 50% of the time. That's enough to make villain's strategy -EV.

Big draws with overcards also give you plenty of equity to continue with the hand.

Once these are added to your raising range (and maybe remove some of the worst pair hands to balance it), you should be able to play back profitably.

The problem here is that your variance shoots way up as a result; but that's the way the HU cookie crumbles.

Man, that PQL is awesome! I'm definitely looking it up when I get back from work. It doesn't look all that complicated to learn either. It's mostly SQL syntax with custom functions. Really nice!

It is pretty neat.

It was nice to see it essentially confirming baudib1's advice; though he's refined it to be a bit more +EV by "slowplaying" monsters.

Okay, lets say you've adjusted a little bit to villain and started being a little bit more selective in your raising hands, choosing only the top 55%. Villain notices that and starts playing more hands from the BB:

A little bit better, but still very swingy. The query doesn't take into account folding equity, though.

Also, there are different types of villains- some will call a shove with any two high cards, some are more selective, therefore we have more FEq with them.

And when we add slowplaying a set or better, we definitely get into the .65 equity range. And that's crushing, IMO.

I feel like 55% is way too tight, but I'd drop the bottom 20% of hands, more if they start 3-betting you. Also, I'm not sure if this is theoretically correct, but when they 3-bet me I fold the worst offsuit Aces and prefer to continue with suited face cards.

Vs. the guy who donks 100%, I suppose we could refine our actions based on how often he check-folds turn. I'm not sure that floating 100% is ever going to be profitable, because he's going to hit something once in a while, and maybe we have enough air built into our floating range (backdoor draws/K-high. A-high is a value float) but those are hands where we mostly have at least 20-30% even if he hit the board. Maybe we can say we'll float the driest of boards (paired boards?) with ATC, where they will be afraid of slowplayed monsters the most.

OOP, I usually c/r the flop if it's a dry board. But I've found it doesn't get folds often enough to compensate.

As per when facing a 3bet, I tend to fold weaker aces as well. Broadways are a lot more playable than small aces, IMO.

Can we talk about applying Yeti theorem to HU?

I feel like this type of line works 99.9% of the time at low BI levels. It's not really a bluff in this case cuz K-high is the nuts anyway (we also block the nut flush draw), but still...


No Limit Holdem Tournament
2 Players
Hand Conversion Powered by weaktight.com
$5 + $0.25 Heads Up Sit & Go

Stacks:
Hero (SB) (3,470)
BB (2,530)

Blinds: 10/20

Pre-Flop: (30, 2 players) Hero is SB 9♣ K♥
Hero raises to 40, BB calls 20

Flop: A♠ A♥ 2♥ (80, 2 players)
BB checks, Hero bets 20, BB raises to 300, Hero raises to 580, BB folds

Final Pot: 960

same guy as last hand, more Yeti


No Limit Holdem Tournament
2 Players
Hand Conversion Powered by weaktight.com
$5 + $0.25 Heads Up Sit & Go

Stacks:
Hero (SB) (3,160)
BB (2,840)

Blinds: 10/20

Pre-Flop: (30, 2 players) Hero is SB 2♣ 6♣
Hero raises to 40, BB calls 20

Flop: 10♠ 6♠ 10♦ (80, 2 players)
BB checks, Hero bets 40, BB raises to 80, Hero raises to 160, BB calls 80

Turn: 7♥ (400, 2 players)
BB bets 20, Hero raises to 120, BB calls 100

River: K♦ (640, 2 players)
BB checks, Hero checks

Final Pot: 640
BB shows
4♦ 8♠
Hero shows two pair, Tens and Sixes
2♣ 6♣

Hero wins 640 (net +320)

BB lost 320

When I find the $5 HU player who has it, he's going to crush me.

yeah, not just broadways but I'd rather have J8s than A2o. 3-bet ranges HU tend to be dominated by AX.

when they min 3-bet, I pretty much min4-bet or fold.

OK yeti discussion went nowhere.

Do you guys have a donking range vs. PFR?

I don't know what yeti is

And not really - which probably isn't a great thing. The only thing I can think of that might be good is a d-bet to induce.

i gave examples of yeti above

This theory was created by a long-time 2+2 poster. It stats: A player who 3-bets on a dry, paired board is [almost] always bluffing.

Ah okay, yes that makes sense.

So in both hands we're not really bluffing when we're 3-betting are we? I guess that's sort of the point then. That could be good when we're up against certain opponents but to some people the 3-bet is still going to be scary.

Aaargh, shit like that tilts me to the extreme. I've been playing this guy for well over 45 minutes, grinding him down to less than 1K chips on several occasions, and he still manages to come back from it. Then this happens:

Grabbed by Holdem Manager
NL Holdem $160(BB) Replayer
Hero ($4,416)
Donkey ($1,584)

Dealt to Hero Q♣ Q♦

Donkey calls $80, Hero raises to $480, Donkey raises to $1,584 (AI), Hero calls $1,104

Hero shows Q♣ Q♦
(Pre 89%, Flop 83.5%, Turn 90.9%)

Donkey shows T♣ Q♠
(Pre 11%, Flop 16.5%, Turn 9.1%)




Donkey wins $3,168

They should have sent a ****ing poet!


Now, on a more serious note: the Yeti thorem. I haven't really applied it in practice, because in order for it to be of any use to you, you have to be bluffing too, and I haven't really found a great spot to 4bet bluff, given the stacksizes with most HUSNGs. I might try it soon in a home game though

Quite frankly, it terrifies me to 4bet bluff. Given the fact that if I'm wrong, I'd be out of the tourney in which, most of the time, I know I'm a big favorite anyway.

Q: What kind of tourneys do you guys play? Regulars, turbos, hyper-turbos? I personally find the double stack HUSNGs at FT most to my liking, given that I come from a cash bg.

Some guy just pwn3d me when i 3-bet bluffed a paired board lol.

Don't know about you, but ever since I learned about the Yeti, every other flop is paired, lol (Just like when I read Barry Greenstein's book, I always seem to notice when an Ace hits the river )

Btw, I think I found why the loose aggressive style works against most people. See, when you raise every button and 3bet quite often, taking stabs at every flop and cbetting almost everything (depending on board texture, of course, but most flops, because your opponent just plays fit or fold), your opponent just waits till he has a decent hand to play back at you. But, and this is where it gets interesting, now you have relative position.

Say you raise pf from the button, villain calls and checks the flop. You do your usual stabbing at around half pot (because he folds no matter what the bet is) and he check-raises you all-in.

Now you look down at your cards and either see trash and fold it or see the nuts and take his stack. Sound right?

First of all, he has no info about your hand. You're doing this both with pure trash, as a semi-bluff and when you have the nuts. So he's fed up with you stealing every single pot from him and decides to put you to the test (or at least so he thinks).

But he's at a disadvantage here, because once he goes all in, he has no control over the hand. Since people rarely do this with the nuts and more often than not with something as vulnerable as top pair, he's at our mercy, basically. And it's a pretty easy decision for us, from here on.

P.S. I don't think I have ever 4bet bluffed before. Maybe because I've been playing only micros, but I don't think I really have. I may have done it a few times live though. Afair, it always ended badly.

though that my style has change alot since i made this video.....enjoy

gg

gg, DoT.

How do you guys color-code people in the poker client? In cash games, I have 8 categories (from Loose/Passive/Passive to Tight/Aggressive/Aggressive), but I don't see much use of this in HUSNGs.

Maybe just a few categories: winning player, losing player (fish), winning reg, losing reg.

I just had an Aha! moment: at first I thought the Yeti theorem could be applied only to paired boards. It states in it's definition, though, that it applies to all dry boards. So why would it only apply to dry boards? Well, because if the board is wet, villain will want to 3bet his hand against potential draws and such, given that a lot of people like raising with draws.

And since the board is dry, there are no draws possible at least on the flop, there is no reason to push off weaker hands like top pair type hands or worse. He would definitely want us to continue firing the turn and river, if possible, setting the trap. This way also allows us to catch up on the turn and improve to a hand that can stand a second barrel. He also allows us to second barrel on good turn cards, thus inducing very well.

I'm just explaining it to myself, basically, but I like trying to explain ideas that are new to me.

P.S. Last night I owned a poor guy's soul on two paired boards. I don't remember the second one, but the first one was TT8, and I had T8 (one of my favorites I love hands with T in them; except T2). I bet, he raised and I 3bet him, he then 4bet shoves (my image was very wild and he was a tight player, so I assumed he had a good hand. Plus I had folded to some of his 4bets before). We did this two times, once the board was, as I said, TT8, and the second time it was something like JJ4, both times I had a boat and he had pocket AA. Lol! Just... :X
P.P.S. That's what I meant in my previous post about having a wild image and relative position.