Modern Short-Stack Play (Part 4)
This is the fourth and final part of this article series about modern shallow-stack play from the blinds. This time we're focusing on playing blind vs blind situations in the big blind with a stack of 20 big blinds or less.
Part 4: Shallow Blind vs Blind Play from the Big Blind
When you're facing a decision in the big blind in a blind vs blind battle , there are three possible scenarios depending on the small blind's move:
- The big blind could face an all-in shove from the small blind.
- The small blind could raise to (typically) 2-3x the big blind.
- The small blind could open limp, giving the big blind an option to raise.
Let's take a detailed look at how to play back against each. Please keep in mind that this article focuses solely on short-stack situations (20 big blinds or less).
Facing an Open Shove from the Small Blind
Much like in the previous article, I'm not going to waste your time by filling this article with push-fold charts. As a serious MTT player, it's your responsibility to know the approximate calling ranges against blind vs blind shoves at every stack depth up to 20 big blinds. Instead, I want to talk a little bit about being exploitable, variance, and assumptions.
Before we begin, please read the following sentence, and think about the answer for a while without clicking to see the chart below.
In the previous parts, I described even a +0.20 edge as significant. It usually is, but I believe that statement to stand the test much better in spots where you're the one doing the shoving. When it's possible to make an unexploitable shove, you can't really ever screw up the hand completely by doing just that. But when you're calling a shove instead, there's much more room for error and many things to consider:
- What if your opponent is shoving tighter than what the programs suggest?
- Is your opponent's strategy to only open shove or fold, or is he splitting his ranges?
- If the table is otherwise very soft and your future expectation to chip up is great, calling shoves with marginal hands becomes sub-optimal.
- Calling off for your tournament life with a speculative hand is never fun.
- The more you subject yourself to coin-flips, the more variance you invite into your life.
The most important point is the first. There are few players out there who shove exactly according to push/fold charts, and many who shove significantly tighter. The chart above is based on the assumption that our opponent shoves the exact Nash range of 60.4% of hands (including hands like 63s, T4s and 76o). But what if we give him this shoving range instead?
...Now you can only call with:
That's almost half of the calling hands sniped off! A few specific observations:
- A20 made 1.48 big blinds against the so-called "correct" shoving range (which can sometimes be sub-optimal for the pusher.) But now, despite him still shoving well over a third of all hands, the same hand is almost breakeven.
- Against the "correct" shoving range, calling J-9o is a breakeven play. Against this slightly tighter version it's losing 2.27 big blinds.
- Q-To made us 1.5 big blinds by calling in the first example, but in the second it loses us 1 – and this is assuming that the small blind is still shoving hands like K-5o and 9-8s!
This is why I think it shouldn't be first priority to memorize every Nash calling range against shoves at every stack depth. Should you have a pretty good idea about them? Sure. But it's much more important to have a good grasp of calling ranges against different shoving ranges on the whole.
What do you call a 15-big blind shove with if you think that the small blind only shoves 20%? What if he shoves every hand? You need to put the hours in and train yourself by using programs like Holdemresources calculator to find out. You're not going to learn this stuff by staring at some push/fold chart, because you're not playing against computers. You're playing against humans, who decide their shoving ranges based on all kinds of factors, and many of them have nothing to do with science (such as emotion).
When you're facing a shove, your job isn't to have a stock answer for what range you need to call to not get exploited. Your job is to:
- Estimate his shoving range
- Know how to respond against that estimated range.
Lastly, let's discuss being exploited. In the above example, how much do you need to call at the very least so that your opponent can't make a profit by going all-in with any two cards?
The answer is roughly 29.1% of hands. That's right – if you're folding K-6s, Q-9s or K-8o to a 15-big blind small blind shove, the small blind can profitably shove 7-2o in your face without hesitation and laugh on his way to the bank.
Now, I assume some of you are thinking: “Holy shit! I really need to start calling wider and stop my opponents from owning me!” But that's the exact opposite of what I want you to take away from this. Sometimes, being somewhat exploitable is totally fine.
Playing Exploit-ably Vs Actually Getting Exploited
Is it good to have a rough idea of how much you “need” to call in each spot? Sure. But here's the thing. I've played MTTs for a living since 2010, making $600,000 in the process. During this time, the thought “is my opponent exploiting me here if I fold?” when facing a blind vs blind shove has entered my mind approximately twice. In most cases, I'd say don't worry about it. Sometimes it's actually okay to let your opponents exploit you a little bit.
Your opponent is not going to know what you're folding, and thus he's not going to know that you're exploitable. He's only going to sit there for a few orbits, since this is MTTs we are talking about, and during that time he might not even get another chance to open blind vs blind against you. Chances are that you two never play that specific stack depth blind vs blind again. And even if you did, what if you just folded 7-2o? He can't know.
Being exploitable and actually getting exploited are two different things, and it's very hard for someone to knowingly exploit you due to the nature of MTTs. Even if you're both regs who play the same tournaments most every day, you're just not likely to get into the exact same situation against that opponent.
Even after months and months of playing against the person, he'd still have to have a huge sample size to know for sure that you're folding too much. AND then he'd have to be capable of knowing how to exploit. See where this is going? It's just not happening very often at all.
Even if he was able to exploit you, so what? Of that $600,000 I've made, I'd guesstimate that $550,000 has come from fish and the rest from regulars. MTTs are the softest form of poker left (and always will be). Your job is mostly to take money from the poor players, not to push some 0.02BB edge against a good regular.
As stated above, you subject yourself to a lot more variance when you make thin calls to avoid being exploitable. Unless you have an unlimited bankroll, it's just not worth it.
If your estimate of his shoving range is wrong, the speculative calls become massively losing ones (as shown above), and I wouldn't take that risk just to not get exploited. Obviously you still need to defend quite a bit, and I don't advocate making absurd folds for no reason. But I think that in most scenarios, you can safely dismiss something like a 0.1BB edge if you're playing in a soft tournament.
Facing a Non-All-In Open Raise From the Small Blind
Playing against 2-3 big blind raises with shallow stacks is a much more interesting subject than dealing with open shoves. In the previous part of this series, we established how much the big blind needs to defend against a raise from the small blind to prevent the small blind from being able to auto-profit. As a reminder, here are those percentages against each different open-sizing (9-handed, 10% antes):
- 2x BB minraise: 61.6%
- 2.25x BB raise: 58.9%
- 2.5x BB raise: 54.6%
- 3x BB raise: 49%
As you can see, you need to defend quite a lot. More importantly, we're always getting the odds to call with most of our hands against small sizings. The math is quite simple:
(amount you need to call) / (pot before calling + amount you need to call)
For example, if the table is 9-handed at blinds of 500/1,000/100 and your opponent minraises from the small blind, there's now 3,900 in the pot (his raise to 2,000 total, your 1,000 big blind, 900 from antes) and you need to call another 1,000. Written out as a formula:
1,000/(3,900+1000) = 0.204
You need just 20.4% raw equity for a call to be theoretically correct. As you can see, this basically means any two cards unless he's turning pocket aces face up.
We have now established the following two really important things:
- Against a small blind minraise, we need only 20.4% equity to call
- To avoid him being able to raise with any two, we need to call at least 61.6% of hands.
QUESTION: Now, imagine yourself in this spot. Your opponent minraises in the small blind, and you're holding T-3s in the big blind. What's your play? Click "Show" below when you're ready for the answer.
The MTT-playing popularity dramatically under-defends their big blind against shallow small blind raises. They think that they can't call because they're only 12 big blinds deep, and preserving your stack and yadda yadda. But as we established in the previous parts of this series, the shallower you are, the more equity you actually get to realize.
If you're 40 big blinds deep, how often will you get to showdown with 8-7o when the board runs 7-Q-2-K-A? Not very often. Your opponent typically either has you beat or blasts you off your hand. But what if you're only 12 big blinds deep?
When you defend against a minraise 12 blinds deep, the pot is ~5 big blinds on the flop and you have 10 big blinds behind. When your opponent c-bets, you have multiple winning options.
- You can (probably) shove your middle pair and make a good profit against most opening ranges (for what it's worth, this would be a bit of a silly play)
- You can call, leaving yourself less than a pot-sized bet behind. It's pretty hard for your opponent to blow you off your hand with no equity when his only play is to move all-in.
Flopped 8 out+ draws are also hands that benefit from being shorter. Imagine defending Ts3s 40 big blinds effective. The flop comes Qs8s2h. Your opponent bets and you call. If the turn is a brick, you're going to be in a tough spot no matter how the hand plays out (unless you bink a free river card.) Now imagine the same situation 12 big blinds deep. With around a pot sized bet behind, you can happily shove all-in against your opponent's flop c-bet. It's not the most sophisticated play in the world, but it'll make a profit.
The shallower the stack-to-pot ratio on the flop becomes, the easier it is for you to get to showdown, and thus realize your equity.
Now, let's go back to the initial decision we have to make when facing a shallow small blind open. Besides folding, we can both shove and flat. If you've paid attention during this article series, you probably already know what I'm going to say next. It's key to know approximately what you can shove against different opening ranges at each stack depth. You have to put the hours in using software created for solving these exact situations to have a good grasp of how everything works.
Anyway, here are a few simulations against different opening ranges in a spot where we're facing a 2.5x open 15 big blinds deep (10% antes, 9-handed). To make things a bit simpler, I assumed that the big blind is calling correctly against our shoves. This also assumes that our opponent never open shoves, which is a pretty ridiculous assumption, but it's just impossible to account for all the variables within the frame of one article. Anyway:
Here's a range that I think is a bit more realistic for a half-decent, tightish regular. Let's assume he 2.5x opens this range:
I know that this looks like a huge mess at first glance, but let me elaborate.
- We assume that he's open-shoving small pairs and weak aces, since those make so many big blinds as shoves, as we learned in this last article.
- Offsuit broadways like K-Qo and K-Jo can go either way, so I assumed he shoves half of them and opens half of them. I think hands like K-Qs make enough as raise/calls not to be shoved.
- I added K2o-K5o as bluffs (assuming he shoves K7o-K9o), and same with weak suited kings and queens.
- We have some weak hands that are really sensible opens such as T-8o and 7-5s.
This is by no means perfect, and constructing a range in a spot like this where you have to decide which hands your opponent opens and which hands he raises small is actually pretty damn tough. But it's a fine example for our purposes.
QUESTION: So, look at the above opening range, and especially take into account how it doesn't include most A-x hands and small pocket pairs, and ask yourself this: What percentage of hands can I profitably shove here? Click "Show" below when you're ready for the answer.
But should we shove all of these hands? Of course not. Instead, you should figure out the mandatory shoves- the kinds of hands that aren't very playable post-flop and that have a lot of margin for error- and put those into the shoving range. On top of that, you can call with most hands that boast some post-flop playability.
In this example, just with 10 seconds of thinking, I'd do something like this:
- Shove any pair (occasionally trap with Aces)
- Shove any Axo
- Shove virtually any Axs (occasionally flat something for deception if I'm feeling fancy)
- Flat some broadway stuff like JTo despite them making a lot as shoves
- Flat virtually every single suited hand apart from the absolute worst ones
- Shove a bunch of Kx and flat the rest of them
- Flat Q6o+, J7o+, T7o+, 97o+, 86o+, 75o+, 65o.
Thus, intuitively, my playback range (both shoves and flats) would be something like this:
Why am I not defending any two, despite only needing 25.4% raw equity against the open? If you've read this article series carefully, you may already know the answer.
It's because while we get to realize a lot of equity in these situations, we still don't get to realize all of our equity. If I flatted 7-2o, I still wouldn't get to the river with it every single time. It's impossible to estimate exactly how much equity you'll get to realize, but I'm confident about defending the above in this scenario. I think some less-experienced players would be wise to drop quite a few hands, and instead only play back at this range (shove the same hands as above, but fold some of the flatting hands):
That's 50.5% of hands. Less than theoretically optimal, but I still think it's completely fine. The small blind being able to exploit you a little doesn't matter all that much – you're still close enough to the so-called minimum response. If he's going to own you a little, good on him. There will be a bunch of fish in the field who you can own a lot harder to make up for it.
Also, it's noteworthy that if you don't really know what you're doing in post-flop situations, it's actually smart to tighten up even more. Playing 60% ranges without the initiative with two pot-sized bets behind really isn't the easiest thing in the world. The first playback range is what I feel comfortable playing now after my years of experience, the second one is what I'd have felt comfortable with maybe 3 years ago.
I wouldn't even be shocked if I was still overestimating my abilities, and I should actually still defend a little bit tighter. There's no shame in being slightly tighter than you theoretically should. If you repeatedly screw things up post-flop, then defending too widely might become more costly than letting your opponent run over you pre-flop. Punting away stacks in bad spots will cost more than the one big blind you could have surrendered.
(Note: Ready to take your poker game to the next level and win more money? Check out the Upswing Lab by clicking below! The Lab is the most comprehensive and efficient way to move up in stakes and grow your bankroll. Don’t take my word for it, see what our members are saying!)
Big Blind Vs Small Blind Open Limp
As mentioned earlier, limping in the small blind has become commonplace over the last couple of years, but it's also pretty hard to do while staying somewhat balanced.
With stacks of 20 big blinds or less – and yes, this is a major generalization – I'd say that the field is unbalanced towards weakness. You can get away with shoving and isolating over their limps a little more than you theoretically should.
I encourage you to toy around with programs like HoldemResources calculator that can simulate these scenarios. But quickly; shoving over a limp in this spot is almost the same as shoving from the small blind when folded to, except that there's now an additional 0.5 big blinds in the pot. Sure, your opponent now doesn't have his worst hands, but they probably don't have a bunch of their best hands either.
It's noteworthy that isolating limps too often creates additional variance, and it's certainly wise to stop and consider what a great spot you're already in when you're facing a limp – you're getting a free flop with whatever you're holding.
Especially if you're a strong post-flop player, I'd embrace these spots and happily go to the streets with a lot of hands, even if I figured that shoving them would make a tiny profit. And I'd be extra careful with using 2.5-3x raises over limps, because that re-opens the betting and allows your opponent to shove over your raise. Against many sharp opponents that's lighting money on fire if you don't choose your hands properly.
In these situations, a pretty good response would be to raise your best and worst hands (make sure to use a 3x type sizing that he can't call against very often, since min-raising with crap hands would lead into post-flop disasters!), and check the middle part of your range. This way, you never really waste a hand – when you use something like 5-3o for a total bluff, it doesn't particularly hurt when your opponent limp-shoves and you have to fold. But if you do that with 8-7s, that's a lot of equity hitting the muck.
Here's an example of a reasonable iso-range vs a SB limp:
Internet Poker Tournament, 17 BB Effective Stacks
folds to the sb, Small Blind limps, Hero...
Let's say the open limper is some generic regular. This is what I'd do:
- Go all-in with the mandatory shoves. I'd start by shoving hands that have trouble post-flop such as small pairs and A-x hands.
- Shove some other hands that make a small profit. I'd also probably select a few hands of each category, such as 8-6s and K-9o, and jam those as well to ensure that I'm not too easily readable.
- Isolate with a relatively small range. Some complete junk that I don't really mind raise-folding, like 72o or 83o, and some really strong hands to raise-call with.
- Happily check with over half of my hands and see a free flop.
So, in a nutshell: I think that in current games, the field as a whole limps more weak hands and less strong hands than they should to be balanced. You can exploit this by raising a bit more than you'd think.
Again, I strongly recommend running simulations, tweaking ranges and asking yourself questions. The more questions you know the answer to, the more prepared you will be for the spots as they come up. For example: If he limps 50% but only calls the top 15%, how many hands can I shove with 15 big blinds effective? (The answer is any two cards.)
Lastly, when facing a limp you should always ask yourself "how is he expecting me to react?"
In many cases it's a bit blurry, but think about a spot like a final table where the small blind is the chip leader and you have 18 big blinds. He's virtually never expecting you to isolate wide because of the setting, and thus isn't planning to limp-trap to shove over your small raise, because that raise is not expected in the first place. Thus, if your opponent shouldn't expect you to isolate light, what's your best response? Damn right. Isolate the shit out of him.
Blind vs Blind Post-flop Play from the Big Blind
When you play a post-flop hand in the big blind with shallow stacks, sometimes you'll be the aggressor, but most of the time you will be the caller/player who checked. Let's take a closer look at each.
Big Blind Aggressor vs Small Blind
When you've isolated the limper and your opponent has called, the hand now plays much like the pots where you're the small blind raiser and your opponent has flatted. You have all the strong hands and your opponent doesn't, and thus a few types of board textures favor you by default:
- Ace-high boards like AJ2
- Bricky boards like 552
- Broadway-heavy boards like KJ6 (to a lesser extent compared to the ones above)
Of course, your opponent sometimes hits them too, but you should be winning the vast majority of these. When your opponent has limp-called, his range is also somewhat face up – the majority of it consists of suited middle-card hands like J-9s or 9-7s. It's pretty hard for the limp-caller to hit a board like K-2-2, because there are so few hands containing a deuce or even a king that people play by limp-calling. But on boards like T-9-6, he has a boatload of hands that can continue, and it's wise to let him have his share when you've missed yourself.
Big Blind Takes Passive Pre-Flop Action vs Small Blind
The most common scenario. This is pure repetition of the last article, but I'm going to say it again because it's just so true: People always spazz out in blind vs blind spots. Even more so in the big blind, I see all kinds of ridiculous stuff every day.
People flat the big blind, and then raise some ace-brick-brick flop, thinking something along the lines of well it's blind versus blind, I doubt my opponent has anything, and it's hard for him to continue without a hand on an ace-high board!
Newsflash: it's not that hard. Any competent player is going to snap you off when you're repping nothing. It doesn't mean that it would be a terrible play to shove something like a flush draw on A-2-6 suited, even though you're not repping much else (since you don't have any sets or two pair combinations in your range). If you have two pot-sized bets behind, the beautiful thing is that it just doesn't matter all that much, because with a flush draw you simply have decent equity that you want to realize (I'm still calling you off with king high, though, and we're off to the races).
The general point I'm trying to make is this. As we established before, we often need very little raw equity to call a small blind raise – 20.4% against a minraise, 25.4% against a 2.5x, 28.9% against a 3x (10% antes, 9-handed table). This is what warrants us to call with a lot of hands, but you also have to understand that we really aren't meant to win a whole lot of these pots.
You didn't call with J-3s to pull off some stunt on K-8-7 rainbow. You called to simply realize your equity, and now that you got a terrible flop for your hand, you fold and move on. When you're this shallow, there simply isn't a lot of room for bluffing. It's just a game of realizing equity.
Of course, you can sometimes bluff, but it should be more along the lines of flatting a continuation bet with 9-7 on a board 6-5-2, and shoving on the 4 turn against a check. If you float 9-7 on A-K-2 you're just lighting money on fire.
Blind vs Blind HUD
Let's take one last look at the same blind vs blind popup HUD from the last article:
As you can see, this also works quite well in the big blind. This player seems to be playing a mixed strategy – he's raising 35% of his hands and limping 19% of hands. This is quite tight, and when we take into account his stats in the big blind (not directly relevant to spots where he's the small blind, but just seeing how tight he is in general), we can probably establish that he's a pretty straightforward, tight player. With these stats, my strategy against said player would be the following:
- When he raises, I'd flat a few more hands that I'd shove against looser players with, say, 20 big blinds. I'm confident that I can play many hands well in position against him, because he seems quite straightforward, and since his opening range is tighter than most players, I don't want to shove speculative hands and risk running into a monster too often.
- When he limps, I'd attack it with reckless abandonment. While I'm sure he's not limping exactly his 36th to 55th percentile hands, I'm very skeptical of him having enough strong hands in his limping range to withstand pressure. Against a player who's 3-bet 0 out of 6 times from the big blind, I'd be happy to isolate to 3x instead of shoving with a lot of junk since I'm not expecting him to limp-shove all that often.
And that's it for this article series. I hope you guys enjoyed it, and I truly hope that you won't put these tricks to good use, because I hate it when people play back at me correctly from the big blind. In case you missed any of them, these are the 4 articles:
If possible, please remove everything you've learned from your memory like the main character in the wonderful movie Eternal Sunshine Of The Spotless Mind, and go back to letting me exploit you. Thanks!
If you have any questions, feel free to hit me up in the comments box below or on Twitter @chuckbasspoker.
(Note: Serious about improving your cash game, tournament and/or live poker skills? Check out the Lab, a comprehensive poker training course developed by Doug Polk & Ryan Fee. Our members are crushing everything and you can too!)