Summary of Poker's 1% by Ed Miller

Pokers 1% is all about frequency-based play. What is frequency based play? I think it's best understood in contrast to hand-strength-based play.

Consider an example. The board is A44r. Your opponent cbets. What hands do you continue with?

A nitty opponent might think:

Meh, 22 and 53 really just aren't strong enough for me to continue. But I guess I'd be cool peeling a street off with TT.

A more straightforward opponent might think:

This is a super dry board. I normally wouldn't like small pocket pairs and naked gutshots without anything else going for them, but in this case I guess I'd call with them. I wouldn't have the balls to call with king high though.

Now, watch how frequency-first approach would work:

I need to defend with 70% of my hands here. I've studied this spot before and I know that I only have about 45% pair or better hands. So I definitely need to continue with my backdoor flush draws. Probably a few king highs as well.

Do you see the difference?

• The hand-strength-first approach looks at the board, and asks what hands are strong enough to play on that board.
• The frequency-first approach first asks what frequency they want to continue at, and then asks what hands they want to select to meet that frequency.

That's the big idea in this book. To use the frequency-first approach.

Why frequency-first is cool

Frequency-first poker is really where it's at. For a bunch of reasons.

Reason #1: Without it, better opponents will eventually crush you.

Say you move to Vegas with a dream to make it big as a poker pro. Here's how life will look if you take a hand-strength-first approach, like so many others do.

You'll start off grinding $1-2. You'll study, improve, increase your winrate from $5/hr to $10/hr to maybe $15-20/hr. Then maybe you move up to $2-5, and go through a similar process. If you're really sharp, make you can take this all the way to $5-10. But eventually, your wits will fail you, and you'll hit a wall.

You'll be upset. Frustrated.

"Why am I not winning? What's going on?"

What's going on is that the way you're thinking about the game has failed you. It's not that your opponents are smarter than you, or harder working. It's that hand-strength-first can only take you so far. To compete with the 1%, you need frequency-first.

Reason #2: It's inexploitable, so worst-case scenario, you'll be breaking even.

When you take this frequency-first approach, something a little big magical happens: you become inexploitable. If you somehow got your frequencies perfect, no one would ever be able to beat you.

You'd even be able to tell them exactly what you're doing! "In this spot I'll be calling with all of my pairs, all of my backdoor flush draws, and KQ-KT. Whatcha gon' do 'bout it?"

They wouldn't be able to do anything about it.

• If they tried to exploit you, they'd fail, and you'd profit off of them.
• If they tried to get their frequencies right, they'd probably make mistakes, and you'd profit off of them.
• If they happened to be elite players who also got their frequencies perfect, you'd just break even.

To me, this still feels a little bit magical. Like, how exactly does this work out to be true? I have some understanding, but I wouldn't say that I fully understand it. The book provides some intuition, but that's really not where it's focus is. I think it's focused a lot more on the "what" rather than the "why".

The cool thing is that in order to be good at poker, you don't really need to understand the why. You just have to know that you have to meet certain frequencies, not why you have to meet them.

(Personally I get an itching discomfort when I don't understand the why, so I'm going to have to get it figured out at some point, but that's neither here nor there.)

Reason #3: It'll help you exploit your opponents.

The frequency-first approach will often tell you to defend vs your opponents bet/raise something like 70% of the time. That'll be inexploitable, but it won't always be the most profitable. That's a very important difference.

For example, suppose your opponent fires a third barrel on the river. You're sitting there with top pair no kicker. A bluff catcher. Maybe it's part of your 70%, you're technically "supposed" to call. But does that mean you should call?

Well, if you know your opponent isn't bluffing enough, the better choice would be to fold. Miller likes to say that once they break the rules — by not doing what frequency-based play says to do — then you can (and should!) break the rules too.

Again, this is a very important point. As poker players, we want to maximize our profits, so we want to find these mistakes that our opponents are making, and we want to make them pay.

But! In order to do that, we have to first know what the rules are.

Maybe we think they're betting 60% of their range when they should be betting 70%. We have to know that the 70% is what they should be doing before we actually know they're making a mistake.

Well, if they're betting like 10-20% of the time, then it's just obvious that they're making a mistake. That they're a nit and have the goods when they bet. But errors that large aren't too common. You want to be able to exploit the small and medium-sized errors too, not just the big and obvious ones. To do that, you need to know what the correct numbers are.

This is why studying frequency-first play will also help you at games like 1-2 and 2-5! I think this point gets often gets lost or misunderstood when people read books like this and talk about frequencies. When people think frequencies, they usually are thinking about inexploitable, balanced poker. They’re not usually thinking about exploiting other people. But, again, knowing the frequencies will help you to find the mistakes that your opponents are making and use it against them.

Disclaimer: Trying to exploit your opponents can backfire! Make sure that you actually know that they're making a mistake.

Reason #4: It's a great thing to fall back on when you're a little lost.

Poker players love to pretend that they know everything, but the reality is that we're all pretty clueless. There's tons of spots where we don't know what the hell we're doing.

For example, maybe you make $10/hr playing 1-2. Not bad, but there's still tons of room for improvement. Maybe you take advantage of your opponents by playing a strong range preflop and value betting hard with the goods, but maybe you never know what to do on monochrome boards, or when the flush completes, or when facing a minraise. Chances are that there are some spots where you feel lost. When this happens, it's nice to be able to fall back to the frequency-perfect play, and know that you're at least not getting exploited.

Another situation where this becomes useful is when you move up to a higher stakes game and run into some trouble. It's nice to be able to play frequency-perfect and not get exploited while you get situated, and then start to deviate from that as you get reads on what your opponents are doing wrong.

Let me give you an example of this. I recently moved up to NL50 online. I haven't been doing as well as I'd like. One thing that worries me is that whenever I check, I feel like I get put in tough spots and have to fold too often. What I'm going to do moving forward is to make sure I get my frequencies right. That way I'll be inexploitable for the time being. Then, once I have a better feel for what my opponents are doing, I can start exploiting them.

What the frequency-first approach really is

Ok, that's been a little overview of why the frequency-first approach is cool. Now let's talk about how exactly to use this frequency-first approach. We saw the A44r example, but there's more to it. The book talks all about it, but essentially, there are two rules that you'll need to follow:

1. When they bet, you'll want to be continuing about 70% of the time as a default to a full pot bet. And maybe something like 75% to a half pot bet.
2. When you bet one street, you'll want to continue betting again on the next street about 70% of the time as a default.

Those two rules are crucial! Like, in some sense, that's all there is to it. The rest is just caveats and exceptions; some bigger and more important than others. But if you ignored those exceptions and just followed the two rules above, you'd be pretty damn good at poker!

To visualize the 70% thing, check out the pyramid below. The idea is that you narrow your range on each street. Eg. as you get further and further up the pyramid. The lower the percent, the steeper the slope.

This pyramid analogy is pretty damn awesome. I think it's really helped me improve my game. Don't underestimate the value of a good mental model.

The pyramids really make it clear when people have exploitable frequency errors.

Along similar lines, it helps you to think about your own strategy and to erase the vulnerabilities in your own game. Keep those slopes smooth!

Before we get to the exceptions, here's another pretty essential thing: when you're betting, you basically want to have a 1-to-2 value-to-bluff ratio on the flop, 1-to-1 on the turn, and 2-to-1 on the river. Since this is a summary and is already really long, just roll with it for now, and check out the book for more information.

Ok! Now on to the exceptions!

• In multiway pots, you basically share responsibility with the other players. Suppose that it's three-way and someone bets. You don't have to defend 70% of the time. You and the other guy have to defend 70% combined. So maybe you each defend 35% or something.
• Say the cutoff opens 2bb and you flat from the BB. You were getting 3.5-to-1 pot odds, so you flat pretty widely. Then the flop comes and they cbet full pot. You don't need to defend the typical 70%. Since you didn't invest much to get here and you have a weak range, you can fold more often.
• Similarly, if you have been the bettor and decide to check at some point, you don't have to continue the normal 70% amount if they bet. The idea is sort of that they had to pay on previous streets to get this opportunity to bluff you when you check.
• The 70% defense thing is really just an average over all scenarios. Maybe you go 80% when you're in position and 60% when you're out of position. 80% when the turn improves your range, and 60% when it improves their range. Stuff like that. But be careful! Most of the time these factors will only tweak things a little bit. It's only when it's like the worst card in the deck or something that you can defend 20% instead of 70%.

The book talks about a bunch of other things that I don't think are worth diving too deeply into in this summary. I'll try to be brief:

• The hands you choose really don't matter too much, as long as you're somewhat intelligent about it. Like, it's not too big a deal whether you defend with 22 or with a gutshot. But still, the book provides some advice on how to choose hands for your different buckets (check-valuebet-bluff, or fold-call-raise).
• The call vs. raise decision is complicated and beyond the scope of the book, but Miller still provides a few guidelines.
• When you play too many hands preflop... you just totally screw yourself. You have a bunch of junk, and you have to do something with that junk. You either have to fold it out on some street, and thus overfold. Or you call down with it and end up paying people off with something stupid like second pair. There's no way out of the dilemma. Don't play junk preflop. I've always been pretty tight preflop, but this was a pretty big insight for me. I felt something click.

Throughout the book, Miller gives a bunch of examples of how you actually use this frequency-first approach. First of all, the work is done off the table. That's where you put your hours of study in. When you're actually at the table, you maybe spend a few seconds thinking, but you’re largely just using the intuition that you developed off the table.

Here's an example of how studying works:

• Your opponent opens from MP, you flat from the CO and go heads up.
• Step 1: Figure out what range you're flatting with and entering the flop with.
• Flop comes A44r. Villain cbets.
• Step 2: Ask yourself how often you want to be continuing. Maybe it's 70% of the time here.
• Step 3: Ask yourself what hands you want to continue with. Maybe it's all pairs and all backdoor flush draws.
• Step 4: Ask yourself how you want to distribute your continuing between calls and raises. Suppose we don't want any raises here. We're flatting our whole continuing range.
• Turn comes an offsuit two. Villain fires a second barrel.
• At this point, we repeat steps 2, 3 and 4 for the turn. Maybe we want to defend 70% of the time again, which means continuing with all of our pairs, and flatting our whole range.
• Same thing for the river. Say it's a ten. Maybe they bet and we want to continue 70% of the time, which ends up meaning that we defend all of our Ax, and fold all of our pocket pairs.

Again, Miller goes through these sorts of examples numerous times throughout the book, in much more detail, and it's really useful to watch him go through the process. There's also a bunch of cool little nuggets of insight where he talks about common errors people make and how to avoid + exploit them.

But yeah, that's the gist of this book! He emphasizes that the book is similar to a workout DVD: it’ll inspire you and show you what to do, but you have to actually put in the work. You have to go through the process of developing your ranges for all the different spots you come across.

He also emphasizes many times that this book is really just the tip of the iceberg. The truly elite players spend a bunch of time pinpointing the precise numbers. Like, maybe the correct number in one spot 66% instead of 70%, and 73% in another spot. Miller is very clear that the numbers he uses should be roughly correct, but definitely not perfect. And that if you progress enough to need better numbers, you can check out stuff like Janda's Applications of NLH, Chen and Ankennman's The Mathematics of Poker, and the software like PIO Solver.

But despite the fact that the numbers in this book aren't perfect, one thing is very clear: if you get your frequencies close to them, you'll have a ton of success. You'll become a significantly better poker player. Most people — even at higher stakes — make many mistakes with their frequencies, so the bar isn't being set very high. If you get your frequencies roughly correct, the poker gods will certainly reward you.