This in-depth guide explains how to play exploitative poker. It should be of interest to any player looking to boost their win rate through understanding exploitative strategies.
We will consider the following areas and you may feel free to read this guide in any order of your choosing.
Table of Contents
Chapter 1: What is exploitative poker?
Chapter 2: Exploitative poker vs GTO
Chapter 5: Exploiting under-aggression
Chapter 6: Getting Data for Exploits
Chapter 7: Exploitative play online
What is Exploitative Poker?
Exploitative poker means actively looking for mistakes by our opponent and taking advantage, by using counter strategies.
Exploitative Poker - Looking for weaknesses in opponents’ games and employing counter strategies
Counter strategies revolving around specific weakness are typically referred to as exploits. As a simple example, imagine a Villain (opponent) who is folding significantly too often when facing continuation bets on the flop.
We can generate an exploit in that scenario by continuation betting relentlessly as a bluff.
The majority of exploits can be grouped into four categories which we’ll discuss later in the guide.
- Aggressive exploits: Exploiting high folders.
- Aggressive exploits: Exploiting low folders.
- Defensive exploits: Exploiting over-aggression.
- Defensive exploits: Exploiting under-aggression.
Exploitative Poker vs GTO
What is GTO poker?
GTO stands for “game theory optimal”. It’s an approach to poker that involves looking to make mathematically correct decisions in every situation. A perfect GTO strategy is unexploitable. This fact means that it would be impossible for our opponents to generate any sort of counter strategy against us.
Sounds great, right? There is one huge drawback. Playing GTO means that we are no longer exploiting our opponents’ mistakes either. Changing our strategy to target our opponents’ weaknesses would be a deviation from a perfect GTO strategy.
We should also keep in mind that a perfect GTO strategy remains an ideal rather than a reality. The best computers are still not able to solve a game as complex as No-Limit Hold’em.
So, although we have some ideas regarding what GTO play looks like, the perfect GTO strategy is as of yet, unknown. Even if a ideal GTO strategy were known, it would likely be too complex to be implemented accurately by a human player.
Is GTO poker or exploitative poker better?
Put simply, exploitative poker always makes the most money.
Despite this being mathematically extremely clear, there is still a lot of debate on the topic. Some players turn into stern GTO advocates while others focus purely on an exploitative style.
Such divisions are the result of a fundamental misunderstanding.
The goal of any poker player is always to take lines that generate the highest amount of profit– and the only way to do this is by aggressively exploiting opponent errors.
In fact, GTO poker itself is a type of exploitative style! It happens to be the most profitable counter strategy against another opponent, who is also following a GTO style! In other words, if we knew our opponent was playing a perfect GTO strategy, the most profitable counter strategy is to take a GTO approach ourselves.
However, since almost any opponent deviates significantly from a theoretically perfect strategy, we can increase our profits by following an exploit-driven approach.
Playing GTO for the sake of it against all opponents unnecessarily leaves profits on the table.
GTO Poker → Theoretically perfect. Impossible to exploit but misses out greatly by not exploiting opponents.
Exploitative Poker → Adapting strategy to target opponents’ weaknesses. The most profitable way to play poker.
Does learning GTO have value?
Does this mean that learning GTO poker is pointless and a waste of time? Absolutely not.
Having a solid understanding of GTO play allows us to generate stronger exploits. We’ll have a keener awareness of where our opponent is deviating from a theoretically perfect strategy and how to exploit it.
As we discuss the different types of exploitative line, we’ll begin to see how a GTO understanding of the game carries value.
Exploiting high folders
When our opponent folds too much to aggression, we can exploit this by playing more aggressively. The concept is simple, but how much is too much? A solid understanding of GTO play allows us to spot scenarios where our opponent is folding too much, but GTO play is a skill that takes time to acquire.
Can we generate a shortcut?
Understanding break-even points of bluffs allows us to potentially spot situations where our opponent is overfolding (folding too much).
The following formula can be used to calculate the break-even threshold of a bluff. (We shouldn’t worry if we don’t know what this means just yet.)
Break-Even Threshold of a Bluff -
Break Even %= % of the total pot invested (including our bet)
To put this into context, imagine the following -
We make a flop continuation bet of $50 into a $100 pot as a bluff. What is the break-even threshold of the bluff?
Well, what percentage of the total pot are we investing?
- Total pot including our bet = $150
- Our investment = $50
- % investment = ($50/150) * 100 = 33.33%
Ok, so what does this mean in practical terms?
If our opponent folds more than 33% of the time to the continuation bet, it will be directly profitable.
Now the truth is, from a GTO point of view, our opponent may be “allowed” to fold more than 33% of the time and give us a small amount of profit. A folding frequency of 35% might not be remarkable in this spot. But what if our opponent is folding 50% or even 60% to our $50 bet?
There is an excellent chance that our opponent is folding significantly too often, and we can exploit this by bluffing relentlessly.
Key takeaway → Be alert for scenarios where our opponent is folding significantly above the break-even threshold on a bluff.
Exploiting Low Folders
Our opponent doesn’t fold as much as he should when facing aggression. This scenario is essentially the opposite of the previous exploit and involves two key counter-adjustments.
- We avoid bluffing since our bluffs won’t work often enough to be profitable.
- We look for opportunities to value-bet wider.
To get a perfect idea of whether our opponent is not folding enough in particular scenarios, we should again base this on a solid understanding of GTO theory. But, as discussed, this takes time to acquire, so it makes sense to have a shortcut until we develop a well-rounded comprehension of GTO principles.
We can once again use our knowledge of break-even thresholds to help us generate some basic estimates.
We make a river bet of $75 into a $100 pot, what would be the break-even point if we were bluffing?
Same maths as before, just slightly different numbers.
- Total pot including our bet = $175
- Our investment = $75
- % investment = ($75/175) * 100 = 42.86%
Our opponent should generally be folding somewhere around this frequency if he is playing correctly (usually slightly above). The precise game theory response depends on the exact situation and considers several crucial variables.
However, if our opponent folds significantly below this threshold (for example he only folds 20%), we are likely faced with an exploitative opportunity.
The key exploit is that we never bluff against someone who is folding below the break-even threshold. One mistake that many players make is that they continue to bluff regardless, often citing balance as the reason. However, the best response is to shut down all bluffs since our opponent is not folding often enough for any of those bluffs to be profitable.
We should also look to expand our value range where possible. As a rough guide, we need to be good over 50% of the time when called, for a value bet to be profitable.
Note: We are assuming Villain’s low folding frequency is caused by the fact that he is calling too wide. It could also be caused by the fact that he reaches a specific spot with a range that is significantly stronger than it should be. In such a case, we still shut down all bluffs, but value-betting wider is not going to be a viable exploit.
Key takeaway → Be alert for scenarios where our opponent is folding significantly below the break-even threshold on a bluff.
If our opponent is being aggressive with an overly-wide range, there are potential exploits we can generate.
The simplest example is on the river. Take a look at the following question.
Our opponent makes a $50 bet into a pot of $100 on the river. Half of the time he is bluffing, half of the time he is value-betting. Is our opponent exploitable?
Answering this question requires a basic knowledge of game theory – but the maths itself is not complicated.
Let’s think about the scenario from our perspective. We are getting 3:1 or 25% pot odds on a call.
- Total pot after projected river call = $200
- Our investment = $50
- Pot Odds = ($50/$200) * 100 = 25% pot odds (aka 3:1).
This calculation means that we can profitably call our opponents river bet if we can expect to win more than 25% of the time. Assuming we have a pure “bluff-catcher” that only succeeds when Villain is bluffing, how often can we expect to win here?
Our opponent is bluffing 50% of the time, meaning we would win significantly more often than the 25% required by our pot odds. We can exploit our opponent here, by calling all of our bluff-catchers.
If our opponent wanted to be unexploitable here, he should bluff at a frequency which is identical to the pot odds we are offered (25% of his range should be bluffs in this case).
Spotting over-aggression on the earlier streets is a little tougher. Unfortunately, there is no simple calculation that tells us how often our opponent should be betting in a particular spot. Knowing correct betting frequencies is generally a result of experience and spending time working with tools known as “GTO solvers”.
As an example, GTO solver seems to indicate that the correct frequency for c betting the flop out-of-position in a heads-up single-raised pot is around 35%. Despite this many players c-bet 60% or even 70% when out of position on the flop.
When a player is overly aggressive early on in the hand, there will always be repercussions later on the hand. For example, what happens if we raise against our opponents excessively high c-bet frequency?
If he tries to continue at the correct frequency vs our raise his continuing range will be too broad (allowing us to generate an exploit). If he tries to continue with solely a profitable range (probably his best option), his folding frequency will become overly high, allowing us to generate a bluff-raise exploit.
Key takeaway → Look for scenarios where our opponent is betting too frequently with an overly wide range.
Our opponent is betting or infrequently raising in a specific spot, usually indicating that his range is stronger than it should be when he takes the aggressive action.
For river scenarios, this type of issue is crystal clear.
Think about the following question -
Our opponent makes a $50 bet into a pot of $100 on the river. He is bluffing around 10% the time. Is it possible to generate an exploit?
We ran the maths earlier in the guide, and we know that our opponent needs to bluff around 25% of the time here if he wants to be balanced. A bluffing frequency of just 10% allows us to generate an exploit since it means our opponents betting range is overly strong.
Any bluff-catcher we hold will not win at the required 25% frequency. We can hence exploit our opponent by folding all of our bluff-catchers. We will only continue with hands that win above 25% of the time, which implies we would need to beat a particular portion of our opponent’s value-betting range for calling to be profitable.
Regarding earlier streets, we once again need some experience to spot overly low betting frequencies. There is no simple formula – it’s a product of experience and working with GTO solvers.
As an example, GTO solvers seem to recommend flop raising frequencies (in single-raised heads up pots) of around 15%. If we encounter an opponent who is raising only 5%, we should generally assume that we have some opportunities for an exploitative counter.
The response might not be especially exciting; we simply fold a lot when Villain raises! The exploit occurs because our opponent doesn’t get any action with his strong hands.
Key takeaway → Look for scenarios where our opponent is betting/raising an overly tight range and exploit by folding a lot.
Getting Data for Exploits
It’s all very well knowing how to exploit an overly low flop-raise %, but how do we get access to that information in the first place? We need to be able to access data on our opponents’ tendencies before we can generate exploits.
So where do we get the data?
The answer depends on whether we are playing live or online.
Live players will only be able to pick up data by carefully observing their opponents over a long period. It’s hard to get reliable data because the sample sizes will be small in live games. Even if we remember that our opponent has fired a continuation bet in the last three flop opportunities, that doesn’t guarantee that he’s c-betting too much or even more than average.
As such, live players may gravitate towards basing their exploitative decisions around physical tells. If a specific facial tick indicates that Villain is bluffing, a live player may act on that without having a sample size of data that demonstrates that Villain bluffs too much.
Online players are deprived of physical reads. Gathering data is a far more technical, math-driven and majorly-automated process relative to data in live games. Tools, known as “tracking software”, are used to automatically store up data on opponents for a wide variety of spots.
It is, therefore, possible to not even remember a particular opponent but have access to his cbetting frequency over the last 50,000 hands. We’ll discuss making use of online tools in the “exploitative play online” section of this guide.
Exploitative Play Online
A HUD (heads up display) is a tool which overlays stats onto an online poker window. The HUD is connected to a database that stores information on all opponents. Poker tracking software creates such a database as part of its install process and is also generally the engine that runs the HUD itself.
By knowing our opponents’ frequencies and tendencies in every spot, we should be able to generate on-the-fly exploits based on the techniques that we have discussed. Of course, skill is still required to be able to interpret the data displayed by a HUD.
A HUD is not a magic good luck charm that boosts winrate just by having it running. Many players run a HUD but don’t have the technical knowledge to create relevant exploits based on the visible data.
Coloured tags allow us to label individual opponents based on their tendencies. In an environment that is becoming increasingly automated, coloured tags are part of an online player’s toolkit that is still used manually. It is especially helpful in situations where running a HUD or tracking software is prohibited. (As an example, at the time of writing 888poker allows players to run a HUD on regular tables but not on snap tables, so using the coloured tags at snap poker is helpful.)
Coloured tags allow for quick recognition of opponent types. It’s easier than looking at a wall of numbers. As an example, imagine we spot an overly-tight opponent at our tables. He never bluffs, hardly plays any hands, and always has the nuts when he starts betting or raising. We could set the colour orange to be our “nit” tag and label such a player.
Now, every time we see that player at our table, we instantly notice the orange tag. Exploitatively, we know that we should try and steal a lot preflop and postflop but get out of the way as soon as Villain starts betting or raising.
Population analysis is the technique of working with a database of hand histories to establish default tendencies of the population. This feature allows us to generate exploitative strategies for dealing with an unknown Villain. (It turns out that the majority of opponents have remarkably similar leaks).
In the past, players assumed that it was necessary to follow a GTO approach against unknown opponents and then switch into an exploitative style after picking up information on their tendencies.
However,it is now possible to play exploitative poker out of the gate by assuming our opponents’ mistakes conform to the common mistakes made by the entire player pool. In some cases, we’ll be wrong vs an individual opponent, but on average we’ll be right.
This setup completely eradicates the need to play a GTO style by default. We should only look to play GTO strategies ifhypothetically, our opponent was playing a perfect GTO strategy himself.
The rest of the time, exploitative play makes the money. In fact, we could go so far as to say that exploitative play is the whole point of poker.