06 Oct 2024 Advanced This article is for experienced players GTO hand charts ranges MDA Are you using GTO preflop ranges? If you’re a professional player or anything close to it, chances are that GTO ranges are the base of your strategy. That’s great. But what if there was something better? Preflop strategies that are even more profitable than GTO strategies? How is that possible you may ask? Aren’t GTO ranges literally the most optimal ranges you can play? Well, yes and no. Yes, because they are indeed optimal ranges – Game Theory Optimal. And GTO strategies in general cannot be exploited. Which is a very powerful base to have. So using GTO ranges will never be wrong. BUT, Game Theory isn’t real life. In real life, there is a way to use strategies that are even more profitable than GTO. The reason is that most other players don’t actually play GTO in practice. Definitely not postflop and not even pre-flop. How can you know that? From working with MDA. MDA is short for Mass Data Analysis and describes the work with large amounts of hands gathered in a database. By analyzing the data that large amounts of people have played over long periods of time, you’re able to look at population trends and tendencies. And those clearly show that people on average are not playing GTO ranges preflop. Which opens up a great opportunity for you. As said, GTO strategies are not exploitable, so if most people played GTO, you wouldn’t be able to exploit them. But when they don’t, you can adjust your strategy to maximally profit from their mistakes. How? Well, that’s what we’re going to find out in this article. Step 1. Explore Players Pool Ranges In Hand2Note What do you need to find out how population plays and how you can exploit them? Well, first, you’ll need a large number of hands. Then you’ll need a poker tracking software. Hand2Note is by far the best software for working with MDA. Which is why I used it to create the exploitative preflop ranges for my course Preflop Xploits. In Hand2Note, you want to filter your stats for specific positions and spots so that you can see the ranges that your player pool actually plays. One way to do that is to go to the “Reports” tab. There, you want to click on the field on the top left, which is the player field. Here, you can choose to analyze a specific player or multiple players. When looking at a player pool, we want to analyze multiple players. You can set specific filters to fit the player pool you want to analyze. For example, it makes sense to distinguish between regulars and fish, which you can do on the right. For this one, we want to look at the regulars. To make sure the players have a decent sample size, we’ll also require them to have played at least 100 hands in this database. And to make sure we really target regulars, we’ll filter for players whose VPIP and PFR differ no more than 10%. Also, you might want to single out a specific poker site. Because tendencies can differ based on the dynamics and especially on the rake structure of a site. Let’s look at Pokerstars in this example. We also want to select the correct game type, but other than that, we don’t need any more filters. So we’ll hit apply. Now we see stats for the player group we selected. What we want to look at is the preflop ranges. So we’ll click on “Preflop Range”. And now, we’ll need to filter the spots we want to analyze one after another by clicking on “Filter”. For example, we want to examine how the population plays their BB against a BTN open range. We go to “Positions”, and select “BB” for our position (because that’s the position we want to see the player pool’s stats for) and “BTN” for villain’s position. Under “General”, you can also limit the hand selection to specific stakes if you want (“Big Blind”) and initial stack sizes. The latter makes a lot of sense, because that way you prevent showing situations where much shorter or much deeper stacks are considered. Here, we want to look at stacks around 100bb, so we limit the selection to 80bb – 120bb. Lastly, we want to go back to “Quick Filters” to select in which spot we want to have the stats shown. Being the BB against the BTN, we want to start by looking at the population’s range when they’re facing an open raise and are 3-Betting. So we click on “3-Bet”. Later, when analyzing the calling range against a BTN open, we would select “Call Open Raise” instead. Lastly, we want to make sure we select “Villain is Reg” as we want to look at play vs other regulars to make sure the preflop sizings are not too out of the ordinary. Then we hit apply. Make sure you’re still in the “Preflop Range” tab and you’ll be able to see the according preflop range for the spot you selected! Well, to be exact, it is not really their preflop range, but rather the selection of hands that went to showdown in the selected hand sample. However, what we need in order to create exploitative counter strategies, are ranges that we can input into a solver, including frequencies for every hand. We cannot simply copy and paste the ranges we see here, as they are simply numbers of times a hand went to showdown, and not a real range with frequencies. So we’ll have to transform those numbers we see in H2N to a preflop range that we can input into a solver. This is a process that consists of 2 components. The first is the frequency. By clicking on “Results”, we’ll see a clear value for the frequency that BB 3-bets against a BTN open. Here, summed up over multiple stakes, that value is at 12%. Which means the average player in the BB will 3-bet 12% of their hands against a BTN open raise. Provided the sample you’re working with is big enough, this value is quantitatively reliable and accurate. The other component is the range composition. Which is based on the distribution of hands that perform the action we’re looking at. And that will not be a quantitative component. This means it will not be 100% exact, but rather a qualitative approach to approximate the range composition that the population plays in this spot. Step 2. Transfer Ranges To HRC And Calculate The Optimal Counter-Strategy So what we’ll do is take the frequency value, 12%, and look at the hands that were shown down. Then we’ll distribute those hands as well as we can, limited by the frequency, to build a range in the solver, which in this case will be the preflop solver HRC (Holdem Resources Calculator). To build the range, we’ll use the hands that were shown down to get an idea of their relative frequencies. For example, if we see AQo being shown down much more often than AJo, the former will get a higher frequency in the range we build. However, in that process, we also need to keep a few things in mind: Offsuit hands have 3x as many combos as suited hands. Weaker hands usually go to showdown less often than strong hands. Variance also plays a role, so small differences between hands might not mean that much. All of those mean that our common sense and real-life experience have to be included in the transformation process. Also, when in doubt, I like to include the GTO approach and mix in what would be GTO. After having done the same for the calling range, what we end up with is a range that is accurate in terms of its frequencies and an approximation in terms of its composition. However, for this spot, we need to do one more thing. We have now researched and node-locked the BB’s reaction against our open raise on the BTN. But, to fully recreate the complete branch of the tree, we also have to research and node lock the BB’s reaction against our 4-betting range from the BTN. Because that massively impacts how our exploitative preflop approach looks like against the BB’s 3-bet. If we didn’t node lock those ranges in HRC, the solver would simply calculate the GTO ranges for those nodes. Which is not what we want. We want to know what the real player pool does. So before calculating the complete branch in the solver, we have to repeat the same research process and convert ranges from H2N to HRC for the node of BB facing a 4-bet by the BTN. How often and with what hands does population call and 5-bet in this spot? After we’ve completed that, we can finally run the solver and let it calculate the best counter strategy to maximally exploit the player pool when we’re playing BTN vs BB. Step 3. Import Strategy To FBR To Manage And Train It Once we’ve received these ranges, we can use them even better by importing them to FreeBetRange. For that, we simply go to HRC and click on “Hand” → “Export Strategies”. Then we select the nodes we want to have in FreeBetRange (those will the BTN’s and the BB’s strategies in this example). This creates a .zip-files which we can then easily import to FreeBetRange. Now, in FreeBetRange, we instantly have the full tree available that we imported. If you want, you can now modify the ranges. For example, you could clean them up and simplify them by rounding the frequencies. For that, you only need to click on “Round now” in the yellow prompt. After the rounding, the range above would look like this: Which is much easier to apply in practice than the previous one. By default, the rounding increment = 25. This means that all the weights in the range will be rounded to the closest number from the [0,25,50,75,100] set. You can also configure the rounding increment in Settings: To make sure you know these ranges in practice, you can train them. Click on “Training mode” and you’ll be able to select which hands you’re going to be asked about in the training ("Dealing coverage"). Because let’s be real, we all know that J4s is not a continue against a 3bet here. So you can select to only get asked about those hands that are comparably close in their action. As you can see, the smart script already selected only hands in range + the closest hands outside the range, for efficient training. You can modify the dealing coverage as you wish. “Start classic training” starts the training mode in which you get dealt a hand and have to choose between each of the 3 options in this spot: fold, call, or 4bet. This makes it very easy to actually remember the ranges when it matters most: in game. Conclusion The above process – researching preflop data in H2N, converting it to ranges that can be input to HRC, and solving for exploitative preflop strategies against the player pool – is exactly what I did to work out preflop ranges in my course Preflop Xploits. In there, I created a full guide including every spot and position and the according exploitative preflop ranges. Does the average regular overfold against a 3bet as BTN vs BB or do they overcall? Do they 4bet too much or too little from the CO against the SB? All that and much more is covered in the course. So that you’ll know exactly how the player pool plays and how to exploit them. So if you want to save yourself the time of going through each and every position and extracting the ranges from a large database of hands yourself, join the course and instantly enjoy the video explanations that I made for each situation.
