ROI: The Life and Times of a Heads Up Pro

Game Selection

In looking for HUSNG matches, we’re faced with the decision between sitting first and taking on whoever decides to play us, or to use various methods to select opponents to sit down with. This need not be an either/or situation though, as we may choose to do either one or the other depending on the circumstances. As a professional poker player, you will need to set your own criteria to decide if and how you’re going to make these decisions.

In order to be properly informed, we need to know the typical caliber of player that sits first and second. This can differ quite a bit by poker room and buy-in, as well as time of day. So we’re going to need to know what type of player we can expect on average by just sitting in, and to do that we can just observe what kind of players are sitting down first and second. So the average caliber of the players sitting in second is our baseline, which we can determine simply by scoping out both players. This is what we need to beat when we are game selecting, or it would make more sense to just sit in and wait for the fish.

One of the potential advantages here is that winning players may tend to avoid you if you’re a winning player yourselves. If they tend to sit first more, as is often the case, then they are less likely to be hunting around for matches. The losing players, however, are less likely to be doing proper game selection and thus more likely to just sit down regardless of your stats. While this isn’t always the case, it often is, which is why determining what can be expected from sitting first is where we need to start. There may be some time considerations here though, depend- ing on how long you have to wait on average for an opponent.

We’ve decided that we can beat the benchmark by selecting our opponents, or at least that we’re going to try to do both, meaning sitting in and at the same time looking to sit with a desirable player. If we do this though we do run the risk of getting stuck in two matches at the same time, so we need to either be comfortable with this or confine this to where the traffic is light enough that the risk is acceptable. Providing that sitting in can be expected to provide a positive ROI, and we’re looking to improve on that, we now need to consider the opportunity cost of the time involved. So efficiency is going to be a concern, and we are going to need a method that is both accurate and timely.


Databases like Sharkscope can give us a quick and dirty idea of a player’s skill level. Time is of- ten of the essence here, and while it’s preferable to have more information on an opponent than this, viewing a players’ results certainly provides us with useful information. Depending on the level of traffic, there may not even be enough time to view these stats and still have a reasonable chance of sitting in, so we may have to rely on other methods such as player recognition if this is the case. It is important to filter these results properly though, so that they are relevant. We also need to be aware of trends, including the player being hot or cold, especially since players who may be on tilt may be easier prey.


It can be pretty time consuming to datamine HUSNGs, as it’s a pretty involving process, but especially if you’re playing the higher stakes, the information gained by this can prove to be pretty valuable. As you collect more and more data, you’ll be able to check out much more detailed stats fairly quickly. This will tell you not only whether a particular player is a winner or not, but what type of player he is, and his potential strengths and weaknesses. As a side benefit, when you do play him, you’ll have more hands on him and can use this to your advantage.
How you’ve done against a particular player is far more important than how he does generally, and it’s a very good idea to track this.

Keep in mind though that you do need a meaningful sample of matches to neutralize the element of luck. However, even more important than the results themselves are the notes you’ll be taking here, and even a single match against someone will give you a very good idea of a player’s skill level and your chances against him in the future. In addition, you’ll be noting reads on him that go beyond the stats.

Tracking your results

Once again we need to be comparing the results of our game selection with how we would do if we were just sitting first. If our selecting results in our lowering our ROI rather than raising it, then we either need to select better or not at all. So we’re going to need to track both to compare, which takes little time but is very important. It will take some time though to get enough of a sample to judge this properly, but we can add in our own observations to this as far as the kind of players we’re seeing with both. In the end, the numbers will tell the story very well, but along the way we surely want to pay attention to the relative caliber of players we’re getting with both, and use that information as well.

Benefits of Game Selecting

Obviously we’re looking to make more money selecting our opponents and if we do this successfully, then a higher ROI will follow. This increased success will also pad our Sharkscope stats so that it we are mixing in sitting first, it becomes more likely that we’ll get a higher percentage of fish, as the better players will be more likely to avoid us. Also, since our winning percentage will be higher, we’ll reduce our variance, as the frequency and magnitude of losing streaks will be reduced. We can also more comfortably play higher stakes, since we’re in control here and won’t be subject to players sitting with us that we may not want to play. Also, if the waiting period sitting in tends to be longer, this can get us more matches potentially, provided that there are acceptable opponents available, further increasing our profit per hour.

Drawbacks of Game Selecting

The biggest one is the risk of putting ourselves in inferior situations relative to sitting first, which include both running into tougher players and making mistakes in our selection. There’s also the risk of having juicy players available but not getting to play them if someone else beats us to the punch while we’re making our decision. This is a particular concern if there’s a waiting list for the next available table, where you’re up against players who aren’t selecting at all. As well, if you’re always looking for the weakest players, you may not be challenged enough and your development may suffer. However, you may see this choosing a more gradual path of improvement as being acceptable if you need to manage your risk more closely due to bankroll considerations. You may also end up playing less matches per hour than if sitting first, as selection takes time, although this isn’t necessarily the case. Meta-game and History
One of the great things about game selecting is that by having experience against an opponent, you can gain a great deal more insight on him than just looking at stats. This need not even be direct experience, and particularly at the higher stakes where the competition increases, it’s often worthwhile to watch players play each other to look to gain reads. What we’re looking for here is particular playing styles, in terms of what we want to see and what we may not want to see. An opponent could be a winning player and we may still find him a very desirable opponent based upon his play and the reads we get on him. On the other hand a losing player may play a style we may not find ideal to play against and we may want to avoid. So while results can be very helpful, they are far from the whole story, and our own observations and reads are even more important as far as deciding who to play.

As is the case with all forms of poker, we want to single out players who have flaws in their game that we can exploit. The great thing about heads up poker is that since both players are involved in every hand, we can identify leaks in their game much more quickly. It always comes down to a player doing one thing or another too much or not enough. In the case of how this all relates to game selection, if we know what we doing then we have a good idea of the differences between solid play and exploitable play, and we’re looking to rule out the solid players and focus more on the weaker ones, regardless of how they may have done against others.

Niche Selecting

Will you play regular-speed or turbo games? How about super-turbos? Deep-stack games? Personally, I prefer the regular speed games. I am pretty good at playing 60-75BB deep, and I don’t really feel pressured to make moves early.

Maybe you’ll play cash. A lot of successful players play cash as well as SNGs. It’s really the same game, underneath it all. Only a few things change, like how good your hand needs to be to get all in once the stacks are 250BB instead of just 75BB.

What about stakes? Will you play at the top of your bankroll, or can you make more money playing lower-stakes games?

How many tables will you play?

You’ll have to answer these questions when deciding which games to play. Focus on the games that bring you the highest $/hr, to the best of your knowledge. It’s also fine to trade a small piece of your $/hr for a much lower variance game, if you have that option. For example, I make nearly as much money four-tabling the $30s as I do 2-tabling the $100s. There’s nothing wrong with playing the $30s in cases like this, especially if you’re playing professionally and have to weather large downswings with your bankroll intact. Whatever you pick, make sure you put in a solid sample before switching limits or games.

Proper game selection is something that we always want to consider doing, with a view on looking to increase our advantages. This ultimately will have us looking to increase our win rate per hour over what can be achieved by simply sitting in, and in the end the numbers will tell the story in terms of what particular game selection strategy will work out the best for us. It’s mostly just a matter of exercising good judgment though and making sure we’re always looking to put ourselves in the best position relative to our goals, which include both making money and looking to improve our game.

Bankrolling in HUSNGs

I’ve seen so many poor systems for bankroll management. Like any other poker decision, you want to make the right one. Being too tight will slow your opportunity for growth and advance- ment. Being too loose will have you moving up and down between limits too fast.

What is the purpose of a bankroll? It’s because you don’t win every game. Of course, it’s possible to lose 100 games in a row, and at the end of a run like that, the only person still solvent is the guy who had the 100 buy-in bankroll. This is unrealistic. Of the more than 10,000 heads-up SNGs that I played, the worst downswing I ever went on was 20 buy-ins.

The proper bankrolling requirements are simple to calculate. There’s an equation called the Kelly Criterion which was invented by economists to answer this exact question. We know our ROI, and we know our bankroll. The Kelly Criterion says we can rationally risk any amount of money up to our ROI. This means if our ROI is 5%, we can play with a 20-buyin bankroll.

Now, there’s nothing wrong with being a little more conservative than this. If you calculate your Kelly Criterion bankroll requirements by estimating your ROI at half what it actually is, you should be very safe.

Remember, the more aggressively that you bankroll yourself, the more comfortable you must be with moving up and down in limits. You will shoot through the limits when you’re hot, but when you’re cold, you will likely have to drop down a limit or two. This serves an extra purpose, though. You will play in bigger games when your confidence is at its peak, and you will play in lower games when you are more likely to tilt.

Using the Kelly Criterion to rationally decide how much of your bankroll to risk is a much better solution than just picking a number out of thin air, don’t you think?

There’s one final point that should be made for the micro-stakes grinders out there. NEVER play in a game where the rake is larger than 5%. Switch sites instead, or instead of playing the $1s, play the $2s. You will never earn a bankroll if you pay more than 5% rake.

Analyzing Your Own Play

One of the keys to improving your poker game is to incorporate a regular and systematic method of continually reviewing your own play. There are various techniques that are effective in doing this, and the key is to never let yourself become too complacent as to not devote enough effort to looking to analyze your game and seek out po- tential improvements. In this lesson I’ll discuss some proven approaches and also point out some of the things you need to be thinking about here.

At the foundation of any successful program of self-analysis is a comprehensive system of note taking. We’re not talking about taking notes on your opponents here though, as these notes will instead be on you. I’ve found this to be extreme- ly helpful over the years and it’s not something a lot of players do well.

There are two phases to this, notes written during live play and notes collected away from the table. The live notes do not need to be very long or comprehensive. What we’re after here is making note of situations that you either could have played differently or a good play that may not be standard for you that you want to rein- force. So in a nutshell, you’re noting both your good and bad plays, with the object of looking to eliminate the bad or questionable moves and reinforce the better ones. These notes can be directed toward a certain category of opponent, but they also need to be general enough to be useful on a broader scale.

Taking live notes will present the advantage of having the hand and your current thinking very fresh in your mind, which is why you want to take them as the hands actually occur. The notes you write after your session will increase your ability to reflect on the situations more and look at the bigger picture. While tracking software can be very helpful in providing analysis and insight, they are somewhat limited in their scope and we also want to use our own brains to spot things to change as well. In fact our own observations, unaided by software, can be con- siderably more powerful in discerning potential improvements.

So by combining both these forms of note taking, we will be constructing a poker diary of sorts, and we can add all of our poker insights to it. It’s important that we edit it frequently though, and this is best done by organizing the minutia by topic, organized around a set goal. Then, once the goal is consistently achieved, you can then archive the detailed notes on the topic, although it’s a good idea to keep your ongoing and com- pleted goals in a prominent place so that they can be easily reviewed.

If you want to do this properly, get a copy of Mindjet MindManager. It’s free to try for thirty days. MindManager is organizational software that will let you quickly group your thoughts by category.

Your Poker Diary

When you first get started on effective note taking, there may be a lot to work on, but as you progress you will find that as the bigger leaks get better resolved, which means you won’t take as many notes. Like all things though, you will get what you put into it, but the benefits that can be gained here are very significant. As long as you are well organized and consistent, you will find that your efforts here are time well spent.

Note taking provides the foundation for my entire program of self-review. A combination of live notes and after-session notes, kept in a poker diary, is the first step to improving your game as a poker player.

For after-session notes, we’ll need to go over the session afterward, selecting hands that at least appear worthy of review. You’ll want to look at hands that made and lost significant amounts of money, as well as any other hands of interest, including those you have marked for review during play. It’s best to look to view common situations together if you can, in order to better identify patterns in your play and what can be improved upon.

From there you can hone in on particular situations by applying filters to have your software look at specific situations. There are two main types of filtering you can do. The first is to find a selection of hands that have the certain characteristics you want, so that you can have a look at them and look to find better lines. The second is filtering for situations, like after you raise when you have a draw, to look at overall profitability, so you can discover whether certain plays make or lose money for you. The second is the foundation of my Hold’em Manager Analysis coaching package (available for $80).

There also are add-on programs like Leak Buster that you can use to help find leaks in your game. I don’t recommend these programs. First, they are not meant for heads-up players, so none of the information will be relevant to you. Second, it’s far more effective to learn to analyze your own play. In the end there’s no substitute for putting in the work here yourself as it allows you to develop your better poker playing skills.

Another popular tool that a lot of players use is Poker Stove, where you can plug in ranges and get the program to assign equity percentages. This can be helpful with some situations, and it’s a good educational tool at times, but it’s not something I’d spend a great deal of time with. Some players love it, and will play around with it at great length. While it has its place, you always have to look at your study time in terms of get- ting the most done, and single hand studies can certainly be overdone. If you’re a very advanced player, PokerStove is an invaluable tool.
One of the most powerful means of improving your game, and one that most players really don’t take advantage of properly, is taping your own sessions and then going over them with other players. In other words, you’re pretending that you’re making a training video, with your live comments included in the audio. You don’t have to make a lot of these videos, but they are certainly worth the effort. The benefits aren’t so much watching yourself play as looking at your play from the view of instructing, which can cause you to think about what you’re doing from a third party perspective and provide insights you would not have gained otherwise.

The same benefit can be gained from writing instructional articles, and what happens here is that you push yourself to think a little more deeply about the material than you would have otherwise. It doesn’t matter if the work ever gets published, as it’s the process itself that provides the benefits. This endeavor is highly underrated generally and regardless of your level of advancement it’s something definitely worth doing. While this isn’t reviewing your own play in a technical sense, it is reviewing your own thoughts and knowledge about the game, and that’s a very important part of the process.

Finally, having someone else to go over your own play with you, whether it’s a friend, a peer review group, or a coach, is very important as well. Ideally, you get yourself in a situation where you both describe and defend your thinking, and that in itself can be of significant benefit, in addition to getting the ideas and opinions of other players. This can be in the format of discussing actual hands or it can be more of a theoretical discussion. Very few players have much interest in discussing poker theory, but this is their misfortune, and if you can find some folks who have opened their minds enough to try to get beyond standard play and look for better approaches to the game then you’re really on to something good.

Improving at poker is something that we all know we must do, yet there are so many players who are so focused on playing that they end up depriving themselves of the enormous potential benefits of looking to improve through work away from the table. Reviewing your own play effectively is the cornerstone of this and must not be neglected. While profit per a given time period is important, it’s critical to look at the long run. Spend the right amount of time on improving, and do so in the most effective manner possible.

One Comment

  • MrNeverdie says:

    This article just abounds with good advice, it is hard to say which are of greater importance. Game selecting is obviously very important for achieving your maximum ROI and good hourly rates, as you will be far better of playing softer competition. Playing against good regs can lead to huge swings in your bankroll, and sometimes you will not even be able to beat these people. Trying to find soft spots, especially in higher stakes games is a good advice to follow. And analysing your play constantly, trying to find what you are doing wrong, where you are losing money, or where you could and should be making more money is quite self-explanatory, yet it is to easy to get into the mindset that you are playing well and just running bad. Always find time to think about your game and check your plays, it will be very valuable investment of your time in the long run.

Leave a Reply