The Modern HUD Realities You Can’t Ignore

Dennis  «Dennis_Stets» 
06 May 2024
Holdem Software
06 May 2024

Hey guys, welcome to the Red Chip Poker podcast! Today's episode is all about the modern HUD. I don't see too much training content on creating and effectively using HUDs, especially in the modern era. Everything has become very solver-focused. 

Most training content just appears to be a coach clicking through a pre-solved tree, talking about various mixes.

And as a result, usage of a HUD has perhaps slipped into the background. And yet, it remains one of the most effective ways of generating an edge over our opponents. Because not too much focus is placed on HUDs, we find that there's a very big gap between players who are using a HUD to its full potential and players who are essentially still living in the Dark Ages. 

When we look at some of the main trackers on the market, they've barely improved in the last 5+ years. They still offer the exact same features they did like 5+ years ago. Combined with the fact there isn't too much training content on creating effective HUD setups, we get players who essentially have been using the same HUD setup.

For the last 5 years, maybe even the last 10 years, the objective of this episode is to help bring you out of the Dark Ages and help you to ask the question: am I making the most effective use of my HUD setup? We'll talk about the types of stats that the best players use in the modern era. We'll also talk about some of the features that are available as part of a modern poker software.

Average vs Modern HUD

First, let's compare the Average vs Modern HUD. The main difference between the average HUD that a player would use 5 to 10 years ago and the type of HUD that an advanced player will be using right now is more specific stats. An example that I like to use, simply because there are a lot of questions on this specific topic, is revolving around the stats like WTSD (Went to Showdown), WWSF (Won When Saw Flop), and WSD (Won at Showdown). 

We didn't have a lot of stats to choose from in the earlier days of trackers, so that's part of the reason these stats were popular.

I wouldn't say they are especially useful, but they do have their place in traditional analysis because these were some of the main stats players would use to try and figure out how much their opponents were bluffing, how much they were folding.

However, in the modern era, there are much more specific stats that we can use. A simple example of this is if we take something like One WISF (Won When Saw Flop). Players might use this to try and figure out how much their opponent is bluffing. Generally, more aggressive players have a higher One WISF. So if a player has a One WISF well above 50, we could make the assumption that that particular player is fighting for a lot of pots post-flop. 

So when we do find ourselves facing a river bet from said opponent, we can assume he's bluffing wider. We can bluff catch wider.

Sounds logical, right? But there's a bit of a problem. You see, a player with a high One WISF could also just be a player with a very nitty preflop starting hand selection.

Imagine a player who only ever played pocket aces pre-flop, for example. What does their One WISF (Won When Saw Flop) look like? It's actually very high. That's because they always have an overpair or top set post-flop. Imagine we just looked at their One WISF as the only stat. In terms of playing back against our opponent, we see that it's high. We start fighting back for pots post-flop. We start bluff-catching wide. 

As you can imagine, that's a problem. We're making the assumption based on a very non-specific stat that our opponent is very aggressive post-flop. And though he's aggressive, it's not because he's bluffing a lot; it's because he always has aces. So, this was the problem with trying to use a fairly non-specific stat to draw specific conclusions about our opponent. A certain stat could have more than one meaning.

So, where does that put us in a modern perspective? Let's say we want to figure out how often our opponent is bluffing the river. Well, now we can just have a stat that tells us how often our opponent is bluffing the river. And we can be even more specific with that. So, if we wanted to know how much our opponent is bluffing the river specifically in the triple barrel line out of position in a single raised heads-up pot, we can check that stat specifically. 

We can see our opponent's bluff-to-value ratio in that scenario.

If you don't have access to such stats, then you don't have a modern HUD setup. Many players are still in the Dark Ages. They're still trying to use stats like WTSD, WWSF, and WSD to analyze or make guesses at how often their opponent is bluffing.

Just to give you some other examples, in the earlier days of HUDs, a player would often have a stat like C-bet flop. In that era, it would just be a single stat: C-bet flop.

How often does our opponent fire a continuation bet after being the preflop raiser? And that CBet (Continuation Bet) stat would encompass every possible CBet situation. It would be a CBet flop in single-raised, three-bet, or four-bet pots. It would be a CBet flop in position and out of position. It would apply to a whole range of different situations. It was wonderfully unspecific. 

Whereas now, advanced players on their HUDs have very specific CBet stats. So, they'll have stats such as CBet flop in position, single-raised heads-up pot. They'll also have another stat for CBet flop out of position, single-raised heads-up pot. They'll have separate stats for CBets in three-bet pots and four-bet pots. We have very specific data points. This is how HUDs have evolved. If you're still making use of unspecific stats, then you're missing out. 

In this case, another example of the extra specificity here is that stats take into account the following board textures.

So, we don't necessarily even just have stats for CBet flop in position, single-raised heads-up pot. We might have CBet flop in position, single-raised heads-up pot on an Ace-high texture or an Ace-high dry texture or mid-range connected board.

For example, we can also take into account bet sizing and runout. So, instead of just CBet flop in position, single-raised heads-up pot on a high dry texture, we might have something such as CBet flop in position, single-raised heads-up pot on a high dry texture, button versus big blind, for 1/3 pot sizing. Sounds kind of intense, right? But this is what we have available to us with a modern HUD. We can create stats to that level of granularity.

The big advancement that made this possible was the advent of custom stats.

You see, in the early days of trackers, we had to basically just use whatever shipped with the tracker by default. It could be something like 3 to 500 stats; players might make use of 50 of those stats. Whereas now, a good modern tracker will allow us to make use of custom stats, and that opens the possibility of having millions of different stats, potentially covering a wide range of possible situations.

Now, in some cases, even with the biggest trackers out there, it's necessary to purchase add-ons in order to create custom stats. As you can see, there's some friction or resistance to being able to use custom stats in the first place. Many players out there perhaps just never purchased an add-on to their HUD, or they're not using a tracker that allows custom stats. So essentially, they're still in the Dark Ages and are missing out on this opportunity to create very specific stats for targeting our opponents.

I'd like to focus on one subset of custom stats, and that is range.

Range Based Stats

Range-based stats are extremely powerful. In fact, they're so potent that they're not allowed on all networks. For example, on the Stars Network, although HUDs are still allowed, range-based stats are banned. However, we can use them off the table for villain analysis without violating the terms and conditions. But, these stats won't load up on the table in-game. If they did, we could find ourselves getting into trouble with the room.

So, what is a range-based stat? It's a statistic that's not based on a frequency, such as how often our opponent c-bets the flop, but rather on a range.

For instance, how often our opponent shows up with a flush draw when they do c-bet the flop, or how often they show up with a gutshot when they raise.

Stats based on range rather than frequency allow access to a few different powerful features. One of them is heat maps. In the early days of HUD usage, many players would have something like a three-bet stat, for example, but it would just be a frequency. Now, it's very common on modern HUDs to have specific stats for different positions, like three-bet from the big blind, three-bet from the small blind, and even position-versus-position three-bet stats.

You definitely want those kinds of stats on your HUD, but moving beyond that, we don't just get frequencies in our modern HUD setup.

We get a snapshot of the range itself via a heat map. In other words, we can hover over our opponent's three-bet stat and see a grid detailing all of the hands that player has three-bet over the sample. 

We can access those heat maps for specific situations. So rather than just being able to see, "Okay, our opponent is three-betting 12% big blind versus small blind," with a modern HUD setup, we can then access a heat map and see all of the hands that he did three-bet in that situation.

Another example of a very powerful range-based stat feature is the ability to create bluff frequencies. As we mentioned earlier in this podcast episode, an example of such a stat would be assuming our opponent bets the river. How frequently does he show up with the bottom pair or worse? 

In other words, this stat tells us directly how often our opponent is bluffing in a certain situation. It then becomes very easy to figure out if we should be defending very wide with bluff catchers or simply folding our bluff catches as an exploit because our opponent is under-bluffing relative to a theoretically correct bluff-to-value ratio. This capability is incredibly potent.

We can also see range distribution information for other streets.

So, if our opponent bets the turn, for example, it's possible to see a breakdown of all the different categories of hands our opponent bet on the turn in a certain situation. This feature provides valuable insights into our opponent's hand range composition and aids in making more informed decisions throughout the hand.

Sizing Patterns

In a modern HUD, one of the key features is the ability to detect sizing patterns. In the earlier days, HUDs didn't really have any sizing information. For example, we would just have a c-bet flop stat. And even if you were especially advanced, you may have broken that down into c-bet flop in position, single-raised pots, heads-up, for example. But you wouldn't have any sizing information.

So, if our opponent was c-betting like a maniac and then occasionally sizing up only with his stronger hands, we wouldn't necessarily know that. In fact, we would just see the larger sizing. We would check the c-bet frequency, see that a player is c-betting too much, and we would probably end up defending very wide against that c-bet, despite the fact that if the opponent has sized up, he always has a strong hand.

A modern HUD now has detailed sizing breakdowns for every scenario.

So, for every stat we have, we also have a sizing breakdown. When our opponent c-bets the flop, we can see exactly how often they used 1/3 sizing, half pot, 2/3 sizing, how often they overbet. For every single part of the game tree, and also for each of those individual sizings, we can have a range-based breakdown.

For example, we can see 1/3 pot. How often did our opponent use that sizing? Is it weighted towards strength or weakness? We can detect sizing tells this way. We might see that some players, every time they size up past half pot with their flop bet, they always have a strong hand. They never have trash. We might find the opposite true with some players. You might find they like to trap with the smaller sizings.

A modern HUD makes it very easy to detect bet sizing patterns, which can be exploited. If you're still using a dark ages HUD from 5 to 10 years ago with no sizing information, you have no reference.

Spotting Sizing Tells

If you want to spot a sizing tell, you have to manually spot that over a large sample. It's going to be very difficult to keep track of potentially hundreds of different opponents. Whereas with a modern HUD, you can detect a sizing pattern just by bringing up a relevant popup. So when sizing information is also combined with range distribution, it can be very powerful.

Now, we often find that if we were trying to detect a sizing tell in a very specific scenario, for example, let's just say we were looking at a river probe bet, and we can access how often our opponent is using different sizings, we can access the range distribution on those sizings. The problem is we might lack sample size. So it becomes hard to detect patterns, especially when we're thinking about the lower frequency spots.

So although the general purpose or functionality of a modern HUD is to make use of more specific stats, there is still room for general data points.

Think about something like turn bet, for example. So we're not breaking it down into a specific situation. We're not talking about turn c-bet or turn probe bet or delayed c-bet on the turn or turn float bet, just turn bet. Any single time a player bets the turn.

Now, that's a very non-specific stat, but the advantage of a non-specific stat is it's going to have a larger sample size for all of the different sizings that a player might use. We can then access the range distribution for each of those sizings. Because we have that larger sample size, it could potentially be easier to spot bet sizing tells.

Now, I'd like to just run through some of the aspects or features of a modern HUD. This is just to open your eyes to what's available. If you're not maximizing the value of your HUD setup, if you are already using a selection of these features, that's fine. This is really to help players who have not advanced their HUD game in the last 5 to 10 years.

Dynamic HUD

Let's think about some important features of a HUD. Now, this first feature—full disclosure—I don't actually use this, and I have reasons why I don't like a dynamic HUD. But it's good to be aware that it's an option, and this may work well for some players.

The idea behind a dynamic HUD is that the HUD panel will change based on the situation that we're in.

So, for example, it could only show preflop data. Then when we head to the flop, the HUD panel changes and now it shows flop data. As we go to the turn, it now changes and shows turn data. Or if our opponent's in position against us, it just shows in position data. Or if he's out of position, it just shows out of position data.

I think this is especially beneficial for a player who doesn't like to have lots of stats available at any given time. The disadvantage is that we lose context for some of our stats. So what I mean by that is, anytime we're looking at, let's say, a turn c-bet frequency, it helps to analyze that within the context of how much did our opponent c-bet the flop initially and how much did they open raise pre-flop or three-bet pre-flop initially. So, good HUD usage often involves interpreting one stat in the context of other stats or scenarios that occurred previously.

Obviously, if we only see turn data on a HUD panel, we can't reference the flop play or the preflop play.

Of course, I guess it depends on how you set the HUD panel up. It could be that we have a static component and a dynamic component. So, an example of how you might think about using a dynamic component is any time our opponent open raises, the HUD panel loads up his fold to 3-bet stat from that position. 

So rather than have his overall fold to 3-bet, or I guess you could have that as well, but we have a dynamic component to our HUD which loads up his specific fold to 3-bet from that position. As you can imagine, there's a lot we can do with this depending on your style for using HUDs. You may be able to unleash some dynamic features which can really help you with finding the relevant information quickly.

Versus Hero Stats

Another feature of modern HUDs is versus hero stats. On a modern HUD, for every single stat that we have, there should also be a relevant versus hero stat. So if we have a three-bet stat on villain, we should also have a versus hero three-bet stat on villain. So, how much does he three-bet on average? How much does he specifically three-bet against us? This applies to every stat. So, if it's a c-bet flop, how much does the villain c-bet? How much does he c-bet against us? Versus hero stats were fairly rare on earlier HUDs. There are still many players without any versus hero stats on the HUD setup. Whereas in the modern era, every single stat we have should really have an equivalent versus hero version for our opponents.

Of course, one of the things that we can use a modern HUD for is range research or population analysis as well.

It is possible to do range research and population analysis with Holdem Manager 3, PokerTracker 4, but of course, the main tool that players use for range research is Hand2Note. It has a few more built-in features that are very useful for range research. But other trackers are also looking into ways they can provide range research or population analysis functionality. So if you don't do any range research or pool analysis, once again, you've been left behind a little bit by the trend here.

Custom Stats

Custom stats, as we've already mentioned, offer the ability to create any stat based on a bunch of preset filters. Hand2Note has built-in custom stat creation.

Holdem Manager and PokerTracker allow the creation of custom stats through an add-on which is referred to as NoteCaddy.

Although PT4 does technically have some built-in ability to create custom stats, there are other trackers now offering custom stats, for example, DriveHUD. I'm not really trying to promote any individual HUD here, just giving you a rough overview of which trackers offer custom stats. I'm sure there are some that we haven't mentioned on the list as well.

Mass Cloning

Mass cloning, this is a feature that is uniquely offered by Hand2Note at this stage. What it allows us to do is quickly create a range of additional stats based on other stats. For example, imagine I created a whole bunch of stats for single raised heads-up pots. Let's just say I invested a bunch of time in it and now I'd like all of those stats recreated for three-bet pots. With most trackers, you'll have to manually recreate every single one of those stats. 

And what this means is to make a large HUD setup, there's a lot of work involved, and we often find that the upper limit in terms of the number of stats is capped.

So back in the day, a large HUD setup would have maybe had 300 to 500 stats. That was considered huge because the amount of work required to make 500 individual custom stats, well, it's a lot. 

With mass cloning, you can just take all of those single raised heads-up pot stats, and in the space of about 30 seconds, you can just clone them but just specify that they are specifically for three-bet pots. You could then do the same thing for four-bet pots. Within 2 minutes, you have an extra few hundred stats on three-bet pots and four-bet pots. That's the power of mass cloning.

In the modern era, a large HUD setup is now something that has over 10,000 stats.

Besides, we often find that a modern tracker may automatically generate additional stats based on custom stats that have been created. A simple example of this is every time we create a custom stat in Hand2Note, it creates an auto generated popup that breaks that stat down by a range of different sizings, by ranges, it adds a versus hero stat, adds some next actions information. So for every stat we create, there's an extra 50 plus stats created automatically by the tracker itself.

Although a large number of custom stats for HUD setups is considered something like 10,000 stats plus, if you factor in all of the automatically generated data from a good tracker, we're talking about HUD setups with well over a million data points available for use in-game.

So, as you can imagine, comparing that to a Dark Ages setup where maybe there are 200 stats in the popup, which was considered a massive HUD setup back in the day, 200 stats versus 1.5 million stats, for example, you can see there's a big gap between players who are maximizing the value of the HUD and players who are not.

Color Coding

Color coding is something that's been around for a long time. The idea is we can color code dynamically. So, the simplest example of this is if we had a VPIP stat, we could set it to a certain color if our opponent is very loose and another color if our opponent was very tight. But we could also have some kind of neutral color to say that our opponent is playing within an acceptable boundary. This has been around since the Dark Ages, so color coding of dynamic stats is something that many players like to make use of. 

Once again, I don't really make use of dynamic colors. I prefer that each group of stats has its own unique color associated with it. A simple example is I always know my VPIP stat is blue, so it's very easy for me to locate stats in a fairly large panel by their coloring.

Obviously, if colors are changing dynamically, there are some advantages to that in the sense that it allows us to perhaps just look at our opponent and see very quickly whether he's a very tight player or a very loose player. But the problem is stats are changing color all of the time.

If you have a very large HUD panel, I've seen players really struggle to identify their own stats. This is something I see very commonly, even among advanced players. They may have 50+ stats on their panel and constantly search around to figure out which stat is which. Even high-stakes players, not just recreational ones, can struggle with too many stats on their HUD. They haven't learned their HUD properly, and I think dynamic colors just make this harder. 

When you're trying to repeatedly find a certain stat but it keeps changing color, it can be frustrating.

That's why I don't use dynamic color coding much. Full disclosure, I do use it on VPIP and PFR, but that's about it. I prefer my stats to have solid colors so I can identify them quickly in a large panel. But this is just personal preference. The point is, this feature is available, and you may be able to leverage its power to improve the speed and efficiency of your opponent profiling.

You might quickly see what type of opponent you're dealing with based on the colors reflected back to you on your HUD panel because you're making use of dynamic coloring.

Note Taking

Regarding note-taking, my personal viewpoint is that it's now obsolete for me. Note-taking is something players used to do 5 to 10 years ago, but I don't think it's something a good player should be doing in the modern game. I don't take any notes on any opponent, simply because the power of what I can do as a human writing notes is completely dwarfed by the power of automated data gathering on our opponent. 

For example, if my opponent c-bets the flop for 2/3 pot and shows up with a value hand, I could go to the notes and write "villain uses 2/3 pot sizing, showed up with a value hand", right? But if I bring up the popup, I can see the last 100 times he c-bet the flop. This level of detail and data availability makes note-taking seem outdated and inefficient by comparison.

I can see how often he used different sizings and his precise range distribution when he used those sizings. You can see the power difference between my one-liner, which doesn't really help that much, fails to detect an overall trend, and the power of my popup, which in a very short space of time visually reveals to me my opponent's entire trend across a range of different sizings and allows me to quickly spot sizing tells.

Note-taking has just become completely pointless in my opinion, and players still do it.

Players do all of the things we've discussed in this podcast episode: they take notes, they use unspecific stats like WTSD, one win, so flop; they have HUD panels that only offer, let's say, 200-300 stats including everything in the popups, because that was considered a big HUD setup back in the day. But if they think carefully about it, they haven't actually upgraded their setup in the last 10 years or so. 

Final words

And that's really the purpose of this podcast episode: to think carefully about your current HUD setup based on the topics we've discussed. Are you personally maximizing value from your HUD? Is your HUD an ultramodern setup with, let's say, over a million data points, or is it something that's a little bit more reminiscent of the Dark Ages of Poker?

So hopefully, it's given you something to think about. You can definitely generate a very, very healthy edge by using HUD stats and your HUD setup better than your opponents, so it's well worth looking into. Thanks very much for listening, guys. This was Coach Weasel, and this was the Red Chip Poker Podcast.

There are no comments here yet, you can be the first!