Join Our Poker Discord Server! >> Discord
We recommend CashGameHero as the best course: CashGameHero
We play at GGPoker: >> https://bit.ly/ggpbonus <<Get a massive deposit bonus by using this link!
One of the first things that I personally thought when I had seen a HUD(Heads Up Display) was that “This is cheating” and “It’s not fair for someone to play in these games without using a HUD!”. But oh was I wrong. A HUD is not Superman. It will not give you very accurate and precise information on your opponents unless you have a massive amount of hands on them which is very hard to get, you have to be playing the same limit for a very long amount of time, and our main goal is to go up in limits, but when you reach your peak limit for the moment and you see that you’re breaking even on it and it’s not going your way anymore, you will probably put a massive amount of hands at this limit and study your opponents very much in detail, figure out their leaks and eventually beat them.
Studying away from the tables becomes very important when you go up to midstakes because the player pool becomes much smaller and you can focus on individuals that you meet very frequently at the tables, figure out what their leaks are and then exploit the hell out of them, and also don’t forget to adjust to their adjustments!
You see, the problem with more in-depth stats is that, the less frequent the event takes place, the more hands this stat will need to be accurate.
Let’s take, for example, VPIP. It relies on the event that a hand has been played. This is the most often played event at the poker table so every hand matters.
Now, let’s take for example BB Defend vs SB Steal. First of all, the event that you have to be in the BB matters, so once in 6 hands. Now, everyone has to be folding for the SB to steal. Normally, on average, a player is going to be opening 15-20% EP/MP, 25% CO and 40% BTN. This means that you have to dodge all these numbers so everyone can fold and the SB has to steal.
To give a math perspective on this, let’s think.
Probability of EP opening = 15%, probability of EP not opening is 85%;
Probability of MP opening = 15%, probability of MP not opening is 85%;
Probability of CO opening = 25%, probability of CO not opening is 75%;
Probability of BTN opening = 40%, probability of BTN not opening is 60%.
To figure out how often, on average, SB is getting an opportunity for a clean steal, we have to multiply all the probabilities that EP/MP/CO/BTN are folding. We arrive to the number:
85% x 85% x 75% x 60% = 32.51% which is roughly 1 in 3 times.
Now, we also have to take into consideration SB’s opening range which is, on average, somewhere around 40% so 1 in 2.5 times.
Now, to figure out the end result, how often does our BB get stolen, we take into consideration these 3 results.
1 in 6 hands in BB multiplied by 1 in 3 times that it gets folded to the SB multiplied by 1 in 2.5 times that SB actually decides to steal = 1 in 6x3x2.5 = 1 in 45 hands.
First conclusion is that, to have a relevant sample of BB Defend vs SB Steal, we’d need at least 100 situations like this, which makes up for 4500 hands. Not very often doable.
Now you have to understand that variance can influence this stat very much, 100 situations is a decent sample but a lot of people will consider this stat over 300 or 500 hands, considering this amount to be a relevant sample size for any stat, which we’ve just proved not to be true.
Now, following that the estimation of 100 situations happened is fairly relevant, let’s try on some other stats and see on how many hands are they mildly accurate:
PFR – Same as VPIP – You can always raise so then 100 hands is enough.
C-bet flop – Here it depends on the player’s PFR, so if we have a player who has a PFR average of 20%, let’s say 20% of the times when he raises he gets 3-bet so we’ll take into consideration the fact that he gets to flops around 16% of times. This is roughly 1 out of 6 times. This means that C-bet flop is going to be accurate on 600 hands. Now, for a 15/13 player, he’s going to get to a flop about 1 in 10 times so this means that we need 1000 hands on him to have an accurate C-bet stat. Pay a little bit of attention here, though. If someone is raising 13% pre-flop they are going to have a stronger range than the 20% guy so they should be prone to c-betting more because they just have more value in their ranges.
The stats that I include in my basic HUD are some of the stats that you need the lowest amount of hands for them to be accurate. You will never see in my HUD very particular stats and it’s worked out very well for me. My general thought process here is that I include the players that I don’t have many hands on in as particular categories as possible, and then I play with their distinct weaknesses to a point where it becomes a cat and mouse game.
So, to figure out how many hands you need for a basic stat in your hud to be moderately accurate, you need to multiply how many hands are needed on average for this event to occur at least once with 100. If you want less accuracy but still an okay sample, multiply it by 50.
My HUD is as follows:
Name / Nr. Of Hands / Tot. Sqz / WTSD
VPIP / PFR / Tot 3-BET / Tot. Fold to 3-bet / Tot Steal
Limp-Fold / Flop donk % / Postflop Aggression Factor / C-bet Flop / Fold to C-bet Flop / Total Fold to Steal
All of these stats are very generalistic and give me a vague idea of where the player might be too tight or too loose and I figure out his exploitation points and then I press on them. You learned how to do this earlier when you read the stealing chapter, for example.
We play at GGPoker: >> https://bit.ly/ggpbonus <<Get a massive deposit bonus by using this link!
