Working out ranges

OK, if I want to work out how often I should be defending my blinds against a button raise how do I go about it?

Is it as simple as putting in a range for villain (say 30%) on Equilab and then selecting a range for myself that has more equity than villain's range? Then deciding what hands I should call/3bet with?

For example, if we say villain opens his button 30% and his range is;

22+,A2s+,K7s+,Q7s+,J8s+,T8s+,98s,87s,76s,65s,A8o+, K9o+,Q9o+,J9o+,T9o

Then our range can be;

22+,A2s+,K9s+,Q9s+,J9s+,T9s,98s,87s,76s,65s,ATo+,K To+,QTo+,JTo

And Equilab says we are 52/48 against villain's range. Is that too close? Should I narrow down my range?

Once we have figured our playing range, we then break it down into our calling/3betting range right?

What stakes are you at mate? I play 2NL and would only defend with any pocket pair, A, K and sometimes any Q, some suited connectors sometimes as well. My 3bet is 11bb and it seems to work more times than it fails. People are generally opening the BTN as a standard steal all the time in the micros but if I get 4bet I would fold unless I had a really good hand to play back with. I don't really think to deeply about the BTN range when they are opening, this might be a leak in my game but it seems to be working for me. This might not be the in depth answer you are looking for but hopefully someone more experienced than me will give a better answer

It depends.

What does the button do when we 3-bet?

If the button raises too often but then folds to 3-bets too often then obviously we want to make more 3-bets than usual. Sometimes I'll even randomize it to force myself to do it. For example, sometimes I'll choose a suit like spades and then force myself to 3-bet a button raise if the highest card in my hand is a spade.

@MrFerguson - I'm not actually playing online atm. I'm mostly playing live.

@Eric - I'm not so much thinking about our 3betting range as opposed to our whole range. Like, how often we should be defending our blind and what our range as a whole should be.

Ah ok mate no worries, my insight is pretty irrelevant then lol

It depends on a lot of different factors. One place to start is that you can't just use the logic that you'll play a range that has 50% equity against his range because that would mean you have hands in your range that are less than 50% against his. You're also not accounting for position and the fact that equity isn't the same thing as EV.

I would suggest starting with the 3-betting game. There are threads in the SSNL digest that show how to do this as well as in my strategy anthology sticky in SSNL.

OK Spoon, I will take a look at 3bet ranges.

How does this help me construct an overall range though?

Shouldn't a 3bet range be decided after I know what my whole range should be?

Originally Posted by Cobra_1878

OK Spoon, I will take a look at 3bet ranges.

How does this help me construct an overall range though?

Shouldn't a 3bet range be decided after I know what my whole range should be?

Nope. The size of your range as a whole is going to be a function of the number of hands you want to 3-bet for value (among other things). Try this:

1. Start with finding the hands in your value 3-betting range. Use the size of that to determine the size of your 3-bet bluffing range (but not the hands in it) by deciding what percentage of your 3-bets you want to be total bluffs.*

2. Call with some reasonable number of hands that aren't strong enough to 3-bet with.

3. Choose the hands for your 3-bet bluffing range that aren't good enough to be in step 2, but keep the range itself to the size you found in step 1.



* Here's a quick example of what I mean by this. Suppose for whatever reason your value 3-betting range is QQ+. That's a total of 18 hand combinations. If you decided that you wanted 75% of your 3-bets to be bluffs, then you would need to bluff a total of 54 hand combinations to achieve that. I'm not sure how mathematically inclined people will be who read this, so I'll offer a quick shortcut formula on how to find that number of combinations of bluffs.

# of bluff combinations = (bluff %/value bet %) * # of value bet combinations

In this example:

# of bluff combinations = (75/25) * 18 = 54

OK, I've been doing this backwards then, damn.

I'll work on those steps and will present what I come up with here.

Is the % of bluffs I want in my 3bet range entirely down to me or should other factors influence this?

Originally Posted by Cobra_1878

OK, I've been doing this backwards then, damn.

I'll work on those steps and will present what I come up with here.

Is the % of bluffs I want in my 3bet range entirely down to me or should other factors influence this?

It more or less comes down to how exploitable you want to be and how much you're trying to exploit your opponent.

As a general rule, a balanced strategy in no-limit hold'em will have about 2/3 of your 3-betting range as bluffs. If you think your opponent is calling down more than he should, then you'd want to bluff less than that to exploit him. Along similar lines, if you think your opponent is folding a lot more than he should, then you'd want to bluff more than that.

If you're going to be out of position, then you might want to bluff a little less as a general rule, but that's mostly just because your bluffs (really semi-bluffs) that get called and see a flop will have less value than if you acted last.

Edit: I went and found a couple of related items from my strategy column:

• Simple 3-Bet Bluffing for Not-So-Advanced Players
• How to Decide Which Hands to 3-Bet for Value Pre-Flop

I might work out an example later if I get time.

Alright, I've read both of those articles and here is what I have come up with;

3bet value range; TT-AA, AQs+, AQo+ - 52 combos.

3bet bluffing range; A2s-A5s, A8s-A9s, JTs, T9s, A2o-A5o, A8o-A9o - 104 combos

Calling range; 88-99, ATs-AJs, KTs+, QTs+, ATo-AJo, KJo+, QJo - 100 combos

The problem I had is that KTo, QTo and JTo don't fall into any of my ranges and they seem way too strong to fold - can these kinds of hands go into my calling range against a BU open?

I thought about taking out some of the o/s Ax (A8o-A9o) but I figure the A blocker is pretty important for my 3bet bluffs.

Thoughts on ranges and comments?

Another thought; KQs could be 3bet for value? Most people are 4betting/jamming AK and sometimes 4betting AQ as well so we're not often dominated when we do get a call.

Originally Posted by Cobra_1878

Alright, I've read both of those articles and here is what I have come up with;

3bet value range; TT-AA, AQs+, AQo+ - 52 combos.

3bet bluffing range; A2s-A5s, A8s-A9s, JTs, T9s, A2o-A5o, A8o-A9o - 104 combos

Calling range; 88-99, ATs-AJs, KTs+, QTs+, ATo-AJo, KJo+, QJo - 100 combos

The problem I had is that KTo, QTo and JTo don't fall into any of my ranges and they seem way too strong to fold - can these kinds of hands go into my calling range against a BU open?

I thought about taking out some of the o/s Ax (A8o-A9o) but I figure the A blocker is pretty important for my 3bet bluffs.

Thoughts on ranges and comments?

I'm interested that you don't have something like 76s+ in your calling range, but that's probably a minor issue.

KQs probably dominates a lot more if you call with it.

Its not quite correct because there are hands in that range that have less than 50% equity against the villians hand, jsut like spoonitnow pointed out. You have to play hands that have more than 50% equity against villian's range. The 3betting range depends on the villian's fold to 3bet in btn. The higher this statistic is, the more hands you have to 3bet, because you have higher FE, but you have to be more careful when you do get called or 4betted.

Originally Posted by spoonitnow

I'm interested that you don't have something like 76s+ in your calling range, but that's probably a minor issue.

KQs probably dominates a lot more if you call with it.

Thanks Spoon.

I'd be more inclined to 3bet with a hand like 76s, it seems a little weak to just call with?

Originally Posted by Cobra_1878

Thanks Spoon.

I'd be more inclined to 3bet with a hand like 76s, it seems a little weak to just call with?

It depends on the situation. You probably don't want to call with it out of position against a single opponent, but in position it's typically going to be worth a call because you have so many opportunities for aggression.

Ah OK, I was thinking of my range in terms of BB vs BU only.

Originally Posted by matiusaa

You have to play hands that have more than 50% equity against villian's range.

Having greater than 50% equity IS NOT THE SAME as having greater than 0 EV.
(I'm resisting the urge to add bold and underline.)

The fact that there's dead money in the pot means you ALWAYS need LESS THAN 50% equity to be +EV.
Exactly how much less than 50% equity will vary, but is simple to figure out.

If you base your ranges on equity and not EV, you will be folding a lot of hands that are +EV to play.

E.g.
It's folded to the button, who opens to 3 BB, and the SB folds, with you in the BB.
There is 4.5 BB in the pot, and you face a bet of 2 BB.
bet/(bet + pot) = 2/6.5 ~= 30.8%

So you only need ~31% equity to have 0EV.
If you're folding hands between 31% and 50% equity, then you're folding +EV hands.

Originally Posted by Cobra_1878

Ah OK, I was thinking of my range in terms of BB vs BU only.

Oh okay. To clarify then, when OOP against a single opponent, I'd be more inclined to 3-bet 76s than call with it.

Originally Posted by MadMojoMonkey

E.g.
It's folded to the button, who opens to 3 BB, and the SB folds, with you in the BB.
There is 4.5 BB in the pot, and you face a bet of 2 BB.
bet/(bet + pot) = 2/6.5 ~= 30.8%

So you only need ~31% equity to have 0EV.
If you're folding hands between 31% and 50% equity, then you're folding +EV hands.

This isn't always true. One example is a spot where villain has the nuts or air(that is currently ahead of us) on K88 and we have a flush draw. Villain can give us better then needed odds to call if no more money went into the pot on future streets. However if villain is shoving his entire range when the flush misses on the turn giving us a bad price on the turn we were actually just burning money on the flop since we probably never actually had the price to realize just 1 street of equity.

Originally Posted by Icanhastreebet

This isn't always true. One example is a spot where villain has the nuts or air(that is currently ahead of us) on K88 and we have a flush draw. Villain can give us better then needed odds to call if no more money went into the pot on future streets. However if villain is shoving his entire range when the flush misses on the turn giving us a bad price on the turn we were actually just burning money on the flop since we probably never actually had the price to realize just 1 street of equity.

This is a fair point, but in no way contradictory of what I said above.

You may have assigned too much equity for a single draw. If you face a bet on the flop that is not all-in, then it is a mistake is to count the equity for 2 draws instead of 1 when estimating your EV for that call. (This might be a spot to semi-bluff shove, btw.)

When I say you need X% equity to call a bet, that's total equity on all later decisions. This is something that is difficult to estimate, even as an intermediate level player, but it's an important idea in the long run. It's something to be mindful of, while you direct your study to more profitable areas.

Thank you for clarifying this point, though.


Afterthought:
In your example above, you'd assign villain a range, based on all previous actions you've observed. You'd examine the stack sizes and make a guess as to how much he'd bet on future streets, given various board runouts. Then you'd apply math and determine the "true" equity of your call. Obviously, this is complicated and relies on a lot of assumptions. The math is all simple algebra, but there's a lot of it to keep track of, and again... at the beginner level of poker, you have much more profitable ideas to learn about.

Originally Posted by MadMojoMonkey

This is a fair point, but in no way contradictory of what I said above.

You may have assigned too much equity for a single draw. If you face a bet on the flop that is not all-in, then it is a mistake is to count the equity for 2 draws instead of 1 when estimating your EV for that call. (This might be a spot to semi-bluff shove, btw.)

When I say you need X% equity to call a bet, that's total equity on all later decisions. This is something that is difficult to estimate, even as an intermediate level player, but it's an important idea in the long run. It's something to be mindful of, while you direct your study to more profitable areas.

Thank you for clarifying this point, though.


Afterthought:
In your example above, you'd assign villain a range, based on all previous actions you've observed. You'd examine the stack sizes and make a guess as to how much he'd bet on future streets, given various board runouts. Then you'd apply math and determine the "true" equity of your call. Obviously, this is complicated and relies on a lot of assumptions. The math is all simple algebra, but there's a lot of it to keep track of, and again... at the beginner level of poker, you have much more profitable ideas to learn about.

You're misusing the term equity as relates to poker. Equity is explicitly used as the percentage of the pot that would belong to a player on average via a hand vs hand comparison if the cards were rolled out in this game.

Sorry. I do that. It's a pain in my ass, too.

Originally Posted by MadMojoMonkey

Sorry. I do that. It's a pain in my ass, too.

It's a dumb convention for poker imo.