PLO HH Review : One Hand, Four Streets

Posted by

You’re watching:

PLO HH Review : One Hand, Four Streets

user avatar

Zachary Freeman

Elite Pro

Video Player is loading.
Current Time 0:00
Loaded: 0%
Duration -:-
Remaining Time 0:00
  • descriptions off, selected

Resume Video

Start from Beginning

Watch Video

Replay Video

10

You’re watching:

PLO HH Review : One Hand, Four Streets

user avatar

Zachary Freeman

POSTED Oct 22, 2014

Zach analyzes a $2/$4 PLO hand street-by-street using PokerJuice in order to sort out the most profitable option on every street.

29 Comments

Loading 29 Comments...

trondeez 10 years, 4 months ago

On the turn, you got rid of Villains AA calling combos by saying QQ-KK instead of QQ-AA. Any reason for this or just an oversight?

Lex Kukkaruukku 10 years, 4 months ago

I like the video concept, but i am little bit disappointed how we constructed villains range. We forgot villains flop raise bluff range which leads villain to be folding more in flop and turn than in reality. I think its safe to assume villain is raising some J+pp under 7,low double pairs and some backdoor stuff. How much villain should be bluff raising on the flop? After flop bet-call, should villain have turn shoving range?

Zachary Freeman 10 years, 4 months ago

strong textThis is a good point. I will see how this affects our FE and EV. Give me a few days as I am out of town right now.
I would add that a lot of villains flop bluffing range would be hands that are in his flop calling range for the current sim. Specifically some of the JX and weaker draws. Accordingly the FE on flop may not change that much but the turn FE might change a lot. I will be interested in seeing the results. Hopefully I can work this just on PPT web based becasue I am camping this weekend and only have my ipad.

Thanks for bringing up this omission Lex

Sro238 10 years, 4 months ago

I'm not 100% convinced that c-betting flop is even best. +1 to villain should be raise bluffing us on the flop some times. In 3 bet pots on paired boards I like to cbet higher pairs as we can represent those more credibly(88-AA) and check low boards(22-77). 7 being right in the middle, I think we should probably bet it some times, and check it some times. It's a spot where if I am villain and I think the 3bettor is auto-cbetting too much will look to raise bluff with hands that dont have much showdown value but can get there like gutters with a backdoor flush draw.

If we do elect to cbet, which is probably fine as we do have a lot of raw fold equity and our hand does need protection, I think our sizing should go down a touch. This is a pretty big lockdown board, and the more lockdowny the board, the less we need to bet me thinks. 2/5 pot sounds about right.

midori 10 years, 4 months ago

Hi Zach,

Interesting video, and cool format! I have a few comments:

a) Preflop, do I understand it correctly that you implied that we can have more JJ than him? I don't think that is true, because we won't 3bet all JJxx, whereas he will raise/call with most of them.

b) On the river, 55+ contains 66 as well because that got counterfeited. A minor difference, but thought I'd point out :)

c) While it is true that we shouldn't be value betting if our equity against his calling range is less than 50%, there are certain exceptions. An example would be folding out his air that can put us into a tricky spot by bluffing when checked to. This is probably not directly applicable to this specific scenario, though.

Zachary Freeman 10 years, 4 months ago

Back home now and am going to attempt to add some content to the video and respond to comments.

First comment by Sro238,

I'm not 100% convinced that c-betting flop is even best. +1 to villain should be raise bluffing us on the flop some times. In 3 bet pots on paired boards I like to cbet higher pairs as we can represent those more credibly(88-AA) and check low boards(22-77). 7 being right in the middle, I think we should probably bet it some times, and check it some times. It's a spot where if I am villain and I think the 3bettor is auto-cbetting too much will look to raise bluff with hands that dont have much showdown value but can get there like gutters with a backdoor flush draw.
If we do elect to cbet, which is probably fine as we do have a lot of raw fold equity and our hand does need protection, I think our sizing should go down a touch. This is a pretty big lockdown board, and the more lockdowny the board, the less we need to bet me thinks. 2/5 pot sounds about right.

Upon my further investigation which I am about to do and provide below we will see if betting flop still looks best. I agree that we should have a mixed strategy of betting and checking this particular flop. What I explain in the video is why this particular hand does best betting due to its vulnerability and lack of blockers of draws. It would be incorrect to interpret me saying that I bet this hand on this flop as saying that I bet my range on this flop. Regarding sizing, first off the ev of betting vs checking is so complex in this spot that and so dependent on villains aggression and stickiness on this flop that I think adjusting the sizing from 50% pot to 40% pot will have a negligible effect. with the presence of a FD and str8 draws I prefer not betting smaller than 50% pot. I would size down on a 772r flop for example.

Hi Zach,
Interesting video, and cool format! I have a few comments:
a) Preflop, do I understand it correctly that you implied that we can have more JJ than him? I don't think that is true, because we won't 3bet all JJxx, whereas he will raise/call with most of them.
b) On the river, 55+ contains 66 as well because that got counterfeited. A minor difference, but thought I'd point out :)
c) While it is true that we shouldn't be value betting if our equity against his calling range is less than 50%, there are certain exceptions. An example would be folding out his air that can put us into a tricky spot by bluffing when checked to. This is probably not directly applicable to this specific scenario, though.

Now to respond to Midori,
1) No, I wasn't implying that we have more JJ than he does. I was instead saying that the JJ combos arent a landslide in his favor. Like on a 733 flop we have far fewer 77 combos. Whereas on J77 he has more JJ combos but we still have a considerable amount.

2)Yes.

3) Makes sense. I haven't explored these scenarios but more importantly for a exercise such as this where we are already having to make many assumptions to arrive at conclusions a outlier scenario such as the one you bring up here wont be significant enough to consider.

In general when we have a range advantage we should be doing more of the betting and when villain has a range advantage we should be doing more checking. But quantifying range advantage is tricky. Becuase its not just whose range has higher equity although thats part of it; its also who has more top end strength combos and by how much. And the importance ratio of top end strength vs equity is a on sliding scale dependent mostly on stack depth. With 1PSB equity reigns supreme but as you get deeper nuttiness is more important.
What I stated in the video and you guys seem to agree is that we should have a mixed strategy of betting and checking on this flop. What I believe is that if we are to have a betting range then this hand should be in it.
Who has the range advantage here? My guess in the video was villain but not by much. Lets look into it some more. My assumption is that our range has an equity advantage but he has more top end combos. We are in a 3b pot so stack depth isn't very high but SPR still is far over 1.

Lets look at some sim data below where we use hero's 3b range as 12%vr,$0g$ss,$1g$ss,$2g$ds,4456+$ds,4556+$ds,4566+$ds,rroo!rrr
and villains pf range as 90%vr!3%vr:

Lets notice that hero's range has a very significant equity advantage so as stacks get shallower this implies hero has a range advantage.

(Not listed in data above) KTT9 is the best 5card hand on flop 55% frequency. Yet, it only wins at SD 54% so that means that more often than not it gets outdrawn and loses. ie this hand benefits greatly from protection betting.

Our range flops best current 5card hand 72.9%

Villain flops trips 12.7% while hero flops trips 8.5%. This race goes to villain but not by much. Interestingly though if you look at boat+ on flop you will see that hero actually flops a boat+ more often than villain. This could be due to the fact that hero's range has more connectivity and the flop is somewhat connected as J77 and/or because hero's range is more PP heavy. Overall though, if you look at top end strength its basically a tie.

Lastly if you look at the winning hand type if we just ran out turn and river you can see that the most often winning hand type is just 2-pr which hero has many more better combos of.

These sims imply hero has the range advantage and by a significant margin. Accordingly we should not only be betting this hand but betting more often than I originally anticipated.

Lex smartly pointed out my mistake of not including a bluff raising range for villain and certainly doing so will lower flop FE and EV. That said, many of the weaker combos that I included for villain peeling flop with would be probable candidates for his bluff raising range, namely naked JX combos, and pair+GS.
The parameters used in the video hero had 52% FE. Flop sizing was 50% so hero shows an immediate profit betting even if he never won when called if his FE is 33.3%+. For hero's FE to be under 33.3% villain would have to be bluff raising flop 19%+ frequency with hands not already listed in the original flop calling range.
Hero is going ot have a x/f range which means his betting range is already stronger than his pf 3b range. His full pf range is going to flop trips+ 18.4%, and he won't be folding (QQ-AA):ss, etc. I wont model the specifics but it looks clear to me that betting flop will still be the best option by far.

midori 10 years, 4 months ago

Zach,

Thanks for your very detailed reply.

As for 3), I was referring to the river scenario because you were looking at our equity vs. his range and said you should have checked. While I agree with that, I wanted to point out that having <50% equity against his calling range doesn't mean we always have to check. I just realised I haven't said it was about river, and I'm sorry for being unclear. At any rate, I don't think they are too different except that ranges are less polar and equities are more dynamic on earlier streets.

That said, I completely agree with what you have said about the range advantage and breakdown on flop, and how they should affect our betting/checking decision. With that in mind, I can think of two reasons for betting this hand. Firstly, we have a range advantage on this flop, and this specific hand is, of course, a part of our overall range. Secondly, we don't have a hand that plays super well by x/c'ing, even against his bluffs. In some sense, this can be thought of as an extension to my argument above; we don't mind him folding his air, because a) his air usually has a significant amount of equity against us, and b) if we check and he decides to (semi-)bluff with those, we will have a hard time calling down. In short, we deny him of hot and cold equity AND bluffing equity by betting.

This becomes interesting when we are facing an opponent who raises a ton on paired flops. That's probably a very complicated topic, but I still think we have to bet/fold this hand and check other types of hands instead (x/f, x/c, x/r).

midori

trondeez 10 years, 4 months ago

Zach,

I have Odds Oracle with the Poker Juice macros and when I plug in Villain's preflop range, 90%!$4b3, I get a ton more combos than you did. (like 50,000 more)

Any ideas as to why this is?

Also, why do i get a different number of combos if I write 90%!$4b3, as opposed to 90%!($4b3)

Zachary Freeman 10 years, 4 months ago

Firstly,
Do you mean you have the macros saved in ppt? If so as an aside, is there a way to import macros? I'd love to have them if you don't mind sending them.

Are you looking at absolute combos or reduced combos given the flop? Could that be the culprit? Unfortunately I no longer have access to PJ so I can't go re check my work.

why do i get a different number of combos if I write 90%!$4b3, as opposed to 90%!($4b3)quote

I don't know why you are getting different results for the above. Could it be that the b3 which tells ppt that we are considering 3handed rankings applies to the 90% part as well when not using the parenthesis?
Try 90%3h!($4b3). I doubt that's it. Sorry I'm not more help.

midori 10 years, 4 months ago

Zach, you can get the ranges here:

http://help.pokerjuice.com/hc/en-us/categories/200116138-PokerJuice-Preflop-Ranges

trondeez 10 years, 4 months ago

"Are you looking at absolute combos or reduced combos given the flop?"

Yes, this was it. Thank you.

As far as the ranges, just follow the link Midori posted above. Click on the download all ranges and follow the instructions. You can then use the Poker Juice syntax in Odds Oracle to get the Poker Juice cleaned up ranges. ($FI40, $3b6i, $4b3, etc.)

Londonjonathan 10 years, 4 months ago

It may be fine to bet 1010 occasionally from a GTO perspective.
When you are OOP, mixed strategies come into play. It's false that you should have a hard line between value hands and checking hands. That is only true in position. I did the full GT analysis of betting after the third draw at deuce to seven when both players draw one, and it's impossible to be unexploitable unless you employ a mixed strategy with made 9--> Q OOP, for example; you can calculate exactly the correct frequencies for each hand for betting and checking if the ranges of both players are perfect information (which is why 2-7 is a good toy game for this - ranges are clear after early heavy action and both sides taking 1 card on second and third draws)

MorePower 10 years, 4 months ago

Nice video,
about the flop bluff raising range. I think you forgot to put in his value raising range as well when you calculated how often he is folding. Maybe you did on paper, but if you put in the combos like that you get the combos he is continuing with, with a call. And that does not mean that he is folding the rest of it. So folding = preflop range - calling rang - value raising range - bluffing range

Zachary Freeman 10 years, 4 months ago

You are correct and I apologize for having multiple mistakes in the video. I will try to not have that happen again. I will rectify it below.

The value raising range is
90%!$4b3:7!(j7,77,jj)
which is another 22906 combos or 12% of his PF range.

As I mentioned when discussing villains bluffing range I think most of the bluffing range would be comprised of hands that were included in the calling range already; namely, naked JX and hands like 889, 998, 88T etc.
Lets estimate his bluffing range is
90%!$4b3:(889,88t,998,99t,j!(89,t8,t9))!(ss,7,qq-aa,89,9t,t8,jj,cc)
15983 combos or 9% of PF range

Pulling those hands out of the calling range makes his calling range now
88223 - 15983 = 72240 combos or 39% of PF Range

Results:
Calls 39%
Value Raises 12%
Bluff Raises 9%
Folds 40%

Adding in his value raising range and assigning a bluff range brings our FE down to 40%. We have a +ev bet on flop still even if we realize 0 equity.

DirtyD 10 years, 4 months ago

I think the implications of the flop and turn analysis you do in the video are a bit murkier than you made them out to be initially (although the subsequent discussion has led to a lot of really good points). Basically you break down his range into a few groups and find that if he plays more or less honest-straightforward, you have a profitable bet. But on a paired board it's not surprising that you can get away with a bet if he plays honest. If we turned the tables and did a similar analysis from his perspective against your putative strategy, I suspect we'd find that he has a profitable raise against your cbet with a4c. So it seems to me that this type of analysis is just the beginning of the story for developing a gameplan against a tough opponent.

Zachary Freeman 10 years, 4 months ago

I never pointed this out but to clarify the intention of the video was not to construct an approximation to a GTO strategy for this board. Nor was it to play this particular hand optimally vs a villain playing ~GTO. I find it more relevant to discuss how to play hands and ranges vs average villains' strategies. That's how I play poker. I assume with even no reads that this player is under attacking these boards to some degree because that is my experience. We can then adjust if that universal read is altered. So I say that because if I am exploitable to a any-4 flop raise that doesn't concern me given the average villain does not over-bluff these boards.

I do regret having the errors I made in the video because it detracts from the content which was intended to be a consideration of our best options across each street with this hand. Any time we look at multiple streets with a dynamic board giving players full reign with calling, folding, raising, checking, betting the tree and analysis become VERY complex and to attain anything useful we need to simplify.

I believe the raising, calling folding ranges I prescribed above in the post do approximate a fairly balanced strategy though and likely a more balanced and bluffy strategy than most mediocre regs employ.

You state that villain probably has a profitable raise with A4c. I assume you mean A4cc? Either way you are saying he has a profitable raise with a low equity hand. That might be the case as mentioned above which Im okay with. But I'm not even convinced that's the case.

Let me do some really rough estimations just to gauge how exploitable our bet is. Our full PF range has trips+ 19% and assuming we c/f our weakest hands on flop than once we bet we will have trips+ 25%-30% given we should be betting trips+ more than checking them.
If our 3b% is 12% and we 3b all AA and half of KK, then we will have KK-AA 31% of our full PF range. For simplification lets just assume we split our decision to check or bet those hands so we will have KK-AA when we bet ~31% (not considering A4 blocking AA)
There are other hands (big draws, QQ+friends, etc) we wont fold and some we will re-bluff but just those numbers above have us calling a flop raise
56%-61% of the time.
If pot is 1, and we cbet 0.5, he raises to 1.5 he is risking 1.5 to win 1.5. SO he has a profitable raise with any-4 if we fold >50%. Taking a quick look it seems we will be calling enough that he can't exploit our bet.

MorePower 10 years, 4 months ago

i think a4c, is supposed to mean any 4 with clubs (32cc+).
This well could be the case that this is +ev, but like you said you evaluate your play to this specific villain and the ranges you put him on. I think there is absolut no need for appliing a gto strategy here.
In Jeans last video he exactly did the same on the A77cc flop hand and pure pointed out how he thinks the average villain is playing his range against him and that he profitable can exploit this line.

DirtyD 10 years, 4 months ago

Sorry it took me awhile to respond. By a4c I meant "any four cards," but it seems I just confused everyone. Oops.

In any case, I agree that preparing to beat a typical villain's strategy is more important than beating a theoretical GTO villain. The range you assigned villain for continuing on the flop didn't immediately strike me as nitty, either -- no obvious continuing hands left out. I think that's a good reminder that in villain's shoes, against a tough opponent, we should consider mixing in some bluff raises.

Going over your numbers, I don't see how you get trips+ for us 19%. A random hand flops trips about 17%, and our 3bet range should be more high-card heavy and therefore make trip 7s somewhat less often. I do, however, agree with your overall conclusion that we can continue with enough hands to deny villain an automatically profitable bluff raise. For example, if we continue with AA, KK, flush draws, and T98 we could be folding as little as 1/3 of our range. I think I underestimated how much heat our range can take here.

Are you auto-peeling all AA and KK to a raise on this flop? I'm not sure I am.

Zachary Freeman 10 years, 4 months ago

I got we have trips 19% from the Sim above on Odds Oracle. It shows boat+ 8.83% and trips 10%. Vs a random opponent with no read I'd probably over fold a bit and b/f naked KK, maybe AA.

Panthea 10 years, 3 months ago

Thank you Zach for your efforts.
Even if you made some mistakes, you did over compensated those with your responses here in the forum.
If you are continuing this as a series than I would like to see how a GTO strategy would roughly look like for cbetting and defending vs cbets and making some analysis about how to change the strategy (which hands to put in which range) based on villain defending too often (a-calls too often b-raises too often) or defending not enough.
It might be also cool to use another approach and some how go from river backwards:
How often villain has to defend => which hands he should call with, than analyzing how his turn calling range should look like etc...

Be the first to add a comment

You must upgrade your account to leave a comment.

Runitonce.com uses cookies to give you the best experience. Learn more about our Cookie Policy