Title: 400nl river line Post by: pleno1 on July 28, 2012, 08:20:56 PM what does my range look like?
what bluffs do I have that take this line? whats bottom of my value range? what will his calling range look like? http://weaktight.com/4880029 Title: Re: 400nl river line Post by: GreekStein on July 28, 2012, 08:22:30 PM wtf
Title: Re: 400nl river line Post by: pleno1 on July 28, 2012, 08:32:37 PM fixed.
Title: Re: 400nl river line Post by: pleno1 on July 28, 2012, 09:39:13 PM Is it just an auto jam with our whole range?
Title: Re: 400nl river line Post by: skolsuper on July 28, 2012, 10:06:27 PM You range looks like houses or air, and vs stats like yours even with only 50-100 hands on you I would be calling with my entire river value betting range (overpairs, straights and flushes). No, it is not an auto-jam with your whole range, what makes you say that?
Title: Re: 400nl river line Post by: pleno1 on July 28, 2012, 10:36:10 PM You range looks like houses or air, and vs stats like yours even with only 50-100 hands on you I would be calling with my entire river value betting range (overpairs, straights and flushes). No, it is not an auto-jam with your whole range, what makes you say that? well we never have air by the time we get to the river but never have a great showdown hand, we either have a bad showdown hand that would turn into a good bluff or top of our range such as big flushes and houses that we would jam. Title: Re: 400nl river line Post by: skolsuper on July 28, 2012, 10:39:49 PM You range looks like houses or air, and vs stats like yours even with only 50-100 hands on you I would be calling with my entire river value betting range (overpairs, straights and flushes). No, it is not an auto-jam with your whole range, what makes you say that? well we never have air by the time we get to the river but never have a great showdown hand, we either have a bad showdown hand that would turn into a personally I would call and expect to see this a lot Title: Re: 400nl river line Post by: pleno1 on July 28, 2012, 10:41:58 PM yeh, that is problem.
what would you say our calling range on riv should look like? Title: Re: 400nl river line Post by: skolsuper on July 28, 2012, 10:49:33 PM yeh, that is problem. what would you say our calling range on riv should look like? I would say it should look a lot like 2 pair. Doesn't especially matter which ones. It's villain dependent obv, but if you're asking for an answer from a game theory perspective, we should be rarely folding here. Title: Re: 400nl river line Post by: pleno1 on July 28, 2012, 10:55:02 PM really? i kinda disagree or at least thought in game it would be a really terrible time to bluffcatch. On this run out I think we are 3 barrel bluffed very infrequently.
btw sorry to sound condescending but you do realise its a paired board right? its really interesting to hear though because your thoughts were the polar opposite of mine and regarding "id think you'd turn hand into bluff" theres still a big bunch of value combos rather than if i did it on some really random obscure board like q449j thanks for thoughts though and definitely interested to hear what others thinks. Title: Re: 400nl river line Post by: Honeybadger on July 29, 2012, 12:40:49 AM It all depends on what you think you should be doing with the nutted part of your range on the turn. By this I mean full houses/quads and, to a (much) lesser extent, nut flushes.
If you think you should raise most or all of these on the turn then your river range as played is going to be almost entirely mid-strength hands/bluff-catchers (even the 'strong' parts of this range like lower flushes become bluff catchers given your opponent's bet-bet-bet line). The only nutted hand you'd have in your range is 99. Which means that you cannot have too many bluff raising combos on the river - you only have 3 combos of 99 to balance them out. In this case, because you have so few value-raising combos on the river, you should hardly have any bluffs either. If you are going to slowplay your fullhouses/quads and nut flushes on the turn then you have a LOT more nutted value-raising hands on the river, and thus can bluff the river more often. I randomly got interested in this and started a lengthy analysis using flopzilla. Unfortunately I also started drinking and now my head is fuzzy, so I am going to have to give up with it. I got as far as constructing your range for getting to the turn. This is roughly: JJ-22,A8s,98s,87s (i.e. a pair or better) - 61 combos AcQc-A9c,KcQc-KcTc,QcJc,QcTc,JcTc,Jc9c,Tc9c,7c6c,6c5c (i.e. a FD) - 14 combos AcQx, AcJx,AdQd,AdJd,AdTd,Ad9d,KdQd,KcQx,KdJd,KdTd,QdJd,QdTd,JdTd,Jd9d,Td9d (i.e. overs plus backdoor FD) - 24 combos AQo (without Ac - i.e. just overs with no BDFD) - 8 combos. This assumes a preflop BTN flatting range of: JJ-22,AQs-A8s,KTs+,QTs+,J9s+,T9s,98s,87s,76s,65s,AQo-AJo,KQo (this is 12.4% of hands. You might disagree with the exact construction of it, but it a pretty good theoretical 'starting point' from which to adjust based on image, history, CO's tendencies, texture of players in the blinds etc) This flop flatting range (and it assumes we are not raising any of our continuing range BTW, which I think is reasonable although I can't prove it is correct) comes to 107 combos from 157 flop combos, which is the GTO solution to prevent opponent being able to auto-cbet 12 into 27 profitably with the worst part of his range. The turned 7c reduces our 107 combos to 102 combos (we can no longer have the 7c in our hand which removes 8c7c and 7c7x). Villain bets 34 into 50. So we need to defend approximately 60% of our range to prevent him betting ATC. This means we need to defend about 61 combos out of our 102. And at this point my head started to spin a little from the booze. But I think I have seen ahead far enough to give an answer that is likely going to be correct. Basically, the next step is to decide whether we are going to have a raising range on this turn. This is going to be a MASSIVE factor in what our range looks like on the river as played, and thus in how much we can bluff. I am actually a bit gutted that I have to stop here because I was enjoying doing some in depth flopzilla work for the first time in ages. FWIW, although I have not finished the analysis, my gut feeling is that we can reasonably slowplay most of our big hands on the turn. Which means that when we get to the river, we are able to turn our opponent's flushes, even nut flushes, into pure bluff catchers. However... and this is important... our opponent has all combinations of full houses/quads in his range too. So we are not allowed to be overbet bluff raising willy-nilly. The maths of the overbet river bluff is pretty unforgiving when our opponent has an uncapped range, as is the case here. So we cannot be taking too many liberties, and Pleno's suggestion that we can perhaps bluff raise our entire river range is definitely incorrect. Sorry I can't add much more to this. Is PaintingByNumbers in the house? He doesn't post much, but he is the guy to ask on this sort of stuff (unless he has been drinking too!) Title: Re: 400nl river line Post by: Honeybadger on July 29, 2012, 01:48:41 AM But to give quick answers to Pleno's questions:
what does my range look like? Your river value-raising range looks like full houses/quads. Raising the nut flush might be a little thin IMO, given that your opponent's range is uncapped (obv we adjust from this depending on opponent's tendencies/history etc, but this is a theory question, right?). Whether you actually do have these hands in your range depends on how you play the turn (see above), but this is what it should look like to your opponent. Full houses and bluffs.what bluffs do I have that take this line? I haven't finished my analysis, as I said. But instinctively I feel that you don't need to be turning pairs into bluffs here. The reason for this is that in order to defend competently vs your opponent's turn barrel you must float the turn with some combos of Acx hands, and so you can use these combos to bluff the river with in order to balance your nutted value-raises. I'd have to do the analysis to be sure, but that's what it feels like to me right now anyway.whats bottom of my value range? I've pretty much already addressed this when I said that raising the nut flush on the river would be thin. So either the nut flush or 77 would be the bottom of your value raising range, depending on whether you think it is profitable to raise the nut flush (and I am talking theoretically here, not exploitatively).what will his calling range look like? As mentioned above, all flushes become pure bluff catchers. And he should not call with all his bluff catchers, or else you'd never bluff. So his calling range is likely nut flushes and full houses/quads.Just remember that all this is just a guess. A guess that is likely to be approximately correct, but not definitely so since I haven't done the analysis properly. Also, remember this is just theory stuff - i.e. sometimes irrelevant when actually playing, and not necessarily how you might actually play in practice (but still important/interesting to know). Title: Re: 400nl river line Post by: Honeybadger on July 29, 2012, 02:03:50 AM I have just realised something that seems extremely counter intuitive, but is very likely correct...
I said your opponent should call with his nut flushes and full houses, and fold all his other flushes. This is very likely incorrect. In fact, your opponent should probably call with his King high flushes (and perhaps his Queen high flushes) and fold his nut flushes! A sticker for the first person to correctly explain why this is the case. ;topofclass; (It's actually not that difficult if you read, and think about, the post above ^^^) Title: Re: 400nl river line Post by: cambridgealex on July 29, 2012, 05:58:28 AM cos of blockers??!
Title: Re: 400nl river line Post by: AlexMartin on July 29, 2012, 08:40:26 AM cool hand, definitely in the camp of never raising turn. Honey, awesome input; but that last statement is pretty goddamn sweeping.
Title: Re: 400nl river line Post by: pleno1 on July 29, 2012, 09:58:58 AM Hey all,
Yeh I think raising turn is quite bad in a vacuum. Pretty much because we get 3barrel bluffed very infrequently here meaning its cheaper to double float and bet small than to raise turn and even commit with our draws etc Title: Re: 400nl river line Post by: Honeybadger on July 29, 2012, 10:11:45 AM cos of blockers??! Can you explain in more detail? Otherwise, no sticker ;)Title: Re: 400nl river line Post by: Honeybadger on July 29, 2012, 01:38:03 PM Ok... I woke up and decided to finish my analysis. It's actually pretty simple, but it looks complicated. Nevertheless, I would advise anyone looking to learn how to do this sort of stuff to take the time to wade through this post slowly since I have tried to write in a pedagogical manner which teaches the methodology involved. Knowing how to do this sort of stuff is really useful, and doing a bit of it yourself helps 'train your intuition', if that makes sense.
BTW, it is very likely I have made the odd mistake in what follows. It won't be anything important, but it might be a simple calculation error or missing a combo or two out. This won't be important to the overall conclusion, and it is the method that is important not the exact details. I have checked it through briefly, but am not going through it with a fine tooth-comb. So please expect to find some small oversights/errors in this analysis, but nothing of any real thematic importance. So here goes... The turned 7c reduces our 107 combos to 102 combos (we can no longer have the 7c in our hand which removes 8c7c and 7c7x). Villain bets 34 into 50. So we need to defend approximately 60% of our range to prevent him betting ATC. This means we need to defend about 61 combos out of the 102. Actually, we need to defend a little more widely than this (especially if we are only flatting and never raising) since our opponent always has some equity with his bluffs. So perhaps make it 64 combos. [N.B. I did not mention the concept of having to defend fractionally wider when I did the flop analysis. But I did in fact add in a couple of extra combos without mentioning it, just to be complete] Let’s assume that we are not going to have a raising range on this turn. This does make sense as Pleno and Alex Martin have said. And it makes the analysis easier too lol! So now we can try to work out the 64 or so combos that we need to be continuing with on the turn. Let’s do the nice easy ones first; the full house, flushes and decent pairs: 22 , 88 , 77 = 7 combos AcQc-A9c, KcQc-KcTc, QcJc, QcTc, JcTc, Jc9c, Tc9c,6c5c = 13 combos JJ-99 (overpairs) = 18 combos 98s, A8s, 87s = 9 combos So this is an easy 47 combos right away. We only need to find around 17 more. Note how we had to include a ton of ‘light’ floats on the flop to defend widely enough (e.g. stuff like Jd9d and KcQx for nothing but overs and a backdoor flush draw), yet by the turn we hardly have to find any ‘air’ hands to float with. Even though we have defended very widely on the flop our range narrows a lot by the turn and is now comprised mostly of pretty strong hands (although hands like 98, A8 and 87 are actually really marginal here). So now let’s find around 17 additional combos to continue with. IMO it is clear that we should fold 66-33 now, since the equity of these underpairs is worse than that of the overcards with a flush draw. So we need to find 17 combos of overs. 13 of these are easy to find: AcQx, AcJx, KcQx,Td9d = 13 combos But we are now struggling to find the final 4 combos. It is up to you exactly which ones you choose from the two gutshots (JdTd/Jd9d) and the 4 combos of AQ without the Ac. Let’s arbitrarily choose the two gutters and half the combos of AQ no club. This will make a difference given that the river is a 9, but this is not an exact science. JdTd, Jd9d, AQ no club(half the combos) = 4 combos And that’s our 64 combos for continuing with on the turn. It’s probably not perfect, but it’ll do for the purposes of this analysis. It looks like this: 22 , 88 , 77 AcQc-A9c, KcQc-KcTc, QcJc, QcTc, JcTc, Jc9c, Tc9c, 6c5c JJ-99 98s, A8s, 87s AcQx, AcJx, KcQx, Td9d JdTd, Jd9d, AQ no club(half the combos) The river is the 9h and we still have 64 combos, since we did not have this card in any of our continuing range. Villain bets 78 into 118. We have to defend with ~60% of our range to prevent villain being able to bluff profitably with his entire range. This means we have to continue with 38 combos. First, we need to decide which hands we are able to value raise on this river. This is where you have to decide whether to raise with the nut flush, or just the full houses/quads. I’d be interested to hear what people think here, and whether the nut flush is a good value raise given opponent’s bet-bet-bet line and his uncapped range. The more I think about it, the more I keep changing my mind. Either way, it is going to be close... it is either going to be a thin value raise or a very nitty flat. In game, you’d decide based on things like your image and opponent’s tendencies of course. But we are talking pure theory here. I guess since it is close then it does not really matter either way. The one thing to note is that if we are NOT raising with our nut flushes then it actually means we will have to make some slightly lighter call-downs on this river. The reason for this is that we are not able to bluff quite as much if we are not raising nut flushes on the river. Looking ahead, if we are raising the NF on the river then we can likely fold all combos of TT for example. For the sake of this analysis I am going to say that we are NOT going to raise with our nut flushes, and the reason for this is that Pleno’s raise is actually a big overbet and this makes the nut flush a much thinner raise. I could well be wrong in this though, and I am happy to be convinced either way. Ok, so... 22 , 88 , 77, 99 = 10 combos These are the hands which we obviously want to be value raising. Now we need to decide on our raise size. I actually like Pleno’s overbet line since we are targeting all our opponent’s flushes for value, and these 10 combos are legitimately strong enough for an overbet. Anyway, I’m not going to get into bet-sizing here and am just going to accept that we are over-bet jamming. So now we need to work out how many bluff combos we are allowed to have in order to balance our value jams. This is nice and easy. We need to make villain indifferent to calling on the river. This would mean that his EV = 0, and thus we have theoretically ‘won the hand’ no matter what villain does and no matter whether we are bluffing or not (the pot after his river bet is 196 and if he has an EV of 0 facing our jam then our EV is the entire 196. Does this explanation make sense? Not sure I’ve explained it clearly, but moving on...). Villain is facing a 428 bet and the pot is now 624. So he has to win 40% of the time in order to be indifferent to calling (which is what we need to achieve to ‘win the hand’). This means that we need to have a 4:6 bluff to value ratio, i.e. we need to be bluffing 4 combos for every 6 value combos. Given that we have 10 value combos, we need 6.66 bluff combos to achieve this ratio. Since we can’t have 6.66 combos, either 6 or 7 will be just fine (this is not an exact science after all). What we do now is leave deciding which combos to bluff with until the end, and instead work out our calling range. The reason we need to work out the number of bluff combos we will have first (even though we are not allocating them) is so that we can work out the number of combos we must call with. We have to defend 38 combos on the river, so removing the 10 value raises and the 6 (or 7) bluff raise combos leaves us with 22 (or 21) flatting combos. It is super easy to find more than half of these, since we have 13 combos of flushes: AcQc-A9c, KcQc-KcTc, QcJc, QcTc, JcTc, Jc9c, Tc9c,6c5c = 13 combos So now we need to find another 8 or 9 flatting combos: JJ is 6 combos. Which means that we should call with around half our combos of TT and fold the other half (probably choose the combos that include the Tc since that blocks some of villain’s flush combos). This means that on the river JJ/TT become the absolute bottom of our flatting range and are now pure bluff catchers. So in practice we should usually fold JJ and TT here since most villains do not treble barrel bluff with high enough frequencies. Or against maniacs who we know barrel ‘too much’ we can widen our calling range to include all combos of TT and perhaps even a few slightly weaker made hands. But I am going off the point here; these are exploitative adjustments and this is a theory post. So, all we have to do now is to look through the rest of our range and pick out 6 or 7 combos to bluff with. And exactly how you want to contruct this range is up to you. You may pick the AcQx and AcJx type hands since these have blockers to opponent having the nut flush. Or you may instead want to use the 8x and 9x hands to bluff with since these contain blockers to villain having a full house. My initial thoughts are to use the full house blockers rather than the nut flush blockers (especially 98 and 87 which have double blocker effects). This does in fact mean that we are turning pairs into bluffs, contrary to my initial instincts in the post above which was made before I had done this analysis. But it is going to be close either way, and I could easily be persuaded that the Acx hands are the best hands to be bluffing the river with. Either way, we need to pick out either 6 or 7 combos of these hands, and BOOM we are balanced! So, our river range looks like this: Value raise: 22 , 88 , 77, 99 = 10 combos Call: AcQc-A9c, KcQc-KcTc, QcJc, QcTc, JcTc, Jc9c, Tc9c, 6c5c, JJ, TT (~3 of the combos) = ~22 combos Bluff raise: Acx, AcJx, 98s, A8s, 87s (choose 6 or 7 combos from these hands) Ok, this looks pretty complete to me. Hand solved lol! Final note regarding which combos to use as river bluffs: If we are turning pairs into bluffs on the river (i.e. full house blockers) then opponent should likely call with his nut flushes and fold his weaker flushes. If we are instead using the Acx hands to bluff with then opponent should probably fold his nut flushes and call with his next best flushes. It is pretty easy to work out exactly what villain’s calling range should look like on the river in both these circumstances. All it takes is an estimation of villain’s range to get there, a calculator, and a little bit of thinking. But I am done now, so if anyone else wants to do this for the sake of completeness (and as an interesting learning exercise) then please go for it. Not holding my breath though lol... Edited to say: Noticed one mistake already... We actually have JdTd in our range on the river for a straight. So that's one extra combo to flat with (which means we should fold more of our TT hands). As I said, it makes very little difference to the overall picture... and I am sure there are other mistakes in there too. It's not important though; this is about the method, logic and thought process. Title: Re: 400nl river line Post by: SuuPRlim on July 29, 2012, 01:43:19 PM cos of blockers??! Can you explain in more detail? Otherwise, no sticker ;)Because. The Ac makes up (in theory) the vast majority of our (as in pleno in the hand theoretically) river bluff raising range. If we have the Nut Thrush ourselves it drastically reduces the liklihood of our opponet raising the river as a bluff, if indeed he isn't going to/doesn't need to turn pairs into bluffs (as Stu suggested) Is this right, do I get a sticker :) BTW Stu, incred posts, makes me wanna just go get a job lol Title: Re: 400nl river line Post by: Honeybadger on July 29, 2012, 01:50:24 PM cos of blockers??! Can you explain in more detail? Otherwise, no sticker ;)Because. The Ac makes up (in theory) the vast majority of our (as in pleno in the hand theoretically) river bluff raising range. If we have the Nut Thrush ourselves it drastically reduces the liklihood of our opponet raising the river as a bluff, if indeed he isn't going to/doesn't need to turn pairs into bluffs (as Stu suggested) Is this right, do I get a sticker :) BTW Stu, incred posts, makes me wanna just go get a job lol BOOM! Yes this is spot on Dave. It could be termed 'reverse blockers' in fact (i.e. we don't want to have blockers in our hand). Well done pal... impressed at your logic! Sticker to follow. However as I mention in the very long post above, after finishing my analysis I am coming round to the idea of turning pairs into bluffs (i.e. full house blockers) on the river rather than the nut flush blockers. Which negates the point. But it is still an interesting theoretical concept (that we should call with K high flushes but not nut flushes) even if, in this case, it turns out to be wrong. And maybe someone can come on here and show that actually my first instinct was right and the Acx hands really are the best hands to be bluffing the river with - in which case this counter-intuitive idea becomes true. Title: Re: 400nl river line Post by: cambridgealex on July 29, 2012, 03:33:01 PM Gold star Stu. Awesome stuff.
|