This series needs more love imo. My only complaint besides the pages on pages of notes I wrote down is the fact that this wasn't done before I read NLHE:TAP about 800 times and some things still don't make sense lol. Thank you for doing this.
This episode of Mathematics of No Limit Holdem again addresses Pot odds and Implied Odds, and this time introduces Reverse Implied Odds. These concepts are illustrated using hand examples examining preflop, flop and turn play. This video also gives an overview of some valuable software to help determine our pot equity - twodimes.net and pokerstove - and finishes off with some mental shortcuts to help you figure out math at the table.
WiltOnTilt will discuss key concepts related to the mathematics of No-Limit play using Powerpoint. Begin with the basics: probability and pot odds. Then follow Wilt to more advanced arenas: implied odds and reverse implied odds, software tools and mental shortcuts for equity calculations, complex EV calculations, and an exploration of fold equity. And watch this series conclude with a discourse on the ultimate in professional poker math: hand frequencies, valuebetting, and G-bucks.
Premium Subscribers can download high-quality, DRM-free videos in multiple formats.
This series needs more love imo. My only complaint besides the pages on pages of notes I wrote down is the fact that this wasn't done before I read NLHE:TAP about 800 times and some things still don't make sense lol. Thank you for doing this.
Hi partytime, thanks for the comments. Hopefully after you watch my videos some of those concepts in NLHE:TAP will make some more sense. If not, please be sure to post any questions in these threads if I'm not explaining something well enough in the video.
Regards,
WoT
Nice video, but ... what you really know about math?
I really liked the example with the reverse implied odds. I aways struggle on how best to take that into account.
This is my first post here on DC and I guess you and this wonderful series deserve it. Thanks for this, I'm learning/refreshing/getting misunderstod concepts corrected a lot.
One thing just caught my eye though:
In the multiway pot example (we have T
9spade: in the BB on a ten handed table) where you threw us a "curveball".
First I can proudly say that I got it once you said that there are 36 ways to make overpairs that this cannot be true as we are holding a T and a 9.
So now there's only 30 ways he can make an overpair and 16 ways he can have AK, meaning he will raise 30 out of 46 times as you explain.
However I think it is even slightly worse for us as there is one AK combo (Aspade:Kspade
which he will most likely raise aswell making the hands he reises 31.
Alright, I might be nitpicking here and by no means am I critisizing your work. Like I said: great series!!
just caught my eye.
Sugar Nut
I understood the basics of poker math before, but not as clearly or completely as I do after watching this series. Now I feel confident in my hand analysis where prior to this series it was a struggle for me to "do the math" and feel certain I hadn't made a mistake. Great Video, Wilt! It has inspired me to go back and re-read NL:TAP and ToP armed with my clearer understanding of the concepts.
Great series, this is stuff I kinda knew but now I know it properly.
I don't have the "winning %" button on PT though? I have a check box "Win %" and if I check it, it displays the % chance of winning underneath the players' names.
Gotcha, it's possible you have a different version of PT than i had when making the video series. Either way as long as you can get a hold of he win % that's the important thing.
WoT
For these Math-oriented series', which are basically lecture-style with power-point slides, would it be possible to upload files other than video? Powerpoint, .txt, etc? To go along with the video, not replace it. Thanks.
For these Math-oriented series', which are basically lecture-style with power-point slides, would it be possible to upload files other than video? Powerpoint, .txt, etc? To go along with the video, not replace it. Thanks.
Covered this in a few other threads, but there's no plan to release the powerpoint file I used to make the series. Sorry guys.
WoT
Covered this in a few other threads, but there's no plan to release the powerpoint file I used to make the series. Sorry guys.
WoT
Bugger. I was hoping that the WoTs answer would be the opposite for us ducklings!
Omaha
The Reverse Impled Odds calculation is incorrect. It should be
EV(TT-AA)=(1/2*[-279.23]-61.13+86.59+86.59+86.59)/4.5
or
EV(TT-AA)=(3*[-279.23]+6*[-61.13]+18*86.59)/27
which is about EV(TT-AA)=13.12
The Reverse Impled Odds calculation is incorrect. It should be
EV(TT-AA)=(1/2*[-279.23]-61.13+86.59+86.59+86.59)/4.5
or
EV(TT-AA)=(3*[-279.23]+6*[-61.13]+18*86.59)/27
which is about EV(TT-AA)=13.12
good point!
what i am struggeling with:
on the flop you calculate with the *8 rule (2 streets). However, that assumes that you are gonna call the turn as well. You can often expect villain to bet on the turn as well. Lets say 50% of the time. Shouldn't you then use a *6 rule instead of *8 for calculating if you want to continue on the flop with a draw?
i feel like im in a college physics lecture. i dont know maybe its me but i immediately get sleepy within 3 minutes no make that 2 minutes.this is extrememly dry material which i think could use some simplification into lay mans terms . i think its to "mathy". i know u put a lot of work into it and i like the series but i cant stay awake why listening to it. thank u though i just am trying to be constructive, have u seen some of royalflush clubs videos. they keep it simple . they probably dont make as much cash as u guys at the tables but they keep it simple. and i find it to be pretty easy viewing. thye basically tell u your pant and hand odds and why they r playing the hand. thats it.
ed
didnt mean to sound ungrateful. i have an engineering degree and cant fathom the interest to follow along
If you can't follow along with it, it may not be the series for you (it's already complete and was recorded in our first season). MoNLH is one of the more popular series we have on DeucesCracked but is generally considered to a series that you get back what you put in -- it's not for everyone. Basically -- it's definitely dense, definitely dry, but will definitely pay off if you work your way through it.
All that said, you may want to check out some more live play and more basic videos if you felt it was too mathy, given that the title really says it all here.
Rob
Hi Ed, Thanks for the feedback.
If you found episodes 1-3 too complex, you'll definitely want to steer clear of episodes 4-8 as the intensity level raises a few notches.
WoT
Loved the 5,5 hand. Before I was 3-betting that hand more randomly but now I got a good foundation to understand when it's correct to 3-bet and when it's correct to call.
The Reverse Impled Odds calculation is incorrect. It should be
EV(TT-AA)=(1/2*[-279.23]-61.13+86.59+86.59+86.59)/4.5
or
EV(TT-AA)=(3*[-279.23]+6*[-61.13]+18*86.59)/27
which is about EV(TT-AA)=13.12
This makes kinda sense. Anyone can approve this?
Especially your second explanation makes me think this should be right, but I'm not sure and I don't want to start off with the wrong theory.
Thx in advance
This makes kinda sense. Anyone can approve this?
Especially your second explanation makes me think this should be right, but I'm not sure and I don't want to start off with the wrong theory.
Thx in advance
i just started watching the series and came here to post about this very mistake but agentus beat me to it.
agentus way using the combos leaves no room for mistakes like this (which can easily occur and im not bling WoT).
one could also go by the fractions using EV(TT-AA)=(1*[-279.23]+2*[-61.13]+6*86.59)/9 since TT is 1/9th of villains range, JJ is 2/9th and QQ+ is 6/9th, but thats pretty much the same thing as the combomethod.
Time Link to 01:06:06
I realize no one is ever going to check this thread, but I'm just now watching this series and I'm throwing out my equation and guess as for the final example.
The equation I used was -220 = .46x - .54x
x being his bet. 46% of the time we win what he bets, then 54% we lose so we have to figure out for what X makes the equation equal 220.
.46 - .54 = -.08
-220 = -.08x
220/.08 = 2750
If he bets more than 2750, we will be making less than the dead money in the pot offers us in the long run.
Now I'm going to watch the next eps to see if I'm right lol.
Hi Tecmo,
Your setup is incorrect because it doesn't account for us winning the $220 already in the pot when we make the best hand. I set the problem up as:
0 = 0.54 * (-x) + 0.46 * (220 + x)
Where 0 is the breakeven point (our EV is 0)
0.54 is the chance we lose
x is villain's bet size (note that we lose it 54% of the time so the first one is negative)
0.46 is the chance we win
220 is the money already in the pot
Solving the equation:
0 = -0.54x + 101.2 + 0.46x
0 = -0.08x + 101.2
0.08x = 101.2
x = 1265
We can also check this. Let's assume villain bets 1265.
54% of the time we lose our 1265 call
0.54 * (-1265) ~ -683.10 (N.B. NEGATIVE 683.10)
46% of the time we win villain's 1265 bet plus the 220 already in the pot
0.46 * (1265 + 220) ~ 683.10
So calling here when villain bets 1265 is neutral EV.
Time Link to 00:56:35
FWIW I think you made this problem WAY too hard.
We have to call 365, which I will define as one unit.
There is 725 in the pot, which is conveniently approximately two units (365*2 = 730).
Therefore when the action is on use there are ~3 units in the pot (the two already in there and the villain's bet). Again, we are looking at calling one unit (because of the way I set the problem up), so it's easy to see we're getting ~3:1 here.
I realize your point was to explore easy mental math tricks for approximating these types of situations, and I'm fine with the methods you used in general. But I also think it's important to recognize situations where we can use an even easier method.
Hi Tecmo,
Your setup is incorrect because it doesn't account for us winning the $220 already in the pot when we make the best hand. I set the problem up as:
0 = 0.54 * (-x) + 0.46 * (220 + x)
Where 0 is the breakeven point (our EV is 0)
0.54 is the chance we lose
x is villain's bet size (note that we lose it 54% of the time so the first one is negative)
0.46 is the chance we win
220 is the money already in the pot
Solving the equation:
0 = -0.54x + 101.2 + 0.46x
0 = -0.08x + 101.2
0.08x = 101.2
x = 1265
We can also check this. Let's assume villain bets 1265.
54% of the time we lose our 1265 call
0.54 * (-1265) ~ -683.10 (N.B. NEGATIVE 683.10)
46% of the time we win villain's 1265 bet plus the 220 already in the pot
0.46 * (1265 + 220) ~ 683.10
So calling here when villain bets 1265 is neutral EV.
Thanks for the help Pygmy. Will have to rewatch this later just to drill it into my head.
Hi Tecmo,
Your setup is incorrect because it doesn't account for us winning the $220 already in the pot when we make the best hand. I set the problem up as:
0 = 0.54 * (-x) + 0.46 * (220 + x)
Where 0 is the breakeven point (our EV is 0)
0.54 is the chance we lose
x is villain's bet size (note that we lose it 54% of the time so the first one is negative)
0.46 is the chance we win
220 is the money already in the pot
Solving the equation:
0 = -0.54x + 101.2 + 0.46x
0 = -0.08x + 101.2
0.08x = 101.2
x = 1265
We can also check this. Let's assume villain bets 1265.
54% of the time we lose our 1265 call
0.54 * (-1265) ~ -683.10 (N.B. NEGATIVE 683.10)
46% of the time we win villain's 1265 bet plus the 220 already in the pot
0.46 * (1265 + 220) ~ 683.10
So calling here when villain bets 1265 is neutral EV.
y = money already in pot.
Ok, so my equation was 0 = y + .46x - .54x.
Your equation was 0 = (y + x).46 - .54x.
I was close! In my equation, I am adding y as if we win that amount no matter what happens (win or lose). In your equation, it accounts for the fact that we only win y 46% of the time (when we win). That makes sense now. Is this correct?
I was close! In my equation, I am adding y as if we win that amount no matter what happens (win or lose). In your equation, it accounts for the fact that we only win y 46% of the time (when we win). That makes sense now. Is this correct?
Yeah, that's right. Basically I think of your equation as:
0 = 100% * 220 + 46% * x - 54% * x
In that setup we win the money in the pot 100% of the time.
HI WOT,
been working through this series really like it, however when looking at reverse implied odds on the hand where we have the OESD on x4T 4 facing a turn bet, you have done some calculations and in each one you say we are 18.2% to hit our OESD, but on your chart in episode one it said 8 outs on the turn is 17.4% to hit?
also when we have just 4 clean outs to the straight you say we are 9.1% to hit our hand, yet in week 1 you said 4 outs has an 8.7% chance hit by the turn?
Is there something I am missing or doing wrong?
thanks
FWIW I think you made this problem WAY too hard.
We have to call 365, which I will define as one unit.
There is 725 in the pot, which is conveniently approximately two units (365*2 = 730).
Therefore when the action is on use there are ~3 units in the pot (the two already in there and the villain's bet). Again, we are looking at calling one unit (because of the way I set the problem up), so it's easy to see we're getting ~3:1 here.
I realize your point was to explore easy mental math tricks for approximating these types of situations, and I'm fine with the methods you used in general. But I also think it's important to recognize situations where we can use an even easier method.
wow this is an awesome trick and makes it much easier to calculate the math. Thanks, a lot pygmy
Texas Donald,
Stuggled with this a bit too - you've probably now worked this out but for anyone else who gets stuck - it's because he's counting the two cards the opponent has as 'seen'.
So there are 52 in the deck, we've seen our 2 hole cards, 4 on the board and the opponent's 2 (say QQ).
52-8 = 44.
There are 8 cards which will help our hand (8 to make the OESD)and 36 unseen that won't. So 36/8, which means we're 4.5:1 to improve.
Convert to percentage: 1/5.5 = 0.181818181 or 18.2%.
Without knowing (or estimating) the opponent's holding, it's 17.4%.
Think this is right but then again I couldn't convert a ratio to percentage or back before I started this series yesterday, so it could be nonsense
Thanks Wilt btw - it's brilliant and fascinating.
maybe im confused but...
i downloaded all the videos for this series in regular and ipod versions.
in other series the ipod versions are smaller in file size than their corresponding regular versions (this makes sense.)
In this series all of my ipod versions are BIGGER that the regular versions. I double checked that i didnt mix them up. sure enough the ipod versions are of lower resolution/quality whatever but they're still bigger than the regular versions. did someone encode them wrong or something?
maybe im confused but...
i downloaded all the videos for this series in regular and ipod versions.
in other series the ipod versions are smaller in file size than their corresponding regular versions (this makes sense.)
In this series all of my ipod versions are BIGGER that the regular versions. I double checked that i didnt mix them up. sure enough the ipod versions are of lower resolution/quality whatever but they're still bigger than the regular versions. did someone encode them wrong or something?
might want to email support about this... i dont think others have had this problem but I'm not sure on the tech aspects of this stuff.
Time Link to 01:01:57
I'm way too bad at remembering math calculations WoT.Information exits my head quicker than it enters it.So please believe me WoT when i say this has been the easiest way EVER to make sure it stays in my head & for that i'm eternally grateful!! ![]()
according to 1:06
hey, I made this stuff:
1.
x-the unkown size bet
P-pot not including the bet
Eq- equity in % (%chance to win)
x=Eq*P/(1-2Eq)
2.
Same but in ratio data:
x- the unkown size bet we are looking for
P-pot not including the bet
F- (F:1 Under DOG, so if we are 3:1 dog, F=3)
x=P/(F-1)
Seems ok but good if some1 would recheck it if its all right.
great videos WiltOnTilt, really enjoying them ![]()
cheers
great series,
really a session that makes you work, but it's worth it.
1.
x-the unkown size bet
P-pot not including the bet
Eq- equity in % (%chance to win)
x=Eq*P/(1-2Eq)
johnyraku, I checked your solution and it is correct.
Here's how i did it:
Eq * (P+x) = (1-Eq) * x => Eq*P + Eq*x = x - Eq*x => Eq*P = x - Eq*x - Eq*x => Eq*P = x * (1-2Eq)
x = Eq*P / (1 - 2Eq)
Great series WoT
Time Link to 00:29:11
Your reasoning to fold here is very thorough, what i don't quite get is why you would even call in the first place given your not likely to flop a better hand with 10 9s than a flush draw. Is it because the villian UTG you described is raising here and some others would just flat SB's lead?
Your reasoning to fold here is very thorough, what i don't quite get is why you would even call in the first place given your not likely to flop a better hand with 10 9s than a flush draw. Is it because the villian UTG you described is raising here and some others would just flat SB's lead?
I wouldn't expect the small blind to lead into the field all that often.
If we thought that UTG wasn't going to play overpairs fast here, I would likely call the flop lead. Given the read of this guy fast playing overpairs, our pot odds are hurt.
I have a question for Wilt, or anyone who can answer me otherwise. When he is doing the EV calc for reverse implied odds when opponent has pocket JJ (at about 15m49sec), he mentions that 9.1% of the time (on turn) we will hit one of our four dity outs and get stacked ( easy to understand). HOWEVER, in the next part of the equation he says that 4.5 % of the time we will hit one of our good outs and win his stack... my question is....
if in an 8 out draw 4 of our outs being bad gives us 9.2 % chance... why would the other 4 good outs not also give us a 9.2 % chance to hit our good outs...
Was this just a mistake or am I seeing things wrong. seems like both ends of the draw should have the same % chance of occurrence.
Thanks!
I have a question for Wilt, or anyone who can answer me otherwise. When he is doing the EV calc for reverse implied odds when opponent has pocket JJ (at about 15m49sec), he mentions that 9.1% of the time (on turn) we will hit one of our four dity outs and get stacked ( easy to understand). HOWEVER, in the next part of the equation he says that 4.5 % of the time we will hit one of our good outs and win his stack... my question is....
if in an 8 out draw 4 of our outs being bad gives us 9.2 % chance... why would the other 4 good outs not also give us a 9.2 % chance to hit our good outs...
Was this just a mistake or am I seeing things wrong. seems like both ends of the draw should have the same % chance of occurrence.
Thanks!
there are only 2 jacks left in the deck yet there are 4 sixes. remember, he has 2 of the jacks in his hand!
I loved this one. Stuck with it like you encouraged me to. I must have watched and rewatched the reverse implied odds part for like an hour trying to figure out how to get the specific %'s and playing with Pokerstove to do the math until it finally started clicking.
It took me some time to figure out how to come up with the JJ = 4.5% until I remembered you saying last video that when you want to come up with the EV for making the hand just on the turn only do it for one street like the turn to the river. So I stoved QQ vs JJ on the same Flop and turn to get the % for hitting a J on the river and voila, 4.5%. (For those who are wondering why I'm so happy to figure this out math is a pretty weak subject for me and Wilt told me to keep at it and not give up so it's kind of exciting that I actually did manage to finally understand what comes fairly easy to a lot of ppl)
If I did it correctly (?) then if you take out QQ-AA and only leave a specific range of 10 10 and JJ for that same hand you would have a -EV of -$100.38. Is that correct?
Aha! yes! Thanks Wilt!
Time Link to 00:15:15
I am a little confused about how you plug the pocket tens ev calc into the calculator. I know if it was 20% I would multiply 1629 (-140 + -765)by .2 right? would I do .18 for 18%? + 140 x.8 correct? Also the final -269 or whatever it is seems really low to me given that 18% of the time you lose over 1k. Please help me to understand this. Thanks for making this video.
I am a little confused about how you plug the pocket tens ev calc into the calculator. I know if it was 20% I would multiply 1629 (-140 + -765)by .2 right? would I do .18 for 18%? + 140 x.8 correct? Also the final -269 or whatever it is seems really low to me given that 18% of the time you lose over 1k. Please help me to understand this. Thanks for making this video.
Got your PM saying you figured it out
Did we ever get an answer to the problem at the end of this video?
Did we ever get an answer to the problem at the end of this video?
if you watch ep 4 and 5 you should be able to do it no problem ![]()
Time Link to 00:06:01
Hey WoT,
I'm a little bit confused on the first pot odds example. What I'm confused about is where are you getting the 4.9:1 odds amount with the OESD. If you base it off of pot odds alone we only require pot odds of 2.2:1 or better to break even or to profit. Even if this was based on just seeing the turn card only, we would need 4.7:1 or better. I'm just wondering where you came up with 4.9:1...
ya weird, must have been a mistype when i made the powerpoint and then when I was talking about it I just didn't catch it. you are right about 4.7:1
ya weird, must have been a mistype when i made the powerpoint and then when I was talking about it I just didn't catch it. you are right about 4.7:1
Okay, thanks for clarifying Wilt, this series is so freaking badass and you put so much work into it, just wanted to say great job man.
Okay, thanks for clarifying Wilt, this series is so freaking badass and you put so much work into it, just wanted to say great job man.
Really appreciate it, thanks!
Hey Wilt,
with implied odds, your example has your pot odds at 340 : 140, and because your opponents cash stack is 765 deep you add that to get an implied odd of 1105 : 140. So i take it that we assume he has a chance of betting his 765 into the pot in the future, but wouldn't that mean we would definately have to call 765 to be in the pot making our implied odds 1105 : (140 + 765) ? How does implied odds actually work, i 'm a little confused. thanks !
Hey Wilt,
with implied odds, your example has your pot odds at 340 : 140, and because your opponents cash stack is 765 deep you add that to get an implied odd of 1105 : 140. So i take it that we assume he has a chance of betting his 765 into the pot in the future, but wouldn't that mean we would definately have to call 765 to be in the pot making our implied odds 1105 : (140 + 765) ? How does implied odds actually work, i 'm a little confused. thanks !
Good question. The way implied odds work is basically we assume that when we hit, we will have the winning hand a very high % of the time. Therefore, if we have to call say 140, to try to hit the next card and then have a lock on the hand, giviing us the opportunity to win that final 765. Usually implied odds are used when you think your opponent has a strong hand and you can draw to a stronger one such that when you hit you can win a lot of his remaining money. There are other factors to consider though, for instance, maybe your read was off and he doesn't have a strong hand. This would mean your implied odds go down because when you hit, you wont get paid off, so if you are calling with a hand to draw where the pot odds aren't good enough alone and you don't have good implied odds for when you hit (because he doesn't have much and you wont get paid off) then you could be making a bad call. However, it cuts both ways. If he doesn't have much, you might still be able to make a call because even though your implied odds went down, you can still profit by bluffing the next street.
Another consideration you bring up with your question is: what about the rest of the money I need to risk to win his last 765. If we draw and hit the nuts, it's not a concern (unless he has some redraws). Although what if we are drawing to a hand that might give us a second best hand often? That is the concept of reverse-implied odds. So when you hit, instead of winning a big pot, you might lose a big pot. You will see this term sometimes thrown around in situations where you might call a 3bet against a tight player with a hand like JTo where yes, you might have really good implied odds because if you hit 2pair+ you will get paid off however what happens when you (more often) flop a medium strength top pair against this tight player? If you aren't disciplined then you might lose a lot, and therefore your reverse implied odds are quite high.
Let me know if you have any follow up questions.
Aaron
Hey Aaron, awesome series.
I have a question about the mental shortcuts part, in example 2 you are talking about fudging it to make the math easier. I think in this case the fudging it makes the math harder.
When you see that someone puts out about a 1/2 potsize bet, and to be thinking in terms of UNITS, it is much easier to see real quick that if his betting 365 is the one unit, you can see he is betting 1 unit into 2 units, to make your pot a total of 3 unit's having to call just 1 unit.
Best regards, and thanks for a 4 year old series that is still super awesome.
Hey Aaron, awesome series.
I have a question about the mental shortcuts part, in example 2 you are talking about fudging it to make the math easier. I think in this case the fudging it makes the math harder.
When you see that someone puts out about a 1/2 potsize bet, and to be thinking in terms of UNITS, it is much easier to see real quick that if his betting 365 is the one unit, you can see he is betting 1 unit into 2 units, to make your pot a total of 3 unit's having to call just 1 unit.
Best regards, and thanks for a 4 year old series that is still super awesome.
Good point! Thanks for the post and I'm glad you're enjoying the series.
Wilt question here are all the calculations in the chart at the beginning off because I'm getting all different answers.
Mazein, can you post your work? I don't think the chart is wrong...
Time Link to 00:29:26
a tricky play would be to make a small raise here and then utg would have to worry about us having flopped a set so he would be inclined to just call our small raise at which point we probably have gotten the right implied odds to draw and have taken the lead in the hand. The action would be re-opened for the SB but that shouldn't be harmful unless he has a set or strong draw and re pops it, but given utg's super tight range the SB shouldn't be leading those hands...then again, he shouldn't really be leading any hands and he did so who knows...
a tricky play would be to make a small raise here and then utg would have to worry about us having flopped a set so he would be inclined to just call our small raise at which point we probably have gotten the right implied odds to draw and have taken the lead in the hand. The action would be re-opened for the SB but that shouldn't be harmful unless he has a set or strong draw and re pops it, but given utg's super tight range the SB shouldn't be leading those hands...then again, he shouldn't really be leading any hands and he did so who knows...
good point
Home → Poker Videos → Mathematics of NL Hold'em → Episode Three