Toy Gaming (part 1)
Posted by Sauce123
Video Player is loading.
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Resume Video
Start from Beginning
Watch Video
10
Loading 44 Comments...
Like in the dark!
Sauce Theory Optimal
i miss STO, long time no see, where toygaming part 5?
Good coverage of the basics of GTO Sauce :) You be take over Phil Galfond's title as the best poker teacher ever by the end of the series. I sincerely hope that the comments section doesn't turn into a debate about the terminology of GTO (however, don't expect your vid to solve qualms about the concept as experience shows that some people will never understand regardless of how sound the explanation)
Nice vid, no mistakes so far. But I'll keep watching!
:)
Superman makes mistakes?
just what i wanted from you ben, thanks, youda best!
im only 15min into the video yet but just thought of something and want to ask before i forget. So from what i heard/think to know thus far, lets say in a hu game, there can be several GTO strategies right? I guess that means strategies with different frequencies and/or betsizes for different situations but that all in itself are balanced so that they are unexploitable.
So again lets say we would play hu and would know all possible GTO strategies and want to play one of them, while our opponent doesnt know GTO. Even though we are then playing GTO, could we not choose a strategy that we think tilts this specific opponent a bit more, for example we figure out this opponent reacts emotional to overbets on certain streets and gets tilted easier, so we would choose a GTO strategy that has alot of overbets in it. So my question is, when there are several GTO strategies, do we still have room to increase our ev vs a certain opponent by choosing the one that he can handle the worst emotionally and will lead him to make more mistakes than vs our other GTO strategies?
Yes under the conditions that:
1. There is in fact more than 1 GTO solution to HUNLHE
2. You know both GTO solutions
3. The opponent is not playing GTO
One of the GTO strategies will perform better than the other.
b) If we did, it would be trivial to exploit our opponents provided we had any sort of inference on their deviations
since you plan to cover more material from MOP can you share the chapters you plan to discuss so i can do a little reading in advance - thanks
Mike- I'm not quite sure yet, I haven't written a syllabus :p
ben i know that we dont know GTO, my question was just hypothetical thought that was clear sorry. Just want to know that IF we knew GTO, if we then still could increase ev vs a non GTO playing opponent by chosing the GTO strategy that keeps him off balance the most but yea you answered that with b)
ben, i know weve never met, but ide go gay for you. that is all.
Great first vid on the basics - thank you for it!
One general question which I spotted and has not rly to do with the topic:
When you work personally with CREV, don`t you set up in the options stuff like %-Rake, Rakecap and Rakeback,
so that the EV-results will get even a more correct picture?
Paying a 100 bucks a month for an elite sauce video to watch this is seriously dissappointing!
Derek,
I'm sorry you didn't like the video. If you can tell me why not, I might be able to do something differently next time which you'll like more. If you just chime in to say you're disappointed, that doesn't do much other than make me feel shitty. So if you don't want to offer constructive criticism, vote with your money plz.
I really enjoyed this video Ben.
As stated, there is a lot of information and misinformation floating around the boards about GTO and I'm glad you're making videos to clear it up.
Looking forward to future concept videos, and would really enjoy some sort of vid about utilizing CREV for us Noobs
Please always use headset when you record , sound quality is so much better.
I wasn't trying to offend you in anyway, I love your videos but was disappointed with this one as it was more of a beginners video, you are an elite video maker which means elite videos? I pay my money to watch you and learn, When it's a beginner one it feels like money lost. Maybe I could suggest if you make a beginners video you can do 3 vids a month rather than 2? For your information I went through a stale part in my poker playing and you changed it back around for me after watching your vids so I'm grateful for that.
Thanks for explaining. I agree with you, for people who are already familiar with game theory and poker, or MoP, then this is a beginners video. So much so that an expert in these topics is going to be bored, that's why I said in the video something to the effect of "if you're already familiar with GTO, skip this one."
I also agree it's my role here at RIO to make expert level videos. The issue is that being expert in GTO poker, and being an expert level (i.e., very skilled) poker player are different things. For example, lots of players beating nosebleeds would learn from this video, and lots of players beating $.50/$1 wouldn't.
As a teacher here my goal is to bring my viewers (who are probably mostly expert or advanced players since they're paying $100/mo) up to speed with important concepts in nlhe and poker theory. My view is that if someone doesn't know anything about GTO poker in 2013, they're missing the boat, and it's my job to explain it to them. Furthermore, it tilts the hell out of me when I'm posting on the forums and people haven't the faintest idea about basic concepts like the exploitative/optimal strategy distinction, and because of that, we end up talking past each other more than I'd like. I want to fix these things.
Since no one has yet made a good video series on RIO discussing GTO poker, I felt like I had to make one to get the word out. As the initial video in this series, I had to lay a bit of simple groundwork before getting into more applicable content, and I tried to do that as quickly as possible. I don't claim to have done the best job, but my hope is that this vid will forestall a lot of misunderstandings later. As things continue, we'll get into some content which is more advanced, but I gotta tell you guys up front, this series isn't going to be primarily cutting edge content. I'm talking about concepts from MoP which are 5 years old, and which I think are criminally overlooked by pros because they take a bit of work to understand, but that doesn't change the fact that I'm primarily explaining existing work instead of presenting work of my own.
So, what I can do for you is promise to release only one Toy Gaming vid each month. If it doesn't teach you much, then, congrats, your consolation prize is that you're already in the top few % of elite members on this site in your understanding of GTO poker.
"Furthermore, it tilts the hell out of me when I'm posting on the forums and people haven't the faintest idea about basic concepts like the exploitative/optimal strategy distinction, and because of that, we end up talking past each other more than I'd like. I want to fix these things."
One thing to note is that the poker community seems to have adapted a terminology in game theory that is somewhat contradictory to what's used in other branches of game theory and mathematics in general. I'm confused as to why this happened, since I think the term GTO is from MoP originally, and the authors certainly have a clue or two about mathematics.
It confuses me to no end though. The concepts are clear, but the way the terminology is used in the poker community is confusing. If you want to use this series to clarify and add value in that area, that would be awesome. This first episode basically just repeated stuff that is already out there. I think you can do better.
I guess sometimes it would be useful to unlearn previously learned concepts (the Yoda way) but the brain isn't really well adapted for forgetting things.
I could go for some toy gaming every week personally! Great video Ben.
The audio quality is terrible. When I pay premium price, I expect premium production quality. The content is good though.
"So, what I can do for you is promise to release only one Toy Gaming vid each month. If it doesn't teach you much, then, congrats, your consolation prize is that you're already in the top few % of elite members on this site in your understanding of GTO poker."
Please consider doing more than just one Toy Games vid per month. I think the people who want to see less of it is in minority.
Please also consider taking it to a slightly more advanced level. Even for an introduction video this was a wee bit basic. Even for someone like me who's a total newbie in all things GTO. I'm not expecting the latest and greatest of GTO in this series, but rehashing stuff from MoP (that came out in 2006) has been done so many times before.
If you're going with the MoP route, at least consider defining and solving the AKQJ game :)
Great video!
I was messing around with a similar toy game a couple months ago which gave some interesting results:
I extended the case study in the video to a case where the OP player has the option to either call or fold vs our bet instead always calling our bet. And solved our GTO strategy as IP player with AA and 55. So, what is our GTO betting range?
Assumptions: Pot is 1, IP can bet 0,50$ or check. OP can call or fold with KK on 22334.
So, two example scenarios:
(1) We bet all our AA and check all our 55. The optimum strategy for the OP player is to simply fold to all our bets, which has an EV of -1. But wins vs all our 55, which has an EV of 1. Thus, this strategy would result in 0EV overall
(2) We bet all our hands. The OP player can now always call, which also has 0EV overall.
Now, imagine the case we are betting all our AA combos and 1 combo 55, and check the remaining 55 combos. OP can not call our bet since it will lose to many times vs AA, so it has to fold to all our raises and we will turn a profit. (In the case of 100 hands, 50 times AA, 50 times 55, we will make a profit of 50 + 1 - 49 = 2.
Thus, this indicates that there is an optimum between scenario 1 and 2:
My initial guess for the optimum was all betting all AA combos and 50% of the 55 combos,
1. If OP always calls, we win 100 -2*25 -25 = 25.
2. If OP always fold, we win 50 +25 - 25 = 50.
So OP will always decide to call to minimize the losses and the system is not in NE.
However, writing out the equations results in a different optimum:
If we bet all AA combos and 1/3 of the 55 combos and check the remaining 2/3 combos:
1. If OP always calls, we win 100 -2*16.7 - 33.3 = 33.3.
2. If OP always fold, we win 50 + 16,7 - 33.3 = 33.3.
In other words, neither player can improve there EV by changes strategy, thus it is in NE and we get maxEV -> GTO.
So, we can conclude that a value/bluff ratio of 3:1 is GTO is this scenario. Interesting to notice is that the OP player gets 3:1 odds to calls. Which is exactly the same ratio as our optimal value/bluff ratio! This is not a coincidence. If we bet 1 into a pot of 1, the OP player gets 2:1 on a call, and working out the math results in a GTO value/bluff ratio of 2:1 also.
Thus, if we translate this to a river situation in real poker and we want to decide what our optimal river bet size should be, we should determine how many value combos comparted to bluff combos we have in our range and based on that determine our betsize accordingly.
Hey teunuss. I stopped reading after this paragraph: "(1) We bet all our AA and check all our 55. The optimum strategy for the OP player is to simply fold to all our bets, which has an EV of -1. But wins vs all our 55, which has an EV of 1. Thus, this strategy would result in 0EV overall"
The EV of a fold is always 0. Thus, the EV of OOP strategy in that example would be 0.5 (half the time we win nothing, half the time we win 1 pot = we win half the pot on average)
Hi Ben. A few posts up you say that if we knew GTO poker then 'exploiting our opponents would be trivial' why is this the case when our goal is to be maximally exploitive
"If one player won't or can't adjust then the NE is the maximally exploitative strategy."
Basically, when you are playing a fish who doesn't adjust then a GTO strategy is one that maximally exploits the fish! Cool stuff, loved the video! :D
That's false. If I said that, there was something in the context that made it not mean what it looks like. The primary time the GTO strategy is MES is when our opponent(s) are playing the GTO strategy as well. There might be other times or situations where GTO play is also MES, but considered as a whole, most deviations from GTO that our opponents make are exploited for their max EV by a non GTO strategy.
Thanks for the clarification Ben! Again, great video! I just finished watching Part 2 and looking forward to the next parts.
This explanation make me understand well i think
Hey Ben,
Have you found studying poker bots helpful? They seem pretty terrible but I'm open to the idea.
Those who can, do. Those who can't, teach. Ben, you can both do and teach. Kudos!
Although it was basic I really enjoyed this video. To have good GTO content laid out in this format is great for me. I will keep watching this series! Thanks Sauce
Why is checking 22 ev less than betting?
I agree that at the end (min: 37.00) it's a little bit confusing when you say that checking is less than betting. I suppose you mean checking loses less money than betting, correct? I thought first it was just a typo, but you say this out loud so I'm not quite sure.
you're The Sauce, you're The Sauce, you're The Sauce. You da man, Sulsky
umm.. your voice is not clear due to lots of echo. long video and hard to keep focusing
Great video, thank you Sulsky
Very begginer oriented video on GTO and toy game, but I liked it.
I suggest the book Play Optimal Poker by Andrew Brokos if you are interested in this topic explained in an easy way to understand.
Be the first to add a comment
You must upgrade your account to leave a comment.