Danny goes over some basic features of StoxEV, including how to construct game trees, how the game trees flow, some important settings, and the intricacies of making sure the game trees you build work.
It was pretty hard to make these videos just because I am talking to myself and have no idea how savvy each person who is watching the videos is at using StoxEV. So I basically tried to explain things in a way that I thought could make sense to a complete noob at StoxEV. (whether or not I succeeded at that, you all will decide)
That said I am thinking people will come away with a fair amount of questions. hopefully some of those are answered in part 2 and 3 of the video which I just finished recording a couple hours ago. But feel free to ask whatever you'd like and if its not something that is covered in part 2 or 3 I will answer as best I can.
I honestly felt after making the 3 videos, that I could have stretched it out to a 10 part series. I really was able to just go over the basics of all the different features because of time constraints. Hope people get some good info from this series!
Great stuff. I was looking to buy the software to model opponents game. It would be great to see how CREV could be used to review a session to model the ranges of a particular opponent based on reads and showdowns, stats vs. ours and see how we performed (I am thinking HU).
Given that if we could adapt our ranges to better exploit theirs and in an ideal world to re-match and see how we perform then.
I guess I am looking for some practical insight on how to use this tool but with some context rather exploring the possibilities of the software.
A second area of interest is on how to construct profitable ranges for specific situations. For example in HUSNGs if I dont want to play push fold between 8-25bb how can I use CREV to construct an SB strategy by evaluating the EV of various sub-ranges and betting lines vs. a set type of opponent.
A final area of interest is how can our model be evaluated vs. various types flop textures. Say I constructed a 3/5 BB HU betting range 100-150bb deep. How can we evaluate this using CREV against various types of flops e.g. XXxr or xxx/2flush, low paired rainbow boards etc.
Have you done much work with deep stack game trees? I think some seemingly simple flop spots may provide for some surprising results due to the amount of flexibility the player IP has, and the inflection points most players have when sitting 250bb+ deep.
Great stuff. I was looking to buy the software to model opponents game. It would be great to see how CREV could be used to review a session to model the ranges of a particular opponent based on reads and showdowns, stats vs. ours and see how we performed (I am thinking HU).
Given that if we could adapt our ranges to better exploit theirs and in an ideal world to re-match and see how we perform then.
I guess I am looking for some practical insight on how to use this tool but with some context rather exploring the possibilities of the software.
Over the next couple days/weeks I will be putting together videos that incorporate stoxev analysis into real play. So for example, breaking down a forum hand in stoxev for a leggo lite, or taking a break during a leakfinder vid to break down a hand in the context of the reads/history we have developed so far in the match. Hopefully that will satisfy your request here
A second area of interest is on how to construct profitable ranges for specific situations. For example in HUSNGs if I dont want to play push fold between 8-25bb how can I use CREV to construct an SB strategy by evaluating the EV of various sub-ranges and betting lines vs. a set type of opponent.
I have probably played less than 50 HU sngs in my life but I think I could potentially do a video on this. I might not be able to show you the best strategy in this situation, but I think I could show how to arrive at that strategy (basically meaning that my assumptions about the ohter player's strategy might be a fair bit off since I dont play HU SNGs). That said, I would probably only do a HU SNG segment/video if there were more people asking for it or leggo wanted me to do it (sorry :/ ).
A final area of interest is how can our model be evaluated vs. various types flop textures. Say I constructed a 3/5 BB HU betting range 100-150bb deep. How can we evaluate this using CREV against various types of flops e.g. XXxr or xxx/2flush, low paired rainbow boards etc.
Well you could input a specific flop and analyze vs that. If you had some type of postflop actions, you could input a postflop condition that used the board texture feature to define the type of board that comes out (this is actually something I feel I did not cover well enough in this 3 part series, so I hope to include how to do it in a future video). However, I am not 100% sure I totally understand what you are asking so if that did not answer your question, then let me know a little bit more about what you meant and Ill try to answer that better if I can
Anyway looking fwd to the rest of the series.
d2themfi,
Have you done much work with deep stack game trees? I think some seemingly simple flop spots may provide for some surprising results due to the amount of flexibility the player IP has, and the inflection points most players have when sitting 250bb+ deep.
I have not done a ton of deepstack gametrees/sims. This is basically because such a high% of my play is 100bb (I think this is true for most players). I also try not to play super deep with people unless I think I have a nice edge in some area, which implies I already know how to exploit them atleast in a general macroscopic sense. But i do agree there is probably a lot of room for exploration in this area with a program like stoxev. Although you may end up dealing with extremely large game trees that are fairly confusing. Hope you liked the video
Great series idea by Leggo and definitely the right person for it.
D2, I would not focus so much on the technical aspects of the software. While the software might seem intimidating to beginners and the learning curve is higher than usual, all the necessary technical features and functionalities are well covered by Scylla in his own technical videos. Therefore, all the technical information is already out there for those that are not lazy to look for it and absorb it. On the other hand, the actual application of the software to away from the tables work is not covered by any of the training sites and many people fail to see or realize the actual benefits of the software. Thus, I hope that Leggo pursues the direction of doing Leggo Lites or full videos on forum hands with STOXEV as this will bring the quality of analysis and discussion to another level, while at the same time helping its forum members a ton. Also, specifically for HU, adjustments, balance and exploitation are great poker theory topics to be addressed in specific or more general scenarios using STOXEV.
I thought that StoxEV=CardrunnersEV?
The former one is gone, so why do you call it here stoxEV?
For me, I called it stoxev for about 2 years before it was rebranded, so I am just used to calling it that. Its also quicker/easier off the tongue.
Great series idea by Leggo and definitely the right person for it.
D2, I would not focus so much on the technical aspects of the software. While the software might seem intimidating to beginners and the learning curve is higher than usual, all the necessary technical features and functionalities are well covered by Scylla in his own technical videos. Therefore, all the technical information is already out there for those that are not lazy to look for it and absorb it. On the other hand, the actual application of the software to away from the tables work is not covered by any of the training sites and many people fail to see or realize the actual benefits of the software. Thus, I hope that Leggo pursues the direction of doing Leggo Lites or full videos on forum hands with STOXEV as this will bring the quality of analysis and discussion to another level, while at the same time helping its forum members a ton. Also, specifically for HU, adjustments, balance and exploitation are great poker theory topics to be addressed in specific or more general scenarios using STOXEV.
I fully understand the need for the application of it to actual play. This is coming from me very soon, I promise!! However, I dont think it would be right to start right off the bat with that without giving a proper introduction to the software. Scylla has put out some technical videos, but virtually everyone I've either talked to about using the program or students I've coached on it, did not like those videos. He obviously knows the software inside and out, and therefore isnt familiar with how it is to approach the software from a complete noob standpoint. Im talking about people who had never heard of a game tree before, that sort of thing. My goal with this vid series is to make it so those people can familiarize themselves with StoxEV so that it is not so intimidating (this is the word Ive heard most when talking to people about using it).
Even for myself, when I first downloaded his software in mid/late 2008, I tried it out and then put it away for about 3 months because I could not comprehend how to use it properly. Then when I started reading up on game theory I came back to it and it made more sense. And even then, it stil took me about 6 months of close to daily use to where I thought I was 100% comfortable with the interface and everything else. So coming from that perspective, I thought there was a need for a "complete noobs guide to StoxEV" series which is essentially what this is. In the coming weeks I will be incorporating StoxEV analysis into most of my videos in one way or another, and the application of the software to real poker scenarios will be done
just recorded a 50 minute video doing as full analysis as I could fit in of a forum hand that was posted a couple months ago. I think it will be exactly what you guys who are a bit more advanced with the program are looking for
edit: no idea when it will "air" that is totally up to the leggo higher ups
just recorded a 50 minute video doing as full analysis as I could fit in of a forum hand that was posted a couple months ago. I think it will be exactly what you guys who are a bit more advanced with the program are looking for
edit: no idea when it will "air" that is totally up to the leggo higher ups
Awesome, I hope you get the green light to make more of these vids. I really like this new directed learning direction Leggo is going and STOXEV analysis of hands and strategies is the ultimate way to tackle the topics. Really looking forward this stuff. Matt speed up the procedures so this idea comes to life soon with plenty of vids!
I have recently bought CREV and I have found your tutorial very helpful - thx for it:)
One question:
When there is a flop involved, then we haven`t to ever increase under run_settings the number os simulations b/c then the math-method can be done and this will show anyways accurate results - is this true?
But what sense does it make then to be able to run both types when there is a flop (math-methode and monte carlo -> you can merge it under "math engine"?
I have recently bought CREV and I have found your tutorial very helpful - thx for it:)
One question:
When there is a flop involved, then we haven`t to ever increase under run_settings the number os simulations b/c then the math-method can be done and this will show anyways accurate results - is this true?
True
But what sense does it make then to be able to run both types when there is a flop (math-methode and monte carlo -> you can merge it under "math engine"?
I mean, idk I have never really thought about it. If you use the math engine, and leave the run settings on default value, the monte carlo takes like an extra second. I also seem to remember the monte carlo doing something extra in these cases, but Idk that for sure
I mean, idk I have never really thought about it. If you use the math engine, and leave the run settings on default value, the monte carlo takes like an extra second. I also seem to remember the monte carlo doing something extra in these cases, but Idk that for sure
