kingLeon

Yeah, it seems GG is kindof a rake trap. I calculated based on your 1.8bb/100 that BBJ is adding 114bb/100 Std dev to your usual Std deviation, so for instance if Your STD dev normally would be 100bb/100 than You would get SQRT(114^2+100^2) = 151 bb/100 std dev instead, which in my opinion is quite massive addition.

At 13:55 you say that Fish under bluff river raises. I think it changes against block bets. Have You looked into it?

At 27:51 Jamming str8 after calling turn IP. I found in a database that it is underbluffed spot. Also I think in theory we usually don't have enough bluffs to use only a big sizing after calling IP and it should give us almost no EV when theoretical villian range should be full of give ups. What makes you think that you definitely overbluffing this spot?
At. 28:25 You say that in this era people have smarthands, you mean a smarthand.pro subscription or something else?

At 36:50 you fold 65s third pair in 3bP on turn. I checked it in Wizard and its +2bb call. From what You're saying looks like you do that because You believe that IP is severely underbluffing, what makes You think that way?

Its a good point that there is bias in showdowns and I was aware of that. Was thinking that it is hard to quantify and such significant equity difference is hard to get without range being weaker. Also if only info we had was frequencies than I think like You pointed out the conclusion would be to think that recreationals 4b range is really strong and I used to think that way till not so long ago. I was thinking about what You said and figured out more unbiased approach(still some flaws there though) - comaparing 4b/call ranges. I found out that for recs it is as strong as top 4.6% of hands and for regz it is as strong as top 3%. Based on that we could estimate the strength of the whole range by dividing the above numbers by call frequencies of 0.79 and 0.59 respectively and will get 5.8% for recs and 5.1% for regs. Obviously I might have been wrong somewhere or have some other flaws out there but given that it is quite unexpected result If you have preflop frequencies in mind. I think it is something worth looking into.

17:10 You say Rec 4b ranges are stronger than they should be, are You sure about that? In my recent database of NL200-1K recs 4b showdown range has 45% equity against Regs 4b showdown range, even though 4b is much more infrequent for Recs.

AT 23:19 You say we should be isoing wide which seems to be a common knowledge that I am not sure is true as I ran couple of sims of isolating recs in simple preflop. What are the reasons for wide iso? What do You mean by wide, 25%?
I include the Iso sims if anybody wants to take a look
https://drive.google.com/drive/folders/1ABmBUbtWHAEaDrqGfdFVMyGzRa6PKyHK?usp=sharing

I love your videos Nachos. I feel like You're sharing all the information with us which I really appreciate. Around 6:00 You say that 4b ranges of recreationals are underbluffed, but in my database of NL 200-1K recs 4b range has 44.5% equity versus regs 4b range. Are you sure that what You said is accurate?

That doesn't really change much, but I think 100bb std dev is a bit to big, You actually can check it in poker tracker, they have a statistic Standard Deviation (BB/100). Also you could download All in adjusted standard deviation from their website which Is usually around 5-10% lower and this is actually a number you should put into pokerdope

Hi Kevin, what is the software that you use for randomizing?

Hey man, great video. I really liked your comment on Av. roi on sharkscope. I noticed that they have systematic error in calculating it because of treating a tournament and reentry as one tournament with double buyin and as you can imagine lower ROI than your actual average. I wonder if this is a place to ask that but I was wondering what are the reasons that on GG network most of the tournaments are really late reg - down to 10bb stacks? I hope you could answer that as you work on tournaments there.

Great Video Daniel, I think when You calculate value of what is rest to play you forgot to subtract 9x44 wich is price for 9th place that was already taken from the price pool by every player which would make a small but significant difference.

I think there is less VPIP in ICM pressure spot with SB 48bb then chip EV cause chip EV solution is most likely for full 1bb ante, am I right?

Hey Daniel, thanks for the video and analysis. At 31 Opening range at 21.3% seems very tight as an optimal range at 25 bb for hijack open is around 28% with 1 bb ante. Granted the ante in the hand is 3/4 so it should go down, but on the other hand you are covering every single player behind you, so it should loosen your range beacause of the ICM tournament pressure. My question is if you still think that agressivenes of the players behind you is enough to tighten your range so much? If so how much more often you think they 3 bet you then they should be?

Hey Daniel, thanks for the video.
At 30:10 you say that Qh boards are gonna be very bad for us, what is it based on?
I am curious cause when I plug in the ranges from preflop solution with 17.5 bb we have 41.3% equity versus 30.4 i 28.3 of bb and sb. Maybe there is something changing because bb and sb are not that short or maybe it is further into the tournament and ICM considerations change something?

Dude, this question is so complex that my mind is overheating right now, but you don't have to have a big understanding of GTO to answer it, cause loose recreational is faaaar better to play against then nit reg.

With the same winrate variance is the same for single Spin and go, so more relevant question for me would be what is possible winrate in these games?

My current understanding is :
Cause bluffs from later street can be effectively treated as value on previous one.

It was definitely like that back then as I used Pio briefly in summer 2016 and then stayed with Simple Postflop. I also much prefer user interface of Simple Postflop. In my opinion most things that You can do in Simple Postflop You can do faster and better than in Pio, but in PIO there are more things You can do.

Is it possible for You to reavel where did You buy recent sample of zoom hands as from what I know it is not possible to datamine zoom hands anymore?

Hey Nick :), great work :), couple questions.
Why did You mix 100 and 200z?
Are hands in the database from this year?
If above true and You can say, how did You obtain those hands?

Hey Phil ;). I think it would be a good idea to have an easy way to report a chat abuse in the client that can be accessed within couple of seconds after someone is berating people at the table. On the other hand some system of rewards for the most chatty and positive people would also be nice.

There is one simple answer wich doesnt require more info: if y/z > x then "actual" 3b% is less than y/z and if y/z< x then "actual" 3b% is more than y/z. This effect is usually quite big for small samples.
For more info you would need more assumptions, but I really don't think you need more than what is above at the tables ;).

@Saulo
It is hard for me to believe that there are some buggs in either of two solver engines, so I would expect that either it is the case that You set up different trees as You said is possible or the nash distance of sollutions was quite big and they appeared as different.
If You could post pictures of CREV subtree wizard and PIO tree building and calculation that would clarify where the problem might have been. Also nash distances would help as well ;).

CREV Solver is quite good after recent update. In my opinion they range explorer funcionality is now nicer than that of PIO.
Also I believe it is a bit quicker than PIOSolver. I don't know How it compares to SImplePostflop though or GTORB. Main disadvantage of CREV is GUI for browsing the tree and much worse treebuilding tools. Main advantages on the other side are by far best in the business locking funcionality and price. With CREV you can do such things like reasonably locking future street actions quite quickly whereas in SimplePostflop, GTORB or PIO it is a lot of work. Also locking only one line or selected combos is way more intuitive.

I used PIO for the moment but chose to go with Simple Postflop wich is 2-3x faster in situation that I compared. From what I know Simple Postflop is at least of the similiar speed as http://jeskola.net/jesolver_beta/ wich You can buy to use with PioViewer. If you use Standalone version of Simple Postflop changing runouts is as fast as in Pio, but you don't have keyboard shortcuts that can do that.

Do not troll DDog, You're better then that ;).

I am suprised that I still had it somewhere ;):
pot is 55 cause i didn't change it from earlier simulations to 60 ;).

1. It would be interesting to see what ranges do You use as in my simulations there is 1-1.1 bb advantage of bu. Also On monotone board both players ranges have high portion of "nuts" and You have to put some nuts into each line or you will be quite exploitable by overbets or high frequency bets in general on later streets. On other board textures usually bu will have more "nut" hands and in general smaller percentage of them and also If you c-bet all your nuts on non flush brds after check some good hands will be promoted to 2 pairs and sets on every turn, but on monotone brds majority of turns will not give you very strong hands unless You check them back.
2. I see quite low x/r. frequency on this brds in this simiulations, but
3. I think in reality people are c-betting to much flushes, flush draws and top pairs on this brds that is merging their c-betting range and making it overall stronger so then x/raising becomes viable option with higher frequency, but more important is attacking their x backing range which will be way to capped. As to not reraising on low stakes there is close to 0% reraising percentage on this brds in simulations( like 2-3%), so I wouldn't look first at exploiting peoples tendencies there.

4 betting AQs would be making your 4 bet range more linear as it has 50,7% equity agianst reasonable calling range of:
100:AA,TT-77,AQs-ATs,KQs,JTs,T9s,50:JJ,66,AKs,A5s,QJs,98s,AKo.
Therefore I am calling unless the btn is 3 betting maniac - like 15%+ versus cu. I think deeper stacked like 200 bb it is viable option as 4 bet ranges will need more board coverage and playability then, but at 100bb with SPR around 2 on the flop polarizing our 4 bet range is the way to go in my opinion.