Brian Space's avatar

Brian Space

239 points

Excellent -- pleasant to watch while thoughtful and informative.

Dec. 21, 2019 | 1:59 p.m.

I see leads fairly often live in such spots, especially if its multiway.

I wonder if the odd equilibrium is always a result of constraints? (Even if one might miss the plays that lead to a more normal looking equilibrium.) Its important to know if these sorts of imbalances can occur as a natural part of equilibrium IMO or if they reflect a player missing a range imbalance relative to the flop or some other spot. I have rarely seen a Snowie spot where the ranges become some imbalanced and it was never a flop spot. It would require a rare run out to the river.

Oct. 20, 2019 | 12:51 p.m.

Are hands like the 632 flop leads when you remove the constraint ? If IP is ✔️ 100% that may indicate OOP is leading a lot. It doesn't seem like a hard spot to find to consider leading as OOP. Maybe with OOP leads, IP can bet when ✔️ to some.

Oct. 19, 2019 | 6:57 p.m.

I am not a PIO fan for various reasons and do some different things. But to use PIO here we have to run several parallel sims with different distribution of bet sizes to see if it appears with a partner bet size as a good approach right? We can't run multiple sizes and see if it used, That is ok to hunt for sizes but is not a true reflection of anything but that exact mixed strategy tested. I am not suggesting you are wrong this is just why I am not a PIO fan. It requires a lot of sims for each spot to get meaningful results. It is so important how nutty combos and good bluffs are distributed in ranges. Don't get me wrong it can be the only way to do certain things. Thank you again. Can't say enough how refreshing your videos are.

Sept. 3, 2019 | 3:06 p.m.

I think the turn spot with QT ♣️ ♣️ might be a spot where we develop an AI range and this combo may fit or mix. Always nice to bet large with a polarized range and these wide range situations short handed can be balanced in some sub ranges reasonably and are hard to exploit because the space of plays is so large. Thoughts ?

And your video style is wonderful. I appreciate the openness and thoughtful work. Thank you.

Sept. 2, 2019 | 2:32 p.m.

The first two videos are fun, insightful and thoughtful. Many presenters are actually a bit too slow -- your speed and timing are effective IMO. The sharing of your thought processes, especially concerning constructing your strategy, is a significant addition to the work. Most players here know the basics of poker strategy. The critical details, like how one even knows what ranges look like to make optimal decisions in post flop spots is rarely discussed. This is likely because most professionals guesstimate and that is hard to explain.

For example,. people jump into PIO too readily IMO and focus on a solution with too many assumptions, especially concerning the opponent. While that has some use, it is more important to obtain actionable information from the thing we can know which is our own range.

I would appreciate even more insight into your choice of bet sizes. There are a lot of profitable options. I like how you think about optimizing EV for range in a practical way and how that balances with EV for the current, vacuum spot.

Wonderful work -- thank you very much.

Aug. 5, 2019 | 11:16 a.m.

Don’t fall into the idea that bets as part of an equilibrium strategy are always classifiable as value or bluffs or anything else. Some combos can e.g. call , fold or raise for equal EV in a particular configuration. Action boards lead to such complex equilibrium with sometimes difficult to classify actions. Often they can be seen to serve a purpose later in the game tree. In any case EV for the range is the optimized parameter. All plays serve just that purpose.

I enjoy your work, thank you.

July 23, 2019 | 4:09 p.m.

I think 84o is a fold. Sure he played it fine. Brilliant even. Still, it complicated the strategy and added a losing combo. Even if it can be played exploitively for roughly zero EV it does nothing to enrich the range or make one tougher to play against. Fold and move to the next hand. Variance is the biggest overlooked enemy in poker.

Worse, river spots like this are very difficult because the IP player is forced to make decisions where IP range is played with bluffs that are unusual. When a lot of money goes in a lot of odd spots will happen IMO. It makes the OOP, even playing optimally, either make assumptions without much evidence or play an approximate equilibrium strategy that may be wrong. Yet more variance.

Fancy play syndrome IMO preflop. The best among us can be perhaps most prone to it.
I don't think Linus play looks just fine with these considerations in mind. Linus plays excellent poker is a true statement.

I like the video very much -- thank you.

April 21, 2019 | 1:47 p.m.

I think your videos are consistently wonderful. Sharing your insights and thought processes provides a lot of value.

A language thing, you say things like "that bet sizing exists at equilibrium." Almost any non-exotic strategy can be balanced to an equilibrium, many with positive EV. I also find RIO and other pros that misapply such logic fairly often. What is a thing in one person's strategy might not be in another and both may be reasonable.

I am not implying that in this case -- I'd just like to hear the conditionals behind your statement. It is easy to infer what you are implying but hearing your analysis would be terrific. How do you think about the multiple bet sizing equilibria and use in the existing metagame?

Thank you again for sharing with the community.

Feb. 26, 2019 | 6:40 p.m.

look at the chart above -- very similar situation.

Feb. 11, 2019 | 8:49 p.m.

same as PLO -- you have a problem with PLO too ?

Feb. 11, 2019 | 7:35 p.m.

Saw WSOP might switch the rankings for straights and three of a kind. Maybe we can get them to look at this. It is pretty convincing.

Jan. 26, 2019 | 6:41 p.m.

These are 6+NLH equities vs regular PLO -- other than flush very similar. I have an article on it coming out in 2+2 magazine probably in February. Note even in PLO it is easier to make straights than three of a kind. I argue it makes sense to promote the flush but change no other hand rankings.


Jan. 26, 2019 | 2:48 p.m.

I very much enjoyed watching you adapt and learn in real time. Thank you very much ! You content is always of the highest quality and incredibly thoughtful.

Dec. 22, 2018 | 3:56 p.m.

The TTxx hand vs quads is a relatively easy fold that is hard to make at times. I find I make those calls when winning / losing more often than just having an uneventful session. When not playing my best I start to look for theory justifications for doing what I want to do emotionally. Thank you for a thoughtful video.

As you suggest a big part of winning is getting away from spots that our opponents will reciprocally play worse than us.

Dec. 17, 2018 | 2:03 p.m.

Poker is just too hard to do as the only vocation for an entire lifetime for almost anyone. I can provide a slightly different perhaps future perspective for young professionals. I am a professional scientist and professor in my early fifties. I play poker at a professional level and can provide some perspective.

In order to survive as an academic scientist, I have to compete successfully with literally the brightest and often hardest working people in the world. My grant proposals compete with National Academy members and Nobel Prize winners and success rates are often in the single digits. Nonetheless, poker is harder than science.

I think the main reason is the human brains is programmed for loss aversion. The wins rarely feel as good as the losses hurt. This is magnified when one’s primary vocation is poker. It is very hard to feel great about yourself when your results are not great or even good over a period of time. Conversely, when you are doing well, you are just a professional that is doing what they are supposed to do. It is a maddening trap that I think most, if not all, humans are programmed to face.

Further, downswings can lead to self-doubt and luck plays a huge role in all aspects of life that is hard to disentangle from performance. The difference is that science, for example, is a highly positive sum game and poker is not. The worst professional scientists can have a huge positive effect of the world by simply spreading the scientific method to many young minds. Just successfully accomplishing this can have a bigger impact on the world than even an Einstein can through individual discoveries. The science world is mostly a healthy intellectual environment where everyone contributes through open discourse, and knowledge builds incrementally together to make world changing discoveries.

Poker is often a cut-throat environment where people rarely share their knowledge and are literally fighting for survival. If I have a bad month at work as professor my pay check still arrives, and my students still grow and make progress. I have a body of work to look back upon, many papers published and former students flourishing in the world. If I make mistakes in poker, or just lose a few large pots in live poker, my entire year of “accomplishments” may be at risk. There is often little to show tangibly for a body of work in poker other than perhaps some highly fungible currency.

Poker is just too hard to do as the only vocation for an entire lifetime for almost anyone. I am not sure I have met a successful professional that did not want to get out of poker in one way or another. They might have a number in mind to “retire” or want another profession so poker can be supplemental income. Worse, we generally see the luckiest folks in poker as our role models. Those who have been fortunate can’t share that magic with anyone. That is not to say that the amazing minds and people that represent, for example, RIO are not remarkable. They are among the best and the brightest. Luck plays a role in all professions. In poker, because only monetary results define success, there is almost no other metric so the top of the heap will also be among the luckiest, at least at some part of their journey. Luck is out of anyone’s control and it is more important than skill. Don’t get me wrong, skill keeps you afloat long enough to get lucky, if you get lucky.

Making things more difficult, I am the kind of person that always feels like they need to improve. This is common among high achievers. I am programmed to focus on things I need to improve on perpetually. I talk to my therapist (something I recommend for all serious poker players) about the asymmetry in personal perception. If I am asked how good I am at something I see my deficiencies. This is not the same as not having a substantial ego or even a realistic view of the world. If I am asked about another accomplished scientist or poker player that I may be better than in some ways, I can see that clearly and answer honestly. Nonetheless, if I am asked how that poker player plays, I would answer honestly that I admire how they play and handle themselves, while still considering my play needs a lot of work. There can be significant logical fallacies in perception that are rooted in our emotions: I can believe that others are terrific while I need a lot of work, and I am somewhat better than they at the same time. The problem is one cannot improve their luck by force of will.

This makes poker especially hard because there will always, by definition, be the luckiest folks playing around you, acting as they are on top of the world. Roughly 5-10% of any player pool will have exceptional results each year. It is unlikely to be you in any given year and over a lifetime may never be. Someone else might be above average in fortune every year. This is how the law of large numbers works, things are normally distributed with significant groups of positive and negative outliers. This makes for a tough way to make a living. In poker we don't see the negative outliers as they disappear.

This also results in income that varies dramatically over time. This is also difficult for people to adapt to and causes stress and confusion. Remember, even people with a regular, defined income often struggle with their finances. This is another significant challenge.

All this said, poker is the most personally fascinating activity I have ever found. It requires intellectual insight and exquisite emotional control. One sees that best of and worst of most people at the poker table at some point as fate drags our emotions up and down. Poker requires a continuous struggle for objectivity and an insane ability to perform at the table. One has to make the next right decision under the duress of distraction, adulation, dejection and boredom. Further, you can always grow as a poker player and I find when I do I improve as a person too.

Life is short. Anyone who is thoughtful and disciplined enough to be successful at poker can be hugely successful in any number of professions. To those in their 20s and 30s especially, I recommend working on developing skills for another vocation to enrich your life. Like scientists, I believe most successful poker players are drawn to the job by more than the money. The personal and intellectual freedom are alluring. Poker is not a way to get rich – if that is your goal and you can win at poker over time, then you are making a serious mistake as there are many far easier paths for the talented, dedicated and ambitious. Lastly, I have not seen that the poker environment can sustain many for a lifetime without seeking other reliable income producing and wellness inducing activities as a supplement.

As an aside, I have an interest in poker theory. I believe that the nature of the optimal solutions is not yet completely understood. It seems to me from an information theoretic perspective that bet sizing should draw from distributions, providing information hiding. Solver work suggests this might be right. I also see an analogy between (statistical) mechanical energy and expected value in that the solution space of poker is a surface of constant expected value. Lastly, I have noted that the game theoretical optimal strategy is the one the requires no information of the opponents play suggesting a Shannon entropy tie in. All of this suggests to me a statistical mechanical approach to poker solutions that I have not formulated. If anyone is interested pursuing this feel free to get in touch.

Sept. 9, 2018 | 4:26 p.m.

the bottom right screen is a bit dark but I enjoy the format and pacing. TY

May 14, 2018 | 9:30 p.m.

works well as a format IMO -- ty

May 14, 2018 | 8:25 p.m.

If I can see the sim properly it looks like OOP blocks 99 and some 77 when it doesn't check them to balance betting things like the overcards. This might change playing the spot meaningfully with one sizing no? It wasn't an insignificant frequency and seems to be an important part of how to play the spot when we have the dry over cards? Any thoughts?

March 23, 2018 | 11:03 p.m.

I very much appreciate seeing this kind of video. Hearing how you think about playing in these configurations & against the crazy mix of strategies one sees live at all stakes is very useful. The stakes and stack sizes change a lot too within live games.

I get the impression from the comments that some online players are not aware that people actually play like this live for significant money on regular basis. It is also not so obvious that it is not alway very easy to develop counterstrategies in a small number of hands without perfect information -- i.e. we don't get to see the hole cards or a lot of showdowns. Also, live recreational players might wildly adjust their strategies for many reasons.

Please continue the series and thank you!

March 22, 2018 | 8:40 p.m.

It seems like the signal in the first sim is on the order of the noise -- the difference in EV was similar to the convergence criterion (if I understand it correctly) and the bet size was changed substantially as well. It would be interesting to see how bad it is to just use the medium bet size as well to see if it is just a flat EV surface here.

Feb. 10, 2018 | 6:43 p.m.

Comment | Brian Space commented on Imminent Changes

the pdf is available on my web site

At the bottom of the page (below the science publications) under, Other Topics heading.


Feb. 1, 2018 | 7:23 p.m.

Comment | Brian Space commented on Imminent Changes

A live grinder friend and I wrote a 2+2 article on live poker player behavior -- worth a read for those embarking on live play IMO.

A ProAm Guide to Live Poker
Phillips, D.; Space, B.
Two Plus Two Magazine Vol. 13, No. 4 2017,

Article archived here (cut and paste the link):

Jan. 29, 2018 | 7:24 p.m.

Comment | Brian Space commented on Exploitive Play

Wonderful insights, thank you very much.

One thing I would like you to address is the issue of how you monitor / implement your frequencies. Clearly relatively subtle frequency changes (from a human standpoint) occur as a response to large strategy changes. As you pointed out, often the same hands are involved while frequencies shift. How do you implement the strategy changes in an effective fashion --- one humans can implement on the fly? Is this further complicated if the frequencies are very stack size dependent? How does one process all this information in a meaningful fashion that still captures some of the quantitive benefit?

Thank you again, the video stimulated a lot of thought for me.

Jan. 25, 2018 | 2:37 a.m.

Can crush 2/5 live and only ever 4! AA -- not saying its optimal but not so far off. Most 3! ranges are so tight (from non-pros) and they are not folding KK and AK or even QQ. But generally just use a linear range for value and decide how to construct it based on your knowledge of the opponent. The 3! strategies of live players are wildly diverse and very rarely balanced / polarized.

Jan. 3, 2018 | 4:47 p.m.

It was terrific to see how you adjust to the multiway pots vs relatively weak strategies. Live games even at pretty high stakes have rec. players with very weak and often gambly and predictable strategies that resemble these tables. Seeing how you react to bet sizing tells in multiway pots was very interesting. Live we almost always know the status of the player be it pro, semi-pro rec. etc. and these kinds of adjustments are the bread and butter of large live win rates.

Thank you for another terrific year of videos. It is a true pleasure to view your work.

Dec. 30, 2017 | 10:17 p.m.

Its hard to follow the real time analysis because of how PIO is constructed IMO. I would still love to see the PIO analysis and sims but after you have processed them as you suggest. Because we are listening and also looking at the results there is a lag for me between your realizations and my understanding the data that makes it hard to follow meaningfully. PIO is especially hard because it pops up so much data as you look around at the results. You know what you are focusing on as you chose but the person on the other end only knows a posteriori. And then its too late to focus.
Thank you -- Brian

Dec. 17, 2017 | 1:11 p.m.

Because of the nature of the program, when you are looking at combos we don't know a priori what your are looking at. By the time you have seen what you are looking for you have moved on and are commenting but the viewer has not yet processed the data IMO. I think real time PIO is not as useful as perhaps first doing it your self & briefly summarizing the conclusions . Then perhaps you can present to the viewer the data and ✔️ if it supports your thesis. This would still be valuable to you and would be easier for us to digest. It is a bit more work but value added for both you and the audience. Just a thought. Thank you for the high quality content.

Dec. 11, 2017 | 8:14 p.m.

Beautiful and pleasant videos. Straightforward, honest and insightful. Excellent -- thank you.

Nov. 26, 2017 | 9:50 p.m.

I thought it was interesting that there is some study along these lines. As an active scientist I find that there is a vast difference between actually performing a demanding cognitive task -- like deriving equations or teaching a complex subject and coming up with new ideas and connections. Many different parts of the brain participate in different tasks based on FMRI studies. I personally have my best insights during light exercise like walking or twilight sleep. And, I could not imagine teaching a complex subject sitting still or with heavy exertion. There was a time I tried to read scientific literature during heavier exercise and found I had poor retention and sometimes no memory.

The challenge -- how to test these things double blind. Great point. Do you not let participants know its a test? Is this a case where placebo effect is desirable in any case? Thanks for the always thoughtful insights.

Nov. 19, 2017 | 8:24 p.m.

Load more uses cookies to give you the best experience. Learn more about our Cookie Policy