Not to dickride doug any further but to me, using analytic tools to see people fold too much to delayed cbets and the like is still a human approach to poker. Memorizing that "solver likes checking backdoor + overs at 33% on this board" is still interesting to me but more like brute force memorizing chess openings.

I remember the 2015 cmu bot challenge. They were manually picking bluffs and calls out of their ranges. It wasn't a solver based style at all. And they won! vs a bot that had like $100k of grant funded cpu hours and a presolved 2tb tree and then 2mins of computation ontop of that. That is insane. I mean 2 years later the humans lost but still.

Nah your info is wrong. It did runtime solving on rivers (unsure about turns?) and it tilted the players because the smaller the pot the longer it would tank. So hands that checked to the river the bot would tank for 2 minutes then check or minbet or whatever. You can see for yourself because its on youtube lol. Also if I remember correctly one of the cmu guys (the other guy, not noam) wrote a paper on whatever solving method it was using on the river.

Big thank you for correcting me!

Claudico (2015) indeed used real-time subgame solving. It was improved in Libratus (2017). https://www.reddit.com/r/MachineLear...omment/drf915q

I can't find a paper about Claudico real-time subgame resolve methods, and whether it was from turn or from river.

Libratus used real-time subgame solving starting from the turn ("third betting round") https://arxiv.org/pdf/1705.02955

Would not have guessed you were 23 in 2013, given that this photo looks like something from the late 1980s.

Also...

2+2 needs more people like you.
No, the internet needs more people like you.
No, the world in general needs more people like you.

There was a time when not knowing something then being told, or stating something incorrect then being corrected, was how people learned. Now, most people take it personally, dig in their heels and hold steadfast on their positions.

I moonlight as the official scorer for a professional baseball team. A few years ago, MLB (or someone) started a rule forbidding managers or coaches from calling up to the press box to question a scoring decision. Obviously, OSs all over the place rejoiced at this new policy. But I hated it. My No. 1 priority is to get the call correct. If that means having a manager ask me to go look at a replay, so be it. I'd rather get a call right because someone complained than to get it wrong and sit in blissful silence simply because the wronged party is no longer permitted to speak up.

Endgame Solving in Large Imperfect-Information Games
Sam Ganzfried and Tuomas Sandholm

I think it was river only. There was a big thread during the bot challenge and I remember either Sam or Noam were answering questions in it. I don't think they ever really published exactly what claudico was doing but my impression at the time was this is the paper with the info on what claudico did on the river.

Really interesting, thanks for sharing your story.

So what’d you do the past decade?

½² Of course. Actually figuring things out is way more fun and profitable back then than memorizing solver outputs.

Only someone with zero actual talent and zero ability to outthink people would prefer we live in a world with solvers.

I mean I get the cat has long been out of the bag so if you want to rely on solvers go nuts but thinking we're better off for them overall is pretty silly.

And even if this guy never made one they'd still exist by now. But if they didn't people with ability would make more money and recs wouldn't get absolutely slaughtered with zero in return for their money online.

So really who is better off for them existing - a bunch of nerds who are absolutely intolerable to play with live? Great.

So why exactly should I prefer a world where poker is less fun and less profitable?

Interesting.

I remember Juan Gonzalez when he was a machine throwing a fit when an official scorer called a ball he hit an error (correctly) instead of a 2 rbi single. It may have been in the year where halfway through the year he was on pace for the rbi record.

My guess is MLB wanted to avoid confrontation with a massive power imbalance especially as you get to the MLB level.

Yeah, it is people memorizing how a program plays. That is one reason there are few high stakes open NLHE games now. It seems like PLO is becoming more popular, partly because it is not as easy to memorize solver lines for it.

Any rec who doesn't play hasn't played plo.-At least give yourself a chance.

The low to low-mid stake NLHE games are still pretty good. 2/2 or whatever PLO is awfully soft, even if no one is totally clueless. High stakes NLHE games have pretty much disappeared, and there is more really high stakes PLO and mixed games.

except thats not what happened lol

Doug had to go open up a poker room and play live poker cuz he couldnt keep up with new era of online poker goats

could u reveal who purchased your omaha solver?

Jungle Podcast with the OP

Still listening but he also created a 2-7 Triple Draw solver specifically for Raul and Trueteller who were relative beginners at the game and end up as the best in the world very quickly using it. The likes of NL had been explored so much online prior to solvers that at least the competition has a decent base level understanding to some degree. Something like 2-7 TD that is relatively unexplored at a wider level must have been like taking candy from a baby. They must have been absolutely crushing souls in it at the time.

I dunno how much of this information was already public but I really dunno how I feel about all the goats just being the dudes who used solvers first

There is something to be said about figuring out the game vs being told how to play and just following that

Enjoyed watching that interview. Perhaps an interesting fact considering that I was apparently one of the very first people in the world to have access to a solver is that I did very little HU solver analysis in my whole career. By the time I got access to Oleg's solver, I could only really get HU action from sauce and ike (jungle wasn't playing on stars at that point in time and this was before Doug rose right to the top so he'd taken a few losing shots at me but didn't really play me). I knew sauce and his dad entered bots in the computer bot competition and I had an inkling that ike might have a solver or at least be developing one so I thought if I used it to work on HU, I'd just be in a solver battle vs sauce and ike, which didn't sound too appealing. I haven't had confirmation either way from them about it, but it sounds like they may have not had access to a solver at that point in time, so maybe it would have been worth my while. I knew about the academic papers posted on solvers though so I assumed some of the other top guys must be doing something in that direction too but perhaps I overestimated how well people were doing with it.

Anyway, I decided to focus on 6 max and did an absolute ton of work with one other guy looking at a bunch of 6max spots. If I remember right we were paying Oleg for each flop studied to some degree and each calculation took days so we were trying to run the minimum spots initially and spent a lot of time and effort extracting concepts, often to find that there was a flaw in our thinking when some other situation didn't play as we expected. We wrote well over a hundred k words of analysis (maybe much more, I can't remember) in total and definitely did pretty well in that period of time before public solvers came out.

It's probably hard to imagine these days, now that so much coaching material has been published using solvers but it was pretty overwhelming initially. All of a sudden we just had access to huge amounts of information and had to work out what to focus on, how to take concepts from it, how to implement any of it in our games. I remember the first thing I did was decide I was going to work things out in detail to get a full understanding of the game and spent a while trying to work out why KhQs was betting so much more than KsQh on the first flop I looked at etc. Quickly abandoned that approach after writing pages on one flop and then realising there were now 40+ turns to analyse and then thousands of rivers just from that one flop. Nowadays you could probably get as much info from an hour long coaching video as we could get from a month of analysis but there were some fairly crucial early insights we had which definitely gave us a nice edge on the competition for a while.

Thanks for sharing that Alex. It mirrors experience I've had with trying to utilise investment model output. Sometimes there's so much that it can be hard to see the wood for the trees. Also very frustrating when it's difficult to explain why the results are what they are or why small changes in input parameters sometimes create big differences in output actions.

As well as advances in solvers, techniques for displaying, abstracting and analysing output continue to improve. Unfortunately its becoming too good - I'm among those who preferred a time without advanced solvers, when there was greater variance in playing styles and a sense of mystery around optimal play.

Did you trust the solver was accurate from the start and, if not, how did you gain that confidence?

That comparison to investment model output is very interesting. I'm actually spending my work time investing these days, using a fairly basic home-made quantitative approach based on academic research. Luckily (sort of) I have a small amount to invest in the grand scheme of things so I get to invest in the stuff which is generally considered too small to be investible. I like it when I see criticism of academic work saying "well of course you got a positive result there, you included all these tiny, uninvestable stocks where everyone knows the results will be really good!"

I can imagine that with more scale you would be trying to refine models and include many more parameters than I do in order to maximise performance though so I can imagine it being very similar indeed to what we were doing with solver output. There were lots of times when we spent a long time puzzling over why certain things were happening or why the results from spot X were so different to a similar seeming spot Y. It was really interesting and rewarding work though, if frustrating at times. I imagine you feel similarly?

I came up through the stakes in the pre-GTO awareness era, never mind pre-solvers, so I reached 25/50 without even knowing what a Nash Equilibrium was. Then I learned about that side of the game and made the breakthrough to being one of the very best HU players in the world by being ahead of the curve in thinking about GTO concepts and trying to apply them the best I could but without having any sort of confirmation as to what was correct. I guess I look back on that time nostalgically as the "best" time for poker but that's no doubt heavily influenced by personal bias.

There was lots more action though because people had very different ideas of what was correct so it was kind of a battle of ideas at the tables and the skillset needed to be one of the best was to think correctly about the ideas and then find ways to implement them successfully while exploiting your opponents mistakes. Memorising ranges was not really my strength or interest so the top players these days no doubt play far more accurately than I ever did or would have done if I continued. I'm sure they think the game is great these days because it suits their strengths, it's all just personal opinion I guess.

I think I was on board with the solver's accuracy pretty quickly, I can't remember all the reasons for that and their weightings but I know I was a bit familiar with the academic research on the topic and I knew Oleg had used that research in the solver. I also imagine it confirmed some of the things that I had been working out independently, similarly to Oleg's description of trueteller and turn leading when the board pairs. Once I got into looking at it and started to see new things and then work out why they were correct theoretically, then confidence is sky high of course. Before this point in time, pretty much everyone had a standard cbet size I think which didn't vary by board texture. Seeing the different sizes, when they were used and working out it had to do with how the ranges interacted with each other was a pretty big learning moment.

Love it so much when we get kings like Kanu and Talal to show up here every once in a while to spread a bit of knowledge or lift the curtain a little bit on the world of ultra high-stakes battles.


I keep hope that someday, someone will try to talk to all of these guys, from Sagstrom to Linus, and make a big, fat book out of it. If it were to be 1000 pages big I'm pretty sure I'd still read all of it in a weekend.

Yep.

The inefficiency in small stocks makes them a great fishing pool but that can be offset by higher trading costs (spreads). Not a problem for long term strategies tho. Some illiquid warrants etc also interesting.

In poker terms its a king and a serf

This is what I'm working on, BTW.

PM me if you want to discuss more.

"In poker terms its a king and a serf"

Haha that's very generous but I'm not sure that's even true in poker terms, I was under the impression that you've been killing it in SHRs etc.

"Some illiquid warrants etc also interesting."

I've actually been considering trying to branch out from the mostly quantitative strategy and really take advantage of my small portfolio size by searching for something truly absurdly priced. My main candidate ideas are trying to gain access to the OTC expert market and searching through illiquid warrants. I already invest in micro/nano caps globally but I imagine to find something that is truly absurd, there needs to be a strong reason that the opportunity exists. I apologise if it's inappropriate to ask and I'm conscious of going off topic but if a sentence or two of advice springs to your mind, I'd be hugely appreciative. In the world of investing there is no question who is the king and who is the serf

Investing skills overlap a lot with poker so you should be well set - intelligence, mathematical ability, emotional stability, being willing to take risk where justified by EV, dealing with incomplete information, not over-reacting to short term results, that tricky balance between confidence and arrogance,... and many more.

In terms of markets to look at, I'm 30 years out of date in dealing in micro-caps etc. In general I found greatest pricing inefficiency in illiquid stuff or investments off the beaten track (eg frontier markets, unusual instruments). Generally investors avoid that which is different or holds greatest uncertainty (even when risk can be diversified away). Obtaining reliable information can be difficult but that's part of what creates the uncertainty and hence opportunity. If investing in small stocks you might be able to meet the company to gain an insight - just message the CFO and ask to video call or visit (small companies usually don't have investor relations professionals). The worst that can happen is they say no.

I liked warrants because you can usually assume the underlying stock is efficiently priced so the required work is limited. Early in my career even some larger European and Japanese warrants were absurdly priced. One thing I looked out for is where there is a big upcoming event for the stock, e.g. an FDA drug approval decision, which could move the stock price dramatically - often warrant prices didn't account for that but were instead lazily priced off historic volatility of the stock. If you're trading warrants you're trading volatility and it is future volatility that matters not historic.

I always avoided what was most popular. If everyone agrees something is a great investment then it almost certainly is not (because everyone already owns it). The big wins come where you find something that everyone agrees is a bad investment but you've done the work and take a different view, as long as you're right more often than not.

I haven't read any recent ones but there are lots of good books on the subject. I liked Market Wizards and One Up on Wall Street. They're old but the principles still apply. I can't speak to the quality of any investment forums etc but generally there is a lot of noise around investment and its good to avoid consensus thinking. If you are taking it very seriously as a career you could consider studying for the Chartered Financial Analyst exam as a way to learn more.

gl in the markets.