i am now interesting in poker again...

i have started playing the Cepheus poker machine from University of Alberta. the #1 poker machine in the world. and i believe this is peer-reviewed or similar.

http://poker.srv.ualberta.ca/

i believe Bryce Paradis, who had one of the great poker runs in history (playing heads-up) is doing his PHD at Alberta.

some i'm wondering, what are the best heads-up books? i remember 2p2 did a book. i recently purchased a book by Mathew Miller (very dense and analytical book, which is probably good)... i believe i am missing one top recent book..

has ed miller written about heads-up? dan harrington?

thx for any responses.... i'd love to get a league going challenging Cepheus. i am usually fighting to get to break-even against it. so i am a loser. Cepheus takes some lines that i find unbelievable and bad but "hey the results don't lie"

two things,

1) the Alberta poker machine is limit...... so if you don't have a hand and you are out of position, you are basically dead (i think)

2) it may be down for maintenance the next few days.

Philip Newall's Further Limit Hold'em, published by 2+2, deals extensively with HULHE. In fact, Cepheus' predecessor, Polaris is one of the LHE-bots analysed.

In my opinion, this is a very good book on HULHE, and it certainly changed my game (which was way too tight, especially OOP, before reading the book).

Also, Cepheus' preflop ranges for both SB and BB are published right there on the University of Alberta web page, and they are both, in a word, wide.

I've played a few thousand hands against Cepheus over the past few years, and had fun with that. Finding a good LHE game in which to use your newly honed skills might be more of a challenge...


Best of luck!
Slim

Slim, thx for the comments

i am just happy to play Cepheus and hone my hand reading (flop reading) etc... i prefer this than putting $1000 on some poker site, never to see the money again or driving to the casino to play slow games. to me, it's like video games.

i was aware of all the pre-flop charts that are on the site. and yes, i consider them very wide too.

the calling a raise pre-flop with some ok cards (J6o), completing missing the flop and folding to the C-bet is very very frustrating in limit. not sure how to combat it... OTOH, it seems to work really well when you are the person in position....... i would be curious as to what cepheus' c-bet frequency is. mine is 100%....

i really need to learn more about flop texture and then reading the turn/river as to big outs hitting... i.e. with JT on flop, Q or 9 on flop is terrible. ace not so bad (as to the hitting the straight)..

seems like with limit head-up that river betting is really key. playing the complete miss or Kx or Ax is so important.

i see some bizarre plays from cepheus.. but i'd say it beats me fairly consistently.... i have had success at times both slowing down and speeding up but i am thinking much of that might have been card luck

once, i had been losing so badly i decided to write down pertinent facts from each hand. and then i went on my best 30 game heater ever as i was documenting the hands. bizarre

lastly, what is tricky about cepheus is it raises almost all hands pre-flop acting first. so you really have no idea what it has, other than not offsuit hands with a 2/3 and another card under 10. that knowledge does help a little though

Further Limit Hold'em also contains a systematic approach to board reading, which will help you address this particular issue.

On certain boards you might find that leading flop, or check/raise flop, lead turn are superior strategies. Despite J6o missing cleanly, your range may not have. Cepheus might not always "have it" either...

In short, get the book, it should improve your HULHE game.


Slim

Slim, thanks for the further comments

for some reason, i buy into the cliche that leading into the pre-flop raiser is "donk betting"... like i said, cliche. but also, heads-up is generally weak hands, full ring early raise is some really good hand. big difference.

i will pick up book. sounds great.

cepheus is back up and online.

i got to within one hand of beating cepheus for 1000 (over 100 hands) twice and got bad beat both times. don't get me wrong, i "bad beat" cepheus a few times to get in that position.

and cepheus has had me close to losing 1000 (in 100 hands) a number of times but we've never hit that figure (even during the season, not just counting the end)..

anyway, it's really improving my hand reading and mostly satisfying my need to play poker.

i feel like you need to be a really good combo of offense and defense (think rafael nadal in tennis) to be competitive with cepheus.

tracking my results is pointless as i stop so many sessions short of 100 hands. sometimes i don't realize i'm stopping the session permanently at the time.

i wish it had a peek option. peek at the computer's hand a couple of times during the session (and i mean, after the hand is over)

is there any heads-up coaching software these days? wilson software was pretty good.

Philip Newall is a member of 2+2? What is his screename? Thanks.

Screen name is philnewall. He has an active thread in NVG called Academic research: Test your probability knowledge.



Slim

I want to write about this book on the website I work for, but the 2+2 store dont deliever in Canada and Amazon.ca is out of stock. Looks like a fantastic book.

HI Jef:

I just checked and on Amazon.CA it says will ship in 3 to 4 days.

https://www.amazon.ca/Further-Limit-...4059311&sr=8-1

Both books by Philip Newall are terrific and I highly recommend them and we're proud to have them as part of the 2+2 family of books.

Feel free to contact me directly if you have any specific questions.

A little side note. When I first heard from Philip I had no idea who he was (and think he was 19 at the time), and told him that we were too busy with other projects to even look at his manuscript. But then a few days later one of our projects fell through so I recontacted him and told him to send the manuscript (which was his first book, The Intelligent Poker Player, for our review, and then when I started to read it quickly realized that this was one of the best poker books ever.

Best wishes,
Mason

Mason, can you encourage Amazon to add prime or fast shipping options to books that are non-current and/or used? i realize it would cost more as part of the book price. probably alot more in some cases.

so often, i want to buy books but i don't want to go into the fog of delivery in the next 6 weeks.......

as an aside, on a huge chinese ecommerce site, they had a valentine's day sale and it often takes 4 months for your item to show up. i've had packages come a few times that i'd completely forgotten about.

+1

hi Mason hope you're doing well.

have played the machine alot since i've posted.

i find i start thinking i'm break-even vs. the computer and then i get absolutely smacked by it over 30 hands. sometimes i smack it, but i'm thinking not as often as opposite.

basically i find many of my sessions are break-evenish and then i'll have a wild one once in 7-8 sessions...... basically kurtosis, and some negative skew in my case.

i have found playing very defensively with mediocre cards and lack of position is the way to go...... and big differentiator is analyzing more difficult decisions. i.e. should i bet? will i get raised?...... should i make hero call on the end?

i see the machine make some crazy plays. i think i'd get massacred if i did the same. of course, the machine's crazy plays are part of its total strategy. if i did those crazy plays they might not fit into my overall strategy...

pretty sure i saw the machine call a river bet with J-high.

or raising with air on the river... to me that makes no sense. if i have a hand, i am calling..if i don't, alberta has no need to raise........ i guess his air could be worse than my air.

EDIT: not sure i made this clear.. but i really toned down the looseness (aggression?)...

I've also written this short paper on the history of humans-vs-machines in HULHE:

https://www.frontiersin.org/articles...018.00210/full

Expert humans have been using the no-capping strategy for much longer than the machines.

Phil, just because you piped up in this thread, i'm going to order your book today.. FYI, from amazon canada.

i find it fascination that the alberta researchers didn't tell the computer how to play and then the computer figured out the best way to play.

apparently, they just told the computer the rules.... and it figured out everything... so the computer had to figure out basic strategy on its own....... very impressive

a bunch of comments from playing the computer more.

1) de-risking seems like the way to go for me.. aggressive game doesn't seem to work for me. and even if it's ok for some, i don't think i can handle the huge downswings. i think i go on tilt.

2) had a few bizarre hands...... can't believe one of the triple-barrelled pure bluffs i ran on AKT9x on board was called by Qx. how can that be?.. i don't triple barrel very often FWIW

3) raising wars seems very inconsistent. because of my play and the machine's.. i would have thought machine is usually right on these, but not my experience..... this is a key factor in your win rate, i'd think... when i say computer is often wrong, bear in mind i'm wrong easily as often.

4) something that surprising seems to work very well is bluffing on a 4-flush board on the river....

5) similar to #4, the computer will surprisingly fold during a raising war on the river. when i put in 2nd or 3rd (and final raise), sometimes it folds. makes no sense to me.. on a similar vein, it makes some comical raises on the river. in contrast to machine, i can barely ever remember folding on river to a raise.

6) bluffing on river with complete garbage might be a +EV play (dependent on board of course)... trying to figure it out.

7) i need to be more mindful of busted draws.... trouble with heads-up is the computer plays 90% of hands, so it's hard to get a read on unpaired, draw, 2nd/3rd/4th high pair for opponent.

i bot the book, but then i didn't like the kindle format. so i will be order the physical copy in the future.

having played the machine alot more,

0) keep the pots small OOP. let the money come to you.. i think this raises my win rate and lowers my volatility. i can't seem to play like the machine. but i think i'm pretty close to break-even this way.

1) aggression and trying to multi-barrel the machine off a pot doesn't seem to work very well. machine calls alot.

2) a little contradictory to #1....... but either second barrel on turn or delayed continuation on turn work pretty well.. triple barrelling almost never works. and the double barrelling has to be very selective and having position works very well

3) check raising river on 3rd flush card works like a charm.. shocked it works well.

4) machine sometimes gets into raising wars on end and then folds.

5) i win and lose alot of raising wars. i don't feel like net loser, but wish i understood things better........ in fact, i'd say i'm not aggressive enough in general rather than too aggressive.

6) probably stating the obvious, but having an ace or king in your hand helps so much on the river.