The Science of Winning Poker
online.wsj.com
online.wsj.com
I find it impossible to believe that concept of a "range" was (a) invented a few years ago and (b) invented by Phil Galfond. This article is disappointing.
Edit: In fairness, the article's sentence is literally correct. But I cringe at its implication.
Galfond's gbucks article, maybe not the first mention of it, brought the concept of equity against a range into the mainstream.
"The more you know about a topic, the more it will annoy you when it is covered by the main stream media."
Once you realize this with regards to topics you're knowledgeable about you start taking coverage of ALL topics with a grain of salt, knowing that someone who knows all about it is probably cringing and / or yelling at how poorly it is being reported.
PS: I was obsessed with online poker a few years back and was an avid 2plus2 forum reader. The 1st person which I recall giving excellent advice about putting people on a range was Cole "CTS" South. He even sold an ebook called Let There Be Range for (initially, from memory) something like $5,000 a pop (it still sells for $1400 on amazon http://www.amazon.com/dp/0982402252 )
Also, in 2006 Chen and Ankenman published The Mathematics of Poker which quite explicitly deals with the analysis of hand distributions (e.g. chapter 8: Playing Accurately Part II: Hand vs. Distribution, chapter 9: Adaptive Play: Distribution vs. Distribution).
So even if we pretend Sklansky's TOP doesn't deal with ranges, Chen clearly does, and came before Galfond's 2007 Bluff article [1].
[1] - http://www.bluff.com/magazine/g-bucks-conceptualizing-money-...
On a shameless self promotional note: If there's any online casual players out there and they want to know their opponents range, check out a lil web app I built: http://www.pokerstoker.com. Keeps track of opponents pre-flop play, which is essentially their range.
I actually want to add a little feature that displays how your (previous) hand stacks up against various ranges across the streets. I'll add it when the poker bug bites me again. :-)
Sklansky's seminal text The Theory of Poker revolves around discussing distributions of hands -- how to guess at an opponent's distribution, how to mix up your own distribution to be harder to read, how to determine optimal (in the game theoretical sense) distributions. TOP was published in 1987. Galfond was 2 years old at that time.
Galfond simply took a concept that already existed and popularized it. He was successful at popularizing because of his idol status in online poker circles, circles in which Sklansky is something of a pariah.
An entire industry has popped up around taking already-existing technical concepts in poker strategy, dumbing them down, and relabelling them. See the rise of video coaching (cardrunners, deucescracked, et al) and more recently poker coaching (wherein people pay 3-figure hourly fees to have pros analyze their play) for examples.
None of these strategy services offers anything new because the math and theory behind poker has not changed. The structure of the game is the same now as it was in the 80s, and the 50s, and 20s, and so on.
The vast majority of modern strategy content is a restatement or application of the content from two canonical texts: the aforementioned Theory of Poker, and Chen's Mathematics of Poker. Even most coaches who charge the three-figure fees admit this. It's just that most people don't want to put in the gritty analysis work, hence the coaches stay in business.
Excepting some of the timeline issues though, the article is reasonable. Computer analysis radically changed how poker is played, largely by making it more aggressive. Read some books from guys like T.J. Cloutier written before everyone had Pokerstove. They brag about laying down QQ in spots where anyone with a Monte Carlo sim would know it was idiotic to do so.
Most people (even among top professional poker players) can't understand all of Chen's book. Most good poker books go much lower level than ToP, which is useful too, as evidenced by the fact that every time Sklansky himself tries to he says something dumb. If the man who wrote ToP can't apply its concepts correctly, then there's value in other people trying to do so.
I was thinking specifically of his personal life and how it bled into the strategy forums. For example the whole Brandi debacle, and before that the Sue revelations (remember the pictures with the little girl and the birds?). Many people who read that stuff were so creeped out by it that they discounted Sklansky's strategy talents entirely. So forum people -- who were a sizeable portion, probably a majority, of strategy-conscious online players at the time -- became unlikely to find out that Sklansky wrote at length about 'ranges', creating a gap for someone like Galfond to fill.
To be technical though, the article doesn't claim that Phil Galfond invented it either. Just that he "published a refinement". He probably posted something on 2+2 that was the earliest example the WSJ's writer found, though he must not have looked very hard if so.
Also computer analysis became a factor before the boom (Chris Ferguson won the WSOP main event largely due to it) but the article is pretty accurate about its impact. It's not too much of an exaggeration to say that nobody was good at poker before the year 2000, or maybe even a bit later. The people who won a lot of tournaments back then would fail miserably if they played the same way today. Some of them do in fact.
This is also easy to understand if you, a novice, are heads up against one of the pros. It's a much better move for you to go all in on any decent hand and just try to get lucky to win, since if you try to play their game, you're going to lose.
I will avoid the fact that a 90% winrate in a format like you describe (heads up sit and go) is not even remotely possible vs anyone who knows the basic rules. In heads up sit and goes 2-5% roi is generally considered great at any non micro level. Poker (especially online) is more of a volume game. However, roi is very flawed which I will get to. Generally the most profitable players are ones who crank out a small winrate over large amounts of volume.
Here is the problem though, we really are not concerned with winning a single game in a format like this. Our top concern is making money. In a winner take all heads up tournament chipEV == $EV so we can just call it EV in this format and not worry about ICM. If we are playing for a living we want to make the most amount of money, so each session we want to maximize our EV (win the most amount of chips). Most internet players worry about their winrate in BB/100 or EV adjusted $/hr.
When we get into a situation which we are a 1% favorite for full stacks vs a player we have a large edge on, we should still take it, its money in our pocket long term. If we lose we pray the opponent accepts a rematch. This is especially true for heads up games which usually only last a few minutes and its possible to play 100-200 games per day. A 3% winrate at $100 stakes with this volume becomes a significant amount of money even before factoring in site bonuses (likely a bit over $1/game) etc.
Not related to your comment, but doesn't deserve another post is that poker really is a great game and its unfortunate how its perceived by so many. Ive been playing for 3 years professionally and coached on a popular training site, yet after transitioning to programming (With a cs degree) I get laughed at by recruiters and seen as a degenerate.
Also, as the article suggests, poker strategies are advancing quickly. I started in 2004, and by 2012 I was still ahead of the curve, but the game had clearly gone light-years forward in terms of general strategy--in particular I'm speaking to HUSNGs, my game of choice.
I would say that by 2008 the average live game was strong enough that I definitely had to practice disciplined game/seat selection, couldn't just sit down anywhere and expect my edge was higher than the rake.
I agree HU strategies advanced exponentially once stoxpoker and some other video sites came on and people started being heads-up specialists.
My ROI was ~14% at 100s when I played professionally, totaling 1000s of games. I agree that at 1ks+ the best players were making ~5% ROI, but there were much better winrates up thru the 500s, which I don't consider "micros."
Also, ROI is generally less at micros, because rake is a bigger issue.
Also, the real reason you're wrong is that if you're heads up against a pro and you go all in, the pro can fold and live to see the next hand.
This is a silly byline. If anything, the best and most mathematical players bluff more than the older generation of typically more conservative players. Bluffing and math are NOT opposed principles.
I remember playing at the toughest game in my area about ten years ago. Two of the best players (both professionals) were heads up. It was kind of a dull hand after the flop - bet, call; bet, call; check, bet, fold.
As the cards were being collected the winner said "Let's try something different next time. How about the guy with the hand bets and the guy who's on the draw calls."
Edit: Oops, too late to edit. Anyway, I should have said subtitle. I think.