Friday, June 27, 2008

Button reraising...or not

I keep thinking about two seemingly competing strategies, both of which are preached by solid pros. One strategy would have you cold call in position with many playable hands, and the other advocates frequent 3-betting.

I remember Samoleus arguing for making more calls in position with hands that have implied odds. He differed with some CardRunners pros who reraise many playable hand types in position.

Dan Harrington writes in his cash game books about calling with a wide range of hands for cheap in position. The deeper-stacked you are, the more hands you want to play from the button, he says.

Finally, I stumbled upon a column by BalugaWhale that lays out a reasonable compromise: 3-bet with strong and weak hands, while call with high implied odds hands. This line of thought makes sense. If you're going to call preflop with low pocket pairs in position, why not do the same with JTs? Sure, you can 3-bet these hands when the circumstances call for it, like when you're up against a player who folds to raises too much or never 4-bets. But seeing flops with drawing hands while not spending too much money seems reasonable.

One big problem with this style comes when you're up against frequent squeezers from the blinds. To balance your range against these types of players, you need to occasionally cold call with premium hands from late position in hopes of 4-betting a squeezer all-in.

Thursday, June 26, 2008

Harrington on Cash Games: How to Win at No-Limit Hold 'em Money Games: Vol I

I think I know what Harrington is trying to do in a large part of his first cash game book: He's demonstrating a strategy that attempts to avoid exploitation by mixing up his play.

The fundamental problem with this approach is that it's almost always better to make the highest percentage play rather than sacrifice value in a misguided effort to obfuscate your hand for potential future profits. Harrington attempts to mix up his play based on the second-hand of his watch instead of balancing his play based on the game situation.

For example: If you hold Ks Kc on a Ac Ts 4s flop, Harrington recommends checking 80 percent of the time rather than making a continuation bet.

I argue for betting the vast majority of the time. In addition, I definitely wouldn't decide whether to check because the second-hand of my watch as Harrington would. He'd have you take the 80 percent action if the second-hand is in the first 48 seconds and check if it's in the last 12 seconds. Does anyone do this in real life cash games?

I would bet because I know my opponent probably didn't cold call with an Ace preflop, I can bluff him off the hand if I think he did cold call with a weak Ace, and I'm just as happy to take down the pot right now. Those are all legitimate, practical reasons for betting completely dependent on what I believe to be the strongest play, and there's no reason to ever do anything different from what I think is right.

A whole chunk of the book is dedicated to these deception plays at the expense of analysis of gametime decisions. It's better to consider whether my opponent is a calling station, bluffer, check-raiser, maniac, regular, weekend warrior, squeeze player, blind stealer or blind folder than it is to try to make ambiguous fancy plays.

At least one hand example in the book illustrates an approach I'm more comfortable with. With ATo from the cutoff, Harrington supports raising a limper to 5X and then continuation betting nearly every time on a Kc Qc 4h flop.

"Given that Player D is a loose, weak player and he checked, this flop probably missed him, and in addition should look pretty scary with the king and queen out there. Make a continuation bet almost 100 percent of the time. Unless Player D caught a piece of the flop, he'll go away."

See? That wasn't so hard. Continuation bet based on the flop texture, your position, your read on your opponent and the previous action. There's no reason to do anything differently for deception or meaningless variation.

I don't think I'm overstating this point. In other hand examples, Harrington goes for variation plays preflop, with random limps as well. That's not going to work in most of today's games, both online and live.

On top of that, the ATo hand is used to make an argument for a terrible "general rule":

"The more outs you have to a big (or winning) hand, the less you want to make a continuation bet."

That rule is not true. When you have more outs, you want to bet because it gives you more ways to win. You'll either take down the pot with a continuation bet, or you build the pot while controlling the action, retaining fold equity and potentially hitting the well-disguised nuts.

I want to say something positive, so I'll say this: I enjoy learning about new styles so that I can better understand what my opponents may be thinking.

But there aren't many new tactics from this book that I'll be incorporating into my game.

Wednesday, June 25, 2008

Self Improvement

I had a pretty awful run at the tables last weekend, and it's taken me a few days to figure out how I could do better next time.

Sure, there were a few coolers and bad beats. But I should to take responsibility for my wins and losses, and I want to learn from my mistakes. By writing them down, I hope to avoid making the same errors in the future.

_ I realized I was giving off some pretty obvious timing tells by betting too quickly. Observant players will be able to read my too-fast bets and abuse me when I'm running on autopilot. Some of these timing tells were created by a reliance on Full Tilt Shortcuts, which is a cool problem but one which I haven't learned to use properly. For now, I've turned FTS off.

_ I got a little bit lax on table selection. I still looked for good tables, but I found myself settling for tables with a bunch of average-type players. I'd rather be at a table with one fish and a bunch of solid players than a table with no fish and a few guys playing their hands for value.

_ I targeted halfstackers and shortstackers too much. I hate these players because they prevent me from playing the type of poker I want to play. When you're forced to decide your hand preflop or on the flop, you're going to end up taking a few too many coinflips. The best defense against shorties is to refuse to sit with them.

_ I stubbornly kept trying to play despite my losses because I believed I was still playing my best. I should have realized that sometimes, it's just not my day. And how can I really be playing my best when I'm yelling "Goddammit!" after every beat?

I'm back on my A game now, and quickly on my way to recovering the weekend's mistakes.

Friday, June 20, 2008

Kinds of bets

I can think of seven types of bets.

Most bets fall under the broad category of value bets or bluffs. In general, you want to either be betting because you have the best hand and you think a worse hand will call, or you think a bet might get a better hand to fold. Some players believe every bet they make should either be for value or a bluff.

Semibluffs fall somewhere between the two. Semibluffing hands have some equity, but they may not be the best hand at the time.

Blocking bets fall under the category of value bets. Players make blocking bets when they want to see a showdown for cheap. They're appropriate especially when you think you have enough equity in the pot to get to showdown, but you don't want to pay off an opponent's expected larger river bet, and you think your opponent won't raise, and you have the discipline to fold on the end when you think you're beat.

Suckbets are blocking bets' evil twin. They're smallish bets on the river meant to look like blocking or weak value bets. A suckbet wants its opponent to raise so then you can call in a larger pot or come over the top.

With information/probe bets, we're starting to get into murky waters. Information bets aren't necessarily value bets or bluffs, and they often return incorrect information. I'm not even certain bets for information have a place in a a solid player's arsenal.

Finally, there are bets to protect your hand. I don't know what to make of these kinds of bets, but I use them all the time.

I usually use them when I'm out of position with a hand like middle pair or a weak top pair. If I were playing my hand purely for value, shouldn't I cold call because my one pair figures to be best against a standard button's steal range, and a raise would only get him to fold worse hands? Or should I donkbet or check-raise the flop, hoping that my opponent will fold or call with a worse hand? If I'm hoping my opponent will fold, then I've essentially turned my hand into a bluff because it no longer has much showdown value in what's getting to be a mid- to large-sized pot. If I'm hoping my opponent calls, I'll probably find myself in a tough situation when my opponent bets, raises or calls on the turn or river. If my opponent calls my flop check-raise and I have a weak pair, usually the best outcome is that my opponent will check it down, but that's fairly unlikely to happen.

It seems like most times when I check-call the flop out of position, I either win a small pot or lose a medium-sized one. I lean more toward betting out or check-raising with these kinds of hands, although I'm not sure how right that is.

Wednesday, June 18, 2008

Shorthanded game selection

Learning to play shorthanded is a necessary skill to become a well-rounded player. The ability to play shorthanded gives you more games to choose from, and these games allow you to play more hands more aggressively. I know this.

Then why do I find myself playing more full ring tables recently?

The explanation is simple: a lot of the people playing shorthanded games have an idea of what they're doing. Many of the true fish who don't even understand the basics of the game test the waters at full ring tables. You won't find a guppy wandering into the shorthanded tables as often as you will find him floundering anonymously with eight or nine other players.

Between full ring and shorthanded games, I prefer shorthanded. But more importantly, I want to play in whichever game gives me the best opportunity to make money.

It's essential to keep your options open so that you can select the most profitable games. Sometimes the best games are shorthanded. But just because the shorthanded games are looser on average doesn't mean they always have the easiest targets.

Sunday, June 15, 2008

What's that new poker game all the kids are playing?

I never knew that Texas hold'em was originally called "Hold Me, Darlin'."

Its origin seems to be a bit hazy, but somehow this new poker variant eventually got twisted around to become hold'em, according to a recent CardPlayer article my Jim McManus.

Hold'em wasn't played until early in the 20th (century) and didn't overtake draw and seven-stud as the most popular game until about 1990. By then, its original name, Hold Me Darling, had long since been abbreviated to Hold Me, and then, via the twanging vicissitudes of cowboy enunciation, to hold'em.

No one knows for sure where and when the first hand of hold'em was dealt. One plausible guess is that a dozen or so Texas ranch hands wanted to play a little stud, but found they had only one deck. The most creative cowboy must've got to thinking: If five cards were shared by all players, as many as 23 of them could be dealt two-card hands. Though every poker variant has roots in the French game of poque, he probably did not drawl, "Voila!"
I also found it fascinating that the powers that be have decided the name of the game is "no-limit hold'em" -- that's with a hyphen and without a space before the apostrophe, according to Short-Stacked Shamus at Hard-Boiled Poker.

The meeting concluded with a “style guide” discussion led by Haley. Some people find such discussions of usage and mechanics at best boring and at worst useless. Not this crowd. These are writers, people who care about words and how they are employed. Have to admit I had a little “I’m-in-the-right-place” moment there as we debated whether hyphens have a place in words like “preflop” (no) or “no-limit” (yes). Haley convinced me, actually, that “hold’em” is in fact a contraction (I have always typed it as two separate words).
Who knew?

Friday, June 13, 2008

Just Giving it Away

It's pretty rare that you find a true gambler at the online tables: someone who is more than happy to get all his money in with the worst of it in hopes of getting lucky. I'll take that action.

This opponent was crazy with stats around 65/45. I knew he would never limp a strong hand from the small blind, so I pushed on the off chance that he would call. Amazingly, he did.

Full Tilt Poker, $5/$10 NL Hold'em Cash Game, 3 Players - Hand History Converter

SB: $685.05

Hero (BB): $1,222.75

BTN: $898

Pre-Flop: A 3 dealt to Hero (BB)

BTN folds, SB calls $5, Hero raises to $1,222.75 and is All-In, SB calls $675.05 and is All-In

Flop: ($1,370.10) 4 2 2 (2 Players - 1 is All-In)

Turn: ($1,370.10) K (2 Players - 1 is All-In)

River: ($1,370.10) 3 (2 Players - 1 is All-In)

Results: $1,370.10 Pot ($1 Rake)
SB showed 5 9 (a pair of Twos) and LOST (-$685.05 NET)
Hero showed A 3 (two pair, Threes and Twos) and WON $1,369.10 (+$684.05 NET)

Then he types in chat something like, "I liked that. Let's do it again."

"OK," I replied.

A few hands later, we were able to get it all in again preflop:

Full Tilt Poker, $5/$10 NL Hold'em Cash Game, 4 Players - Hand History Converter

BB: $770

Hero (UTG): $1,901.80

BTN: $873

SB: $308

Pre-Flop: Q Q dealt to Hero (UTG)
Hero raises to $30, BTN calls $30, SB folds, BB raises to $770 and is All-In, Hero raises to $1,901.80 and is All-In, BTN folds

Flop: ($1,575) 4 A 2 (2 Players - 1 is All-In)

Turn: ($1,575) J (2 Players - 1 is All-In)

River: ($1,575) K (2 Players - 1 is All-In)

Results: $1,575 Pot ($2 Rake)
BB showed Q 2 (a flush, Ace high) and LOST (-$770 NET)
Hero showed Q Q (a flush, Ace high) and WON $1,573 (+$803 NET)

Thanks for the gifts, fishy.

Monday, June 09, 2008

When are you pot committed to calling a preflop all-in with AK?

UPDATE: I'm pretty sure my original calculations were wrong, so I've revised this post throughout to reflect those changes. Thanks to Greylocks for his help, and I apologize for the mistake.

Calling all-in with AK with 100BB effective stack sizes is not a leak in cash games against an opponent who will push with QQ or better and you've already invested only 21BB.

That's my conclusion after running some numbers based on a recent hand in which I had 4-bet raised to $400 preflop (40BB) with AK and then got pushed on in a 5/10 game. I had to call $600 to win $1,400 in the pot. I was up against KK, failed to spike an Ace and lost a buy-in.

The hand got me wondering whether my call was correct against a villain's hand range of AA, KK, AK and QQ.

When am I committed to calling all-in?

AK is about a 60-40 dog against a range of AA, KK, AK and QQ, according to PokerStove:

equity win tie pots won pots tied
Hand 0: 39.594% 19.26% 20.33% 110831016 116968584.00 { AKs, AKo }
Hand 1: 60.406% 40.08% 20.33% 230565960 116968584.00 { QQ+, AKs, AKo }

The break-even point for calling comes when your opponent shoves for 121BB, and you have to call 79BB more:

0.395(121) - .605(79) = 0


Therefore, against this range, I have to call after only putting in 1/5 of my stack (21 BB) with AK.

What if my opponent's range is AA, KK or AK?

Then my equity drops to 38-62, and I become committed after investing 24BB.

What if my opponent's range is only AA and KK?

Then my equity drops to 20-80, and I'm not committed unless I've already invested 60BB.

Calling an all-in with AK is far easier against someone with a wider preflop pushing range. That changes the odds substantially.

These numbers show me how hard it is to get away from AK preflop against anything but tight players. Against a wider range, I lose the option of folding unless I have a strong read or I'm deeper-stacked.

Thursday, June 05, 2008

Holdem Manager

I like the new PokerTracker, but it has a long way to go before it's completely debugged and has all the features I'd like to see. The bugs haven't caused me significant problems, but I know they've been more severe for many players.

So I finally gave in and bought Holdem Manager after hearing so many good things about it. It isn't quite as slick as PT should/will be, but it has lots of features that are just fantastic.

_ You can double click on a player's name in the HUD to flip between their total stats and their current-session stats.

_ You can take notes on players in HM rather than in the poker client.

_ Stats are displayed in a tighter box that takes up less screen space.

_ There are countless popup stats at your fingertips -- far more than PT's default.

_ Easy-to-read hand histories are viewable in a popout window, and they can be exported in several formats (including 2+2 and text).

The heads-up display has always been the most important part of poker database software. Holdem Manager has a clear HUD advantage.

PokerTracker looks like it will be better for away-from-the-table analysis, but it'll take a while for it to catch up with in-game features.

Monday, June 02, 2008

In defense of Full Tilt

With so many poker players dogging on Full Tilt, you'd think it was a bad site to play on. That's not true.

1) Full Tilt has the best software on the Internet, and it isn't close. PokerStars is OK, but Bodog is awful. Every other site runs from unplayable to mediocre.

2) I've had fantastic response times on cashouts, although I understand some people haven't been so lucky with e-transfers. I've cashed out by check five times this year, and I've never had to wait more than a week for my check to travel across the ocean to my mailbox here in Hawaii. My most recent cashout was Wednesday, and the check arrived Friday. That's a fantastic response time.

3) There are many cash games to choose from at the limits I currently play (from 2/4 heads-up to 5/10). I'm able to find fishy games almost every time I log in these days.

4) Full Tilt recently offered rakeback to many players who have been begging for it for years.

That said, Full Tilt has a significant and glaring weakness in customer support.

The site insists on doing business exclusively by e-mail, and they won't permit their customers to ever call them by phone. I've never had a problem with their e-mail support, although I've heard horror stories about their support personnel being unresponsive. Regular poker players pay large sums of rake, and they deserve to be able to talk to a real person when their money is on the line.


Ultimate Bet released inadequate conclusions of its investigation into cheating on the site at high limits. The public still doesn't know the name or position of the perpetrator, how this kind of obvious cheating could be allowed to happen for so long or whether there will be any sanctions from the site's regulatory body, the Kahnawake Gaming Commission.

You're a fool if you ever play on UB or Absolute Poker again.