Monday, July 30, 2007

Poker at the Commerce Casino

While I was at the WPT Ca$h Camp this weekend, I had a chance to play at the tables. I did two different things while I was there: I played two sessions at the $1/$2 NL ($40 buy-in cap) game, and I played 5 $34/$6 SnG tournaments. Each deserves some mention for different reasons:

First, the SnGs. These are played on a special table they call "Lightning Poker." It is the size of a normal table, but embedded in the top is one large LCD screen in the center and 10 smaller screens around the periphery. The small screens are touch sensitive and have a small slot for a special mag-stripe ID card. You take your card to the cashier's booth to cash it up, more or less just like online play. The card has a PIN to make sure it can't be used if it's lost or stolen.

While the novelty of the SnGs was cool, I didn't like them for a few important reasons:

1. The house took a $6 rake off each $40 buy-in. That's a hell of a lot. Easily triple what the same tournament rake would be online.

2. They only paid two places. The same SnGs online pay 3 places.

3. Each player starts with a paltry $500 in tournament chips. With the starting blinds at $10/$20, that makes everybody a short stack. If your first move doesn't work out (standard raise plus a continuation bet), you've probably lost half your stack. Your next hand is all-in, and if you lose that you're done. That's a pretty harsh tournament.

I wound up chopping one out of the 5 tourneys I played, for $170 (the computer doesn't know how to chop them, so once the players agree, they simply go all-in in the dark until it's over, then the winner gives the 2nd place player $34 from the $204 first prize). So net loss of $30. But I don't feel too bad about it. The first two I played I lost because of a 2-outer and 3-outer suck-out, respectively. The third I chopped, then in the 4th I lost a race and in the 5th I wound up all-in with an under-pair to an over-pair. I still think that the crowd was pretty soft, but at the same time, the conditions sort of set you up for higher variance, which doesn't favor good players.

I had two sessions of no-limit. These had a $40 buy-in cap to play $1/$2 NL. That also makes the stacks a bit shorter than they ought to be. One of the things I learned at camp was that the correct buy-in for NL cash is 100x the big blind. That said, we should be starting $1/$2 with $200 stacks. Being so short, it plays a bit more like a tournament, with normal moves resulting in people being pot committed and then all-in.

The first session was a beautiful thing. I wound up winning $20 in a half hour before breakfast. The crowd was quite soft. The second session, I also found myself up $20 and almost stood up and left with it, but decided I wasn't done. I then lost all that and another $20 behind. Bummer. So total for the weekend in cash: down $40.

I'll write more about the camp later.

Wednesday, July 25, 2007

You don't know Jack on the web!

The folks at Jellyvision, once responsible for the "You Don't Know Jack" series of quiz games have sort of moved in a more... corporate direction. Now they want to take their expertise at conversational interfaces and make demos and attract flash for marketing. Nice work if you can get it, I suppose.

But all is not lost! They are making YDKJ snippets available on the web for free - including the Daily DisOrDat. To get you in the mood, why don't you have a taste? The subject of this DisOrDat is particularly apropos.

Just so you know, I scored. Huh huh.

P.s. I tried to embed it on this page, but the flash didn't work correctly. Oh well.

Wednesday, July 18, 2007

Algorithm for evaluating poker hands

I've never actually seen the full description in one place on the net, so I thought I'd do a public service.

First, if you're playing a game with extra cards, like Hold'em or 7 stud, you first use recursion thusly: Iterate through the set of cards, removing one at a time and recurse. From the recursion results, save the best hand and return it.

Once you get down to 5 cards, you first take a histogram of the card ranks. That is, for each rank in the hand, count how often it appears. Sort the histogram by the count backwards (high values first).

If the histogram counts are 4 and 1, then the hand is quads.

If the histogram counts are 3 and a 2, then the hand is a boat.

If the histogram counts are 3, 1 and 1, then the hand is a set.

If the histogram counts are 2, 2 and 1, then the hand is two pair.

If the histogram has 4 ranks in it, then the hand is one pair.

Next, check to see if the hand is a flush. You do this by iterating through the cards to see if the suit of a card is the same as its neighbor. If not, then the hand is not a flush. Don't return a result yet, just note whether or not it's a flush.

Next, check for straights. Do this by sorting the list of cards. Then subtract the rank of the bottom card from the top card. If you get 4, it's a straight. At this point, you must also check either for the wheel or for broadway, depending on whether your sort puts the ace at the top or bottom. I would expect most folks to put the ace at the top of the sort, since it is usually a high card. So to check for the wheel, check to see if the top card is an ace and the 2nd to top card is a 5. If so, then it is the wheel.

If the hand is a straight and a flush, it's a straight-flush. Otherwise if it's one or the other, you can return that.

If we haven't matched the hand by now, it's High Card.

Once you know what the hand is, it is fairly straightforward to compare two hands of the same type. For straights, you simply compare the top card. For flushes, you iterate down through a reverse sort of the ranks until you find a higher card or run out. For the histogram-based hands, you start at the beginning of the reverse-sorted count list and compare the ranks (two pair is tricky - you must always evaluate the higher of the two pairs for each hand first - the histogram won't help, then the lower pair, then the kicker).

Checking for a low (for high/low or Razz) is a little easier. Sort the hand by rank. Iterate through the list. If any card is bigger than an 8 or is the same rank as it's neighbor, it's not a low hand.

Saturday, July 14, 2007

MST3K mark 2

The Mystery Science Theater 3000 guys are back. They call themselves The Film Crew, and their first effort is available via Netflix's Watch Now feature.

The silhouettes are gone, but the commentary is as good as it ever was. There are also fewer skits and more taking apart of the movie.

This effort appears to be direct to video.

Metallica IS Spinal Tap

I've always thought metallica sucked. But I never really cared so much. I sort of ignored them when they were on the radio. But ever since Metallica came out as the sock puppets for the RIAA against Napster, I've actively disliked them. Not that I'm really in favor of illegal music swapping, but the RIAA was (and is) more a part of the problem than not. It's just extra fitting that Metallica, out of everyone, would be the tools that would step up and spout the party line for them.

But this isn't about that, really.

Lately, a new Metallica track has been playing on SquiZZ. It's Metallica's take on "The Ecstasy of Gold." Now, you may not recognize that song title. It's part of the soundtrack of the movie "The Good, The Bad and The Ugly." Remember the scene where Tuco, Blondie and Angel Eyes had a 3-way shoot out in the graveyard near the end of the movie? It's that music. It's one of Ennio Morricone's finest works. And, like John Williams, Morricone is truly representative of what classical music composers historically have been. The fact that the music is the background to a movie isn't really that different from the times when the music was the background to an opera. So I give much props to Morricone's talent. His work will live as long as Mozart's.

So, again, where does Metallica come in? Remember the scene from "This Is Spinal Tap" where Nigel and Marty discuss the minor key and its relationship to classical music... only to then reveal that the complex classical-based composition they've been noodling around is called, "Lick My Love Pump." That's really the only thing missing from the Metallica to Spinal Tap connection with Morricone's tune - they didn't rename it "The Ecstasy of Taking My Meat Hammer Up Your Poop-chute."

Feh. Take that to eleven, beyotch.

P.s. Before anyone jumps down my throat, yes, I am aware that the band has used "Ecstasy" as their warm up music for years. That just drives the point home, as far as I'm concerned. I wasn't aware of it before now because you'd never catch me dead at one of their concerts.

Friday, July 13, 2007

Was this a bad play?

Full Tilt Poker Game #_: $20 + $2 Sit & Go (_), Table 1 - 20/40 - No Limit Hold'em - 19:54:01 ET - 2007/07/13
Seat 1: hero (2,825)
Seat 2: (1,245)
Seat 3: (4,220)
Seat 4: (710)
Seat 2 posts the small blind of 20
Seat 3 posts the big blind of 40
The button is in seat #1
*** HOLE CARDS ***
Dealt to hero [Ah 5h]
villain calls 40
hero raises to 160

Yes, it was a loose raise, but we were 4 handed and I'd been very tight.

Seat 2 folds
Seat 3 folds
villain calls 120
*** FLOP *** [3s 5c 7d]
villain bets 550, and is all in

So he limp-called before the flop. Classic tell of a small or mid pair. At the time, I didn't pick up on it. Now he jumps for his last 550 chips, with over 300 in the pot already. I at least have a pair, and if he's bluffing with crap, then I've got him beat. If he has a better hand than my pair of 5s with an ace, then I'm still in good shape overall. I decided to risk that he was just pulling a stop-n-go with any two cards.

hero calls 550
villain shows [8s 8c]
hero shows [Ah 5h]
*** TURN *** [3s 5c 7d] [5d]
*** RIVER *** [3s 5c 7d 5d] [3d]
villain shows two pair, Eights and Fives
hero shows a full house, Fives full of Threes
hero wins the pot (1,480) with a full house, Fives full of Threes
villain stands up

At this time he starts lambasting me in the chat window as being a donkey.

I called him because he was short, and because I had something and could afford it. If I either beat him or sucked-out, then we'd get that one step closer to the money.

It wasn't the proudest call I've ever made, but I think it was, all in all, more correct than incorrect. What do you all think?

*** SUMMARY ***
Total pot 1,480 | Rake 0
Board: [3s 5c 7d 5d 3d]
Seat 1: hero (button) showed [Ah 5h] and won (1,480) with a full house, Fives full of Threes
Seat 2: (small blind) folded before the Flop
Seat 3: (big blind) folded before the Flop
Seat 4: villain showed [8s 8c] and lost with two pair, Eights and Fives

I went on to go heads up with Seat 3 and win, but I got very lucky to take out the 3rd place finisher.

It's been my lucky week so far, even being Friday the 13th and all. I hope it continues.

A solution to the GPS-less iPhone

I am embarrassed I didn't think of this myself.

One shortcoming of the iPhone is that the google maps app - though it is without a doubt amazing - doesn't have any positional information. That is, the map doesn't know where you are. There's just no room in the iPhone (or juice in the battery) for a GPS chipset.

But who says the GPS has to be built in?

All credit goes to Craigy on MacRumors for the idea. And it's a fantastic one. If you simply kept a bluetooth GPS box in your car, then your iPhone could use it to provide a "You are here" pin in the google maps app, and could automatically step through the turn-by-turn directions as you go. It could even give you ETAs (with real-time traffic, of course).

Yes, it would be better to have GPS built-in, but Apple could create a whole market for iPhone GPS add-ons. And, conceivably, they don't actually have to be bluetooth tethered. There's no reason you couldn't simply connect up a GPS add-on to the dock connector.

But in any event, it's going to require a lot of software support from Apple.

We'll see.

Heads-up position

I haven't played much poker since the victory on Monday. It's as if I don't want to spoil the victory, in a way. But last night I did sit down to a $20 heads-up match at FTP. I won it in 3 hands. I managed to catch on the other two hands and that allowed me to appear to be quite aggressive to my opponent, which probably had a lot to do with the outcome, but in any event, the last hand was a great lesson in heads-up position.

When you're heads-up, the button is the small blind and acts first before the flop, and last after. That little change-up is something quite special in heads-up play.

Full Tilt Poker Game #_: $20 + $1 Heads Up Sit & Go (_), Table 1 - 10/20 - No Limit Hold'em - 20:19:53 ET - 2007/07/12
Seat 1: victim (1,320)
Seat 2: hero (1,680)
hero posts the small blind of 10
victim posts the big blind of 20
The button is in seat #2
*** HOLE CARDS ***
Dealt to hero [Td Ad]
hero raises to 60
victim raises to 120

So he has indicated that he has something. A re-raise normally would tell me that it's a big pocket pair, but heads-up it could be any pair or even a big ace. The minimum raise is often a monster hand tell, but before the flop it's not all that meaningful, particularly heads-up. As we'll see shortly, all he did was make it really cheap for me to call. He saw that I was an aggressive player the other two hands. He might have pressed here to try and take the play away from me. But no.

hero calls 60
*** FLOP *** [2s 3c Ts]
victim bets 200

Here's where position comes in. If we were out of position, I would have made this bet and he would have had to evaluate whether or not I was full of crap. He might have just called a bet given what his hand is.

As for me, I've flopped top pair with an ace. I am only beat by a pocket set or a bigger pair. If he has one of those, so be it.

hero raises to 1,560, and is all in
victim calls 1,000, and is all in
hero shows [Td Ad]
victim shows [9c 9h]

That was a bad call. I raised before the flop and called a re-raise. What hand could he have put me on that 9s would beat? A smaller pair (but not 2s or 3s)? There have only been two other hands. Could he really have believed that I made that big a move with nothing?

Uncalled bet of 360 returned to hero
*** TURN *** [2s 3c Ts] [Kd]

See? If he'd have been in position on me and if I had made the continuation bet and he'd have called it, the king would have been a scare card. He could have represented it and probably taken the pot away from me. That's exactly why I went all-in on the flop. I had what was likely the best hand and didn't want to see any more cards.

*** RIVER *** [2s 3c Ts Kd] [3d]
hero shows two pair, Tens and Threes
victim shows two pair, Nines and Threes
hero wins the pot (2,640) with two pair, Tens and Threes
victim stands up
hero stands up
*** SUMMARY ***
Total pot 2,640 | Rake 0
Board: [2s 3c Ts Kd 3d]
Seat 1: victim (big blind) showed [9c 9h] and lost with two pair, Nines and Threes
Seat 2: hero (small blind) showed [Td Ad] and won (2,640) with two pair, Tens and Threes

Tuesday, July 10, 2007

WPT Ca$h camp

I'll be at the WPT Ca$h camp at the Commerce casino at the end of the month.

Why?

You'll recall that I went to the 2 day "normal" tournament camp in January (thanks to my lovely wife Scarlet), and got some good validation and help tuning my game. I also met Crispin, who helped tune my game further.

Scarlet and I went to Reno a couple weeks ago and while I didn't have any tournament success, I actually was up in the cash games. But that was, for the most part, playing $3/$6 and $4/$8. I made $25 playing no-limit during one session, but I was out of my comfort zone. I know I can succeed at cash games, but my aggressive tournament style is not correct for cash play, and knowing that I am not where I need to be in the cash game keeps me uncomfortable because of the higher risks.

So unlike last time, I am not really seeking validation of where my game is. Rather, I know my game is not where it needs to be and I need to tweak it a little, get more comfortable, and then I'll know I'm one of the sharks rather than one of the minnows.

Monday, July 9, 2007

Victory!

I won this morning's Bay 101 Midnight tournament!

I credit this win almost completely to Crispin helping me tune my game. I played conservative poker for most of it. A key lesson I've learned from my last few failures is that when you get close to the bubble, tighten up. You don't want to spend all that time playing for 11th place. That play took quite a bit of patience, since I was a relatively short stack going into hand-for-hand play. I folded AJo and pocket 5s while we were at 11. Once we were safely in the money, I opened up a tiny bit and managed to get lucky in a couple of key hands and double into the chip lead. Once there, I just waited for good spots to punish the blinds and short stacks. I lost a bunch of match-ups, but for the most part they were situations where the big blind was all-in and it was one more chip to call - you throw down with 27o in those situations.

So all in all, $120 turned into $2990 (minus a $50 toke for the dealers).

Since the last time I won at Bay 101, they seem to have stopped giving away jackets and winner chips. Bummer!

But I got 290 more points towards the Bay 101 open in August. That should put me in fifth place overall at the moment.

Tuesday, July 3, 2007

Little things in iPhone

Here's a to-do for Steve and his team:

1. A mechanism to add additional root certificates to the X509 key store for Safari. Lots of folks use them for intranet stuff.

2. LEAP authentication for wireless. It's what we do where I work. The mac can do it, so should the iPhone.

3. Flash. You're just not going to get your way on this one, Steve. Promise.

4. Voice dialing.

5. Internet IM.

6. custom ringtones from iTunes.

7. A2DP - also known as high quality stereo over bluetooth. Rumored to already be on the way.