Friday, August 03, 2007

Survival of the Fattest

I expected to write this yesterday.

After writing my last post, with opening-night jitters, I expected to write yesterday that everything went okay. But I never dealt a hand on Thursday morning.

During the afternoon and evening of August 1, the casino was open for various special audiences: The tribe in the afternoon, and VIP's in the evening. Then, at midnight, the casino was open to the general public. However, all of the promotional material said that we would open at noon on August 2. So, when I got there at 4 AM, the casino mostly had holdovers from the night before, and not many of those. At the slowest point of the morning—probably around 7:00—there were maybe 25 customers in the entire casino. And all the dealers of the graveyard (excuse me, "sunrise") shift were there. With the other employees, the customers were probably outnumbered five to one or more.

So I never dealt. There were two games going in the pit I was assigned to, when I got there, and two other dealers tapped into those tables. So I went on break, preparing to come back and relieve one of them. I found, instead, that they were preparing to close down Pit 2 completely, and they sent me into Pit 5. When we got there, the pit manager looked at us agog, since he had at least 15 dealers he didn't know what to do with.

We ended up doing makework. In this case, it consisted of setting up decks for the day shift, so that it wouldn't take as long to get tables open for them. This is normally something that the dealer opening the table does, so it truly was makework. At a little after 11:30, with all the tables prepared to open, we were told we were done for the day. People had been coming in pretty steadily, and when I passed the front entrance of the casino, I saw a pretty long line of cars backed up to turn in.

I reported today expecting the casino to likely be rather busy. I never did get a handle on how busy we were, overall, but my pit stayed pretty hoppin’ all morning. Other people told me that they weren't that busy, though, so maybe I was the lucky one. And, bear in mind, it was early in the morning.

I reported to Pit 2 this morning to be assigned rapidly to a mini-baccarat game. This is a game which superficially looks a bit like blackjack, but it really comes down to simply deciding whether the player hand will win, or the banker hand, or that they'll tie. The players have no say in whether either the player or the banker will draw to improve their hands. My job is simply to pull out four to six cards, and then take the losing bets, and pay the winners. "Player" and "Banker" bets pay even money, but a winning "Banker" bet must pay 5% commission on the winnings. In practice this simply means that a player who bets $20 is only paid $19. (Some other casinos keep track of the commission separately, and a player must pay his cumulative commission before he leaves.) The "Tie" bet, when it wins, pays 8 to 1 (or, on some tables, "9 For 1," which means the same thing), and "Player" and "Banker" bets push. Our tables also have a bonus bet (the "Dragon Bonus"), which can pay as much as 30 to 1, although I never paid more than 10 to 1 on it.

Baccarat players skew heavily Asian, as I've stated, and in fact I dealt to quite a number of them. But the three longest-tenured players on my table were all middle-aged white men, all of whom were averaging more than $100 a hand. One of them, who had been at the table about fifteen hours when I left, frequently bet between $300 and $800, with the occasional $1200–$1400 bet. One of the Asians, a woman whom I have to say that I was ruthless to with the cards, made repeated bets of over $1000, including one $1600 winner I got to pay. In fact, most of the Asians ended up losing their shirts at my table. One fellow said it had been a bad night for him, and we all smiled and nodded, and then he brought out three or four (empty) bank wrappers to prove it. That was fairly impressive.

I was indeed nervous when I sat down, but not nearly as much as I would have been the night before. In fact, it had pretty much dissipated after the second hand. Although I know for a fact I overpaid a player at least once, the single biggest mistake I made that was really and truly my fault was on my first "Banker" winner, when I paid everyone 1:1 on their bets (did not take the commission). But the players took them, making it difficult to reconstruct, and I just ended up treating it as "lesson learned" and moving on. Surveillance probably made a bunch of little notes on me, with some of the mistakes that I made, including that one. Other "mistakes" I made were just areas where I was unsure of procedure; often when I didn't know the procedure for something I just did it in my own stumbling way, and I'm sure I'll hear about those eventually.

Overall, I think I did fairly okay today. One of the pit bosses said that she couldn't tell that I'd never dealt before, which was good, particularly with how nervous I was before it happened. To those who told me "you'll be fine," you were right. Fortunately, I knew you would be, and my nervousness therefore never rose to the level of a full-on panic. There were probably others for whom it did, but I never heard any stories about that.

I don't know if I might be confirming a stereotype, here, but almost none of my Asian players tipped me. Occasionally one of them would put out a 25¢ bet for the dealers, but those almost always lost. I think one Asian player put out one $5 bet for the dealer, and it lost. Most of my tokes came from the fifteen-hour guy, who would occasionally toss me $25, and probably toked $200ish while I was there.

For those who are curious about the poker room, there were two tables going when I got there (a $3/$6 limit and a $1/$2 no-limit), but the limit game broke up during the morning. When I left, there were two tables again, but I didn't see whether the second might have been another $1/$2 game (as I assume). Nobody seems to know anything about the Royal-over-quad-Fours promotion I mentioned in my last post, so that seems to be disconfirmed, wasting all of the math that I did (that probably nobody looked at anyway). Not only is the bad-beat jackpot for Holdem seeded with $44,440, but the secondary jackpot was, as well, and the primary and secondary Omaha jackpots, for about $167,000 seeded into those player jackpots. I forgot to find out what the qualifiers are; I still suspect they're rather low.

Oh, and I'm not actually the fattest: There are at least two table-games people on my shift who are fatter than I am. But whichever of them is the fattest, probably survived, too, so my headline still works.

  Wednesday, August 01, 2007

Less than Twelve Hours

Okay, I admit it: I'm nervous as hell.

I know intellectually that I have nothing to worry about. I'm not the only break-in dealer. There are in fact a whole freakin’ lot of break-in dealers on my shift. And I'm as comfortable as I could be, with the game(s) I expect to deal tonight. (Read: I'm no expert.)

I even know that whatever mistakes I make tonight will be expected, and to some degree, tolerated. I won't be the only one making them.

But I'm still nervous.

Perhaps I can attribute some of it to new-job jitters. That wouldn't quite seem to apply, though: I've been around the property for months, and I know the people I'm going to be working with. Those ordinary causes of new-job jitters don't apply.

You know, suddenly I have an insight: I know that I don't like looking foolish in front of strangers. (Weirdly, I mind less looking foolish in front of people I know.) If I am indeed dealing Baccarat tonight (I don't remember whether I posted that earlier, or not), a bunch of Vietnamese players will be strangers twice removed.

I claim some familiarity with poker players as a class. And we're generally pretty good to break-in dealers, as long as we're not paying time, and helping them out when a situation happens that's new to them. Even when a dealer, experienced or not, makes a significant mistake, then once it's corrected, we generally don't have a problem moving on. I have no reason to expect that players of other games, as a class, are any more or less tolerant or helpful than poker players. Intellectually, this should reassure me.

I might be okay after dealing two or three hands of Baccarat or after two or three rolls of the dice, reassured that I know what I'm doing.

But for now, I'm still nervous.


I've noticed something in a couple of people's reaction to learning that the poker tables at Four Winds will be electronic. Both of the people I'm thinking of are over sixty, and thus didn't grow up in the computer age, and this is probably important in an understanding of their point of view.

Both of these people assume that somebody in the poker room knows what cards everybody has. One person, a man who's probably in his early seventies, assumes that "somebody has to be running that computer" and thus that person knows what cards everyone has and is perhaps responsible for giving them their cards in the first place. Another, a woman who's probably in her early sixties, gets there by assuming that whoever programs the computer either gives everybody their cards or, at least, knows what cards they're going to get.

This is a fundamental misapprehension of how a computer works, but I don't think I could educate them in the space of a paragraph even if I was motivated to make such a speech. More, it bespeaks yet another problem with getting people to accept the PokerPro tables. My forecast that those tables will be seen to have failed in a matter of months, I think is reinforced.

I don't believe that I previously posted that the poker room's bad beat jackpot is being seeded with $50,000 (presumed), including $44,000 (known) into the main (primary) jackpot. It's a state secret what hand must be beat, which leads me to believe that the qualifier is pretty low. I'd be surprised if it's less generous than that Aces full of Tens (AAATT) must lose.

I also heard an unverified rumour last night that there is a very specific jackpot in place in the poker room: According to this rumour, if quad Fours (Four of a Kind, Fours) is beat by a Royal Flush, everybody at the table it hits on, is given one million dollars. Of course, as an employee, I'm almost certainly not eligible for this jackpot if it hits, but I have a specific concern about that jackpot. It's large enough that it encourages cheating.

I can think of at least two ways to cheat that jackpot, one of which is barely cheating and one of which is absolutely, positively cheating. First, the tables’ software runs on Windows XP, and they are controlled by some sort of wireless connection. Assuming the latter is just an ordinary WiFi network, one might be able to hack one's way onto the network. I am far from a Windows guru, but Microsoft's products are notoriously vulnerable and buggy, and it might be possible to exploit those vulnerabilities in order to modify the PokerPro software.

Second, the money is enough that it might be worth it to create a syndicate whose sole purpose is to hit that jackpot. If the stakes are low enough ($3/$6 should be fine), and players fold unless they're dealt pocket fours or suited Broadway cards, they can get a whole lot of hands out in an attempt to hit that jackpot. Some people might remember that when PartyPoker's bad beat jackpot was at $750,000, there were four tables of $3/$6 that were attempting to do exactly that (they failed). Hitting the jackpot would probably require months of play, at least, but for ten million dollars, more than a few people would be willing to do it.

Here's my attempt at working out how likely this jackpot is to hit on any specific hand. The likelihood that a given person will be dealt pocket fours is 1 in 221 (4/52 × 3/51). I therefore assume that the likelihood that the likelihood that one person at a ten-handed table will be dealt pocket fours is 10 in 221, or 1 in 22.1. (That may not be a valid assumption.) Given that those two fours are dead, the likelihood that a given person will be dealt suited Broadway cards is 1 in 30.625 (20/50 × 4/49). Given the same dubious assumption I made about the fours, the likelihood that one of the nine remaining players will have suited Broadway is 9 in 30.625, or 1 in about 3.4. Therefore the likelihood that the deal will give two players jackpot-eligible hands is about 1 in 75 (22.1 × 3.4). In some subset of those hands, more than one player will have suited Broadway, and in an even smaller subset, those hands will interfere with one another, making a jackpot impossible. I'm going to ignore all of that. It also turns out not to matter, for purpose of calculation, whether other players have "stopper" cards blocking the jackpot-eligible hands from winning the jackpot. We only need to work out how likely it is that the board will come with the five exact cards you need to hit the jackpot. For that, I get 1 in 1,712,304 (5/48 × 4/47 × 3/46 × 2/45 × 1/44). Multiply the two together (75 × 1,712,304) and you find that on any particular hand, the likelihood that you will hit the jackpot is 1 in 128,422,800. (For the paragraph below, I'm going to call that number J.)

If we were to assume that the syndicate I posited was able to get 200 hands through in an hour, playing as I suggested, it would take 642,114 hours (J / 200) to play that many hands, or about 73.3 years (642,114 / 24 / 365) of continuous play. If the syndicate filled up the 15 non-headsup tables, they could get that down to below five years, but that operation would require at least 600 people (150 players at a time, in three shifts, with another shift's worth of people for backup and days off), and ten million dollars split 600 ways means that each of those 600 people gets only $16,666, for five years’ work. Of course, it wouldn't be exactly five years. In about 2½ years of constant play, the jackpot would be 39.3% likely to hit at least once [ 1 − ( 1 − 1/J)J/2 ]. In five years, 63.2%. Make it ten years and the number becomes 86.5%. In twenty years of constant play, the jackpot would be 98.2% likely to have hit at least once. (For the later calculations just change the exponent).

The long and short of all that math is that the syndicate idea wouldn't work. So, you'll just have to hack the system.