Most of everything is crap. Most food is crap. Most movies are crap. Most books are crap. Most journalism is crap. Most web sites are crap. Most blogs are crap. Most TV and radio is crap. Most cars are crap. Most clothes are crap. Most nations are crap. Most games are crap. Most of the mail is crap. Most politicians are crap. Most magazines are crap. Most music is crap. This is just the way the world operates. Because most people are undiscerning and have no taste in what they eat, buy, wear, drive, and entertain themselves with, the market supplies mostly crap, in large volumes and at low prices.
It's the same with poker publications, both books and magazines, and, as an extension, with gambling publications generally. Most of what's out there is crap.
But sometimes I'm stuck in a poker room waiting for a table, and the only things to read are issues of the worthwhile publications that I've already read, and the crap. So I pick up the crap, and hope that maybe today it will be a little less crappy than the last time I looked at it. Hasn't happened yet, but hope springs eternal.
One of the worst rags is a dual tabloid, with "Gaming Today" in one half and "Slots Today" in the other half. They actually have columnists specializing in how to play slot machines and other casino games. I feel kind of sorry for these writers, because, really, your first installment in the series pretty much has to be, "Playing this game is like flushing your money down the toilet, except that you'll probably lose it a little bit more slowly than you would to the sewers." And then what do you say after that?
The keno columnist
The first time I perused this paper, I was particularly struck by an apparently recurring column on how to play keno, written by one Linda "L.J." Zahm. Turns out that she has also published a book of her ideas.
The other day I picked up the latest issue of GT/ST, and there she was again. She said pretty much the same things that she did the other time I read her column, at least as I recall it. I was going to try to check the online archives of her writing, but I discovered that this nasty little rag actually charges you a membership fee to look at their past stuff, even though they give the thing away at every casino in town. This I don't get: the credible, professional, helpful periodicals, like Card Player and Bluff, have free access to their past material online, but one of the lowliest, crappiest, most worthless pieces of trash in the industry thinks people will pay to see it?
Anyway, Ms. Zahm has two central tenets of playing video keno. First, the winning numbers tend to fall into clusters on the screen, so you should pick your numbers in clusters, too. Second, you're more likely to make a big win shortly after you sit down at a machine or, failing that, cash out and then log back on to one.
The machine re-setting strategy
Let's look at the second premise first. Here's how she states it in the current column:
[T]here's no secret that part of my strategy in playing video keno is the
notion of "re-setting" the game frequently, as virtually all of my big jackpots have come within the first few plays of re-setting the machine.
Once again, "re-setting" means cashing out the EZ Pay ticket and beginning over again, or going back to the starting screen menu (on the Game King machine), then returning to the game....
Oftentimes people ask why this works, if in fact it does work. First, let's look at the opposite strategy: staying on your numbers without a break, that is, playing the game as if it were a live keno game by letting the numbers "come to you."
I know this doesn't work because I've testing it several times--usually at great cost in time and mostly money!
For whatever reason, the keno machine often slips into a cycle in which the numbers simply won't "come to you," no matter what.
Thus, it makes sense that if you feel this is occurring then you should cash out and start again.
She freely admits that she doesn't know how this works or why it should work: "I simply don't know, not being privy to how the keno program works." There is not one speck of evidence or statistical analysis in the column, beyond the claim that she has tried not "resetting" the machine "several times" and has lost money by doing so. QED.
Now, I'll readily admit that, like Ms. Zahm, I have no insider information as to how video keno machines' software works. But I would readily wager my entire net worth on the proposition that her observation is pure hogwash--that is, that over a number of trials large enough to even out random statistical fluctuations, there would be absolutely no difference in a given set of numbers hitting a winning combination on the first game after sitting down (or resetting a machine) and the 100th game.
How can I be so confident? Well, look at it from the perspective of the owners of the casino. They will, within allowable regulatory parameters, set up games to reward the kind of player behavior that makes them the most money. They would much rather have a player stick his ID card in the machine, charge it up with cash, then sit there and play uninterrupted until all the money has gone into the casino's coffers. The last thing they would want is for it to be both true and widely known that a player's odds of winning are greatest within the first few games, because then customers would respond by playing a few times and walking away. If nothing else, the time that the player spends logging off then checking in again, as Ms. Zahm advises, is time that that player isn't gambling. From the point of view of the casino, that customer is wasting space during those seconds.
There is just no way that a casino would choose to purchase machines that were programmed to operate that way, which means that there is no way that a manufacturer would design them that way.
So what explains the author's perception? That's easy. Look at her own language: She tried the opposite "several times." She obviously plays a lot of video keno, which means that all except "several times" she has employed her frequent-resetting strategy. It is therefore no surprise that nearly all of her big jackpots have come within the first few games after resetting, because all (or nearly all) of her games are within the first few games after resetting! If you buy a lottery ticket every week, and always pick 2-4-6-8-10, then if you ever win, it is guaranteed that your winning numbers will be 2-4-6-8-10. It will have nothing to do with whether that particular combination is really more likely to hit than any other, and everything to do with the fact that that's all you ever tried.
If Ms. Zahm had an identical twin in a parallel universe who, for whatever reason, stumbled upon the opposite strategy, she would write her book and columns to encourage people to stick it out for long sessions, having learned from experience that virtually all of her big jackpots came that way. She would write that she had tried the frequent-resetting strategy "several times" and it never worked for her. And her claim would be just as valid as Ms. Zahm's is. In fact, although it's unlikely that the machines have any built-in bias for or against picking winners based on duration of play, just thinking about how casinos want to reinforce players' behavior, you'd have to guess, in the absence of any hard data, that the alternate-universe columnist is at least slightly more likely to be correct.
The cluster theory
What about the central tenet of the Gospel According to Zahm, that of playing keno numbers in clusters? Well, this, too, is clearly just a figment of her imagination.
It's certainly true that if you look at a typical completed keno board (traditional or video), you can spot several clusters of numbers. To understand why, just think about what would have to happen for that not to be the case: The numbers would have to be quite evenly distributed. Even distributions are statistically rare. Randomness is "clumpy" by its very nature. Let me give you three examples of this phenomenon from three completely different fields.
First, a while back I took the time to figure out how many different possible Texas Hold'em boards of five cards did not make a possible straight for the lucky player holding the right two hole cards: http://pokergrump.blogspot.com/2007/10/doyle-brunson-is-wrong.html. I discovered that of all the full, unpaired boards, 1206/1287, or 93.7%, have a straight possible, given the right hole cards. That is, the great majority of the time the cards are "clustered" closely enough together in rank that you can make a straight. It is quite difficult for a random shuffle to produce the relatively few combinations of cards in which the spacing is so dispersed that no straight is possible.
Second, people are always reporting to various government health agencies their fears of a cancer "cluster." Typically this happens when somebody becomes aware of, say, three kids on one block all developing leukemia, or a few co-workers all being diagnosed with brain tumors in some relatively short period of time. But, again, that's just how randomness is--it doesn't spread things evenly. Random things tend to have more clumps or clusters than one would be inclined to think. There are lots of governmentally sponsored web pages that try to explain this to people so they won't leap to the conclusion that there's some scary external force that is behind the cancer clusters--for example, this one: http://www.cancer.gov/cancertopics/factsheet/Risk/clusters. It doesn't help much, though. People see a cluster and immediately convince themselves that it's the electrical lines or the flu vaccine or something in the water or the nuclear power plant a mile away that's the culprit. It isn't, the vast majority of the time.
Finally, consider why people make very poor random number generators. If you ask people to write down a long string of random digits, the result will almost always fail statistical tests for randomness, and almost always for the same reason: there aren't enough clusters. That's because a person will think that something like 4-4-4-4 is too obviously non-random, so even if he puts down "4" twice in a row, he's unlikely to write it a third and fourth time. But real random digit strings are chock-full of sequences that look distinctly non-random. Of course, over several million digits, the occurrence of any particular suspicious-looking string won't be any more than would be predicted. But there are lots and lots of digit sequences that, in isolation, look artificial. The human digit-generator will tend to filter those out, and the results are much more "smooth," statisically, than actual randomness produces. The human thinks, "Hmmm, I haven't written a "9" in quite a while now," so he throws one in. The true random number generator has no memory for what has come before, and nothing in it that is trying to "look" random, so it has no bias to spit out a "9" just because there hasn't been one in the last 20 or 30 digits.
All of which is a long way of getting to this point: When Ms. Zahm sees clusters of numbers on a keno board, they really are there. But the knowledge that they tend to come in clusters is of exactly zero helpfulness in predicting the winners.
Everybody who has played poker even once has seen that some players go on streaks of unusually good luck, while others go on streaks of unusually bad luck. Again, that's the "clumpiness" of statistical variance at work. But knowing that that will happen doesn't help you. Unless you know in advance which seat at the table is going to be hitting the lucky streak, and exactly when that streak will begin and when it will end, the theoretical knowledge that streaks will occur is useless.
It is precisely the same for Ms. Zahm's cluster-keno theory. Her telling you to pick numbers in clusters is just as pointless as if she told you, before sitting down in a poker game, "Be sure to sit down in the seat that is about to get hit by the deck." Gee, thanks for the hint!
Until Ms. Zahm figures out a way to tell you exactly which clusters, of all of the hundreds of possible clusters one could define on a keno board or card, are going to be the ones that hit, her observation is meaningless. I think it's safe to say that if Ms. Zahm possessed this information, she would (1) be so rich that she wouldn't need the income from writing for a cheap, pathetic tabloid, and (2) she would keep that secret very, very quiet.
I wonder if she keeps complete and detailed records as to her wins and losses. It would be interesting to see if she is actually a winning keno player over the long run. I kind of doubt it.
Wrapping up
Well, to my chagrin, I see that I have rambled unforgivably here. This post is already about twice as long as I had envisioned it, and I haven't even gotten yet to the main point suggested in the title! That's a clue to end it here, and start a new entry, I think.
Posts
No comments:
Post a Comment