Video Poker Trucks Run Over Everybody

[Alan Mendelson]

Help me to understand your position, because I can't visualize your bell.

I think of the axes as having increasing denominations. And in video poker, most players would fit the normal range of expected return and only a few would be "lucky" and hit the royals. So again, I think of this: _________/ if the vertical axis is the number of credits and the horizontal axis is the number of players.

You believe in the "math" of the game, so why would there be a bell curve? Why wouldnt the universe of video poker players just chart out as the expected return which is actually a flat line?

[Vegas_lover]

You believe in the "math" of the game, so why would there be a bell curve? Why wouldnt the universe of video poker players just chart out as the expected return which is actually a flat line?

Because not every player will always get the same results at every session. In other words, in a single session a player will not always achieve the expected return of the game. Some players will under perform, others will over perform. That's why there's no flat line. A bell curve is a very common thing in the world of math. It's not something specifically designed for plotting vp results and that's why you don't hear it mentioned a whole lot in vp related discussion. But at the same time, a bell curve is very convenient to plot vp results.

[arcimede$]

Help me to understand your position, because I can't visualize your bell.

I think of the axes as having increasing denominations. And in video poker, most players would fit the normal range of expected return and only a few would be "lucky" and hit the royals. So again, I think of this: _________/ if the vertical axis is the number of credits and the horizontal axis is the number of players.

You believe in the "math" of the game, so why would there be a bell curve? Why wouldnt the universe of video poker players just chart out as the expected return which is actually a flat line?

The y-axis is the number of players. The x-axis is the random variable. Could be credits won, % return or anything dependent on the randomness of the game. It looks like this :

............____
...........|......|
....____|......|____
__|.....................|__

The left bar might represent one player with a 90% return, the next bar represents 10 players with a 95% return and so forth. Way at the end you might have one player with a 110% return.

Edit: It looked good when I built it. The software must compress out blanks. I added some dots.

[Vegas_lover]

The y-axis is the number of players. The x-axis is the random variable. Could be credits won, % return or anything dependent on the randomness of the game. It looks like this :

............____
...........|......|
....____|......|____
__|.....................|__

The left bar might represent one player with a 90% return, the next bar represents 10 players with a 95% return and so forth. Way at the end you might have one player with a 110% return.

Edit: It looked good when I built it. The software must compress out blanks. I added some dots.

Arci, I didn't read the entire thing but did you already explain the basics of standard deviation?

[Alan Mendelson]

The y-axis is the number of players. The x-axis is the random variable. Could be credits won, % return or anything dependent on the randomness of the game.

Whoa. You are now talking something that is way out in left field for me. When I am talking about a graph of video poker results I am looking at two things:

1. number of players
2. cash they win

you lost me with random variable. Can you deposit a random variable in the bank? I am going to have bid adieu from this discussion.
I just don't think that the "winings" of players fits a bell curve. When you start talking random variables you lost me.

[arcimede$]

Whoa. You are now talking something that is way out in left field for me. When I am talking about a graph of video poker results I am looking at two things:

1. number of players
2. cash they win

you lost me with random variable. Can you deposit a random variable in the bank? I am going to have bid adieu from this discussion.
I just don't think that the "winings" of players fits a bell curve. When you start talking random variables you lost me.

Yes, the amount of credits won (not winnings) will fit a bell curve if all the players play the same amount. Or, you can divide by the number of hands played to normalize their results.

You just can't look at "cash" by itself. Maybe that's your problem. You have to be looking at apples to apples. If one player plays a million hands and another player plays 10 hands I suspect you can guess which one will win/lose more money most of the time. That's why you need to do something to make the players results comparable.

[arcimede$]

Arci, I didn't read the entire thing but did you already explain the basics of standard deviation?

I've tried to keep it as simple as possible. And, you can see how difficult it gets even then. Anyone can look up standard deviation on wiki if they are interested.

[Alan Mendelson]

Yes, the amount of credits won (not winnings) will fit a bell curve if all the players play the same amount.

Arc, what is the difference between "the amount of credits won" and winnings? Are you talking about "intra session wins" and not cash outs?

Let's discuss the cash a player has -- or doesn't have -- at the end of their session or trip.

Arc, if you just want to focus on this one statement, I will discuss it. My feeling is that most video poker at the end of their trips will flatline at "zero" for no money won. And a tiny few will leave the casino with a profit. That "curve" would look like this: _________________________/

I don't have any research to back it up. It's just from what I have observed plus the data that shows 95% of casino goers lose money.

So going back to my original request: if you have data that supports your belief of a bell curve, I'm ready to see it.

But do me this favor: plot out what you think are the values of the bell curve you are talking about. I would like to see that. And if you like, just concentrate on $1 players to use round numbers. I would be very interested to see the values in your bell curve. Thanks.

[arcimede$]

Alan, nothing I've said disputes the fact that 95% (or more) players lose. As I've already stated several times the bell curve could be almost entirely in negative territory. It does not by definition provide a view of winners and losers. You really need to try and understand the subject before continually making comments that make no sense.

Now, assume the x-axis is coin-out/coin-in. That would be the actual return for the players being viewed. The y-axis is the number of players at any particular return. For anyone to make money the coin-out/coin-in would have to be greater than 1. Only the right hand tail of the bell curve would have values at 1 or greater. Most of the players would be below 1, probably at least 95%. Take a look at this bell curve.

http://en.wikipedia.org/wiki/File:St...on_diagram.svg

Where you see the 2 sigma value is about where I would expect a value of 1 to be found. Most of the players would be losers and hence their coin-out/coin-in < 1.

Understand?

[Alan Mendelson]

Okay, now you're saying that the bell curve would even accommodate the 95% of video poker players who lose. Gee do you really need a bell curve to express that 95% of video poker players go bust? "Going Bust" to me is a flat line. Whether you lose $10 on your trip or $10,000 on your trip it's still a zero return. So, which part of the bell curve for 95% of the video poker players is "zero"??

Is this the bell curve you are offering us? ______________________________/\___________________ Is that it? That's about "zero" for 95% of the players.

[arcimede$]

Okay, now you're saying that the bell curve would even accommodate the 95% of video poker players who lose. Gee do you really need a bell curve to express that 95% of video poker players go bust?

No, as I said before, the bell curve is a nice tool to help in understanding how random events are distributed. With my previous example, the area under any part of the curve represents the players at any particular return. Those under the far right hand corner present those who win ... the 5%. It's nice because it shows several things with one chart. Not only can you see the winners, you can see how the losers are distributed.

"Going Bust" to me is a flat line. Whether you lose $10 on your trip or $10,000 on your trip it's still a zero return. So, which part of the bell curve for 95% of the video poker players is "zero"??

I suspect I'd find a big difference between losing $10 and $10K. So would most people. The bell curve shows the likelihood of both situations.

Is this the bell curve you are offering us? ______________________________/\___________________ Is that it? That's about "zero" for 95% of the players.

Go look at the reference I provided above. Now, think about the area under the curve. That is the totality of all the players being considered. The ones on the right are the winners (notice there are very few), the ones on the left are the big losers (also very few). The ones in the middle are the ones who came close to the expected return of their play.

[Alan Mendelson]

Very good Arc, I understand everything you say here:

Go look at the reference I provided above. Now, think about the area under the curve. That is the totality of all the players being considered. The ones on the right are the winners (notice there are very few), the ones on the left are the big losers (also very few). The ones in the middle are the ones who came close to the expected return of their play.

Now all you have to do is show me the research that backs this up.

[arcimede$]

Now all you have to do is show me the research that backs this up.

Alan, it's all well established, basic mathematical probability. It doesn't require any research. All you need to do is read the Wiki references I've provided.

You appear to be trying to make more of this than it warrants. No one is claiming exactly 12,472 people are winners and 195,459 are losers. The bell curve is a tool where you can visually see the results of a probability distribution. The probability distribution would be based on the games being played and their associated returns along with the skill level of the individuals playing.

This allows to make predictions without really knowing every detail. Those predictions won't be exact but they will be better than guesses. And, we know the shape of the distribution will be a bell. This is the type of computation a casino might do to make forecasts of their profits in the future based on a particular installed base of machines. They could then modify a forecast with a different set of machines with higher or lower returns and associated changes in their customer base.

[Alan Mendelson]

Alan, it's all well established, basic mathematical probability.

Here's the problem I have with your "basic mathematical probability." Your probability is based on random results. In video poker, players must decide which cards to choose, and how much they are willing to lose, or how long they will keep playing when they win. This is a lot different from coin tosses and throwing dice or scattering of pebbles dropped from ten feet onto a target painted on the floor.

I don't think your math accounts for these decisions made by players... unless you want to say that the decisions made by players are also just random.

Again, I've always believed that statistics should be developed from research. Some day someone will survey video poker players to find out just how they do play: how many bust out, how much they lose, when they cash out when they win, and how much they win when they do cash out. Then we might see a real bell curve-- if there is one.

[arcimede$]

Here's the problem I have with your "basic mathematical probability." Your probability is based on random results. In video poker, players must decide which cards to choose, and how much they are willing to lose, or how long they will keep playing when they win. This is a lot different from coin tosses and throwing dice or scattering of pebbles dropped from ten feet onto a target painted on the floor.

I don't think your math accounts for these decisions made by players... unless you want to say that the decisions made by players are also just random.

Again, I've always believed that statistics should be developed from research. Some day someone will survey video poker players to find out just how they do play: how many bust out, how much they lose, when they cash out when they win, and how much they win when they do cash out. Then we might see a real bell curve-- if there is one.

All the choices do is skew the bell curve left or right and possibly widen the middle. It doesn't change it from a bell curve.

[Vegas_lover]

All the choices do is skew the bell curve left or right and possibly widen the middle. It doesn't change it from a bell curve.

Again, exactly correct. Alan, a large part of the reason why you and Arci keep on talking different languages is because of your lack of basic math skills. That's not an attack on you it just is what it is. So that's why you are probably not the right person to evluate Singer's system because you don't see how far away Singer's claims are from what the math says.

Math research has been done for over a 100 years. VP is based on math, there's nothing new that has to be determened as far as VP goes. That's exactly why math experts will ALWAYS contradict Singer's claims. Now, you might think Arci is an obnoxious jerk because he sometimes attacks you but that doesn't mean his statements aren't correct. Some aspects of the things he tries to explain are just too far above your head and that can be frustrating for both parties.

Personally, I don't see this discussion going anywhere. After a number of pages filled with posts your opinions don't seem to have come any closer.

[Alan Mendelson]

First of all, I respect Arc. Secondly, I know what a bell curve is. And there is a difference between a bell curve and all other charts with various kinds of slopes. All I am saying is "show me." Show me the facts that justify a bell curve. That's all.

[Vegas_lover]

All I am saying is "show me." Show me the facts that justify a bell curve. That's all.

The fact that VP is a math based game justifies the use of a bell curve to plot results. There's no other facts that have to be shown. It is what it is, a basic mathimatical approach of charting results.

That's what Arci has been telling you, there is no need to prove anything else.

[Alan Mendelson]

No, because I am telling you that the results of video poker players (and that is the bell curve I am talking about) is not purely math based. The results of video poker players (make sure we are talking about the same thing) will vary with the decisions made by the players. Want an example? Here is it is: Rob Singer. Rob Singer wins on negative expectation games. He violates the math.

I am not disagreeing with the math. I'm saying that the results of video poker players will differ from the math.

Now, if you can show me that that difference fits a bell curve I will accept it. But you can't tell me that non random video poker play fits a bell curve automatically.

[Alan Mendelson]

Let me just add this, Arc and others: I don't care if there is a bell curve or not. It is meaningless to me, and probably to everyone who plays video poker or any other game. But the point is that your adherence to strict math prevents you from understanding how Singer could possibly win. That's all.