additional information for each game

[jim18]

WebmanCurrently, I am spending all my time playing various free games that you offer.  I do make an effort to get into the top ten scores for each of the games that I play.  Many times this never happens.While looking at the daily scores of one game it dawned on me that out of the top ten scores for that day, nine of them were negative numbers.  On the surface this suggests to me that this is a game that I should avoid playing if I encounter it at a casino.Building on that observation, I thought that it would be helpful to all of us at VP.com if we could see some sort of  average score for each game.(not just the all time best score)  To be meaningful that average score would be based not on today’s play, but a culmative average.One of the shortcomings of the top scores is that I can only see the top ten scores.  Why can’t I see all of the scores for that day, for each game?What is the purpose of seeing the information that I speak about?  Well, I think that what would happen is that I would seek out those games where 50% or more of the players were able to post a positive score.  And I would avoid playing those games at the casino where more than 50 % of players posted a negative score.In today’s world, games are rated based on a pay back schedule.  For example, 9/6 JOB has a payback percentage of 99.5%.  However, for both the recreational gamble and the serious gambler, would it not be of more value to see which games have the highest likelihood of producing a winning session for the player?I would like to know if I am better off playing 100 hand super times pay or better off playing 100 hand quick quads.  Even if the payback schedules were identical, I am almost sure that chances of leaving such  machines as a winner, are not equal.  And I believe that if VP.com were willing to add an average score just under where it has posted the all time best daily score, that this number would be useful to VP players.Is this calculation something that VP.com can do?Is this information something that other VP players want to see?

[Webman]

Well, there are a lot of things to address in this topic.First, I understand where you are coming from, and I think some more statistics would be useful, as long as they are portrayed in the proper context. We'll have to give this some thought.I'll try to address some of your other points, though I admit I will probably fall short of a complete explanation. Maybe some others will chime in and share their thoughts as well.[quote]In today’s world, games are rated based on a pay back schedule.  For
example, 9/6 JOB has a payback percentage of 99.5%.  However, for both
the recreational gamble and the serious gambler, would it not be of more
value to see which games have the highest likelihood of producing a
winning session for the player?[/quote]I believe the answer is that a % return is still the best way to "measure" a game with a single number, and here's why... measuring the likelihood of a winning session with a single number is, simply put, impossible. There are too many other variables that would impact this number: How many coins are you betting, what is your bankroll, how long will you play in this session, when will you determine when you stop? These things can only be evaluated on a personal level. While one could standardize each of these variables and then run the necessary calculations to arrive at a % of expected winning sessions, they would not be particularly useful for anyone else unless they did the exact same thing.  Further, and this is probably the most important thing that gets overlooked: What if you are winning more sessions, but your losing sessions more than offset your wins? Is it still a better game than one in which you only win 25% of the time but have a net return over 100%?[quote]Currently, I am spending all my time playing various free games that you
offer.  I do make an effort to get into the top ten scores for each of
the games that I play.  Many times this never happens.[/quote]I think this is to be expected with the number of players, and the games where you can more easily reach the top ten scores would likely be the ones that fewer people are playing on a given day.[quote]While looking at the daily scores of one game it dawned on me that out
of the top ten scores for that day, nine of them were negative numbers. 
On the surface this suggests to me that this is a game that I should
avoid playing if I encounter it at a casino.[/quote]On the surface, I agree with you. But it lacks a lot of factors. Any game that has mostly negative scores in the top ten is clearly a game that hasn't been played much that day. There are a several games on the site that this is the case for most of the time, since certain games tend to be more popular and get much more play.So, what factors might need to be considered... first, how much have people played that day. Did the players have any idea how to play that game? What strategy are they using? Did they play long enough to get a quad or jackpot that might be necessary for a winning session?Based on these questions, I do think the following information would be useful to display on the site somehow: 1) How many players (gold members) have played the game that day 2) How many hands did they play? Those questions would help you better assess whether the scores you are looking at are a fair sample to make any conclusions, and also how you personally stack up for the day.[quote]Building on that observation, I thought that it would be helpful to all
of us at VP.com if we could see some sort of  average score for each
game.(not just the all time best score)  To be meaningful that average
score would be based not on today’s play, but a culmative average.[/quote]Here is one reason why we decided that would not be a good idea... it isn't a good measure for you to go by in determining whether the game is a good game for you to play personally. I can tell you that the games on the site return about 1-4% less than the mathematically perfect return for each game. The reasons for this should be obvious... imperfect strategy or lack of skill, risk-taking strategy in order to reach a possible top score, and experimentation. On top of this, any % listed would account for a blend of all possible pay tables, and would also be limited to the pay tables offered on the site. In assessing whether you should play a game in the casino, some factors you may consider are:1) How well do you know the game 2) What are the pay tables, and the expected returns 3) What are your objectives for that casino visit, including your enjoyment, odds of winning, and whether this visit is occasional or something you will repeat frequentlyHow these things relate to the games on this site will vary greatly for each player and like I said, looking at other players with the huge variety of play isn't the best way to determine whether a game is right for you. The best thing to do is have a detailed analysis of each game based on your own objectives, strategy, and bankroll. If you care to share some of these things and would like to know the best types of games to play, I'm sure some of our more knowledgeable players would be happy to offer some advice.[quote]What is the purpose of seeing the information that I speak about?  Well,
I think that what would happen is that I would seek out those games
where 50% or more of the players were able to post a positive score. 
And I would avoid playing those games at the casino where more than 50 %
of players posted a negative score.[/quote]Again, this is another reason why sharing more information may actually do people a disservice unless it can somehow be given proper contest. Expectations. That's why discussing such things in the forum can actually be far more valuable than a number.You mention that you would avoid any games where 50% of the sessions are negative. First, consider all the factors mentioned above as we define a session... how many hands did each person play, was it enough to hit any good hands with reasonable expectation, and is that player any good? In the casino environment, add on variables such as bankroll, denomination, and profit objectives. If 100 people play a game in a casino and 60 people win $10, while the rest lose more than $100 each, is it a good game? Conversely, if only 20% of the people win, with the losers each losing $100 but the winners winning as much as $25,000 is it a good game? To make that decision more honestly, you need to look at expected returns and variance (volatility.) This is why those are the two numbers that the experts cite regularly in evaluating a game.A simple percentage of winners, or odds of winning an individual session is inadequate end even deceiving when it comes to the money at stake.Second... your expectations need to be reviewed as it pertains to winning percentages. I will give a specific example: 9/6 Jacks or Better, single line, playing quarters with a $200 bankroll for 500 hands. This is a fairly basic example that someone might go into a casino with.  Good game, 9/6. Do you expect to win at least half the time? The expectation of a winning session is actually closer to 37.22%. So how can it be a 99.5% return game? This comes back to my points above. The most you could lose is only $200, but the most you can win is much, much more. Again, return percentage and variance are the key.These things combined demonstrate why looking at the top scores, or even all scores for a day will not give you a good picture of the nature of the game. A mathematical analysis of the game based on your own parameters is the only thing that can do this with great accuracy. Advanced software such as Video Poker for Winners can calculate this for you.I'm sure I missed some points but I do need to move on this morning after such a long post. I didn't know what length I was getting myself into when I started typing.  As always, thanks for the feedback and I hope some of this information is somehow useful to you and others.

[New2vp]

Second... your expectations need to be reviewed as it pertains to winning percentages. I will give a specific example: 9/6 Jacks or Better, single line, playing quarters with a $200 bankroll for 500 hands. This is a fairly basic example that someone might go into a casino with.  Good game, 9/6. Do you expect to win at least half the time? The expectation of a winning session is actually closer to 37.22%. So how can it be a 99.5% return game? This comes back to my points above. The most you could lose is only $200, but the most you can win is much, much more. Again, return percentage and variance are the key.Excellent post, Webman!!  Maybe you should do a weekly column.The portion that I excerpted above might be good info for some to have on games, given some standard bankroll and number of hands as given in your assumptions.  I say that because to many it might be more instructive info than EV and variance (or at least a helpful supplement).  As you stated VPFW gives that info on many games.Nothing more to add.  I just wanted to have a post of mine in a thread where others' posts are significantly longer for a change.

[faygo]

[QUOTE=Webman.[QUOTE)   Nothing more to add.  I just wanted to have a post of mine in a thread where others' posts are significantly longer for a change.



 

[royal flush]

video poker has a payout that to hit a winning session it is most likely that a player needs a royal flush orother major hit or the player has a losing session but the money usually won on good sessions exceeds the money lost per losing session example 9/6 job 3 sessions lose 200 dollars each session 4 hit royal flush win net $800. 25% session win but net $200 net win for 4 sessions

[jim18]

WebmanThanks for taking so much of your time to read and respond to my posting.  Generally speaking I think that you have made good points as to why my suggestions will not produce pure statistical accurate information.Let me try to clarify a few points just to be sure that we are on the same page.  First, I realize that using daily scores is not a valid way to proceed.  However, by observation, daily scores are all that your web site is allowing me to see.  Second, I realize that a hundred hand game should have a much larger volatilely than a ten hand game would produce.I think that if we could match up games with similar volatility on this site that we would expect that the results would be similar, if those games were played using all the variables that go into determining outcomes.  To illustrate, if a player is not good at 100 hand draw poker, then that player will not be good at 100 hand quick quads.  So the effect on statistics will be neutralized.In your response you have included many variables that are certainly a part of what determines outcome.  And you conclude that with so many variables, it would not be worthwhile to compile statistics on the games.  I do not agree with your conclusion, only because you have included many factors such as size of bankroll or denomination which have nothing to do with the results that are occurring on your website.  In short, you have taken a complex subject and turned it into a puzzle that is impossible to solve.Perhaps I need to read a book to better understand  my quest.  However, I am dealing with your website, and trying to maximize my time on your site, and at the same time have fun playing games.  I want to play games on your site that I have a reasonable chance of doing well at.  For example, using today’s data, 50 play draw poker has all the top ten scores in the positive.  While 50 play quick quads has only the top three scores in the positive. One day’s results do not tell the story fairly.  But, I would start with the assumption that these games would have similar volatility.  Since players on your site are doing better at one game than the other, I might conclude that these games are not so similar.  There may be sound and identifiable reasons why these two games would produce different results.  But for the purpose of comparing these games on your site, we can not look at bankroll, or skill level.  And we have to overlook luck.Thank you for accepting that some additional statistics may be worthwhile adding to your site.  That, is a big first step.You mention that it would be useful if we could see how many gold members played a game, and how many hands they played.  I think that this information would be of some help to me.  Still, daily results are a far cry from the long term results that I am seeking stats on.You mention that statistics of how others have done in a game may not be a good measure of how I would do in this game.  Perhaps that is a fair point.  Then the conversation should be on how to add statistics that deal with the individual player, so that I can see how I am doing at any game over a period of time.You mention that games on this site are running 1 - 4 % less than mathematically perfect play.  How would you know this, unless your site is already capturing this information?  You say that some information would do us a disservice, so for our protection you are not sharing this information.  That sounds so much like what Senator Reid of NV would say, because he believes that people need to be protected by big government!Anyway, thanks for reading this post.  I look forward to seeing what changes occur on this site as a result of my original post.

[Webman]

Thanks for the reply. Just to clarify two points... knowing how many people had played 50 play QQ and how many had played 50 play might help demonstrate why only 3 players had positive scores on QQ with 10+ having positive scores on 50 play. What if there were 25 times as many people playing the normal version than the QQ version. Then it makes sense, even if the games have a similar performance. So I do think more data is a good thing.As for not sharing information to "protect" the user, that isn't what I intended. What I meant was that more information is only better if it can be provided in context. So rather than just showing a number along with the game, say 30% winning sessions, I think those numbers are best shared and analyzed in a place such as the forum, where the context can be discussed. If you have questions about the analysis of a particular game, I suggest starting a topic about it, or ask in the topic relevant to that game (most individual games have their own topic already). I'm sure people would be happy to oblige with some good information.The 1-4% range is based on times we have looked up the return for a game manually in a given day, week or month to see how the games are performing. But we frequently purge data (there are a lot of hands played on the site in the long run) so overall numbers are not retained except in backup that is not easily dealt with. We could show the return on a given day, but again I think it would do the casual player a disservice if they looked at a game with very few players, very little play  and saw a 70% return, coming to the incorrect conclusion that a game that is actually a 98-99% return game is 70% and giving up on it, if they didn't look at better data over time.Thanks again for the comments.