And which app was that? I'm only aware of the app(s) that aided card counting, which is very obviously illegal under the statute that I referenced.Originally Posted by Norm
My app does nothing of the sort.
So, please let me know of the app that does anything similar to what mine does that was covered in the news and pulled by Apple... or I would suggest that you start doing a little research before posting.
Please do not argue here, in a public forum, from which novices take advice, that it would be perfectly legal to use a computer program on your phone to calculate probabilities for the various hands you receive in a blackjack game. In fact, doing so would very likely lead to your jailing, at least in Las Vegas.
In the meantime, without entering the value of every card as it falls, thereby inputting the exact deck composition at the time of play, your program isn't going to tell the proficient counter to do anything different from what he already knows to do, with respect to any holding, simply by applying his count to the situation. Knowing, in addition to the correct play, whether it is 2%, 4%, or 6% better than the next-best alternative may be of anecdotal interest, but it is not exploitable information at the moment of making the correct play. The bet is already made; there is no further gain from knowing your precise edge at the moment you make your decision.
Don
You're absolutely right, my recounting of the story was ambiguous and probably confusing. My friend did not ask me about the hand in order to make a decision. In fact, he asked me about it long after he had played the hand. We weren't even at a blackjack table at the time, just walking around the casino. I would never suggest that people try to use my app to make decisions at a casino and in fact the app has a "How to play blackjack" file that says that you aren't allowed to use cell phones or other electronic devices at or near a blackjack table.
I think there is a lot of confusion on this thread because I am not trying to sell a "gambling system" and I'm not trying to say that my app will help advantage players make better decisions or win more money or anything like that.
I freely admit that I made the app only because I think it's fun and interesting to see EVs. If you aren't interested in EVs, then I guess don't buy the app.
But I never claimed that my app produced any novel or unique or exploitable information so asking me to defend it on that basis is making me kind of depressed.
Seriously, all I wanted to do was offer something amusing to the community for a few dollars.
Actually, I normally run billions when it comes to a single decision. And, that's not billions of random hands, as some generators think are useful. It is billions of hands specific to the decision. Single decisions require far more hands than overall EV, and the lowest accuracy setting in my generator requires far more than 100K hands. (Actually, the number of hands varies dynamically by decisions depending on the standard error, convergence rate, and another factor kind of difficult to explain.) I've made those comparisons more times than I can remember.
The fact that you have almost a couple decades in technology is good. The first Monte Carlo sim I wrote was a heuristic checker program in 1965 on an IBM 7040 mainframe. I know you have knowledge of chess applications. I won't lecture you on chess apps. Try to avoid lecturing people here on blackjack.
"I don't think outside the box; I think of what I can do with the box." - Henri Matisse
My point (and curiosity) about the importance of EV accuracy was hardly a "lecture."
If you disagree with my point then maybe you could provide a modicum of information as to why.
If you've really "made those comparisons more times than [you] can remember" then maybe you can share some data about just one of those times?
Read my books. Or, Don's. Or, many, many others. Or so many, many posts on forums. As I said, read before you write. Nothing snide about that. If you enter a site and start by saying you have something "new" and go on to provide dangerous legal advice, and assume that no one else understands the impact of blackjack statistics, when you have stated you don't understand standard error, you are going to get pushback.
I wouldn't go on a chess site and base a post on the 1911 version of the MCO and get defensive or snarky when people in the field pointed out my errors.
"I don't think outside the box; I think of what I can do with the box." - Henri Matisse
BTW -- I reworked some of my code to do this "experiment." For every player decision, I ran simulations with 100k iterations and then 10M iterations.
I kept count of how many times the 100k EVs indicated the same action as the 10M EVs.
Then I played my game for 200 decisions. At no point did the actions disagree.
Admittedly this is a small sample size and I expect your software is better equipped to measure the same thing for many more decisions, probably automatically. But so far my data does seem to indicate that it's not really a big deal if your EVs are a little inaccurate, as I expected. I would be happy to see data that indicates otherwise though.
AGAIN, please read. I have pointed out twice in this thread that a poor RNG will act EXACTLY as you have just stated. You have again proved, without a doubt, that you are using an inadequate RNG.
"I don't think outside the box; I think of what I can do with the box." - Henri Matisse
Except that you haven't pointed out any of my errors, have you?
You said that my convergence rate was impossible but you literally provided no explanation about how or why it's impossible or what rate you would expect, i.e., what degree of accuracy do you expect if you do a blackjack simulation for 100k iterations? I've provided all the numbers and data that anybody has asked for on this thread, and you have provided literally nothing other than just saying it's impossible.
You also spent a lot of time pointing out all the weaknesses of a random number generator that I'm not using.
You told me that using my app at a casino is illegal but you clearly didn't read the law in question and your only evidence that it's illegal is that a different app that does a different thing is illegal. (Sort of like the random number generator thing all over again.)
And you're also saying that EVs that are "off a tad" is "really bad news" without doing any sort of quantification of that statement. What do you consider "bad"? Because I just did an experiment that EVs that are "off a tad" still almost always indicate the same decisions. You have brought nothing to the table other than saying you've been doing this a long time.
No, you quoted a Wikipedia article about a PRNG that's almost completely unrelated.
Where are your numbers Norm? How often do you expect 100k simulations to indicate different decisions than 10M simulations? Tell us your data or do an experiment or something.
Telling people how much experience you have doing something will only get you so far. At some point you really have to do that thing.
WOW. You haven't learned anything at all.
I am closing this thread because he is still suggesting that it would be legal to use this app in a casino, which exposes this site to liability problems. He can start a new thread in the Disadvantage Forum, after he has studied standard error and agrees that it would be illegal to use this app in a casino. He is NOT banned.
"I don't think outside the box; I think of what I can do with the box." - Henri Matisse
Posting Permissions
-*-
Bookmarks