There are times I've stayed on 16 against a 6 and the guy beside me hits his 10 and gets a 5, sure I'd love to have that 5, but hitting a 16 against a 6 will lose more times than win. You can also double down on a 10 and draw a 2. Crap happens.

I've seen the opposite where newbs at a table double a 12. Then draw a 9. Terrible basic strategy, but sometimes the luck happens.

If you play a game, and play it right long enough, you will be pretty dang close to the house edge.

If you enjoy the game, can afford to lose the house edge as entertainment expense, and can find promos good enough to tilt the -EV a little more your way, I say do it.

This is the non-card counters guide to gambling and blackjack and is about as close to the truth as you will find. No smoke and mirrors. Real life stuff.

Quote:kewljQuote:billryan

He said he played on several sites, including the Wizards. I don't think there is any question that site is legit. 99% of the time when new players get worse results than expected, it is their play that is sub-optimal, not the sites.

Billryan, it is going to take one heck of a lot of mistakes to go from a player disadvantage of 0.5% to 15%.

I mean what are the most common basic strategy mistakes? Standing 16 vs 10 instead of hitting. Standing 12 vs 2 and 3 instead of hitting. Standing soft 18 vs 9 or 10 instead of hitting. Maybe standing soft 18 vs 5 or 6 instead of doubling.

If you were to make these plays incorrectly each and every time, plus throw in a few more even, the house advantage is still less that 1%. He is talking about a house advantage of 15% over his 1000 hands. A few (or even many) basic strategy mistakes just is not going to account for that.

Probably it is just bad variance over a small sample size, but as somebody pointed out, that is 4 standard deviations. That is pretty high. I personally thought it might be lower than that like in the 3 standard deviations area, but I trust the 4 standard deviations number.

You are either talking really bad variance over a small sample size, or something isn't legit. If a guy says he is playing good basic strategy, a few mistakes, would account for nothing.

link to original post

I'd say it's 100% bad variance or a shady site. If it's not regulated I'd be cautious.

I'm also curious if he is flat betting, chasing losses, etc... If you are all out martingaling, your variance from EV can turn greatly negative quickly especially on a small sample size of hands.

Quote:mwalz9

I'm also curious if he is flat betting, chasing losses, etc... If you are all out martingaling, your variance from EV can turn greatly negative quickly especially on a small sample size of hands.

link to original post

You know what? Good question about flat betting. He didn't actually say flat betting, but I assumed and yes I know about "assuming".

Here is the thing about flat betting and using a spread. If a player is using a spread, his total results are pretty much going to be reflective of what occurs when his bigger bets are out. And if you are placing those larger bets by advantage as in card counting the total number of larger or max bets is going to be small in comparison to total rounds played. It is what happens at those large or max bets that will determine or VERY strongly influence your total results. So because there larger bets occur so infrequently maybe 1 out of every 10, 12 hands, it takes a card counter 40, 50, 60 thousand rounds to have accumulated enough large bet rounds or max bet round to begin to be significant. That is why absent a large number of trials results are deemed almost insignificant in card counting.

Now a player flat betting, doesn't need those 40, 50, 60 thousand trials to be significant. 1000 is getting there. 1000 is still small enough that results can be very skewed, 2, 3 even 4 (as posted here) standard deviations from expectation. I would like to see more trials, 3000, 4000 for flat betting, but 1000 is starting to see to get to the point that you can see something and what we are seeing is a 15% loss rate is just not right.

Quote:Jack2022but I've been discouraged by seeing similar patterns for example after 150 hands (tiny sample I know) @ 1$ each on the Wizard's trainer I'm $30 down using only the recommended play.

I just played 100 rounds on Wizard's trainer. Down $7. That is 7% but better than your 20%. lol

But these really are too small a sample size to mean much.

I know this isn't going to help you. YOU have to convince yourself, but I want to share an experience of real live play from just last year.

I count cards and have for 18 years. Casinos were closed here in Vegas from March through June. I was up 12k (Feb & March) when they closed. Upon reopening with everyone required to wear masks, I went to one of the sweatier casinos that I don't normally play too much and played their really good double deck game (trying to take advantage of the mask situation). I lost 30k in about 2 weeks at that one location, playing a good, fairly deeply dealt, double deck game that I should have been crushing. I was SURE I was being cheated.

I stopped playing that location for a little while, but slowly came back to it, playing once a week or so. Low and behold I began to win a bit. And then a bit more. I didn't recoup ALL my losses from those two weeks, but I did recoup a pretty good portion, maybe 2/3's. Still finished in the red at that location for the year, but I had won enough back that I knew I wasn't being cheated. I just had a really horrible run. And like I said at the start. I have been doing this for a living for 18 years. These runs happen. And they can play tricks on your mind and convince you that something is wrong when it really isn't.

Now my example is about card counting and that is higher varinace because of the different size bets, but it is the same with flat betting. You can hit these streak, IN BOTH directions, that seem almost impossible. That is blackjack. :/

Assuming you're not playing a trivially small number of rounds, the chances of being 2, 3 even 4 standard deviations from expectation does not change, no matter how many more rounds you play.Quote:kewlj

Now a player flat betting, doesn't need those 40, 50, 60 thousand trials to be significant. 1000 is getting there. 1000 is still small enough that results can be very skewed, 2, 3 even 4 (as posted here) standard deviations from expectation.

It's really not that hard to learn.

Quote:mwalz9I'll bet you, you can give me 100 hands in an 8 deck game, hitting soft 17, double after split allowed and I make 0 mistakes.

It's really not that hard to learn.

link to original post

And the only mistakes an educated player is likely to make are on "close calls" - decisions in which the ΔEV between options is not very large and situations that do not occur very frequently. When to Split 22 or 33 or when to double on certain soft hands.

Ehhh, how about 1000 or more hands? Like it's definitely a learn able skill but not everyone picks it up quickly.Quote:mwalz9I'll bet you, you can give me 100 hands in an 8 deck game, hitting soft 17, double after split allowed and I make 0 mistakes.

It's really not that hard to learn.

link to original post