Blackjack Imbalance Question?
March 6th, 2016 at 9:28:13 PM
permalink
Hey all. I play for fun in an online casino, and one of the popular complaints is that the blackjack dealer is (at least somewhat) rigged. I know that no matter what the situation, the house should have an advantage, but it seems the dealers are able to pull just the right card a little too often. So I started tracking the information via a spreadsheet, keeping track of my hand total and all the dealer's cards. It's still early enough that there are can be relatively decent aberrations in the data, but I've been seeing some interesting results so far (dealer seems to bust a lot less than expected, draws blackjack a bit more often than expected, etc.)
I've also been keeping track of the dealer's results and I'm not sure if what I've seen so far is kosher or not. So does anyone know what the odds are for each of the dealer's possible ending totals (17-21 & Bust)? Dealer stands on soft 17s, and it's a 6-deck table. Thanks.
I've also been keeping track of the dealer's results and I'm not sure if what I've seen so far is kosher or not. So does anyone know what the odds are for each of the dealer's possible ending totals (17-21 & Bust)? Dealer stands on soft 17s, and it's a 6-deck table. Thanks.
March 7th, 2016 at 7:40:41 AM
permalink
Just search on the Wizard's Odds Site (www.wizardofodds.com) and you'll find his appendixes that have all of this information. i.e. How often the dealer busts with certain cards, odds of busting with any particular hand, getting blackjacks, etc.
Past that I'll caution you 2 fold:
1) Online casinos need to have a reputable name... and even then I'd never play them for fun. There's far too many horror stories of people getting cheated, scamemd, and just plain old told "we're not paying you" even if they miraculously win. Tread carefully when playing online. Use the Wizard's links to approved sites that come with the Wizard's protection.
2) You're going to need not only proof, but a LOT of hands. I'm not sure what you think "a lot" is, but I can tell you now it's not 100 or even 1,000 hands. You need to get tens of thousands of hands and track their results to get to the point where you could make any sort of mathematical claim about cheating. Once you get tens of thousands of hands you can see what the dealers real percentages are with certain hands, getting blackjacks, etc, and then they can be compared to the true values plus or minus STANDARD DEVIATIONS. Just because you're supposed to get a blackjack ~1 in 30 hands, doesn't mean the site is cheating if the dealer gets 5 in 30 hands. Short term results can be quite skewed in either direction. Remember, when you gather this data, you'll more than likely need to video record each hand as well, as proof. Otherwise, someone could just say "Yeah I saw 50,000 hands and the dealer blackjacked 50% of the time!" It takes a lot of effort and proof to prove cheating, which is often why so many online casinos do it... past the "we're just not paying you" bit.
Past that I'll caution you 2 fold:
1) Online casinos need to have a reputable name... and even then I'd never play them for fun. There's far too many horror stories of people getting cheated, scamemd, and just plain old told "we're not paying you" even if they miraculously win. Tread carefully when playing online. Use the Wizard's links to approved sites that come with the Wizard's protection.
2) You're going to need not only proof, but a LOT of hands. I'm not sure what you think "a lot" is, but I can tell you now it's not 100 or even 1,000 hands. You need to get tens of thousands of hands and track their results to get to the point where you could make any sort of mathematical claim about cheating. Once you get tens of thousands of hands you can see what the dealers real percentages are with certain hands, getting blackjacks, etc, and then they can be compared to the true values plus or minus STANDARD DEVIATIONS. Just because you're supposed to get a blackjack ~1 in 30 hands, doesn't mean the site is cheating if the dealer gets 5 in 30 hands. Short term results can be quite skewed in either direction. Remember, when you gather this data, you'll more than likely need to video record each hand as well, as proof. Otherwise, someone could just say "Yeah I saw 50,000 hands and the dealer blackjacked 50% of the time!" It takes a lot of effort and proof to prove cheating, which is often why so many online casinos do it... past the "we're just not paying you" bit.
Playing it correctly means you've already won.
March 7th, 2016 at 8:23:09 PM
permalink
I was browsing his Blackjack appendices and I didn't notice one that seemed to correlate to this. I'll go back and go through them all individually though. I saw something in the Ask The Wizard area that mentioned the average rate of blackjacks, but not anything about other results. I'll look more deeply, though.
1) Don't worry. There's no risk of lost money in this case, at least in the case of myself or any of my friends. If there was actual money on the line, I'd go with something I had as close to 100% faith in.
2) Trust me, I know. I'm rather early in my data-collecting, but I add to it with each hand I play. I know that you need a VERY significant number of pieces of data to minimize the chances of statistical anomalies. I'm still at a point where it could be just a series of lucky runs, but I wanted to at least start getting some hard data to start backing up what a number of people have claimed. I might not get enough data to completely validate their claims, but it's something.
For example, one of the general complaints is how often the dealer gets blackjack. It's felt at times like it's "one for the dealer, one for a player, one for the dealer, one for a player" while playing with anywhere from 3-5 people at the table. That's part of why I started doing this. The last chunk of data I got is about 135 hands' worth. During that time, I received only 2 or 3 blackjacks. A bit low, but it happens. The dealer got 12. That's about double the rate that I recall seeing. So it could be luck, or it could be a bit of weighting. I'm just curious to see how the results trend over time.
Thanks for the help.
1) Don't worry. There's no risk of lost money in this case, at least in the case of myself or any of my friends. If there was actual money on the line, I'd go with something I had as close to 100% faith in.
2) Trust me, I know. I'm rather early in my data-collecting, but I add to it with each hand I play. I know that you need a VERY significant number of pieces of data to minimize the chances of statistical anomalies. I'm still at a point where it could be just a series of lucky runs, but I wanted to at least start getting some hard data to start backing up what a number of people have claimed. I might not get enough data to completely validate their claims, but it's something.
For example, one of the general complaints is how often the dealer gets blackjack. It's felt at times like it's "one for the dealer, one for a player, one for the dealer, one for a player" while playing with anywhere from 3-5 people at the table. That's part of why I started doing this. The last chunk of data I got is about 135 hands' worth. During that time, I received only 2 or 3 blackjacks. A bit low, but it happens. The dealer got 12. That's about double the rate that I recall seeing. So it could be luck, or it could be a bit of weighting. I'm just curious to see how the results trend over time.
Thanks for the help.
