Yesterday I heard from Scott at MobilityWare that the next upgrade of the iPod blackjack app will fix the split naturals problem and bring the game more in line with casino reality.
Great!
Per the caption above, what's not so wonderful is that I have now played more than 5,000 rounds and should have been logging them, primarily to see if a Martingale would have matched the "winnings" achieved with target betting.
Already, the bankroll above is around $50,000 behind the times (this morning, while waiting for my 6:30am coffee to brew, I blasted past the $200,000 barrier).
Guess I'm just lucky...
Regular readers will know that I have never advocated the Martingale as a means to overcome the house edge at casino table games, but it makes a handy benchmark, and while almost unplayable in the real world, is widely misunderstood.
For example, most of its opponents cite the relative frequency of prolonged losing streaks that can take the required bet from $10 to an amount far in excess of the table limit at a low-stakes game.
In fact, the only limit that matters is the house limit, and even then, the assumption that the moment the "green ceiling" is reached it's game over is naive.
Martingale players usually don't bet more than three losers at any one layout, and will simply repeat a wager on those rare occasions when they are prevented from going higher.
Sunday, August 2:
The total 1T21 jackpot is now up to $210,585 with the AV edging up again to -2.87% and the indicated loss (AV x est. action) at close to -$121,000.
As always, no proof implied, just one more illustration of what target betting can do against persistent negative expectation.
The knee-jerk academic mathematician's response to these numbers would go something like this:
You got lucky.
Negative expectation cannot be overcome in the long run.
A sample of 5,477 bets is not representative and results achieved against such a small sample are anecdotal and irrelevant.
In due time (sooner rather than later) a prolonged negative trend will occur that will wipe out your imagined "profit" and replace it with a loss that is in compliance with negative expectation.
The most likely explanation for your apparent success is that you cheated.
Either way, you are an idiot.
Luck is certainly helpful when the goal is to make money at casino games of chance with an inherent house bias.
However, it can never be relied upon.
A sample of less than a trillion outcomes will never be accepted as "representative" and if a sample that large were to show a positive result for the player (a very old person, exhausted after playing non-stop for several million years!) it would be dismissed as anecdotal and irrelevant.
Target betting succeeds because of the cumulative effect of consistent profits from small series of rounds in which more wagers were lost than won.
The primary consequence of all those small victories is that overall, the player wins more when he wins than he loses when he loses.
A prolonged negative trend that brings ruin is certainly possible, but grows increasingly improbable as the bankroll strengthens.
At this point, for example, given a maximum bet of $5,000, the "house" would have to get at least 42 bets ahead of the player to grab back $210,000 in funny money. It could happen. But it probably won't.
Watch this space!
I'm looking forward to seeing the improved version of 1-thumb BJ when the update to the app pops up on my iPod.
I have a couple of other minor quibbles with the game and have suggested the addition of a deck value counter. But all in all, the arrival of an honest game simulation that can be played anywhere from a toilet seat to a mountaintop has to be a leap forward for humankind.
Thanks, MobilityWare, with whom I have no connection beyond a sense of some gratitude!
An important reminder: The only person likely to make money out of this blog is you, Dear Reader. There's nothing to buy, ever, and your soul is safe (from me, at least). Test my ideas and use them or don't. It's up to you.
_
