Where will Bitcoin be on 9/4/2018?

Quote: OnceDear

Who did they get to sit for that portrait of HM Queen Elizabeth II ? (Queen of the Bahamas)

It's difficult to tell at what age she was



But even fairly accurate portraits look a little airbrushed. They seem to have given Viola a bit of an eye-lift.

Bermuda got rid of their portraits entirely.

Quote: SOOPOO

It is only up $1,000 since you bought it. But it is SO volatile after I finish typing this that will have changed.

I'm up 119 bucks in less then 24 hours
So far I am liking this
I send my nephew money for Christmas every year
He said he wants bit coin next year lol

Quote: terapined

I dont understand bitcoin
I just got a text from my nephew who is big on this and follows it closely
He insisted I buy
So I just bought 1k worth of bitcoin on coinbase
Just dipping my toe in
Told my nephew if I eventually sell, will split the profit if it goes up

+0.02935759 BTC
@ $33,562.70 per BTC

Coincidentally, that's about how much I own! 0.02439406 in my Coinbase wallet. I bought when it was about $8k per BTC, so I'm up a bunch, but don't plan on selling it any time soon. In fact I'm hoping for another crash where I can buy in for a bit more. Kicking myself for not doing that early in 2020.

My wife thinks it's folly, even moreso than precious metals. But she was on board with taking a $500 flyer back in 2018. I got $200 BTC, $200 ETH, $50 LTC and $50 BCH. Everything is in the green now except the BCH. Actually LTC is dead even today, but a few days ago it was up.

Because Bitcoin can be broken down into satoshis which are a very small value (one mil ~3 Satoshis), and bots on gambling sites are allowed, the urge to use Martingale is very strong for many people.

I wrote the following essay about Martingale on the just-dice web site. For this study I was paid 0.4 BTC which I squandered a few months later because I was convinced that bitcoin would never go above $1000 again.

=======================================================
At a Monte Carlo casino roulette table on August 18, 1913, black came up a record twenty-six times in succession. The streak was reported around the world along with stories about the mad rush to bet on red. Also in 1913 was the first recorded mention of the famous “monkey theorem”, by mathematician Emile Borel. Simply stated: given enough monkeys and typewriters eventually they would randomly type the complete works of William Shakespeare.

The Bitcoin website Just-Dice.com offers a game with a very low 1% house edge (50.5% chance of losing to double your stake). The owner, dooglus was looking at an unfortunate string of 27 losses in a row from a player. A lively debate ensued questioning if this streak was unique since no one had reported seeing anything similar on the forums. No doubt inspired by visions of monkeys typing out “to be or not to be” dooglus accepted a wager that this was not the most extreme run in the database of 1.2 billion bets at all odds. The most popular bet is the “double your stake” bet.

In order to understand the analysis it is helpful to look at a simpler problem. Let's eliminate the house edge temporarily and simply look at coin tosses. Arbitrarily choose one side as a loss (tails), and consider a run of losses of at least three in a row. Since the odds are 1:8, a first guess is that throwing the coin 8 times will be an even chance of getting at least one run of three losses in a row. Laying out a table with all 256 combinations and counting all the instances results in 107 (107/256=41.8%). Note that our guess was wrong!

A very important statistic in analyzing streaks is the expected number of trials before you get a streak of a given length. The problem naturally lends itself to a recursion formula. In order to get a streak of a given length, you first have to get a streak of a length one less. So E(k+1)=(E(k)+1)/p and E(0)=0. E(k) is the expected number of trials to get a run of length k, and p is the probability of each event. For coins, p=50%, E(1)=2, E(2)=6, and E(3)=14. So we can expect, on average, to toss the coin 14 times before getting a run of three losses.

The third part of the problem is to look for the number of coin tosses that give us an even chance. In order to do this we have to count the number of instances for each case. There is another set of recursion formulas which I won't go into for this paper, but the results are 107/256 8 times , 9 times is 119/256, and 10 times is 130/256. So the even chance occurs between 9 and 10 coin tosses. It's important to note that even in this simple problem the answer is not an integer. The probability of at least three losses in a row out of 14 tosses is 64.8%.

Now we apply the same analysis to dooglus's problem. In order to win the bet he needs at least 28 losses in a row. Following the example of the simpler problem we first calculate the 1 over the probability of losing 28 times in a row starting from scratch ( 50.5%^28 = 203,161,501 or 200M). The next step is to use the recursion formula 28 times to get E(28; p=50.5%)=410,427,273 or 400M.

Like the 3 loss coin toss problem, the first calculation (200M) represents the number of trials you need to get roughly a 40% chance of getting 28 losses in a row. The second calculation (400M) represent the number of trials you need to get well over 60% probability of 28 losses in a row. The even chance can also be calculated by formulas not reviewed in this paper (350M).

Part of the reason that I am glossing over some of the detailed calculations is that in the statistics of very rare events variance is naturally huge. As an example, set at the California Casino and Hotel in 1979 at Las Vegas, the longest reported number of dice throws in a pass line bet in craps was 117 throws. The thrower was dubbed the man with the golden arm. Then in 2009 a grandmother who was a novice at craps playing at the Borgata was celebrated around the world with a run of 154 throws of the dice before she got a 7-out. The second streak is over 200 times less likely than the first, yet no report of anything in between was made in 20 years.

Since we expect to see very few long streaks over the data universe, the bet odds only change by a few percentile if the universe increases or decreases by a few tens of millions of trials.

How big was the number of trials in dooglus's database after choosing only the double your money bets? There were 375M trials, so odds were in his favor, but probably not by the margin he wished. Did dooglus win his bet? There were no other streaks of losses of length 27, there were 4 winning streaks of length 27. There was a winning streak of length 30, and a losing streak of length 32 (good thing the player abandoned Martingale).

As expected, variance is huge, and there were no runs of losses of length 28-31.

Consider the player who won 30 times in a row. The recursion formula tells us that the expected number of trials to win 30 times in a row is 1.2 billion. The probability of winning the Powerball is only 375 million to one. As the player was either using Martingale, or had programmed their bot to use Martingale, they made the minimum bet every time and the winnings were less than a nickel.

Should be knocking on $60000 by Selection Sunday if not tomorrow. Was around $4000 a year ago this week. Talk about wishing I had doubled down!

Quote: AZDuffman

Should be knocking on $60000 by Selection Sunday if not tomorrow. Was around $4000 a year ago this week. Talk about wishing I had doubled down!

You were correct. It is trading at $60,640.

I have to guess that allowing you to make purchases on PayPal has a lot to do with this run up. If you add tens of millions of new Bitcoin owners at once, it has to make a difference.

Quote: Pacomartin

You were correct. It is trading at $60,640.

I have to guess that allowing you to make purchases on PayPal has a lot to do with this run up. If you add tens of millions of new Bitcoin owners at once, it has to make a difference.

It is helping it but much is driving it. A few big boys are taking every mined bitcoin. Those are the ones we know about. Even Jamie Dimon has come around on it.

Historically we should see $250-300,000 by late fall then it falls back down to about the levels we are seeing now. However, before we never had the same large players buying so we could see a super-spike to as much as $1MM. What I really need to learn is what it means to "build on top of" when it comes to crypto. I think I get it but I am not sure.

Quote: AZDuffman

Should be knocking on $60000 by Selection Sunday if not tomorrow. Was around $4000 a year ago this week. Talk about wishing I had doubled down!

You were correct. It is trading at $60,640.

I have to guess that allowing you to make purchases on PayPal has a lot to do with this run up. If you add tens of millions of new Bitcoin owners at once, it has to make a difference.

Bitcoin is continuing to go down slowly once it's gotten below $30,000. At 5:30 p.m. EST it's at 29600. It was a 29400 briefly last night. We'll see what happens tonight.