-2

In craps the house has a 1.45% edge (51.45% chance to lose). If you martingale+1 (double up + 1 unit, every time you lose) and do this for 15 outcomes, then double up the initial bet for 30 more outcomes you can almost 4x your money

The odds of 4 losses in a row would be .5145^4 = about 7%

But what are the odds I will make it 40 rolls(outcomes) without losing more then 4 in a row?

  • Here is a link to what might be the formula. I’m just not smart enough to work it out :( Please help :)

    https://math.stackexchange.com/questions/602123/what-are-the-odds-of-getting-heads-7-times-in-a-row-in-40-tries-of-flipping-a-co

    – Mike Kelly Mar 02 '21 at 04:23
  • Welcome to MSE. Please use MathJax to format your posts. To begin with, surround math expressions (including numbers) with $ signs and use _ for subscripts. $x_1$ comes out as $x_1$. – saulspatz Mar 02 '21 at 04:40
  • Can you also do this for 4 losses in a row? Thank you so much – Mike Kelly Mar 02 '21 at 06:13

1 Answers1

0

We can treat this as a Markov chain. We have $6$ states. For $k=0,1,2,3,4$ state $k$ means that we currently have a string of $k$ losses. State $5$ means that at some time, a string of $5$ losses has occurred, so state $5$ is absorbing. If the probability of a loss is $p$ then the transition matrix is $$\begin{bmatrix} 1-p&p&0&0&0&0\\ 1-p&0&p&0&0&0\\ 1-p&0&0&p&0&0\\ 1-p&0&0&0&p&0\\ 1-p&0&0&0&0&p\\ 0&0&0&0&0&1 \end{bmatrix}$$ and the initial distribution is $$X=\begin{bmatrix}1&0&0&0&0&0\end{bmatrix}$$ We simply have to compute $XP^{45}$ and read off the answers.

I wrote a python script to do this:

import numpy as np

p = (1+.0145)/2 P = np.zeros((6,6)) for i in range(5): P[i,0] = 1-p P[i,i+1] = p P[5,5] = 1 X = np.array([1,0,0,0,0,0]) Y = [email protected]_power(P, 45) print(Y)

This produced the output

[0.2337404  0.12075448 0.06238393 0.03222866 0.0166499  0.53424264]

so the probability of $5$ losses in a row is $$.53424264$$

saulspatz
  • 53,131
  • If this answer solved your problem, please accept it clicking the check mark, so that it no longer shows up as unsolved. I don't quite understand your comment, but if you think that you have a positive expectation by following this strategy, you are mistaken. – saulspatz Mar 02 '21 at 06:14
  • The house edge is 1.45%, so the odds of losing would be 51.45%. Could you also do this for 4 losses in a row. If you have your program still setup. Thank you so much – Mike Kelly Mar 02 '21 at 06:15
  • I get $0.80915933$ for $4$ losses – saulspatz Mar 02 '21 at 06:16
  • That sounds right. Thank you – Mike Kelly Mar 02 '21 at 06:20
  • One more :) I’m trying to get it under .667 will I get there if I reduce the tries to 30 outcomes? Thank you for all your time and patience – Mike Kelly Mar 02 '21 at 09:29
  • Good guess! With $30$ tries the probability of a run of $4$ losses is $ 0.65820036$ – saulspatz Mar 02 '21 at 12:54
  • I think I would have asked this a different way. Is my question the same as, what is the probability that I get less the 4 in row out of 30 tries? – Mike Kelly Mar 02 '21 at 21:45
  • The probability that a run of $4$ losses occurs sometime in a string of $30$ plays is about $0.65820036$. The probability that no such run occurs is approximately $0.34179964$. – saulspatz Mar 02 '21 at 22:15