Since the bat shit about signatures campaign is at least as useless as the shitstorm of math in thread, I allow myself to reply to dooglus' post here. Again, pure technicalities.
I made an obvious mistake. I had another piece of code in mind (if nicolaennio allows me, I'll publish it).
)
the rest of the discussion rely on the definition of event and all the IF's rely on the available information or the information we decide to include to our model(s). In mathematical terms, it translates to conditions.
P( Event ¦ Information ) (conditional probability)
In my previous analysis, I mentioned three(+doog's) of such expressions. For me (I seem to be alone), the probability I) is the first that came into my mind, seemed the most natural to me. Others might have different opinion. Nobody's right or wrong, we just have different views on how to approach a problem with an open question, a non mathematical question.
I made an obvious mistake. I had another piece of code in mind (if nicolaennio allows me, I'll publish it).
Now about dooglus' simulation, the probability he finds is the answer (I think) to this question :
III) Given that
- all bets are continuously updated at the maximum bet size (which is 0.5% of bankroll) all the timeat most 60k times,
- betting stops if bankroll reaches 15% of initial BR,
what is the probability to win 85% of the initial bankroll ?
III) Given that
- all bets are continuously updated at the maximum bet size (which is 0.5% of bankroll) all the time
- betting stops if bankroll reaches 15% of initial BR,
what is the probability to win 85% of the initial bankroll ?
I don't think so. I didn't include the 60k bets limit at all. First, I don't know where it came from, or whether it's really how many bets he made, and second I don't know that it matters.
Indeed there's no stopping condition on the number of bet in this model. If the random number generator was biased (evenly distributed around 0.495) or if the criterion was "< 0.5", the betting maybe could go forever (Nerds: with what probability?). It would be indeed be interesting do modify the code to capture the number of bets until bust either side (a.k.a. survival time). With max bet size, I expect it to be very short. (Nerds: what is the distribution of that number? )When asking "how likely is it that what happened was for real?", we need to decide what aspects of "what happened" are significant.
That's right ! And all P( Event ¦ Information ) (conditional probability)
In my previous analysis, I mentioned three(+doog's) of such expressions. For me (I seem to be alone), the probability I) is the first that came into my mind, seemed the most natural to me. Others might have different opinion. Nobody's right or wrong, we just have different views on how to approach a problem with an open question, a non mathematical question.