I did run this many times by adapting the OP's own code (and vectorizing it, which he asked someone to do). Here it is:
import numpy as np
import pandas as pd
n_subjects = 100
n_trials = 1000
start = 10.0
win = 1.5
loss = 0.6
prob = 0.5
results = np.ones((n_trials, n_subjects)) * start
for trial in range(1, n_trials):
wins = np.random.binomial(1, prob, n_subjects) == 1
results[trial, wins] = results[trial - 1, wins] * win
results[trial, ~wins] = results[trial - 1, ~wins] * loss
ax = pd.DataFrame(results).plot(legend=False, figsize=(18,10), logy=True, linewidth=0.5)
ax.plot(results.mean(axis=1), color='red', lw=2, linestyle='--')
The mean always trends to 0 and every single player eventually loses. There are never any winners at all past around 500 trials or so. Not sure how you're getting a different result as I have never used Julia and can't tell what your code is doing (except apparently something different).
