The final amount would be depending on the number of players left until the last day of the contest. I found some reference posts for the 1st contest. This might help you make up your decision on wether or not it will be profitable for you
Since your expected losses aren't in a lump sum, you can do some data analysis off a range of a few days and estimate the number of people that are dropping out: you can have a maximum of 333 participants and still break even, provided you're playing on blackjack with a 0.5% edge. Halve the amount (and then some) if you're going baccarat, and even more for roulette. 120 days x 5 mBTC x 1/37 edge is a whopping 16.21mBTC cost of participation, worth over $800 at current prices. Bear this in mind.
It's the doubling down to chase losses that killed my balance once or twice, so tempting when you think "I've lost 6 hands in a row, surely a win's coming?".
Incidentally, in blackjack games, you
could argue that martingale is slightly more effective than flat-betting, since you are more likely to lose with excess low-cards, which would raise your odds in subsequent rounds.