Get a job as a software engineer at a large company. Rent a small room, don't own a car, don't eat out, don't go to Hawaii for vacation, whatever it takes to save $250k in three years.
Then go to Vegas and put everything on red twice. (Metaphorically speaking -- I don't know if they'd allow a $500k bet in Vegas. But in the wonderful world of contemporary financial markets, there are instruments to make any kind of bet one might desire, and probably the house's cut will be even lower than in roulette.)
The obvious downside of this strategy is that it only works with the probability slightly less than 25%, otherwise you end up broke. But quite frankly, I doubt other ideas presented in this thread (make something people want, get on a TV show etc.) have better success rates. If there was a sure-fire way of getting from $0 to $1M in three years everyone would be doing it.
If you're OK not only with ending up broke in case of a failure, but even being in debt, then you can increase you odds of success even further by using leverage.
Not only does it make sense, but it allows you to see exactly what your odds of success are. Everyone acknowledges that both the "get famous" and the "do a startup" method have an element of luck, but this method allows you to quantify it.
Also, if you did it there a non-zero chance that you could get famous and become rich though your fame (even if the Vegas bets failed).
Your method works, but even if successful, you'll end up at $1M pre-tax, out of which $750 are gambling income. Gambling income is taxable in the US. If you win that much you might have to pay estimated taxes on that, so you'd only actually have $1M for a few months.
To defend myself (feebly) I can say that you can max out the after-tax 401(k) to the $49k a year limit, then fund a traditional IRA with $5k per year, then roll it over to Roth IRA after quitting your job (or even right away, if your employer's 401(k) plan supports it). That way $162k out of $250k will be tax-sheltered. (Of course, you can't take the money in Roth IRA to Vegas, but as I said with carefully tuned investments you can have the same risk-reward line-up). So you will only own taxes on $264k so the taxes will be in the neighborhood of $100k. So I guess you have to save a bit more than $250k or accept a bit worse odds.
One problem with money in Roth IRA, of course, is that you cannot take the earnings out without paying the tax and penalty, until you're 65. But if you plan to use your $1M to live modestly off of it (as opposed to buying a tropical island), then you can get around this by starting a SEPP program.
I hope this post and the upvotes were sarcastic. Trying this method 4 times means that you have a 68%* success rate. You'd have to try it 8 times to even get to a 90% success rate.
* To calculate this, chance of success after 4 tries = (1 - chance of failing 4 times in a row) = 1 - 0.75^4 = .68
Then go to Vegas and put everything on red twice. (Metaphorically speaking -- I don't know if they'd allow a $500k bet in Vegas. But in the wonderful world of contemporary financial markets, there are instruments to make any kind of bet one might desire, and probably the house's cut will be even lower than in roulette.)
The obvious downside of this strategy is that it only works with the probability slightly less than 25%, otherwise you end up broke. But quite frankly, I doubt other ideas presented in this thread (make something people want, get on a TV show etc.) have better success rates. If there was a sure-fire way of getting from $0 to $1M in three years everyone would be doing it.
If you're OK not only with ending up broke in case of a failure, but even being in debt, then you can increase you odds of success even further by using leverage.