Is it ever wise to take a bet just because the chances of a profit is really high (even if expected value is negative?)
I know if you keep doing this in the long run you will most likely fail but what about just to do it once?
Like let's say there is 95% chance that you will get $1but a 5% you could lose $20. Would you ever take this bet? If so, how many times?
Sorry my stats background is really bad.
| +14 | Natural Gas Analyst Path to Trader | 3 | 1d |
| +9 | CRE to S&T lateral opportunities | 5 | 1d |
| +6 | Exiting Sell Side FI Trading | 1 | 3d |
| +6 | S&T Outlook and Pivot from Buy Side | 2 | 2d |
| +6 | EM Resources / Study Material | 1 | 1d |
Career Resources
Get instant access to lessons taught by experienced private equity pros and bulge bracket investment bankers including financial statement modeling, DCF, M&A, LBO, Comps and Excel Modeling.
In this case the expected value is negative so you would never take the bet. Well almost never - imagine you need $30 for a bus ticket back home, but only have $29. In that case, it might make sense to take the bet once, but it becomes a question of utility. If the EV was 0, then you would still never take it due to the very small amount of risk involved. If the expected value was positive at all then you would take the bet as many times as possible. Assuming the bet size is small compared to the money you have to gamble with, your risk is negligible and the system basically gives you free money, since the risk is diversified across many events. On the other hand, if your house is worth $500K (and it is all you have) and the EV of betting is $10k but the cost to play ONCE is $500K, then you probably would not take the bet, since the risk is not diversified. But you would take the bet if the bet size became 1,000 times smaller, due to diversification. As you do the same experiment (bet) n times, the expectation grows linearly, but the standard deviation of the outcome only grows by sqrt(n). Look at the variance formula to figure this out - it is quite simple.
anyone who sells naked puts would disagree with you. although most of them end up exploding quite fantastically. (but they do take in those small profits for quite some time)
You hit the nail right on the head. Selling options is an income generating strategy with an inverse risk reward ratio. It works very well if you know options.
In my opinion if your new to trading I wouldn't advise you to make any trade with those #'s until you FULLY grasp the concept of risk management. The fact that you even asked this question leads me to believe that you are new to the game.... so no, don't do it because you will most likely make the same mistake twice.
Good Luck Trading!
No. No. No. They are not at all comparable at all. Selling options is NOT a Negative EV bet unless it is mis-priced, and in general this is not the case. If you know someone who will make negative EV trades, let me know, I'd love to have them as a counterparty.
same concept as a casino... if you go to a casino and think you'll make money (without counting cards, having extra info, etc to give you an advantage over the house) then you're an idiot.
Yes but casinos make money because people don't stop playing so they lose in the long run. However, people who play only a little and quit do sometimes make money (making only a couple of high percentage bets like above).
You are right in the sense that there are bets you can make in a casino that give you a chance >50% of making a profit. But so what? I am not sure what your point is here... The point that playing only a little and quitting while you are ahead is a legit strategy in a casino is obviously flawed (and it sounds like you understand that).
If I ever open up a casino, please come play at it...
Yes. If you are a Pittsburgh fan and like losing money.
I just paid off my bookie, very depressing feeling.
sportsinteraction.com
Well I wouldn't take your bet, but it gets more interesting when the probability of loss is very small, i.e. 5%.
For instance, take a coin flipping game where you get paid $2 if you flip HT, $4 if you flip HHT, $8 if you flip HHHT, $16 if you flip HHHHT, etc. Basically you're getting paid 2^n for flipping n heads in a row.
What's the expected value of the game?
How much you pay to play it?
How much would you accept to let someone else play the game?
To play it, I would probably pay a couple hundred bucks.
To let someone else play it, I would accept $50, maybe less.
Interesting disconnect there. Good interview question by the way. Who wants to explain the disconnect? No googling, will pay with bananas.
St Petersburg Paradox, EV is infinite. Solution is based on marginal utility instead of payoffs (Bernoulli), and credit constraints of the counterparty.
Good.
You wouldn't take that bet: its EV is negative, and the big payoff is your loss.The expected value determines whether you play: if it's negative, you don't; if it's zero, you do but only if you're not risk-averse; and if it's positive, you do (assuming that you don't need the money, of course). If it were the opposite - if there were a good chance of a small loss but a small chance of a big win, you might play. That's the lottery: the expected value is negative but the winning payoff is astronomically high. If you're walking home and the Powerball is $200 million, with odds of winning equal to 1 in 500 million, the $2 dollar ticket might be worth taking a flyer on.
If you've read The Greatest Trade Ever, JP took the other side of your bet up there and made a killing.
When would you play this game? Assuming you are not risk-averse, what is the point? Legitimately curious. Could a fund ever benefit from adding free volatility? I am thinking Bernie could have benefited from some to make his returns more realistic, but any other instances?
Be careful here; "not risk-averse" does not mean risk-loving. It means risk-neutral or risk-loving.
Even if you're risk-neutral or risk-averse, it can make sense to add volatility, particularly if it is anti-correlated or even uncorrelated with your volatility.
Which "bet"? I described a game. Also, describing the EV as "negative" is not a good answer.
welcome to the world of short vol investing...merryweather has made a career out of taking these kind of bets every day and then just opening up a new fund everytime the 5% comes in. So the answer is that you dont take the bet unless you get paid performance fees quarterly and you have another set of suckers ready to recapitilize you when you blow up...
would you play the lottery? its 5 bucks to play and you have a chance at winning millions...
Larry Hite said it best
There are good bets, bad bets, winning bets and losing bets.
The worst that can happen is winning a bad bet.
There are several conditions under which you would make a negative EV bet.
Two have been outlined already:
A third condition is when the time scale of the bad event happening is much greater than once every lifetime. Unfortunately shit happens in markets way too often to make bets on this basis, unless you're running a fund or bank. But in everyday life, we do this all the time. When you cross the road you are implicitly making the bet that you won't die in some freak accident. The upside is finite, while the downside is very, very large. I would argue the EV is quite negative, but since the bad event probably occurs once every 10,000 or more lifetimes, you take the bet.
Quis quidem sunt quisquam non aperiam vero vel. Ratione laudantium deleniti dignissimos voluptatem quae quaerat ad eaque.
Est cumque fugiat sint inventore. Iste cupiditate accusantium saepe qui pariatur. Eius maxime sed vel soluta earum.
Quae ut ipsam soluta asperiores autem. Placeat praesentium cupiditate sed porro voluptatem.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...
Possimus expedita et et vel aut dolorem dignissimos. Molestias harum laborum consequatur nisi qui veritatis.
Error quisquam architecto qui. Aut autem atque est. Quisquam placeat molestiae laborum saepe nihil voluptatem necessitatibus atque.
Perferendis quas aliquid magni ea dicta non praesentium. Nobis et voluptas et nostrum. Unde sit voluptate officiis non id aut recusandae. Est in sed quae. Eaque consectetur voluptatem velit dolorem.