One of the early highlights in Gary Stevenson’s book “The Trading Game” is the description of how he won a trading game organised by Citigroup to land a coveted summer internship. He describes how he employed second level thinking to beat all the chums with elite university educations who applied their theory knowledge from economics classes to a simulated trading market. And, he discusses how he eventually got hired by continuing to back himself and double down on his convictions even when the ‘rules of the game’ had changed and he lost big time.
Something you said here jibes strongly with something that's been dawning on me: the only massively successful traders are the ones who played strategies with "chaotic" returns that make a ton of money when they work and lose even more when they don't… their returns are no better than anybody else's overall, they're just "clumpier"… the wins and losses from their strategies come in inconsistent runs, fits and starts, sometimes lasting far longer than one might expect from random chance. When their wins are coming fast and strong and long, they look like geniuses. When the tide turns they pay for it, but they don't get any press anymore when they've been ruined, so you never hear that it all averaged out in the long run to very little. Remember, Jesse Livermore died broke and in misery.
I noticed for instance that if you look at delta as an approximation for odds of an option ending ITM, turns out that higher-risk strategies that pay the most in terms of returns per winning trade often lose the most over time when you factor in the odds of winning the trade. As risk increases your returns, odds of losing increases even faster. There are low-risk strategies where your wins will stack up only infrequently, but losses are very small and the wins will likely add up to much more over time… and those, most of the time you won't get a fill, because, why would a market maker take a bet that's likely to lose them money over time? They can do math just as much as we can. So, pretty much, the only available strategies are A.) win some, lose some, and not make much over time, or B.) win big, make a lot, and then eventually during a sudden turn of events lose even more than the total you've made. That's options but I wouldn't be surprised if it works the same way in stocks. If the odds didn't favor the house, why would there be a house at all?
In my more cynical moments I think that the market makers will accept a certain level of losses because a certain level of visible wins for retail keeps people playing the game, hoping they'll win bigger.
All just a hunch, but one that's been growing lately. Watch your Sharpe ratios.
Thank you very much. I got very curious; in the spirit of the book I will look for a pirated copy online
Read it too and fully confirm what you say.. Stromg buy :)
Something you said here jibes strongly with something that's been dawning on me: the only massively successful traders are the ones who played strategies with "chaotic" returns that make a ton of money when they work and lose even more when they don't… their returns are no better than anybody else's overall, they're just "clumpier"… the wins and losses from their strategies come in inconsistent runs, fits and starts, sometimes lasting far longer than one might expect from random chance. When their wins are coming fast and strong and long, they look like geniuses. When the tide turns they pay for it, but they don't get any press anymore when they've been ruined, so you never hear that it all averaged out in the long run to very little. Remember, Jesse Livermore died broke and in misery.
I noticed for instance that if you look at delta as an approximation for odds of an option ending ITM, turns out that higher-risk strategies that pay the most in terms of returns per winning trade often lose the most over time when you factor in the odds of winning the trade. As risk increases your returns, odds of losing increases even faster. There are low-risk strategies where your wins will stack up only infrequently, but losses are very small and the wins will likely add up to much more over time… and those, most of the time you won't get a fill, because, why would a market maker take a bet that's likely to lose them money over time? They can do math just as much as we can. So, pretty much, the only available strategies are A.) win some, lose some, and not make much over time, or B.) win big, make a lot, and then eventually during a sudden turn of events lose even more than the total you've made. That's options but I wouldn't be surprised if it works the same way in stocks. If the odds didn't favor the house, why would there be a house at all?
In my more cynical moments I think that the market makers will accept a certain level of losses because a certain level of visible wins for retail keeps people playing the game, hoping they'll win bigger.
All just a hunch, but one that's been growing lately. Watch your Sharpe ratios.
Greg Smith: Why I Left Goldman Sachs