Expectancy formula

[earth_imperator]

I'm going to write a similar MC simulation to find the correct formula for my said options problem.
As can be seen on the "P/L chart" (ie. the PnL diagram), to each stock price at expiration there is a corrosponding PnL value (the orange "curve").
I'll generate normally distributed relative change values (ie. pct changes of stock price) (or equivalently lognormally distributed stock prices), millions or even billions of them per test-run.
No need to generate for intermediate days, only the expiration day is important.
I'm hopeful this will lead to the solution of the problem, ie. finding the correct Expectancy formula.

 

[longandshort]

You guys are doing this backwards.

Start with a trading strategy. A simple example would be: “if stock has a beat and raise, buy at open the next day, hold for the quarter”.

If I conduct the analysis across spx stocks I may observe a positive return by doing this trade. That is positive expectancy, and your risk metrics will start to make sense.

From there, record your trading. If that is your strategy, how well does your actual performance track the underlying strategy? Now you can benchmark your performance to your strategy.

No one is testing risk management strategies on a coin flip— this is a losing trade when you have a transaction cost! Doesn’t matter how many times you slice it lol.

 

[earth_imperator]

You guys are doing this backwards.

Start with a trading strategy. A simple example would be: “if stock has a beat and raise, buy at open the next day, hold for the quarter”.

If I conduct the analysis across spx stocks I may observe a positive return by doing this trade. That is positive expectancy, and your risk metrics will start to make sense.

From there, record your trading. If that is your strategy, how well does your actual performance track the underlying strategy? Now you can benchmark your performance to your strategy.

No one is testing risk management strategies on a coin flip— this is a losing trade when you have a transaction cost! Doesn’t matter how many times you slice it lol.

BS

 

[murray t turtle]

No one is testing risk management strategies on a coin flip— this is a losing trade when you have a transaction cost! Doesn’t matter how many times you slice it lol.

%%
LOL so true;
actually amazing some would think business is a coin flip =its not. Nor is it as simple as coin flip;
like Seiko/ if you insist on a coin flip/ its loaded, but not that simple.
Of course if one is just thoughtlessly + unwisely hitting numbers \could be a lot worse than coin flip, add in slippage + bid\ask..LOL
Business is more like a coin toss.
NObody could honestly flip a coin 88% wins for yearly large sample profit, based on JANUARY coin flips;
markets+ capital markets are better+ more complex than a coin flip.

 

[ironchef]

You guys are doing this backwards.

Start with a trading strategy. A simple example would be: “if stock has a beat and raise, buy at open the next day, hold for the quarter”.

If I conduct the analysis across spx stocks I may observe a positive return by doing this trade. That is positive expectancy, and your risk metrics will start to make sense.

From there, record your trading. If that is your strategy, how well does your actual performance track the underlying strategy? Now you can benchmark your performance to your strategy.

No one is testing risk management strategies on a coin flip— this is a losing trade when you have a transaction cost! Doesn’t matter how many times you slice it lol.

BS

@longandshort makes more sense to me.

@earth_imperator, MC your said option problem will only give you a theoretical answer may not be real. I still remember LTCM.

 

[earth_imperator]

@longandshort makes more sense to me.

@earth_imperator, MC your said option problem will only give you a theoretical answer may not be real. I still remember LTCM.

IMO it's the basis. Things are not that easy to explain to anybody as it's an "advanced stuff" requiring some higher math skills...

 

[murray t turtle]

@longandshort makes more sense to me.

@earth_imperator, MC your said option problem will only give you a theoretical answer may not be real. I still remember LTCM.

%%
Long + short had good timing with his nickname also
LTCM great fund/insane leverage market maker warned[caution, caution signs]

 

[SunTrader]

Who knew ... that expectancy didn't involve winning trades and winning percentage plus losing trades and losing percentage - of each particular symbol.

Me and 1000's of others (at least) for decades apparently.

Or maybe it still does and someone is throwing unnecessary curveballs.

 

[smallballtrader]

I found out that in the context of trading there are at least 3 formulas for calculating Expectancy:

Formula
1) Expectancy = (1.0 - AvgWinAmt / AvgLossAmt) * pWin - 1.0
2) Expectancy = AvgWinAmt * pWin - AvgLossAmt * (1.0 - pWin)
3) Expectancy = (AvgWinAmt * pWin + AvgLossAmt * (1.0 - pWin)) / AvgLossAmt
where
Expectancy: multiplying by 100 gives a pct number
AvgWinAmt: $
AvgLossAmt: $, as positive number
pWin: probability for win (a number between 0 and 1). BTW: the saying "50:50" means pWin=0.5, not 1.0

Which one is the correct formula for Expectancy?

Formulas found at these links:
https://www.tradingwithrayner.com/risk-reward-ratio/
https://easycators.com/trade-expectancy-formula/
https://www.iexpats.com/expectancy-ratio/
https://www.dara.trade/blog/2019/10/7/does-your-strategy-has-positive-expectancy
https://tradeciety.com/ultimate-math-guide-for-traders
https://easylanguagemastery.com/rank-your-trading-system-with-expectancy-score/
https://www.netpicks.com/profit-factor-expectancy/

This assumes complete information about future trading is contained in your past track record. Say if you have a massive-fuckup-potential then the formulas dont hold.