Okay - sorry about the Twitter spam today. Last night I talked on Clubhouse with @kfalter to her friends who are just learning about trading so feel compelled to write a post about risk management. The GAMSI index, esp isn't how I trade- so let's talk about risk. Thread

1/ Risk mgmt in trading boils down to 3 questions.
a. What is the most I could realistically lose with my positions?
b. How much money am expected to gain or lose per day?
c. What is my trading plan for different volatility regimes that arise in markets?

2/ Max Loss a. Take every stock in your portfolio. Find the 10 most related things. For example, AT&T is like Verizon, Comcast, Tmobile. AT&T's max loss [Y] should be the min daily % change of [VZ,CMCSA,TMUS] over past 10 yrs. Don't let Y exceed 2% of your account per stock

3/ Max Loss b. Take your entire portfolio and simulate its return. Take its max drawdown over the past 3 years (=X). Then, take the peers you generated in 2 and run a monte carlo simulation swapping portfolio w peers (=Z). Average X and Z. That's your max loss. Keep it below 4%

4/ Are you lost? Let's be real. What you don't know can kill you. Do the work. Learn how to run a monte carlo simulation in python or excel. Learn how to generate peer sets. You're smart enough, don't make excuses. Buy books on risk. Hedge funds think you're too dumb to do this.

5/ Expected gain a. Once you know how much you're risking, how much are you trying to make. You need some way to quantify expected gain, whether that's a target price, a target move per stock, a backtest showing daily avg gains for a strat. And you need to track it.

6/ Expected gain b. You should roll your expected gain up to your portfolio and get a $ amt you expect to make - a median #. You can to compare this # to your max loss to get a payback period. Making $1k a day and max loss is $5k, that's a 5 day payback period. Keep above 10 days

7/ Gains / Risk. Every day you should have those 2 numbers. Expected gain and max risk. Your 2 KPIs should be the ratio of those 2 numbers over time (gain/max loss), and how well you're estimating them (expected vs actual). When things go wrong you need to know why /when.

8/ Most ppl when they blow up trading do so because they haven't quantified how much risk they're actually taking, and don't have a clear sense for if their quantification is correct or not. Pros mess this up too. Look at what just happened to Melvin. Doesn't matter how smart u r

9/ Volatility regimes a. There are 4 basic vol regimes. a. VIX sub 14, Kumbaya. b. Vix 14-25. This could end in tears but we still want returns. c. Vix 26-40. *you are here* things have gotten fcking weird but disaster is not obvious. d. Vix 40+ there are genuine liquidity issues

10/ There are 2 other risk regimes. x. Equilibrium - the market is working normally and not defined by weird interaction effects btwn politicians, traders and the Fed. y. Far from equilibrium - the markets are now subject to judgments of a few important people.

11/ You should have a plan for ax,bx,cx,dx. ay,by,cy,dy. Write these plans down. Backtest your strategies for each of those. I think we're in c.y. right now. My plan? I stop market making limit orders, turn equity leverage to 0, and 50% of risk is liquid futures and FX

12/ As a general rule, the weirder things get on the x/y axis -- i.e. entering far from equilibrium - the worse idea it is to run a quant strategy heavily reliant on historical data. There's no real reason to expect far from equilibrium things resolve according to the past.

13/ The key point I want to get across, is I don't just blindly buy the GAMSI index, and I don't think you should either. It's a tool to measure a far from equilibrium situation, and there are strategies around it to ensure you don't blow up.

14/ Finally - if this all seems too hard, and you don't want to bother learning how to trade - it's important to understand that you're in all likelihood going to lose money. But I'm not gonna say "buy my course" or "don't do it". F that. I encourage you to do the work. Good luck