Thread by @Quantdom1

1/n Lessons from the book Super Forecasting by Phillip Tetlock a top researcher from Wharton School. This book was recommended by @GiganticRebirth one of the best traders/thinkers in #CT.

wsp.wharton.upenn.edu/book/superfore…

2/n Traders are, by definition, professional forecasters. Successful traders are successful forecasters. Let´s dig deeper into what thinking habits can improve our forecast ability.

3/n First, let´s understand that markets are largely unpredictable. Consider a dice with the number 1 on 5 sides and the number 0 on the remaining side. An intelligent trader would bet on the outcome 1 100% of the time and still be wrong 16.66% times. Losses are unavoidable.

4/n Let me repeat that… even the most successful trades lose many, many times, given that the market has an unpredictability that no amount of analysis can reduce. In the example I gave before, a trader knew exactly the properties of the dice and still lost 16.6%% of the time.

5/n A more optimistic way of looking at this is that a trader can become a millionaire, even after being wrong many times. Heck, a trader can become rich with a batting average of 50% or lower. Jason Shapiro, a market wizard, mentioned that he has a batting rate of ~ 30%.

6/n How is this even possible? He wins 300 and loses 700 out of each 1,000 trades. The losses when he loses are smaller than the wins when he wins. Risk management is, of course, crucial. Feel free to DM me if you have questions about the math.

7/n We have established that markets are predominantly random and that it is possible to be a profitable trader even by being wrong many times. Now, let´s dig deeper into the habits and modes of thinking that can improve our forecasting ability.

8/n The best forecasters are flexible. They change their mind when new facts or information arrives. A dogmatic view of the world (one that does not change with facts or new info) makes a person, almost by definition, a lousy forecaster (and trader).

9/n A $BTC maximalist cannot be bearish at 68k, and a perpetual BTC bear like @Nouriel, or @PeterSchiff cannot be bullish at 3k. Top forecasters are fluid and agnostic: they can be long or short $BTC when circumstances and information change.

10/n @GiganticRebirth and @Pentosh1 are the best examples of fluid, probabilistic thinkers I can think of from #CT.

11/n Top forecasters measure. What to measure? 1. overall batting average + 2. batting average by strategy (i.e., shorting when Bollinger bands reach a certain level) + 3. batting average by market circumstances (how each strategy works in different market environments).

12/n Top forecasters improve over and over, and they use precise language. What were the mistakes in the analysis? Where exactly did I go wrong? A losing trade does not mean bad logic, and a winning trade does not mean good logic. Focus on improving your reasoning.

13/n Top forecasters understand that the world is complex: few things are 0 or 1. Most of them are in the middle (i.e. in my opinion there is around 60% that BTC touches 20k before touching 18k). The size of each trade should be a function of the degree of confidence.

14/n Higher confidence -> larger size. @GiganticRebirth does this over and over. He posts only his high-confidence trades, and his batting average has been excellent in those (close to 100% accuracy). He is definitely a super forecaster.

15/n Top forecasters are data-driven. How do you determine the probability of price going in a certain direction? You start from the past to get a baseline: “Price action like the one we are observing has preceded BTC increasing 10% over the next weeks 50% of the time”

16/n This is incredibly rare in CT, where most people use their gut feeling to make forecasts without mentioning past data.

17/n Top forecasters break down problems that seem too complex into tractable sub-problems. They decompose the problem into their knowable and unknowable parts. This is the only way to produce sensible forecasts.

18/n Consider the forecasting challenge that @GiganticRebirth tackled back in Dec 2021: what are the $alts that will drop the most over the next 6 months? This seems like a hard question to answer as there are 1000s out there. Let´s dig deeper and generate a reasonable forecast.

19/n The price of an asset is a function of its demand and supply. If supply increases, then price is likely to drop. What are the assets that will see an increase in supply between Dec 2021 and June 2022? Those that have unlocks scheduled!

20/n So, the logic was: more supply at a time when overall demand will be most likely decreasing -> dramatic drops in price. This proved to be a fantastic forecast.

21/n the result of this forecast was @GiganticRebirth big short that generated millions of dollars in profits.

https://twitter.com/GiganticRebirth/status/1464278150121197589?s=20&t=M76b_jc3RBhdUwHJqYaEqw

n/n If you enjoyed this threat please share it and feel free to DM if you have any questions. Thanks for your time!