
Caladrius
1.1K posts

Caladrius
@i_caladrius
A human. A fixed income nerd. In reverse order.







Will #Iran close the #Strait of #Hormuz ? The answer is - NO They may announce it but they won’t be able to do it.


My conversation with Marc Andreessen (@pmarca), co-founder of @a16z and Netscape. 0:00 Caffeine Heart Scare 0:56 Zero Introspection Mindset 3:24 Psychedelics and Founders 4:54 Motivation Beyond Happiness 7:18 Tech as Progress Engine 10:27 Founders Versus Managers 20:01 HP Intel Founder Legacy 21:32 Why Start the Firm 24:14 Venture Barbell Theory 28:57 JP Morgan Boutique Banking 30:02 Religion Split Wall Street 30:41 Barbell of Banking 31:42 Allen & Company Model 33:16 Planning the VC Firm 33:45 CAA Playbook Lessons 36:49 First Principles vs. Status Quo 39:03 Scaling Venture Capital 40:37 Private Equity and Mad Men 42:52 Valley Shifts to Full Stack 45:59 Meeting Jim Clark 48:53 Founder vs. Manager at SGI 54:20 Recruiting Dinner Story 56:58 Starting the Next Company 57:57 Nintendo Online Gamble 58:33 Building Mosaic Browser 59:45 NSFnet Commercial Ban 1:01:28 Eternal September Shift 1:03:11 Spam and Web Controversy 1:04:49 Mosaic Tech Support Flood 1:07:49 Netscape Business Model 1:09:05 Early Internet Skepticism 1:11:15 Moral Panic Pattern 1:13:08 Bicycle Face Story 1:14:48 Music Panic Examples 1:18:12 Lessons from Jim Clark 1:19:36 Clark Versus Barksdale 1:21:22 Tesla Versus Edison 1:23:00 Edison Digression Setup 1:23:13 AI Forecasting Myths 1:23:43 Edison Phonograph Lesson 1:25:11 Netscape Two Jims 1:29:11 Bottling Innovation 1:31:44 Elon Management Code 1:32:24 IBM Big Gray Cloud 1:37:12 Engineer First Truth 1:38:28 Bottlenecks and Speed 1:42:46 Milli Elon Metric 1:47:20 Starlink Side Project 1:49:10 Closing Includes paid partnerships.




You're not "smart" if you don't know about these 4 formulas - the code is here - Quants now use the Monte Carlo method most often, but there are 3 other high-quality formulas besides it. So you don't trade on Polymarket or another prediction markets as if it were a biased coin. // • 1. Probability assessment (Monte Carlo) \[ \hat{p} = \frac{1}{N} \sum_{i=1}^{N} 1_{\{A_i\}} \] - What it does: Calculates the probability of an event through simulations. - How to use it: If you are modeling an outcome (e.g., a macro event, elections, BTC > 100k), you get a numerical estimate rather than a “70% feeling.” Compare: - Your estimate is 0.68 - the market is 0.61 = if the difference is stable → an edge is possible. • 2. Standard error of estimate \\[ SE = \\sqrt{\\frac{p(1-p)}{N}} \\] - What it does: Shows how noisy your estimate is. - How to use it: If you got 0.68, but SE = 0.02, and the market is 0.66 → your edge is within statistical error. This protects you from entering trades without a real advantage. • 3. Brier score \[ BS = \frac{1}{N} \sum (p_i - y_i)^2 \] - What it does: Checks whether your predictions are really accurate. - How to use it: Write down your probability before entering. After resolution, calculate the Brier score. If your Brier score is worse than ~0.20 → you are not systematically outperforming the market. This is a filter for the illusion of edge. • 4. effective sample size (particle filter) \\[ ESS = \\frac{1}{\\sum \\tilde{w}_i^2} \\] - What it does: Shows how “live” your probability update is when new information comes in. - How to use: If you update the probability based on news/data: Don't react to every price movement; the estimate should change in proportion to the strength of the signal. This protects against: Emotional overreactions and noise trading. // You must study the article by @gemchange_ltd if you want to use quantum formulas.

Of course when a ton of long vol guys are at a conference in NY, the market tanks 😪🤣







I would love for J Sai Deepak and Ranganathan to try & compete in writing a well cited paper making cogent arguments with even history undergrads from a good liberal arts college. Man, the delusions that Indian techbros have about their ability…..

















