I sold a Big Lizard in DRAM this morning, and no matter how high it goes, I still get paid. DRAM went zero to $10B in 7 weeks, a pace only the biggest Bitcoin and gold launches have matched. I went through the holdings file: 3 companies are 70% of this memory chip fund, and it's paying 74.6% implied volatility. Almost nobody trades a Big Lizard. Full trade structure in today's article

I'm honestly shocked by how easily traders get seduced by backtests. Tools like OptionsOmega, OptionStack, eDeltaPro are curve-fitting machines. Tweak delta, move DTE, change width, entries, exits and filters until the curve finally behaves. There is no mathematical reason a cherry-picked in-sample curve should predict future returns. Test 45 independent variations of a strategy with zero real edge and the expected best Sharpe already lands near 1.0. Test 1,000 and it rises to 1.46. Bailey, Borwein, Lopez de Prado and Zhu showed mathematically how selection alone manufactures "alpha." I ran the formula myself. Zero edge. Options are even dirtier. Duarte, Jones and Wang found microstructure biases exceeding 50 basis points per day in some option-return estimates. Yet the backtest assumes you traded at the midpoint. That is why NOBODY from the OptionsOmega crowd has EVER replicated those backtests with transparent live results. Think about it. Post the live account next to the backtest. I'll go first. 51.43% average CAGR over the last 5 years using the Trading Plan. Every drawdown public. No backtests needed. P.S. Here's a second problem I skipped above. A paper accepted at The Review of Financial Studies found that many options backtests quietly throw out price quotes that look wrong, using information nobody had at the time of the trade. The tool curve-fits your entries. It also cheats on your fills. P.P.S. This is why, in our hedge fund, we don't use retail tools like these in the way they're advertised. There's a much smarter process for researching these strategies, closer to how it gets done at the institutional level. Happy to get into that in the comments if people want it. Has anyone here found a vendor, in this space or elsewhere, that puts live results next to the backtest instead of just the curve?

You've already seen my personal portfolio and the hedge fund book side by side, and I post many trade ideas from both my personal account and our audited fund. However, what happens after is what matters more: how to roll, how to adjust and transform trades, how to neutralize delta, how to hedge, how to recenter as the market moves, and how to repair trades that went wrong. That's what the real Model Portfolio is going to show. You already pushed me toward this idea in the previous survey. Now I'm funding a real account at a real broker and running it completely in the open. Summer engagement and market participation both run lower here, and a launch like this needs everyone paying attention. So I'm targeting mid-September. How I deliver all of this in real time is still something I'm working out. I'm exploring whether I can provide premium subscribers with read-only login and password, allowing you to log in anytime and see the actual portfolio directly: positions, P&L, buying power, Greeks, and adjustments. I still need to determine the safest and most practical solution. Two decisions left before I lock it in. Question: How big should the real Model Portfolio be? Vote for the size that teaches you the most. Bigger doesn't automatically mean more useful to watch.

Many people here trade options on leveraged ETFs. I simulated TQQQ back to QQQ's inception in March 1999, with financing costs and the fund's expense ratio built into the model. Almost every leveraged ETF discussion eventually runs into assumption 3x daily leverage should produce something close to 3x the long-term return. The pre-2010 series models 3x daily exposure to QQQ, with financing costs on the borrowed notional and the fund's expense ratio subtracted daily. Starting in 2010, the simulated series is spliced directly into TQQQ's real, traded adjusted price history, so everything after that point comes from actual market data. The results, $10,000 invested in March 1999: - QQQ: $10,000 → $167,265 (11% annualized) - TQQQ (simulated pre-2010, real data after): $10,000 → $24,569 (3.4% annualized) Maximum drawdown over the same 27 years: - QQQ: -82.96% - TQQQ: -99.98% A -99.98% drawdown means every $10,000 fell to $2. TQQQ carried far more risk the entire way and still finished with $14,569 in total profit against QQQ's $157,265, under 10% of the unleveraged return. The volatility drag (beta slippage) scales with the square of the leverage multiple. So double the leverage and the drag roughly quadruples, triple it and the drag runs close to nine times larger. That is why I don't trade options on leveraged ETFs. You would be layering theta and IV risk on top of an instrument that is already decaying by design and has never been tested by the environment that would break it.

Since 1991, REITs and the S&P 500 have delivered almost the same total return, but the engine was completely different. Most options traders never notice this. REITs are boring. Theta is not. I used this week's earnings-volatility spike to get paid extra for buying real estate at my price, instead of whatever price the market hands me on a random Wednesday. Active trading builds the war chest. Boring trades like this help build your net worth. You need both.