The personality. 58-year-old Istanbul-based US-tech mega-cap investor, on X since November 2011 (~15-year account), 56,177 lifetime tweets at ~10/day, profile bio: “US Equities & Tech Investor. 4 YouTube videos a week. I do not give recommendations.” Founder of the paid Skool community “Haddini As Kulübü” where pre-earnings analyses are paywalled to members. Disclaimer-first persona: “I share thoughts, not advice.”
The two public claims under audit (as of 2026-05-22):
• +99.10% since March 7, 2025 (vs SPX +29.00%, NDX +45.00%)
• +33.30% YTD 2026 (vs SPX +8.25%, NDX +15.02%)
Stated portfolio shape: largest position NVDA (“PEG 0.99, cheapest mega-cap”), long-term tech spot, “small options book,” ~14.6% cash, no hedge, no Turkish equity, no crypto. Last 21-day mention distribution (n=111): SPY/SPX 20, NVDA 17, QQQ 16, GOOGL 5, MU 5, META 4 — consistent with the stated NVDA-anchored mega-cap concentration.
The test. Build five mechanical “beta-control” portfolios for the same window as each claim, run them against the claim, and report which allocation lands closest. If a stock-selection edge exists, the claim should beat the best mechanical mix. If not, one of these portfolios will reproduce the claim almost exactly — meaning the returns are explained by allocation, not selection.
| Allocation | Total Return | vs Bora's +99.10% | Note |
|---|---|---|---|
| Bora's claim | +99.10% | — | Self-reported, 15m |
| NVDA-only buy & hold | +94.85% | −4.25% gap | Single name explains 95.7% of the claim |
| 60/20/20 (NVDA+META+GOOGL) | +81.05% | −18.05% gap | Stated portfolio shape underperforms |
| MAG7 equal-weight | +45.10% | −54.00% gap | Diversification destroys the claim |
| QQQ | +46.21% | −52.89% gap | Bora's NDX +45% claim is close to this |
| SPY | +30.83% | −68.27% gap | Bora's SPX +29% claim aligns |
The +99% claim decomposes almost entirely into NVDA buy & hold.
NVDA-only return over the same window is +94.85% — 95.7% of the +99.10% claim. The residual edge is +4.25% over 15 months, or roughly +0.28% per month. For comparison, the stated “small options book” could trivially generate that residual at low option-weight (3-5% of NAV in OTM calls on the same NVDA, MSFT, GOOGL names). Stock-selection edge after beta-control is statistically zero.
Solving exactly: w * NVDA + (1-w) * QQQ = 0.991 gives w = 108.5% — meaning a portfolio of 108.5% NVDA and −8.5% QQQ matches the claim. The 8.5% over-allocation is the option leverage on top of the cash equity book.
| Ticker | Window 1 Return | vs SPY (+30.83%) | Note |
|---|---|---|---|
| $GOOGL | +124.00% | +93.2% | 1Y leader; Gemini cycle + Pelosi-bought tape |
| $NVDA | +94.85% | +64.0% | Bora's largest position; AI capex still bid |
| $TSLA | +59.08% | +28.2% | FSD/robotaxi narrative back online H2 2025 |
| $AMZN | +34.74% | +3.9% | Barely beats SPY |
| $AAPL | +28.25% | −2.6% | Slight SPY laggard |
| $MSFT | +7.60% | −23.2% | Capex compression, multiple repricing |
| $META | −2.53% | −33.4% | Only MAG7 name negative in the window |
Only NVDA, GOOGL, and TSLA beat the basket benchmark single-handed. Everything else — MSFT, META, AAPL — would have dragged a real diversified portfolio significantly below +99%. The 60/20/20 stated-shape portfolio comes in at +81% in our replication: still 18 points short of the claim. The only way to land at +99% with the stated equity book plus “small options” is to be heavily concentrated in NVDA with a small option overlay — which is exactly what Bora describes.
| Allocation | YTD Return | vs Bora's +33.30% | Note |
|---|---|---|---|
| Bora's claim | +33.30% | — | Self-reported, 4.5 months |
| $GOOGL alone | +23.09% | −10.21% | Highest single-name YTD — still short |
| $AMZN alone | +18.53% | −14.77% | Solid Q1, decel late April |
| QQQ | +16.68% | −16.62% | Claim is 2.00× this |
| NVDA-only | +16.24% | −17.06% | Half the claim. Q1 dip + recovery |
| 60/20/20 (NVDA+META+GOOGL) | +13.66% | −19.64% | META drag −6.5% YTD |
| MAG7 equal-weight | +5.79% | −27.51% | MSFT −11%, META −6.5%, TSLA −4.6% |
| SPY | +9.01% | −24.29% | Broad market reference |
+33% YTD does not reconcile with pure equity.
The strongest pure-equity outcome in our beta-control set is GOOGL alone at +23.09% — still 10 points below the claim. NVDA-only YTD is +16.24%, less than half the claim. QQQ matches NVDA at +16.68%, also half. There is no mix of cash equity in mega-cap names that produces +33% YTD without significant option leverage or a different entry date than Jan 1.
Two scenarios reconcile the claim:
(a) Option-heavy book. 25-35% of NAV in OTM calls on NVDA/GOOGL during the Q1 dip rally would generate the additional 15-17 points. This contradicts the stated “small options.”
(b) Cherry-picked window. Entering at the Q1 dip (late February / early March 2026, post-correction) rather than Jan 1 would make +33% trivial — NVDA rallied ~40% from the Q1 low to May 22. This is a measurement-convention issue, not a performance issue, but it materially changes the apples-to-apples comparison.
Window 1 is honest, Window 2 is harder. The +99% over 15 months reconciles cleanly with NVDA buy & hold + small option overlay. The +33% YTD over 4.5 months does not reconcile with any pure-equity NVDA-heavy book. The most parsimonious explanation: option weight is materially higher than “small,” or the YTD measurement window is entered post-dip rather than from Jan 1. We have not found a third explanation that fits the stated portfolio shape.
This is not a fraud claim — it's a transparency claim. Bora has never published a brokerage statement or live trade tape. The +99% claim is consistent with stated portfolio shape. The +33% claim requires assumptions that are unstated. The audit conclusion is not that the returns are fake — it is that the stock-selection edge implied by the claims is overstated, because the bulk is mechanical NVDA beta available to anyone via NVDA buy & hold.
1. If you like the thesis, just buy NVDA + GOOGL.
NVDA buy & hold from Mar 2025 returned +94.85%. Adding GOOGL exposure would have lifted the basket meaningfully without paying for a Skool subscription. The “edge” the claim implies is sub-+0.3%/month residual that you would not capture by mirroring publicly available picks anyway.
2. Treat the Skool subscription as education, not signal.
Pre-earnings analyses inside the paid community may be high-quality learning material. They are not the source of the +99% number — that number comes from holding NVDA. Pay for the framework if you want to learn long-term mega-cap investing; don't pay expecting actionable picks that beat NVDA buy & hold.
3. Don't take the +33% YTD claim at face value.
It does not reconcile with pure-equity allocation on any mix we tested. Either the options weight is materially higher than stated, or the YTD entry is post-Q1-dip rather than Jan 1. Until brokerage statements are public, treat YTD as the unverifiable claim and the 15-month number as the verifiable (NVDA-equivalent) one.
4. Risk-control reminder.
NVDA at 108.5% effective weight is a single-name volatility book. A 30% NVDA drawdown takes the implied portfolio down ~33%. The +99% claim window did not test that scenario in real time. If you want NVDA-heavy exposure, size it for the drawdown you can stomach — not the upside Bora has already experienced.
+99.10% over 15 months = NVDA buy & hold (+94.85%) + ~+4% small-options residual. Implied portfolio: 108.5% NVDA-equivalent, −8.5% QQQ — i.e., NVDA + small leverage. Honest reading of the stated portfolio shape.
+33.30% YTD does not match any pure-equity mega-cap allocation we tested (max +23% on GOOGL alone). Reconciliation requires either >25% NAV in options or a post-Q1-dip YTD entry. Stated “small options” contradicts the former.
After beta-control, the 15-month residual edge over NVDA-only is +0.28%/month — in option-overlay range, not selection. Pay for the framework, not for the picks. The picks are NVDA + GOOGL, which everyone can buy directly.
Report generated 2026-05-23 by the fxcryptobots research desk. Source: @BoraOzkent public X posts and YouTube content; price data from Yahoo Finance daily bars. Beta-control allocations are mechanical (static weights, daily rebalance, no transaction-cost overlay). “Implied NVDA weight” solves w · NVDA + (1−w) · QQQ = claim. Per-name returns are total return including dividends via auto-adjusted close. This is research and educational analysis, not investment advice and not an accusation of misrepresentation; see our risk disclaimer.