Trimming GRAIL, Inc.
−12.8pp in GRAL this quarter — the book's largest single weight change.
Big move
GURU PROFILE · Q1 2026 · 13F-HR Standard
PI
Paul Isaac — concentrated balanced contrarian
Paul Isaac runs Arbiter Partners, an eclectic value and special-situations shop. The book is concentrated, with very high risk and contrarian readings.
Arbiter Partners Capital Management
- Positions
- 42
Avg Annual Return, 5 yr
+34,1%
Alpha vs S&P 500, Annual
+15,6%
Max Drawdown
−39,4%
Sharpe (5 yr)
0,87
DNA Fingerprint
6 axes · Q1 2026Q1 2026 (latest)
Style Traits
-
- Value Value / Growth
- 61 +32 since Q1 2025
Growth Value
-
- Risk Risk Appetite
- 97 +2 since Q1 2025
Low High
-
- Concentration Portfolio Focus
- 68 −12 since Q1 2025
Diversified Focused
-
- Turnover Trading Activity
- 60 −27 since Q1 2025
Long-term Active
-
- Quality Profitability
- 3 −3 since Q1 2025
Speculative Profitable
-
- Contrarian Contrarianism
- 92 +58 since Q1 2025
With Crowd Against Crowd
Methodology
Each axis is a percentile rank from 0 to 100 relative to the profiled guru cohort; 50 is the cohort midpoint. The higher the value, the stronger the trait. The dashed overlay is the median shape of the guru's behavioral peer group, so you can read where this investor is unusual within their own style. The Trajectory arrows trace the drift from four quarters ago to now — an illustration of the method over a short window, not a statistical claim.
- Value (Value / Growth) — how cheaply the portfolio's stocks are priced versus the market, by valuation multiples (e.g. price-to-earnings). A higher score means cheaper, value-style picks; a lower score means pricier, growth-style bets the market expects to grow fast.
- Risk (Risk Appetite) — how much price volatility and business fragility the portfolio carries. A higher score means bolder, more volatile holdings (think biotech or meme stocks); a lower score means defensive, steady names like utilities and staples.
- Concentration (Portfolio Focus) — how much of the portfolio sits in its few largest positions. A higher score means a focused book of big convictions; a lower score means bets spread thinly across many names.
- Turnover (Trading Activity) — how much the lineup of holdings changes from one quarter to the next. A higher score means an active trader rotating names in and out; a lower score means a patient, buy-and-hold roster.
- Quality (Profitability) — how profitable and financially solid the businesses in the portfolio are. A higher score means durable, profitable companies; a lower score means speculative or pre-profit bets.
- Contrarian (Contrarianism) — how often the guru's trades go against what the other gurus are doing. A higher score means buying what the crowd sells and vice versa; a lower score means moving with the consensus.
Extended DNA
Beyond the radar's six: two further behavioral reads — whether new buys add to existing holdings or start fresh, and the size of companies the book favors — on the same 0–100 cohort-percentile scale.
- Follow-On Buying — Of the quarter's buys, the share that add to tickers already held rather than open fresh positions. A higher score means doubling down on existing convictions; a lower score means spreading into new bets.
- Size Bias — Whether the book tilts toward large, easy-to-trade companies or smaller, thinly-traded ones — gauged from each holding's trading liquidity, which stands in for company size rather than measuring market cap directly. A higher score means large, liquid mega-caps; a lower score means smaller, less-liquid names.
-
- Conviction Follow-On Buying
- 61 +45 since Q1 2025
New bets Adds to conviction
-
- Size Size Bias
- 3 −5 since Q1 2025
Small-cap Large-cap
DNA Anomalies
style stable · basis Q1 2024–Q4 2025Dot colour — type of deviationConcentration / contrarianMomentumOther positions (no deviation)5% of portfolio
Backtest · Cost of Delay
real 13F delay vs S&P 500Guru (2 years)
+89,3%
Copier (13F)
+126,3%
S&P 500
+44,6%
Managers with assets above $100M disclose positions in Form 13F-HR within 45 days after quarter end. The backtest runs three $100,000 portfolios: the guru rebalances at quarter-end prices, the copier mirrors the same weights on the actual filing date, and S&P 500 is the benchmark.
The hatched area is the cost of delay: the difference in cumulative returns between the guru and the copier in percentage points. The delay is not always a disadvantage — if a stock fell between quarter end and the filing date, the copier buys in cheaper.
The guru curve is also a model: 13F shows only long positions in US equities at quarter end, without intra-quarter trades, shorts, or cash. Commissions, slippage, and taxes are not accounted for.
Notable Signals
synthesis metricsBig bets against the crowd
Concentrated positions in names the cohort largely avoids — high-conviction contrarianism (concentration 68/100, contrarian 92/100 vs peers).
Contrarian
Swinging for the fences
High risk appetite with low quality scores — the boom-or-bust end of the cohort (risk 97/100, quality 3/100 vs peers).
Aggressive
Top 10 Positions
of 42 · Q1 2026| Ticker | Company | Δ for quarter | Weight |
|---|---|---|---|
| GRAL | GRAIL, Inc. | −12,8 | 31,6% |
| TSLA | Tesla, Inc. | +4,4 | 10,9% |
| GEF-B | Greif, Inc. | +2,7 | 10,1% |
| ODV | Osisko Development Corp. | +2,4 | 4,0% |
| CRBG | Corebridge Financial, Inc. | −0,8 | 3,3% |
| STXS | Stereotaxis, Inc. | −0,2 | 3,1% |
| SNDA | Sonida Senior Living, Inc. | −0,1 | 2,9% |
| OPY | Oppenheimer Holdings Inc. | +0,9 | 2,9% |
| CBL | CBL & Associates Properties, Inc. | +0,3 | 2,5% |
| NAVI | Navient Corporation | +1,5 | 2,5% |