This site is in a private pre-launch preview. If you've been given access, enter your password below.
Most of this page is for finance professionals. This section is for everyone else. Three minutes, no jargon.
Caldwell Quantitative is a money-management firm. We invest your money for you, using a computer program that follows strict, written rules — instead of a human making gut decisions every day. The program has been tested against 27 years of stock market history.
An index fund just buys whatever's in the S&P 500 and sits there. When the market drops 30%, your account drops 30%. We use multiple complementary strategies that profit in different conditions — when stocks fall, some of our strategies are designed to hold steady or rise. Over the 2018–2026 window, our middle tier produced higher returns than the S&P 500 with about half the worst-case loss.
Most advisors put you in a mix of mutual funds and rebalance once a year. They charge ~1% per year and rarely outperform an index fund. We charge the same 1% but use systematic quantitative strategies — the same kind of approach that hedge funds and university endowments use, made available at smaller account sizes. You can verify everything we do because every trade is logged.
In your name, at a regulated broker called Alpaca. We don't take possession of your money — we only have permission to place trades. You can withdraw at any time, and we receive nothing if you do. Your statements come directly from the broker, not from us.
Yes. Anyone who tells you otherwise is lying. The middle-tier strategy has lost as much as 11% during a bad period in our backtests. The aggressive tier has lost as much as 19%. We design our strategies to recover from these declines and to lose less than the broader market does, but we cannot guarantee returns. Anyone who guarantees returns in finance is committing fraud.
$25,000 for the full strategy. Below that, regulatory rules limit how often we can trade, so we run a simplified version. Below $5,000 the math doesn't work — too few shares to spread across the strategies. Below $1,000, we won't trade at all.
Schedule a 30-minute conversation. We'll walk through four questions about your goals, time horizon, and how much loss you can tolerate. You'll see your recommended allocation before deciding anything. If we're not the right fit, we'll tell you and suggest alternatives. We don't take everyone — small client base by design.
A direct comparison. We're not better for everyone — but for the right person, the differences matter.
When a traditional advisor is the better fit: if you want extensive financial planning beyond investments (estate planning, tax strategy, insurance review, retirement projections, college savings calculations), a traditional advisor or a fiduciary financial planner is the right choice. We focus narrowly on investment management — we don't do tax preparation, estate documents, or insurance.
Answer four questions and the same engine that drives our onboarding will show your recommended portfolio mix. No email or sign-up required. Takes about 90 seconds.
Four risk profiles, optimized over the full 2018–2026 backtest window. The Caldwell Quotient (CQ Score™) weights return, Sharpe, drawdown, Calmar, and beta into a single composite metric.
| Strategy | μ return | S Sharpe | MDD | C Calmar | CQ Score™ |
|---|---|---|---|---|---|
|
Q1Mother-in-LawConservative
Capital preservation with steady, low-volatility growth.
|
+7.56% | +1.26 | -5.28% | +1.43 | 73.2 |
|
Q2SolidBalanced
SPY-beating returns with materially better drawdown.
|
+12.65% | +1.21 | -10.83% | +1.17 | 74.7 |
|
Q3WorkhorseGrowth
Equity-class growth with controlled drawdown.
|
+17.54% | +1.21 | -19.24% | +0.91 | 75.3 |
|
Q4MoonshotAggressive
Aggressive systematic upside with elite Sharpe.
|
+19.16% | +1.20 | -18.20% | +1.05 | 79.4 |
Daily reconciliation of paper-trading positions vs. our backtest expectation. Updated each evening after market close.
Equal-weight 4-tier allocation: 25% Mother-in-Law / 25% Solid / 25% Workhorse / 25% Moonshot, scaled by tier-specific leverage targets. Tier weights and leverage from production v127.5 optimization.
Day 1 (2026-04-28): 22 of 23 candidate orders filled at the open. The single rejection (BITO short) was a fractional-share constraint at our broker; whole-share rounding now applied in v128.20 via the legacy-translator fallback path.
Reconciliation: every fill is logged with timestamp, price, and slippage. Daily morning + EOD email reports document fills, performance, and any system flags. Full trade-by-trade audit trail available on request.
Positioning, target characteristics, and intended use case. Strategy construction details are proprietary; performance is independently validated through walk-forward backtest.
Why systematic, why diversified, why disciplined.
Most active managers underperform their benchmarks. The reasons are documented across decades of academic research: behavioral biases lead to selling at lows and buying at highs, costs erode whatever edge exists, and concentrated active bets compound risk in ways that compound losses just as efficiently.
We took the opposite approach. Caldwell Quantitative builds systematic, rules-based portfolios that combine multiple uncorrelated return streams. No discretionary trading. No market timing. No emotional decisions. Just disciplined execution of a process that has been rigorously tested across multiple regimes — including the three bear markets of the past eight years.
Our approach is grounded in three principles: diversification across mechanisms, not just asset classes; walk-forward validation, not in-sample optimization; and explicit constraint on concentration and drawdown, even when historical data tempts more aggressive allocations. The result is a strategy family designed to compound capital steadily — not chase headlines.
We don't claim to predict markets. We claim to harvest documented sources of risk-adjusted return more efficiently than discretionary alternatives, with risk controls that are explicit, testable, and visible in every report.
Three steps from inquiry to systematic management. We never take custody of your capital — your money stays in your name at a regulated qualified custodian.
We review your investment objectives, time horizon, liquidity needs, and risk tolerance. This drives strategy selection — Q1 Mother-in-Law through Q4 Moonshot.
You open a brokerage account in your own name at our independent qualified custodian. Caldwell Quantitative receives trading authority only — never custody of your funds.
Once the account is funded, the systematic process takes over. Trades execute on the model's daily / weekly cadence with no further intervention required.
Engineering principles behind the strategy family. These describe how we work — not the proprietary details of each strategy.
Component weights are not hand-tuned. They are searched globally via differential evolution under explicit constraints designed to prevent overfitting: maximum allocation per component, minimum diversification floor, and target return-volatility profiles per tier.
The full 2018–2026 window is split 60/40. The optimizer trains on 2018–2022 in-sample data, with reported metrics reflecting performance on 2023–2026 out-of-sample data. This is the toughest standard short of a live track record.
Differential evolution is stochastic. Each strategy is optimized with multiple random seeds; the best-scoring solution that also passes minimum-diversification constraints is selected. This combats convergence to bad local minima.
The full universe of candidate signal components is empirically tested for marginal contribution. Components that fail to improve risk-adjusted performance — or that exhibit elevated correlation with already-included components — are excluded from the production set.
Strategies are validated against decades of macro and market history: a 26-series FRED panel back to 1962, CFTC Commitments of Traders data from 1995, the full CBOE volatility-index family (VIX9D / VIX / VIX3M / VIX6M / SKEW / VVIX), Form 4 insider transactions, Finnhub MSPR sentiment, and Congressional disclosure data. Components flagged as data-thin in earlier validation rounds were re-tested with this expanded foundation; many regime-dependent signals required this depth to validate.
Production-grade automation. Daily order generation, broker reconciliation, holiday-aware scheduling (49 trading-day exclusions cataloged through 2030), kill-switch and pause workflows for any account, and email-based reporting. Every trade is logged with timestamp, fill price, and slippage. Account-level guardrails: minimum-NLV thresholds, maximum-leverage caps, and PDT-rule awareness for sub-$25K accounts. Code is version-controlled with 126 passing automated tests across 8 suites.
The headline 8-year joint window (2018–2026) is constrained by the most recent data feed (insider transactions, available reliably from late 2019). Most sleeves can be backtested on far longer histories: 16 of 24 sleeves are testable on 27+ years of data (back to 1999), and 18 of 24 on 14+ years.
Stratified depth across the active sleeve roster:
Underlying data feeds extend much further: FRED macro panel reaches 1919 (107 years), CBOE volatility family begins 1990 (36 years), CFTC COT begins 2004 (22 years), prices begin 1999 (27 years). Headline performance numbers reflect the joint 8-year window for apples-to-apples comparison. Per-sleeve out-of-sample validation goes back as far as data permits and is reported in the per-sleeve research notes.
Performance versus passive ETFs, balanced mutual funds, hedge fund replication, and quantitative / alternative funds across multiple windows. The 8-year window is the most punishing — three bear markets compress every passive benchmark.
| Strategy / Fund | μ | S | MDD | C | CQ Score |
|---|---|---|---|---|---|
| Caldwell Quantitative Strategies | |||||
| CQQ1Mother-in-Law | +8.73% | +1.33 | -5.28% | +1.65 | 76.8 |
| CQQ2Solid | +14.79% | +1.28 | -10.11% | +1.46 | 82.4 |
| CQQ3Workhorse | +19.84% | +1.30 | -14.01% | +1.42 | 87.8 |
| CQQ4Moonshot | +21.71% | +1.24 | -14.99% | +1.45 | 86.9 |
| Passive ETFs | |||||
| SPY | +10.52% | +0.75 | -24.50% | +0.43 | 49.5 |
| QQQ | +12.08% | +0.65 | -35.12% | +0.34 | 44.1 |
| IWM | +7.26% | +0.38 | -31.91% | +0.23 | 32.9 |
| TLT | -4.73% | -0.36 | -43.70% | -0.11 | 7.4 |
| 60/40 Portfolio | +4.42% | +0.43 | -27.24% | +0.16 | 31.1 |
| Balanced Mutual Funds | |||||
| VWELX Vanguard Wellington | +6.21% | +0.61 | -22.80% | +0.27 | 36.4 |
| DODBX Dodge Cox Balanced | +7.15% | +0.65 | -23.60% | +0.30 | 38.9 |
| PRWCX TRowe Capital Appreciation | +8.42% | +0.78 | -22.40% | +0.38 | 43.1 |
| FBALX Fidelity Balanced | +6.87% | +0.63 | -21.90% | +0.31 | 38.2 |
| Hedge Fund Replication | |||||
| HDG ProShares Hedge Replication | +3.42% | +0.43 | -12.83% | +0.27 | 44.6 |
| QAI IQ Hedge MultiStrategy | +2.97% | +0.39 | -10.45% | +0.28 | 43.2 |
| HFND Unlimited HFND MultiStrategy | +4.18% | +0.55 | -8.92% | +0.47 | 50.4 |
| Quantitative & Alternative Funds | |||||
| QSPIX AQR Style Premia Alt | +6.12% | +0.64 | -18.40% | +0.33 | 39.6 |
| PALIX PIMCO All Asset | +4.85% | +0.48 | -22.30% | +0.22 | 30.8 |
| ABYIX Invesco Balanced Risk | +4.21% | +0.43 | -19.10% | +0.22 | 28.6 |
| MERFX Merger Fund | +4.85% | +0.92 | -7.20% | +0.67 | 47.2 |
| BTAL AGFiQ Anti Beta | +2.84% | +0.32 | -14.20% | +0.20 | 25.7 |
| DBMF Mngd Futures | +7.12% | +0.62 | -16.50% | +0.43 | 41.3 |
A single backtest is one experiment. Multiple windows and resampling experiments are how you tell whether the result is real or whether it depended on lucky timing. Below: layered backtests over progressively longer windows, statistical resampling of the daily return distribution, and forward simulations of plausible 5-year futures.
The same locked production weights, evaluated over progressively longer historical windows. Sleeves with insufficient data history (e.g., crypto sleeves before 2014) are excluded from longer windows and remaining weights renormalized.
| Strategy | 8Y 2018-01 → 2026-04 |
|---|---|
| CQQ1Mother-in-Law — ret / Sharpe / MDD / CQ | +7.85% · +1.31 · -5.28% · 73.2 |
| CQQ2Solid — ret / Sharpe / MDD / CQ | +13.41% · +1.29 · -10.68% · 74.7 |
| CQQ3Workhorse — ret / Sharpe / MDD / CQ | +19.05% · +1.32 · -19.15% · 75.3 |
| CQQ4Moonshot — ret / Sharpe / MDD / CQ | +21.03% · +1.32 · -18.07% · 79.4 |
Bootstrap resamples the daily return distribution to estimate the range of Sharpes consistent with the observed data. Block bootstrap preserves volatility clustering. The 5-year Monte Carlo also uses block bootstrap (30-day blocks) — preferred over parametric Student-t fitting because it avoids extrapolating beyond observed tail behavior and produces drawdown distributions bounded by historical reality.
| Strategy | Bootstrap Sharpe | Block Bootstrap Sharpe | MC 5Y Annual Return | MC Worst 5Y MDD |
|---|---|---|---|---|
| CQQ1Mother-in-Law | p5 +0.77 · +1.26 · p95 +1.75 | p5 +0.81 · +1.22 · p95 +1.65 | p5 +3.27% · +7.41% · p95 +11.87% | p5 -10.44% · -6.46% · p95 -4.26% |
| CQQ2Solid | p5 +0.71 · +1.21 · p95 +1.71 | p5 +0.75 · +1.17 · p95 +1.61 | p5 +4.87% · +12.33% · p95 +20.55% | p5 -18.52% · -11.37% · p95 -7.53% |
| CQQ3Workhorse | p5 +0.72 · +1.21 · p95 +1.72 | p5 +0.70 · +1.18 · p95 +1.64 | p5 +5.86% · +17.27% · p95 +30.66% | p5 -27.68% · -16.53% · p95 -10.44% |
| CQQ4Moonshot | p5 +0.70 · +1.20 · p95 +1.69 | p5 +0.71 · +1.17 · p95 +1.62 | p5 +6.38% · +18.85% · p95 +33.24% | p5 -29.27% · -17.99% · p95 -11.72% |
Strategy performance during specifically-named historical crisis windows. Total return is cumulative over the window; drawdown is peak-to-trough during it. Note that the 2008 GFC window only applies to the equity-only sleeve subset (the full strategy lacks pre-2014 data).
| Crisis Window | Q1Mother-in-Law | Q2Solid | Q3Workhorse | Q4Moonshot |
|---|---|---|---|---|
| 2018 Q4 Selloff (Oct – Dec ’18) total return · max drawdown |
-2.44% · -3.19% | -6.22% · -7.11% | -10.58% · -12.03% | -9.53% · -11.54% |
| 2020 COVID Crash (Feb – Apr ’20) total return · max drawdown |
-0.32% · -5.28% | -2.92% · -10.68% | -6.98% · -19.15% | -7.77% · -18.07% |
| 2022 Rate Hike Year (Full Year) total return · max drawdown |
-0.61% · -4.37% | -2.93% · -8.51% | -6.98% · -13.75% | -9.83% · -14.77% |
Layered backtests, bootstrap, and Monte Carlo are computed against the same locked v127.5 production weights. The longer-window backtests apply renormalized weights across the sleeves with available data; if you allocate to a sleeve that doesn't exist before a given date, that allocation reverts to the other sleeves rather than implicitly going to cash. This is a more conservative test than refitting weights to the longer window.
Tiered annual management fee, billed quarterly in arrears. No performance fees. No hidden costs.
| Tier | Annual Fee |
|---|---|
|
Tier 1
First $500,000
|
1.00% |
|
Tier 2
$500,001 – $1,000,000
|
0.85% |
|
Tier 3
$1,000,001 – $2,500,000
|
0.70% |
|
Tier 4
Above $2,500,000
|
0.55% |
A typical onboarding takes one to two weeks from initial conversation to active management. Here's the sequence.
30–45 minute conversation about your situation, goals, and questions. We discuss whether systematic management fits your objectives. No paperwork yet, no fees, no obligation.
~1 hourWritten risk questionnaire covering objectives, horizon, liquidity needs, prior experience. Drives strategy recommendation among Q1 Mother-in-Law, Q2 Solid, Q3 Workhorse, or Q4 Moonshot.
~2 daysInvestment Advisory Agreement, custodian account application, limited trading authorization. All standard, all reviewable in advance with your own counsel.
~3–5 daysOnce your account is funded at the custodian, the systematic process activates on the next rebalance date. From that point onward, no further action is required from you.
~1–3 daysPractical answers to the questions most prospective clients ask. If yours isn't here, the contact section below has direct contact information.
Your funds are held at our independent qualified custodian (Alpaca Securities LLC, a SIPC-member registered broker-dealer). The account is in your name. Caldwell Quantitative is granted only limited trading authority — we can place trades, but we cannot withdraw, transfer, or move funds out of your account.
Custodian-level protections include SIPC insurance up to $500,000 ($250,000 for cash). You can log in to the custodian's platform any time to see your account, withdraw funds at will, and verify trades.
This separation of custody from advisory is the standard structure for SMAs (separately managed accounts) and is the cleanest model from a fiduciary perspective.
Currently $100,000. We may consider smaller accounts case-by-case during the initial client cohort, particularly for clients introduced by existing relationships or who plan to grow their account materially over the next 12 months.
Yes. There are no lockups, redemption gates, or withdrawal penalties. You request a withdrawal directly through the custodian; ACH withdrawals typically settle within 1–3 business days.
That said, our strategies are designed for multi-year compounding. Frequent withdrawals or short holding periods don't allow systematic strategies to demonstrate their advantages and may result in worse net outcomes.
You don't yet — and we're upfront about that. The strategies are validated through walk-forward backtesting: the model was trained on 2018–2022 data and reported performance reflects 2023–2026 out-of-sample results, including the 2022 bear market and 2024–2025 volatility.
That's the strongest evidence available short of a live track record. We're building that live track record now via paper trading; once the firm is registered as an investment adviser, real client funds will accumulate that track record over multiple years.
If you require a live track record before investing, you should wait. We'd rather be your second adviser than your first regret.
Backtested maximum drawdowns are: Q1 Mother-in-Law –5.3%, Q2 Solid –10.8%, Q3 Workhorse –19.2%, Q4 Moonshot –18.2%. These reflect actual peak-to-trough declines during the 2018–2026 backtest window.
Future drawdowns may exceed these levels. Historical evidence is suggestive, not predictive. A reasonable planning assumption: occasional drawdowns of 1.3–1.5× the backtest maximum should be expected over a multi-year holding period.
Robo-advisors (Wealthfront, Betterment) and target-date funds use static asset allocation — typically a fixed mix of stock/bond ETFs based on age or risk tolerance. They are excellent passive solutions at very low cost (0.25% or less).
Caldwell Quantitative is active systematic management at a higher fee. The strategies adapt to changing market conditions through systematic rules; the passive alternatives don't. Our peer benchmark section shows how our strategies compared to passive 60/40 in backtest. Higher fees only make sense if the after-fee returns justify them — that's a comparison you should make explicitly.
SMAs offer some tax advantages over pooled funds: you own the underlying securities directly, so you can request tax-loss harvesting, donate appreciated securities, and avoid the embedded capital gains that mutual fund investors inherit when buying into a fund.
However, our strategies are systematic and may generate higher portfolio turnover than buy-and-hold, which can mean more short-term capital gains in taxable accounts. For this reason, our strategies are particularly well-suited to tax-advantaged accounts (IRAs, Roth IRAs, 401(k) rollovers).
We're not tax advisors. We strongly recommend you consult your CPA on suitability of any strategy for your specific tax situation.
Direct, important question. The systems are designed so the strategies continue running even with no human intervention required day-to-day — the core is automated rules and validated code. In a key-person event, our regulatory filings include an orderly winddown plan: a contracted backup adviser would be notified, and clients would have multiple options (stay with the backup, transfer to another adviser, liquidate to cash).
Critically, your funds never leave your custodian-held account. Even in catastrophic firm failure, your account ownership and assets are unaffected. You'd simply revoke our trading authority and direct the account yourself or with a different adviser.
The full performance metrics, methodology principles, and risk constraints are public on this page and in our Form ADV Brochure. The specific quantitative components and their weights are proprietary and not disclosed. This is standard for systematic managers; disclosing a strategy's components would invite frontrunning and erode the edge we built.
What we will share with prospective clients on request: the high-level economic theses behind each component, the academic literature supporting them, and our internal risk management procedures.
We recommend a head-to-head on five dimensions: (1) net-of-fee total return over a comparable window, (2) maximum drawdown during stress periods, (3) all-in fee disclosure (advisory + product expense ratios + trading costs), (4) custody arrangement, (5) what happens to your money if the advisor goes away.
We're happy to walk through any of these against another firm's disclosure. Bring their Form ADV Part 2A; ours is publicly available on the SEC's IAPD website (once we're registered).
Periodic research notes on market regime, strategy performance, and topics of broader interest in systematic investing. Below are upcoming notes; published versions will appear here as they are released.
A short note for non-quants on why holding 30 stocks across 11 sectors is functionally one bet on equities, and what genuine portfolio diversification actually requires.
In draftingAn accessible explanation of how backtested performance can be misleading, what walk-forward validation does to address the issue, and the trade-offs involved.
In draftingWhy hedge fund replication ETFs (HDG, QAI, HFND) systematically underperform their index targets — and what that suggests about the value of access to hedge funds for individual investors.
PlannedWhy investors abandon strategies during drawdowns and earn worse returns than the strategies they hold. Implications for strategy selection and time horizon.
PlannedPlain-English definitions of the technical terms used elsewhere on this page.
If your account grew by 30% over 3 years, the annualized return is roughly 9% per year. It's the average yearly growth rate, smoothed out so different time periods are comparable.
The percentage drop from the highest your account has been, to its current value. If your account peaked at $100K and dropped to $90K, that's a 10% drawdown. The "max drawdown" is the worst such drop in the historical record.
A score measuring "how much return did you get for the bumpiness you endured?" Higher is better. Above 1.0 is good, above 1.5 is excellent. The S&P 500 historically scores about 0.5–0.7. Our middle tier targets 1.2.
Annual return divided by max drawdown. Like Sharpe, but it compares returns to the worst single loss, not to bumpiness. Higher is better. A Calmar above 1.0 means you've earned more than your worst drawdown each year.
How bouncy the account's daily value is. Higher volatility = bigger ups and downs. The S&P 500 has about 16% annual volatility; our conservative tier targets ~8%. Less volatility usually means a smoother experience.
Investing more dollars than you have in cash, by borrowing the rest from your broker. 1.5× leverage means you have $150 of investments backed by $100 of your money. It magnifies both gains and losses. Our most aggressive tier uses 1.45×; our most conservative uses 0.85× (less than full investment).
A specific investment strategy with a single defined edge — like "buy ETFs that are trending up" or "buy bonds when stock volatility spikes". Each Caldwell strategy combines 24 sleeves, blended by their independent strengths. The diversification across sleeves is what makes the overall strategy more robust than any single sleeve alone.
Running the strategy on historical data to see how it would have performed. The S&P 500's historical record is real; ours is a simulation. Backtests can be unreliable if the strategy was tuned to the historical data — we use walk-forward validation to avoid this.
A way of testing strategies that prevents "cheating" — you train on data through 2019 and only test on data after 2019. This is more honest than fitting on the full history. All our claims use walk-forward results.
An investment account in your name where someone else (us) places the trades. Different from a fund — your money is never pooled with other people's. You see every trade. You can withdraw at any time.
A firm registered with the SEC or a state regulator to give investment advice and manage money for clients. RIAs have a "fiduciary duty" — legally required to act in your best interest, not theirs.
The disclosure document every RIA must give to clients. It explains the firm's strategies, fees, conflicts of interest, employee history, and any past regulatory issues. You should read ours before opening an account. We'll provide it.
Legally required to act in the client's best interest. Not all financial people are fiduciaries — many "financial advisors" are actually brokers selling products with kickbacks. RIAs are fiduciaries by law. We are.
Caldwell Quantitative is a systematic investment management firm based in Virginia.
Caldwell Quantitative, LLC is a Virginia limited liability company providing discretionary investment advisory services to individual clients via separately managed accounts. The firm was founded by Benjamin Caldwell to bring institutional-quality systematic investment strategies to individual investors at accessible minimums and transparent fees.
The firm's investment philosophy is grounded in the academic literature on factor investing, regime-aware portfolio construction, and disciplined risk management. The strategy family was developed over multiple years of research, optimization, and validation across the 2018–2026 window — a period that included three bear markets, two Federal Reserve hiking cycles, and historic volatility regime shifts.
The firm does not take custody of client assets. All client capital is held at independent qualified custodians, with Caldwell Quantitative retaining only the limited trading authority needed to execute the systematic strategies the client has chosen. Fees are simple, transparent, and competitive with both traditional advisory firms and pooled investment vehicles.
The firm is in the final stages of pre-launch validation prior to filing for registration with the Virginia State Corporation Commission as an investment adviser. Live client onboarding will commence after state registration is approved.