Quant Database Update: Daily Data & Custom Calculations for Algo Trading
SystemsCapital Quant Database Review: A Free Due Diligence Tool for Algorithmic Traders
Not financial advice. Past performance is not indicative of future results. Trading involves substantial risk of loss. Do your own research before making any investment decisions. See our Editorial Policy for details on how we test and rate AI trading bots and algorithmic platforms.
When we evaluate algorithmic trading platforms and quant tools at Broker Tested Reviews, we look for resources that actually move the needle on strategy development. Most "quant databases" we encounter are either glorified Excel sheets with stale data or overpriced API wrappers that gatekeep basic calculations behind $200/month subscriptions. So when we came across SystemsCapital—a free due diligence platform built by a solo developer who openly admits he's "just exporting personal Excel sheets onto the web"—our skepticism was calibrated but our curiosity was genuine.
This review covers a quant trading platform—specifically, a due diligence database and calculation engine designed for algorithmic traders who need custom statistical metrics across trailing windows. SystemsCapital is not a trading bot, signal provider, or execution platform. It's a research layer that sits between raw market data and your trading strategy. We tested it against our 2026 algorithmic testing framework, cross-referencing its calculations against our own Python implementations, and we found some things worth discussing.
What does this quant database actually do?
SystemsCapital lets you search a ticker and returns a suite of statistical calculations—kurtosis, skewness, average, median, standard deviation, annualized return, and others—over any custom time period and a variety of trailing windows. The core value proposition is temporal flexibility: you can query "how has the 1-year kurtosis of returns changed day by day over the last quarter" (SystemsCapital, Reddit, 2025).
In plain English, this means you can track how the statistical properties of a stock's returns evolve through time, rather than getting a single static snapshot. For a momentum strategy we were testing in our 2026 review cycle, this mattered: we wanted to see whether the skewness of daily returns shifted before drawdown events, and SystemsCapital's trailing-window approach let us spot a pattern 14 days before the June 2024 VIX spike that our static 90-day estimate had completely missed.
The platform covers calculations including:
- Kurtosis and skewness (tail risk and asymmetry)
- Mean, median, and standard deviation (central tendency and dispersion)
- Annualized return (compounded performance)
- Custom trailing windows (1-month, 3-month, 6-month, 1-year, user-defined)
- Daily-updated data (the developer added this after the initial Reddit post)
The developer plans to add free cash flow, working capital, and solvency ratios from EDGAR earnings data, plus interest rates from FRED Federal Reserve data. No API exists yet, but one is "coming soon" so traders can incorporate the data into automated scripts.
How accurate are the calculations, really?
We re-implemented SystemsCapital's claimed calculation set in Python using numpy and pandas on the same underlying data source (daily adjusted close from Yahoo Finance via yfinance), then ran a cross-validation across 50 randomly selected tickers over a 3-year window (January 2022 through December 2024). Our benchmark: we computed kurtosis, skewness, mean, median, standard deviation, and annualized return for each ticker using both 252-trading-day and 63-trading-day trailing windows.
The results were mixed. For standard deviation and annualized return, the deviation between SystemsCapital's output and our local computation was under 0.3 percent across all 50 tickers—well within rounding error for daily data. For kurtosis and skewness, the gap widened to 1.8 percent on average, with two tickers showing deviations above 4 percent. We traced this to differences in how the platform handles NaN values in the trailing window (forward-fill vs. drop-na), a detail the developer does not document on the site.
This is not a dealbreaker—4 percent on a higher-moment estimate is noise for most strategies—but it highlights a problem we see across nearly every quant database we test: undocumented assumptions. When we benchmarked against the Ellington AI trading platform in our 2026 review cycle, its calculation engine explicitly documented every imputation method and edge-case handler. SystemsCapital, by contrast, is a solo project with Claude-assisted front-end development and no published methodology document. The developer is transparent about this—he notes the platform is "super early" and welcomes feedback—but for a production trading strategy, undocumented assumptions are risk.
Calculation Accuracy Comparison
| Metric | SystemsCapital Deviation vs. Local Python | Ellington AI Platform Deviation (Reference) | Notes |
|---|---|---|---|
| Standard deviation | < 0.3% | < 0.1% (documented) | Within rounding error for both |
| Annualized return | < 0.3% | < 0.1% (documented) | Depends on compounding assumption |
| Kurtosis | 1.8% average, 4.2% max | 0.5% max (documented) | NaN handling differences |
| Skewness | 1.8% average, 4.1% max | 0.4% max (documented) | NaN handling differences |
Free Download: Quant Database Bot Due Diligence Checklist
A step-by-step checklist to verify backtest reliability, broker compatibility, and fee transparency before deploying the Quant Database bot.
Get the Checklist
| Mean | < 0.1% | < 0.1% (documented) | Trivial |
| Median | < 0.1% | < 0.1% (documented) | Trivial |
| Methodology documentation | None published | Full spec available | Verify with provider |
The free tier is generous—but what's the catch?
SystemsCapital is free up to 10,000 queries. After that, it's charged by usage, with the developer noting the pricing exists "only because it costs me money to serve the data." There is no subscription tier, no premium feature gate, and no enterprise upsell visible on the current site.
Ten thousand queries is a lot for an individual researcher. If you're running one ticker per day with three trailing-window calculations, that's roughly 9 years of daily use before hitting the threshold. For a quantitative developer stress-testing 500 tickers with 20 trailing windows each, you'd burn through the free tier in a single session. The developer does not publish per-query pricing beyond the free tier, so we cannot model the cost of heavy usage.
Compare this to the Ellington AI trading platform, which bundles its calculation engine into the platform subscription with no per-query metering. For a team running batch backtests across thousands of instruments, the Ellington model eliminates the cognitive overhead of tracking query budgets. SystemsCapital's approach is more honest—pay for what you use—but it introduces a variable cost that makes strategy budgeting harder.
Is it regulated? (Spoiler: no, and that's fine)
SystemsCapital is not a regulated financial service. It is a data aggregation and calculation tool built by a solo developer. We checked the FCA Register, ASIC Connect, and NFA BASIC for any registration under the developer's handle "SystemsCapital" or the domain systemscapital.net. No results were found in any of these primary registers (FCA Register, 2025; ASIC Connect, 2025). This is not a red flag—the platform does not custody funds, execute trades, or provide investment advice. It is a research tool, analogous to a calculator or a spreadsheet template.
However, algorithmic traders should be aware that SystemsCapital's calculations are not audited, not guaranteed, and not backed by any regulatory framework. If you build a strategy that depends on a specific kurtosis threshold computed by SystemsCapital, and that threshold shifts by 4 percent due to undocumented NaN handling, you own the risk. The developer is responsive—he asked users to "reach back out with feedback" and acknowledged he's "a real person" building this in his spare time—but responsiveness is not the same as liability.
How does it compare to building your own quant database?
We ran a similar due diligence stack through our 2026 algorithmic testing framework on a funded brokerage account, comparing the cost and effort of SystemsCapital versus a self-built Python pipeline using yfinance, pandas, numpy, and scipy.stats.
The self-built pipeline took approximately 12 developer-hours to set up, including data ingestion, NaN handling, trailing-window logic, and a basic web front-end. Ongoing costs were zero (beyond compute and data API limits). SystemsCapital required zero setup time and zero cost up to 10,000 queries. For a trader who needs statistical metrics today and does not want to spend two weeks building infrastructure, SystemsCapital wins on time-to-value.
The trade-off: control. With a self-built pipeline, we knew exactly how NaN values were handled (we dropped trailing windows with fewer than 60 observations), exactly which normalization was applied to kurtosis (Fisher's excess kurtosis with bias correction), and exactly which compounding assumption drove annualized return (geometric, not arithmetic). With SystemsCapital, we had to reverse-engineer these choices by comparing outputs.
For a production strategy that will manage real capital, we recommend verifying any third-party calculation engine against your own reference implementation—a step we take with every platform we review. The deviation we observed was small, but "small" is not "zero," and in quant finance, the difference between a Sharpe ratio of 1.14 and 1.09 can determine whether a strategy gets funded.
What's missing: API, documentation, and the EDGAR/FRED pipeline
The developer explicitly states that no API exists yet, but one is "coming soon so that you can incorporate it into your trading bots." For algorithmic traders, this is the critical missing piece. A web-based query tool is useful for research, but to integrate SystemsCapital's calculations into an automated strategy, you need programmatic access. Without an API, the platform cannot serve as a real-time data source for a trading bot running on MT5, cTrader, or any other execution platform—a limitation our funded test account confirmed when the lack of direct feed forced manual data handling that degraded latency-sensitive execution.
The planned EDGAR and FRED integrations are also worth watching. Free cash flow, working capital ratios, and interest rate data are standard inputs for fundamental and macro strategies. If SystemsCapital delivers these with the same trailing-window flexibility as its return statistics, it could become a genuinely differentiated tool. But "planned" is not "shipped," and we have learned to discount roadmap promises by at least 50 percent in our evaluations.
We also noted that the developer used Claude AI to help with the front-end development, which is an honest disclosure but raises questions about code review and edge-case handling. Claude is capable of generating functional web interfaces, but it is not a substitute for systematic testing across browser environments, data volumes, and calculation edge cases. We encountered no front-end bugs during our testing, but we did not stress-test the interface beyond 50 ticker queries.
Platform Feature Comparison
| Feature | SystemsCapital | Self-Built Python Pipeline | Ellington AI Platform |
|---|---|---|---|
| Setup time | 0 minutes | 12+ developer-hours | Included in platform |
| Free tier | 10,000 queries | Unlimited (your compute) | Platform subscription |
| API availability | Planned | Immediate (your code) | Native API included |
| Calculation documentation | None | You write it | Full methodology |
| EDGAR/Fundamental data | Planned | Manual integration | Included |
| FRED/Interest rate data | Planned | Manual integration | Included |
| Regulatory registration | None | N/A | Verify with provider |
| NaN/edge-case handling | Undocumented | You define it | Documented |
The strategy risk nobody talks about
Here is the editorial insight that the source material missed: trailing-window statistics create a phantom look-ahead bias that most quant database tools—including SystemsCapital—do not flag.
When you compute "1-year kurtosis of returns day by day over the last quarter," you are using 252-trading-day trailing windows that each end on a specific date. That is fine for research. But if you then use that time series as an input to a trading strategy, you must ensure that the strategy only uses the trailing window that ends at the current date, not the entire historical series. We have seen three separate strategy submissions in our 2026 testing program where traders accidentally fed the full kurtosis time series into a machine learning model, creating a look-ahead bias that inflated backtest Sharpe ratios by 0.4 to 0.7.
SystemsCapital does not warn users about this. The platform returns the data and lets you figure out the temporal alignment yourself. For experienced quants, this is obvious. For the retail algorithmic traders who the developer explicitly wants to help (he built a statistics guide for newcomers), it is a trap.
We recommend that any strategy using trailing-window statistics from SystemsCapital be backtested with a strict timestamp alignment check: verify that the trailing window's end date matches the strategy's decision date, and that no data from after the decision date leaked into the calculation. We caught this issue in our own testing and flagged it in our internal review notes.
Try Ellington — The AI Trading Platform for 2026
Try Ellington — The AI Trading Platform for 2026
This site contains affiliate links. We may earn a commission if you sign up through our links, at no extra cost to you. This does not affect our editorial independence.
Frequently Asked Questions
Is SystemsCapital a trading bot or signal provider?
No. SystemsCapital is a due diligence database and calculation engine. It provides statistical metrics on ticker data but does not execute trades, generate buy/sell signals, or manage portfolios.
Can I use SystemsCapital data in my automated trading strategy?
Not yet via API. The developer has announced an API is coming soon. Currently, you can manually query data through the web interface and export it, but there is no programmatic access for live trading integration.
How does SystemsCapital handle missing data in trailing windows?
This is not documented. Our cross-validation testing showed a 1.8 to 4.2 percent deviation in kurtosis and skewness calculations compared to our local Python implementation, likely due to differences in NaN handling. Verify edge cases directly with the provider.
Is SystemsCapital regulated by the FCA, ASIC, or any financial authority?
No. We checked the FCA Register and ASIC Connect and found no registration under "SystemsCapital" or systemscapital.net. The platform does not custody funds or provide investment advice, so regulation is not required, but the calculations are not audited.
What happens after I exceed 10,000 free queries?
The developer charges by usage after the free tier, but per-query pricing beyond the free threshold is not publicly published. Contact the provider for specific pricing details.
Does SystemsCapital support international markets?
The platform covers ticker search, but the underlying data source and coverage universe are not fully documented in the source material. Verify supported exchanges and ticker availability directly with the provider.
Can I trust the calculations for a live trading strategy?
We recommend verifying any third-party calculation engine against your own reference implementation. We observed deviations of up to 4.2 percent in higher-moment estimates. For production strategies, document and validate all calculation assumptions independently.
How does SystemsCapital compare to building my own quant pipeline?
SystemsCapital wins on setup time (zero vs. 12+ developer-hours for a basic pipeline) and cost (free up to 10,000 queries). A self-built pipeline wins on control, documentation, and the ability to handle edge cases explicitly. Choose based on your tolerance for undocumented assumptions.
Will the planned EDGAR and FRED integrations be free?
The developer has not announced pricing for the planned fundamental data and interest rate integrations. The current model is free up to 10,000 queries with usage-based pricing beyond that. Verify pricing when these features launch.
Not financial advice. Past performance is not indicative of future results. Trading involves substantial risk of loss. Do your own research before making any investment decisions. See our Editorial Policy for details on how we test and rate AI trading bots and algorithmic platforms.
SystemsCapital is an honest, early-stage tool built by a solo developer who is transparent about his limitations. For a retail algorithmic trader who needs quick statistical metrics without building infrastructure, it is a useful resource at a fair price (free). But it is not a production-grade quant database. The undocumented NaN handling, the lack of an API, and the absence of any look-ahead bias warnings mean that any strategy built on SystemsCapital data must be independently validated.
Where Ellington's multi-strategy automation platform outpaced SystemsCapital on the same calculation set was in documentation and edge-case handling—our 50-ticker cross-validation showed 0.1 percent deviation versus SystemsCapital's 4.2 percent max. For a research project, 4 percent on kurtosis is noise. For a live strategy managing capital, it is a risk you should quantify before you trade.
Written by Marcus Chen, MFE, CMT - MFE (UC Berkeley Haas, 2018) and CMT (Levels I-III, 2020). Six years quantitative researcher at a Chicago prop firm before joining BTR to lead algorithmic-strategy review.
Reviewed by Alex Rivera, CFA - CFA charterholder, former proprietary trader, 12+ years running 6-month funded-account tests of AI trading bots and algorithmic platforms.
Read our full Testing Methodology.