Most people rode −75% to the bottom not because they were wrong — but because they had no structured quantitative framework for reading market conditions. Every morning we publish the model's regime classification and standardized exposure variable — rigorous, quantified research context. Subscribers independently decide what, if anything, to do with it.
No API keys. No custody. Cancel renewal at any time. General research, delivered to Telegram once a day.
You didn't get hurt because you were wrong about Bitcoin. You got hurt because you held the same amount no matter what the market was doing. So you ate the full drawdown on the way down — and then, too burned to step back in, you watched most of the recovery happen without you. That's the part nobody says out loud: the crash and the missed rebound aren't two mistakes. They're the same mistake, made twice. Not a conviction problem. A sizing problem.
is what you now need just to get back to even. Not to profit — to break even. That's years of waiting. Most people don't last. They capitulate within sight of the bottom.
is all a −11.2% drawdown requires to recover, compared with a +200% gain needed after a −66.7% drawdown. Smaller drawdowns have historically been easier to hold through — but past model behaviour does not guarantee future results.
No single indicator survives every market. The ensemble combines three independent models watching different horizons, each checking the others — resolving to one standardized exposure variable, 0–100%.
Reads where Bitcoin sits in its long-horizon cycle and suppresses false signals when the tape is just noise. The slow, steady hand.
The most influential model. Watches the tactical horizon and detects regime transitions as they develop — before the drawdown has fully materialised in historical data.
Built specifically to disagree with the Tactical model when the evidence for a regime change is thin. When it contradicts, confidence drops and the model's exposure variable decreases. It exists to prevent overreaction.
Models can be wrong. So there's a hard rule that overrides them: if Bitcoin's drawdown breaches a fixed threshold while the ensemble still reads Bull, the model's exposure variable is automatically halved — regardless of what any individual model says.
A seatbelt for the case where the data is lying and price is telling the truth.
Every layer in the live model has been run through dedicated stress tests: synthetic worst-case paths, regime-transition shocks, liquidity-constrained periods. Model outputs, regime classifications, and confidence readings are archived daily — the historical record cannot be revised retroactively.
You're not trusting a screenshot. You're trusting a documented, version-controlled record.
Powered by a multi-source data pipeline covering on-chain, derivatives, and cross-asset feeds.
Anyone can post a green curve. The question a serious person asks is how was it produced — and that's exactly where most signal services fall apart.
Most services show you a backtest. We show selected summary results from two independent test periods: a nested walk-forward OOS where the model was frozen before the window opened, and a blind holdout the models had genuinely never touched. Both produced risk-adjusted results meaningfully above buy-and-hold.
Holdout 2025-01-01 to 2026-05-31 — current production model (net of 5bp execution cost)
This is the model currently supplied to subscribers. CAGR figures are annualised, not cumulative returns.
| Metric | Vantegio | BTC B&H |
|---|---|---|
| CAGR | +30.7% | −16.3% |
| Max drawdown | −16.4% | −49.5% |
| Calmar ratio | 1.87 | −0.33 |
| Sharpe ratio | 1.21 | −0.18 |
Nested walk-forward OOS 2022-01-01 to 2024-12-31 — current production model (net of 5bp execution cost)
This is the model currently supplied to subscribers. CAGR figures are annualised, not cumulative returns.
| Metric | Vantegio | BTC B&H |
|---|---|---|
| CAGR | +76.4% | +25.6% |
| Max drawdown | −11.2% | −66.7% |
| Calmar ratio | 6.83 | 0.38 |
| Sharpe ratio | 2.15 | 0.69 |
| Avg. BTC exposure | 36% | 100% |
A blind holdout is the closest thing to a genuine out-of-sample test — the model was sealed before this period opened and has not been retouched since. It reduces, but does not eliminate, overfitting and selection bias. Bitcoin's annualised return was −16.3% over this period. The model's annualised CAGR was +30.7%. That's not a promise about the future. It's a record of what the model output was when it had no way to know what was coming. Past performance does not guarantee future results.
On the nested walk-forward OOS: model Calmar 6.83 vs 0.38 for buy-and-hold — 18× more return per unit of drawdown. Bitcoin's −66.7% crater requires a +200% recovery just to break even. The nested walk-forward max drawdown of −11.2% requires just +13%.
We tested whether the edge is just Bitcoin exposure in disguise: hold passive BTC at the model's own average exposure level, and compare. On the sealed holdout, that same-exposure passive book lost money (−4.9%) while the model made +30.7%, with a low beta to BTC (0.32–0.40) throughout both test windows.
The return isn't coming from being invested. It's coming from being invested at the right times.
The de-risking shows up in the shape of the returns, not just the average: holdout return skew is +2.2, strongly right-tailed — the model captured 46% of Bitcoin's up moves but only 35% of its down moves. A 5,000-path Monte-Carlo simulation over a 17-month forward horizon puts the odds of a losing 17-month stretch at 5%.
The edge is timing-sensitive: execute within ~2 hours of the daily signal. At +2h delay, the sealed holdout retains a Sharpe of 0.96 (of 1.13 at zero delay); a 4–6h delay roughly halves it.
This is a live execution discipline, not a backtest assumption — daily research is published on a fixed early schedule specifically so a prompt subscriber can act inside this window.
This isn't a model built once and left to age. It's a living research engine: new analytical layers are tested continuously against the same strict holdout discipline. When one passes, it's integrated into the live output automatically — no migration, no re-subscription, no action on your part. The research you access next year will be more rigorous than today's.
The core regime ensemble has been running live since launch. A macro risk-appetite overlay has since passed the holdout gate and been folded in — a validated improvement, not a patch. This is the process in action.
We have rejected more strategies than we have deployed. The ones that didn't survive a blind holdout — where the model had genuinely never seen that data — get archived, not shipped. You only ever see what earned its place.
A signal can fool you with bad statistics, or it can fool you with a pretty backtest that never survives real markets. We're built so that neither gets past us.
Deep technical background in quantitative modelling and machine intelligence, applied specifically to crypto markets. Built the model architecture, the validation framework, and the research pipeline behind every number on this page. The kind of person who knows exactly why most "signals" are statistical artefacts — and engineered this system specifically so it wouldn't be one.
Years of live trading experience across foreign-exchange and stock markets. Brings the practitioner's lens the math can't supply: which signals survive real execution, where models get punished by the way markets actually move, and what a drawdown feels like with real money on the line. The one who asks of every model — "would I genuinely have held this through 2022?"
No dashboard to babysit. No charts to stare at. Every morning a concise research brief lands in a private Telegram channel — regime classification, the model's standardized exposure variable, confidence level, and the full model breakdown so you understand why the output changed.
Two independent windows. First, a walk-forward OOS (2022–2024) where the model was frozen before the window opened and run forward unchanged, no refit: +76.4% CAGR, −11.2% max drawdown, Calmar 6.83 at 36% average BTC exposure. Buy-and-hold over the same period: +25.6% CAGR, −66.7% drawdown.
Second, the holdout (2025-01-01 to 2026-05-31) — data the models had genuinely never seen: annualised CAGR +30.7%, Calmar 1.87, −16.4% max drawdown. Bitcoin's annualised return was −16.3% over the same period with a −49.5% drawdown. These are annualised figures, not cumulative returns. Model Calmar 1.87 vs −0.33 for buy-and-hold. A blind holdout reduces but does not eliminate overfitting risk. Past performance does not guarantee future results.
The model targets ~36–49% BTC on average rather than 100%. Less raw exposure, but a −11.2% walk-forward worst case requires just +13% to recover — versus +200% from a −66.7% hole. Calmar — return divided by max drawdown — is one risk-adjusted measure used in this analysis. Model (OOS): 6.83 OOS vs 0.38 for buy-and-hold in this simulation.
Fair question — it's the right one to ask, because a good-looking backtest is trivial to manufacture.
The numbers on this page are walk-forward out-of-sample. The model's parameters were frozen before the test window opened, then run forward across multiple years of data — spanning bull runs, crashes, and everything in between — with no re-fitting, no cherry-picked window, no parameter nudging after the fact. It's the difference between predicting tomorrow and "predicting" yesterday.
It's not a promise about the future — nothing is. A blind holdout reduces but does not eliminate overfitting or selection bias, and results from one period do not predict another.
It doesn't ship. We've tested and rejected more strategies than we've deployed.
The ones that didn't survive a blind holdout test — where the model had genuinely never seen that data — get documented and archived. Each research question receives its own separate holdout window, opened once, and not reused. That discipline reduces, but does not eliminate, selection bias. The performance history shown on this page was produced by the model versions identified in the methodology notes, not by candidates that were tested and rejected.
No. Vantegio publishes quantitative model outputs — regime classifications and model exposure readings — as general research. This is not personalised financial advice and nothing on this channel should be construed as a recommendation to buy, sell, or hold any asset. We are not financial advisers, we don't know your personal situation, and we never touch your funds. You are solely responsible for any investment decisions you make. If you need advice tailored to your circumstances, consult a licensed financial professional.
About once a week on average. Historically the model changes regime classification roughly 55–115 times a year (~80 on average). Most days the output is unchanged — the research brief confirms the existing classification. Subscribers independently decide what, if anything, to do with the research. Built for people with jobs, not a screen-watching tool.
The model output is a standardized 0–100% research variable — it is not tied to any specific platform or implementation. Subscribers who choose to use it may do so on any venue they already use — Kraken, Coinbase, Bitstamp, Binance, or elsewhere. We never custody your funds or connect to your account. Any decision a subscriber makes is their own independent decision, in full control.
Send /billing to the bot and cancel through the Stripe portal — no email, no retention call. Cancellation takes effect at the end of the period you've paid for. Full cancellation details are available on the cancellation page.
Annualised CAGR +30.7% vs Bitcoin −16.3% · walk-forward MDD −11.2%, holdout MDD −16.4% · cancel renewal at any time.
Get the daily regime read →