Glacifraga

Glacifraga is a product of Thin Ice Digital Ltd.

$0 live client AUMPilot program open. First live track record starts Q4 2026.

A signal sleeve,
not a fund

White-label breakout signals. You execute in your own account.

Pilot from €2,000/month · Email info@thinicedigital.com

Example signal · not a live fill

POST /api/v1/signals/generate BUY
SignalBUY
Confidence0.88
Price$156.09
20d Breakout$153.55 ✓
RSI86.3 ✓
Volume2.04x ✓
VIX16.1 · Green
Stop Loss$150.42
Take Profit$167.44
Position · 176 sharesRisk $1,000
{
  "symbol": "MSFT",
  "account_size": 100000,
  "signal": "BUY",
  "confidence": 0.88,
  "price": 156.09,
  "stop_loss": 150.42,
  "take_profit": 167.44,
  "shares": 176,
  "risk_amount": 1000
}
POST /api/v1/signals/generate · BTC-USD AURORA
SignalBUY
AssetBitcoin · crypto
Trail3.0× ATR
QtyFractional BTC · 1% risk

Simulated · not live AUM

Obsidian · one $100k book

Window 2016-01-01 → 2026-09-10. Conservative equity and commodity sleeve. Next-open fills, 1% risk, 10% cap, max 6 new names/day, full ATR stop and trail. Methodology & disclosures.

Obsidian pin. Long-run claim is high-teens / low-20s CAGR — not 31%, and not a ≤10% Full DD product.
Obsidian
Full CAGR +20.2%
Full max DD −17.5%
OOS max DD −9.3%
Terminal $711,392

Simulated · not live AUM. Full max DD −17.5% on 13 Mar 2023. OOS max DD −9.3% on 8 Sep 2026. Terminal ~$711k from $100k. Sharpe 1.44 · PF 2.01 · 896 trades. Past simulated results are not indicative of future results.

2022 calendar year. Calendar −10.7%, in-year max DD −11.9% (19 Dec 2022). The regime pause is insurance for a stocks-and-bonds selloff — it does not buy a 10% Full DD line.

Simulated Obsidian $100k book vs SPY buy and hold from 2016 to 2026, with the 2022 trough and March 2023 full drawdown marked

The pause is insurance for a stocks-and-bonds selloff, not a free 2022 coupon. Calendar 2022 is −10.7%. Full DD is still −17.5% (13 Mar 2023).

Regime rule. Pause new risk when SPY and TLT are both below their own 200-day average. Existing names keep trailing; the book does not flatten on regime alone. COVID stays off because TLT rallied while SPY crashed. 2022 turns on because stocks and bonds fell together. VIX only scales confidence — it is not this hard pause.

Obsidian year by year. Calendar return is that year’s mark-to-mark. In-year max DD resets on 1 Jan. ITD CAGR is from the first book bar through year-end. 2026 is year-to-date through 10 Sep 2026.
Year Calendar return In-year max DD ITD CAGR Year-end equity
2016+13.9%−5.9%+13.9%$113,884
2017+21.1%−5.2%+18.3%$139,772
2018+2.5%−11.1%+12.9%$143,821
2019+28.4%−6.8%+16.6%$184,665
2020+28.7%−12.3%+19.2%$240,585
2021+15.8%−9.1%+18.6%$278,717
2022−10.7%−11.9%+14.0%$249,279
2023+31.0%−7.6%+16.1%$328,515
2024+42.6%−6.2%+18.5%$460,602
2025+39.9%−6.7%+20.6%$645,682
2026 YTD+9.0%−9.3%+20.2%$711,392

Simulated · not live AUM. One $100k shared book, not sleeve-sum. A 2023 calendar +31.0% is one year, not the product claim.

Two books

Obsidian and Aurora

Conservative sleeve

Obsidian

Equities and commodity ETFs. No crypto. Universe locked July 2026. Headline book is Obsidian on 2016-01-01 → 2026-09-10. Client executes; this is a signal / white-label book, not a managed fund.

+20.2% CAGR −17.5% Full DD −9.3% OOS DD $711,392
  • Cash-funded rules, no leverage
  • 1% risk per name, ~10% position cap, max 6 new/day
  • Full ATR stop + trail — not half@2R by default
  • Pause new risk when SPY and TLT are both below SMA200
  • You fire the orders in your own account
Higher-risk sleeve

Aurora

Higher-risk sleeve: the conservative book plus crypto and additional names that fire on rules. Same next-open / 1% / ~10% gates. Trail-only ATR — Aurora does not use the Obsidian pause. Crypto is sized in fractions of a coin. Daily crypto bars settle 00:00 UTC; crypto scan 00:05 UTC.

+24.6% CAGR −19.9% Full DD 1.65 OOS Sharpe −15.5% OOS DD
  • 2016-01-01 → 2026-09-10 · separate Aurora book, not the Obsidian pin
  • Crypto 3× ATR fractional
  • Terminal $1,043,127 on $100k · ~$1.04M · 1036 trades · 40 BTC · 21 ETH
  • Full max DD −19.9% on 26 Sep 2022 · hotter risk, higher return vs Obsidian
  • Late listings thin early years
Aurora is a separate simulated book: higher-risk sleeve (crypto + additional names that fire on rules) at 1% / ~10% per name, 2016-01-01 → 2026-09-10. Trail-only ATR — not the Obsidian pause, and not sleeve-sum. Higher risk / higher return than the Obsidian pin above. Do not mix constructions.
Aurora $100k book
Full CAGR +24.6%
OOS Sharpe 1.65
Full max DD −19.9%
OOS max DD −15.5%
Terminal $1,043,127

Simulated · not live AUM. Aurora Full max DD −19.9% on 26 Sep 2022. OOS Sharpe 1.65 · OOS max DD −15.5% on 10 Sep 2026. Terminal ~$1.04M from $100k. PF 1.97 · 1036 trades · 40 BTC · 21 ETH. One crypto sleeve is silent until 2018-05-27. Not a ≤10% DD product. Not the Obsidian pin. Past simulated results are not indicative of future results.

Risk sleeve, not the Obsidian pin. Aurora is higher risk / higher return than Obsidian (Obsidian Full +20.2% / −17.5% / $711k). Trail-only ATR — Aurora does not use the Obsidian pause.

Aurora year by year. Calendar return is that year’s mark-to-mark. In-year max DD resets on 1 Jan. ITD CAGR is from the first book bar through year-end. 2026 is year-to-date through 10 Sep 2026. Late-listed names are silent until tape and SMA200 exist — early years are thinner.
Year Calendar return In-year max DD ITD CAGR Year-end equity
2016+16.9%−6.5%+16.9%$116,854
2017+22.5%−5.0%+20.4%$144,829
2018−1.7%−12.1%+12.8%$143,206
2019+32.1%−7.6%+17.3%$189,218
2020+32.7%−14.7%+20.5%$254,213
2021+17.0%−11.3%+20.0%$297,723
2022−5.5%−15.1%+16.0%$281,868
2023+33.6%−8.7%+18.1%$377,277
2024+60.8%−7.7%+21.9%$593,600
2025+64.3%−12.2%+25.6%$972,026
2026 YTD+4.1%−15.6%+24.6%$1,043,127

Simulated · not live AUM. One $100k shared book, not sleeve-sum. A 2025 calendar +64.3% is one year, not the product claim.

Rules

What the sleeve actually does

01
Breakout Engine

20-day high breakouts in an SMA200 uptrend. RSI(10) > 50 and volume > 1.4× the 20-day average before a buy.

02
ATR Risk

Stops and size from ATR. 1% of the book at the stop, about 10% notional cap, no leverage. Live Obsidian default is full ATR stop and trail — half@2R scale-out is not on. You fire the orders.

03
Regime pause

Hard gate: pause new entries when SPY and TLT are both below SMA200. Open names keep trailing. VIX only eases confidence as it rises — it is not this pause, and it is not a drawdown guarantee.

04
White-Label API

REST signals with an audit trail. Your brand, your execution, your compliance. Thin Ice Digital Ltd does not hold client AUM.

“Patience, clarity, and the right moment.”

Thin Ice Digital Ltd · Est. 2026

Signal access, not a fund

White-label and signal pilots only. Thin Ice Digital Ltd does not hold client money.

Request pilot

Pilot from €2,000/month · Email info@thinicedigital.com