Triton Quantitative Trading
Quant
Date Joined
13th May, 2024
Bootcamp progress
0% Complete
Achievements
Algorithm Sharing
Share an algorithm with the community by making your algorithm public.
2026-02-25 06:17:32
Live Trading
Deploy a live trading algorithm using one of QuantConnect's supported brokerages.
2026-02-20 03:08:15
Python Programming
Choose Python as your preferred programming language.
2024-05-13 19:14:55
Verified User
Verify your identity and become a verified QuantConnect user.
2024-05-15 20:43:59
- Profile
- Backtests
- Community
- Certificates
Biography
Activity on QuantConnect
This section highlights your contributions and engagement across the QuantConnect platform — including backtests, live trades, published research, and community involvement through comments and threads. It reflects your overall activity as part of the QuantConnect community.
Public Backtests (63)
View More144.884Net Profit
47.259PSR
0.736Sharpe Ratio
0.082Alpha
0.314Beta
19.609CAR
20.7Drawdown
-2.51Loss Rate
37Parameters
1Security Types
1254Tradeable Dates
110Trades
0.321Treynor Ratio
2.96Win Rate
168.493Net Profit
61.198PSR
0.869Sharpe Ratio
0.072Alpha
0.757Beta
21.83CAR
18.6Drawdown
-0.26Loss Rate
0Parameters
1Security Types
1254Tradeable Dates
634Trades
0.15Treynor Ratio
0.58Win Rate
62.843Net Profit
10.031PSR
0.268Sharpe Ratio
-0.018Alpha
1.03Beta
10.252CAR
27.9Drawdown
-1.59Loss Rate
21Parameters
1Security Types
1254Tradeable Dates
106Trades
0.039Treynor Ratio
2.18Win Rate
429.311Net Profit
70.748PSR
1.175Sharpe Ratio
0.211Alpha
0.647Beta
39.588CAR
25.5Drawdown
-2.26Loss Rate
11Parameters
1Security Types
1254Tradeable Dates
107Trades
0.382Treynor Ratio
4.56Win Rate
221.476Net Profit
36.095PSR
0.757Sharpe Ratio
0.129Alpha
0.609Beta
26.329CAR
26Drawdown
-1.05Loss Rate
11Parameters
1Security Types
1254Tradeable Dates
34Trades
0.267Treynor Ratio
17.93Win Rate
Community
View MoreTriton submitted the research Filing Language Stability as a Selection Signal
Abstract
This research investigates whether stability in corporate filing language can serve as a systematic stock-selection signal. Using textual similarity across 10-K and 10-Q filings, the strategy ranks 100 liquid US equities and selects the 25 companies whose disclosures change the least over time. The portfolio is constructed using Sharpe-ratio optimization and compared with SPY from January 2020 through June 2026. Results show a modest improvement in risk-adjusted returns, although parameter testing produced mixed outcomes. The study highlights how alternative data from regulatory filings may reveal information about corporate consistency, while emphasizing the need for validation across periods and market regimes.
Triton submitted the research Volatility-Targeted QQQ–TLT Rotation with a Multi-Scale CNN-LSTM Forecast
Abstract
We present an intraday equity strategy that reduces market ambiguity by forecasting short-horizon volatility with a CNN-LSTM, then using the forecast as a regime filter. Using 15-minute OHLC windows, three parallel convolutional branches (kernel sizes 3, 5, and 11) learn multi-scale price patterns, which are concatenated and passed to an LSTM to model volatility persistence and clustering. A final dense head outputs a forward realized-volatility estimate that governs risk-on/risk-off positioning, with an 8-bar cooldown after exits to avoid whipsaws. In backtests from Feb 1 to Jun 1, 2022, the model exited SPY during turbulence and rotated into SH, returning 3.09% during the rate-hike selloff.
Triton submitted the research LPPLS for Bubbles in Speculative Markets
Abstract
This project implements the Log-Periodic Power Law Singularity (LPPLS) model in QuantConnect to detect speculative bubbles in high-liquidity equities and anticipate crash windows. LPPLS captures the shift from near-linear growth to super-exponential acceleration with increasingly frequent volatility oscillations as prices approach a critical time \(t_c\). To improve robustness and speed, the LPPLS equation is re-parameterized so four coefficients \((A,B,C_1,C_2)\) are solved via Ordinary Least Squares, leaving only \((t_c,m,\omega)\) for nonlinear optimization. A dual-EMA regime filter is integrated to gate signals to appropriate momentum states, reducing false positives and improving deployability in modern hype-driven markets.
144.884Net Profit
47.259PSR
0.736Sharpe Ratio
0.082Alpha
0.314Beta
19.609CAR
20.7Drawdown
-2.51Loss Rate
37Parameters
1Security Types
1254Tradeable Dates
110Trades
0.321Treynor Ratio
2.96Win Rate
168.493Net Profit
61.198PSR
0.869Sharpe Ratio
0.072Alpha
0.757Beta
21.83CAR
18.6Drawdown
-0.26Loss Rate
0Parameters
1Security Types
1254Tradeable Dates
634Trades
0.15Treynor Ratio
0.58Win Rate
62.843Net Profit
10.031PSR
0.268Sharpe Ratio
-0.018Alpha
1.03Beta
10.252CAR
27.9Drawdown
-1.59Loss Rate
21Parameters
1Security Types
1254Tradeable Dates
106Trades
0.039Treynor Ratio
2.18Win Rate
429.311Net Profit
70.748PSR
1.175Sharpe Ratio
0.211Alpha
0.647Beta
39.588CAR
25.5Drawdown
-2.26Loss Rate
11Parameters
1Security Types
1254Tradeable Dates
107Trades
0.382Treynor Ratio
4.56Win Rate
221.476Net Profit
36.095PSR
0.757Sharpe Ratio
0.129Alpha
0.609Beta
26.329CAR
26Drawdown
-1.05Loss Rate
11Parameters
1Security Types
1254Tradeable Dates
34Trades
0.267Treynor Ratio
17.93Win Rate
323.914Net Profit
47.452PSR
0.913Sharpe Ratio
0.164Alpha
0.84Beta
33.52CAR
29.6Drawdown
-1.68Loss Rate
14Parameters
1Security Types
1254Tradeable Dates
110Trades
0.251Treynor Ratio
3.85Win Rate
341.628Net Profit
48.354PSR
0.931Sharpe Ratio
0.171Alpha
0.867Beta
34.619CAR
34Drawdown
-4.16Loss Rate
14Parameters
1Security Types
1254Tradeable Dates
45Trades
0.253Treynor Ratio
8.29Win Rate
490.256Net Profit
36.556PSR
0.898Sharpe Ratio
0.241Alpha
1.296Beta
42.666CAR
45.9Drawdown
-2.17Loss Rate
20Parameters
1Security Types
1254Tradeable Dates
493Trades
0.242Treynor Ratio
1.36Win Rate
330.407Net Profit
56.785PSR
0.991Sharpe Ratio
0.157Alpha
0.914Beta
33.927CAR
27.2Drawdown
-2.41Loss Rate
22Parameters
1Security Types
1254Tradeable Dates
205Trades
0.227Treynor Ratio
2Win Rate
224.209Net Profit
79.108PSR
1.102Sharpe Ratio
0.121Alpha
0.409Beta
26.543CAR
16.2Drawdown
-0.39Loss Rate
27Parameters
1Security Types
1254Tradeable Dates
576Trades
0.352Treynor Ratio
0.62Win Rate
642.858Net Profit
66.307PSR
1.204Sharpe Ratio
0.237Alpha
1.095Beta
49.375CAR
41.7Drawdown
-0.83Loss Rate
0Parameters
1Security Types
1254Tradeable Dates
765Trades
0.299Treynor Ratio
1.14Win Rate
302.818Net Profit
77.082PSR
1.153Sharpe Ratio
0.16Alpha
0.466Beta
32.163CAR
14.8Drawdown
-0.92Loss Rate
65Parameters
1Security Types
1254Tradeable Dates
368Trades
0.399Treynor Ratio
1.62Win Rate
545.292Net Profit
77.287PSR
1.276Sharpe Ratio
0.261Alpha
0.476Beta
45.234CAR
19Drawdown
-1.09Loss Rate
12Parameters
1Security Types
1254Tradeable Dates
209Trades
0.603Treynor Ratio
2.3Win Rate
388.366Net Profit
88.089PSR
1.362Sharpe Ratio
0.196Alpha
0.401Beta
37.357CAR
17Drawdown
-0.65Loss Rate
22Parameters
1Security Types
1254Tradeable Dates
1114Trades
0.546Treynor Ratio
0.58Win Rate
1818.772Net Profit
76.665PSR
1.5Sharpe Ratio
0.571Alpha
0.032Beta
80.495CAR
47.4Drawdown
-2.65Loss Rate
26Parameters
1Security Types
1561Tradeable Dates
861Trades
18.139Treynor Ratio
5.85Win Rate
500.064Net Profit
96.537PSR
1.64Sharpe Ratio
0.214Alpha
0.726Beta
43.137CAR
13.5Drawdown
-0.58Loss Rate
12Parameters
1Security Types
1254Tradeable Dates
1151Trades
0.35Treynor Ratio
0.44Win Rate
2634.362Net Profit
97.479PSR
2.05Sharpe Ratio
0.567Alpha
0.729Beta
93.902CAR
25.5Drawdown
-1.3Loss Rate
60Parameters
1Security Types
1254Tradeable Dates
197Trades
0.834Treynor Ratio
5.04Win Rate
349.623Net Profit
61.586PSR
1.046Sharpe Ratio
0.162Alpha
0.9Beta
35.059CAR
27.2Drawdown
-2.31Loss Rate
24Parameters
1Security Types
1256Tradeable Dates
209Trades
0.24Treynor Ratio
2.06Win Rate
2737.393Net Profit
98.628PSR
2.141Sharpe Ratio
0.564Alpha
0.683Beta
95.199CAR
25.5Drawdown
-1.46Loss Rate
69Parameters
1Security Types
1255Tradeable Dates
198Trades
0.889Treynor Ratio
5.1Win Rate
195.691Net Profit
74.118PSR
1.02Sharpe Ratio
0.109Alpha
0.306Beta
24.203CAR
16.5Drawdown
-1.25Loss Rate
21Parameters
1Security Types
1255Tradeable Dates
405Trades
0.421Treynor Ratio
0.95Win Rate
92.425Net Profit
21.712PSR
0.458Sharpe Ratio
0Alpha
0.998Beta
13.989CAR
24.4Drawdown
0Loss Rate
5Parameters
1Security Types
1255Tradeable Dates
1Trades
0.065Treynor Ratio
0Win Rate
9.37Net Profit
14.888PSR
0.006Sharpe Ratio
-0.002Alpha
-0.202Beta
4.593CAR
16.1Drawdown
-0.4Loss Rate
24Parameters
2Security Types
501Tradeable Dates
115Trades
-0.003Treynor Ratio
0.45Win Rate
9.413Net Profit
14.899PSR
0.008Sharpe Ratio
-0.001Alpha
-0.203Beta
4.614CAR
16.1Drawdown
-0.41Loss Rate
24Parameters
2Security Types
501Tradeable Dates
115Trades
-0.004Treynor Ratio
0.45Win Rate
1436.161Net Profit
9.647PSR
0.643Sharpe Ratio
0.043Alpha
0.725Beta
15.111CAR
35Drawdown
-0.09Loss Rate
89Parameters
1Security Types
4879Tradeable Dates
5660Trades
0.126Treynor Ratio
0.19Win Rate
89.51Net Profit
20.469PSR
0.442Sharpe Ratio
-0.001Alpha
0.999Beta
13.633CAR
24.5Drawdown
0Loss Rate
5Parameters
1Security Types
1254Tradeable Dates
1Trades
0.062Treynor Ratio
0Win Rate
20.377Net Profit
8.721PSR
0.095Sharpe Ratio
-0.014Alpha
0.995Beta
6.373CAR
30.6Drawdown
-0.12Loss Rate
91Parameters
1Security Types
753Tradeable Dates
1150Trades
0.015Treynor Ratio
0.19Win Rate
-9.22Net Profit
5.516PSR
-0.295Sharpe Ratio
-0.02Alpha
0.995Beta
-7.012CAR
30.8Drawdown
-0.16Loss Rate
91Parameters
1Security Types
335Tradeable Dates
502Trades
-0.059Treynor Ratio
0.11Win Rate
-1.425Net Profit
20.835PSR
-0.137Sharpe Ratio
-0.224Alpha
0.844Beta
-3.843CAR
9.3Drawdown
-0.09Loss Rate
91Parameters
1Security Types
92Tradeable Dates
90Trades
-0.023Treynor Ratio
0.1Win Rate
101.79Net Profit
87.953PSR
2.025Sharpe Ratio
0.03Alpha
0.952Beta
49.268CAR
12.5Drawdown
-0.08Loss Rate
91Parameters
1Security Types
443Tradeable Dates
350Trades
0.351Treynor Ratio
0.54Win Rate
39.753Net Profit
72.106PSR
1.941Sharpe Ratio
0.38Alpha
0.566Beta
65.112CAR
16.8Drawdown
-0.17Loss Rate
91Parameters
1Security Types
169Tradeable Dates
166Trades
0.791Treynor Ratio
0.22Win Rate
Triton submitted the research Filing Language Stability as a Selection Signal
Abstract
This research investigates whether stability in corporate filing language can serve as a systematic stock-selection signal. Using textual similarity across 10-K and 10-Q filings, the strategy ranks 100 liquid US equities and selects the 25 companies whose disclosures change the least over time. The portfolio is constructed using Sharpe-ratio optimization and compared with SPY from January 2020 through June 2026. Results show a modest improvement in risk-adjusted returns, although parameter testing produced mixed outcomes. The study highlights how alternative data from regulatory filings may reveal information about corporate consistency, while emphasizing the need for validation across periods and market regimes.
Triton submitted the research Volatility-Targeted QQQ–TLT Rotation with a Multi-Scale CNN-LSTM Forecast
Abstract
We present an intraday equity strategy that reduces market ambiguity by forecasting short-horizon volatility with a CNN-LSTM, then using the forecast as a regime filter. Using 15-minute OHLC windows, three parallel convolutional branches (kernel sizes 3, 5, and 11) learn multi-scale price patterns, which are concatenated and passed to an LSTM to model volatility persistence and clustering. A final dense head outputs a forward realized-volatility estimate that governs risk-on/risk-off positioning, with an 8-bar cooldown after exits to avoid whipsaws. In backtests from Feb 1 to Jun 1, 2022, the model exited SPY during turbulence and rotated into SH, returning 3.09% during the rate-hike selloff.
Triton submitted the research LPPLS for Bubbles in Speculative Markets
Abstract
This project implements the Log-Periodic Power Law Singularity (LPPLS) model in QuantConnect to detect speculative bubbles in high-liquidity equities and anticipate crash windows. LPPLS captures the shift from near-linear growth to super-exponential acceleration with increasingly frequent volatility oscillations as prices approach a critical time \(t_c\). To improve robustness and speed, the LPPLS equation is re-parameterized so four coefficients \((A,B,C_1,C_2)\) are solved via Ordinary Least Squares, leaving only \((t_c,m,\omega)\) for nonlinear optimization. A dual-EMA regime filter is integrated to gate signals to appropriate momentum states, reducing false positives and improving deployability in modern hype-driven markets.
Open Quant League
The Open-Quant League is a quarterly competition between universities and investment clubs for the best-performing strategy. The previous quarter's code is open-sourced, and competitors must adapt to survive.
Get this certificate by participating in our Open Quant League
QuantConnect™ 2026. All Rights Reserved