Prev2 / 4Next
TQQQ (leveraged QQQ ETF)

TQQQ VWAP System

Minute-resolution VWAP band breakout with event-day filters

Backtest results

Performance

Core breakout plus event-day filters

Equity Curve

Drag left-to-right to zoom inJan 2015Jan 2026 · normalized to $100k start
$43.2M$32.4M$21.6M$10.8M$0
Jan 15Oct 17Jul 20Apr 23Dec 25

Drawdown

depth from prior peak
39.7%29.8%19.9%9.9%0.0%
Jan 15Oct 17Jul 20Apr 23Dec 25

Performance

Annualized return
71.0%
Sharpe ratio
1.36
Max drawdown
−36.8%

Detail

Sortino
3.46
Expectancy
0.27
PSR
73%
Win rate
21%
Beta
-0.19
Alpha
+54.50%
Avg win / loss
3.72% / -0.74%

Results from a single backtest on QuantConnect. Costs and slippage modeled per the strategy's deployment specification.

About

Universe
TQQQ (leveraged QQQ ETF)
Horizon
Intraday

A one-minute intraday system on the leveraged Nasdaq ETF. The core trade is a VWAP-anchored band breakout with a tight trailing stop. Layered on top of the core, a small set of deterministic event-day rules controls when the strategy is allowed to take new positions and at what size.

Entry: a 1-minute close that pushes outside a VWAP-anchored band, in the direction of the breakout. Exit: a tight trailing stop that returns price back toward the band, with hard liquidation at session close.

The two event-day rules that matter most reduce or zero allocation on scheduled high-impact event days: FOMC announcements and CPI releases. Those dates are taken from the published Federal Reserve and Bureau of Labor Statistics schedules, not from in-sample fitting, and they target the specific sessions on which the breakout trade has historically been most punished.

The backtest starts in 2015 because TQQQ trading volume only reached the level at which one-minute intraday execution can be modeled with confidence around that time. Earlier fills on the ETF would not be representative of the spread and liquidity behavior the strategy depends on today.

The shipped configuration is the product of a multi-phase refinement: each candidate rule was tested independently, the best were stacked, and the combined version was re-validated on a full-period holdout.