Your interactive trading laboratory!
 • 
10 users online

Model - Connors TPS
Created by Q Merlin on 1/6/2019
, last modified on 1/6/2019
Please sign in to save your Models


C# Coded Model


C# Coded Model Tips

These models are programmed in the C# language and utilize the Microsoft .NET framework. A Quantacula model is a C# class derived from the UserModelBase base class.

The Initialize method is called first. Override this method to instantiate indicators or other objects you will need during your model's processing.

The Execute method is called once for each bar of data in the history. Override this method to implement your model's trading logic. You are passed the current index into the historical data in the idx parameter.

Model Name
Connors TPS

Description
You can use Markdown syntax in your description.

Connors TPS from the book 'High Probability ETF Trading' does combine time, price and scaling-in to give the highest percent correct on the ETFs of any strategy they created or ever traded. It basically identifies when an ETF is overbought or oversold and then averages into the position as it becomes more oversold. TPS was first taught to the Chairman’s Club members in 2008 and has since been expanded. As part of this expansion, full protection for the ETF positions and literally tens of thousands of variations of TPS have been published and presented to the Chairman’s Club members. To put this in perspective it would take thousands of pages to publish the complete TPS findings and the many ways to trade it.

  1. The ETF is above the 200-day.
  2. The 2-period RSI is below 25 for two days in a row.Buy 10% of your position on the close.
  3. If prices are lower on the close than your previous entry price, any day you're in the position, buy 20%
  4. If prices are lower on the close than your previous entry price, any day you're in the position, buy 30%
  5. If prices are lower on the close than your previous entry price, any day you're in the position, buy 40%
  6. Exit on the close when the 2-period RSI closes above 70.

Discussions
Attachment

Submit your Post
Cancel

Merlin, you're sandbagging us here! As presented it looks like a very modest strategy with low drawdown and low exposure. But if you double or triple the position sizing (to $10k or $15k) or even change it to 10-15% of portfolio, whoopee! Exposure is still low, but drawdown is exceptionally low and returns are double or triple what SPY yielded over that long period. Thanks for publishing!

Merlin, you're sandbagging us here! As presented it looks like a very modest strategy with low drawdown and low exposure. But if you double or triple the position sizing (to $10k or $15k) or even change it to 10-15% of portfolio, whoopee! Exposure is still low, but drawdown is exceptionally low and returns are double or triple what SPY yielded over that long period. Thanks for publishing!

Of course, it has underperformed on a CAGR basis over the past 10 years, but its risk-adjusted performance is still admirable. And finishing 2018 nearly flat or even maybe a slight gain is not a bad thing either.

Of course, it has underperformed on a CAGR basis over the past 10 years, but its risk-adjusted performance is still admirable. And finishing 2018 nearly flat or even maybe a slight gain is not a bad thing either.

It's not perfect yet. On the one hand, we have to implement the MarketClose OrderType tweak with the "bars + 1" rule, on the other hand the position "scaling-in" need's to be extended to the Connors "10% 20% 30% 40%" rule. Glitch is currently working on both features (MarketClose and Scaling-In on a percentage base). Once it's done i'll update the model.

Obviously the system is constructed for low volatility instruments like country ETF's. But what about stocks? I haven't tested that yet but later, i'll run that over an universe of low volatility stocks which are index independent. Maybe someone has an additional ideas?

It's not perfect yet. On the one hand, we have to implement the MarketClose OrderType tweak with the "bars + 1" rule, on the other hand the position "scaling-in" need's to be extended to the Connors "10% 20% 30% 40%" rule. Glitch is currently working on both features (MarketClose and Scaling-In on a percentage base). Once it's done i'll update the model. Obviously the system is constructed for low volatility instruments like country ETF's. But what about stocks? I haven't tested that yet but later, i'll run that over an universe of low volatility stocks which are index independent. Maybe someone has an additional ideas?
Submit Changes
Cancel
Delete Post
Your Models

You will need to Sign In in order to save your Models.

When saving a Model, provide a short and descriptive name that summarizes its spirit.

Also, provide a brief description that explains the logic of your Model, and any other interesting background material you'd like to include.

When you're ready, please Publish your Model so other members of the Quantacula community can review and learn from it.

Remember, in the free membership, you can save only 3 non-published Models. Publish a Model to open up one of your free Model slots. Premium members can save an unlimited number of private Models.

Data Selection
Select this option to backtest the Model on a single symbol.
Select this option to backtest the Model on a portfilio of symbols. Quantacula organizes groups of symbols into Universes.

Universe
This dynamic Universe eliminates the risk of Survivorship Bias
Upgrade to a premium membership to create your own custom Universes

Benchmark Symbol
The benchmark symbol is used to compare your backtest results against the overall market. Typically a market index symbol such as SPY or QQQ is used.


Data Scale
Backtest Settings

Starting Capital
Margin Factor
Max Open Positions
Position Sizing
% of equity
Date Range
Settings Tips

The settings here control what data is used when you backtest your Model, as well as the position sizing. Settings information is saved when you save your Model.

Backtest

Equity Curve

Performance Metrics
Summary Model Benchmark
Profit
Profit %
APR
Sharpe Ratio
Sortino Ratio
Exposure
Alpha α N/A
Beta β N/A
Profit Factor
Recovery Factor
Drawdown Model Benchmark
Maximum Drawdown
Maximum Drawdown %
Maximum Drawdown Date
Interest and Commission Model Benchmark
Commission Paid
Cash Interest Earned
Margin Interest Paid
Positions Model Benchmark
Number of Positions
Number of NSF Positions (not taken)
Win Rate
Average Profit
Average Profit %
Average Hold Time (Bars)
Winning Positions Model Benchmark
Number of Winners
Winning %
Winners Avg Profit
Winners Avg Profit %
Winners Avg Hold Time (Bars)
Losing Positions Model Benchmark
Number of Losers
Losing %
Losers Avg Loss
Losers Avg Loss %
Losers Avg Hold Time (Bars)

Monthly Performance
Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec YTD
Open Positions
Position Symbol Quantity Entry Date Entry Price Last Price Profit Profit %

Closed Positions
Position Symbol Quantity Entry Date Entry Price Exit Date Exit Price Profit Profit %
Trade Signals
In order to view next day Trade Signals, please upgrade to a Quantacula Premium Account.
Symbol: