Threshold Solver

The Threshold solver compares the Input value against any number of threshold levels. This solver is primarily used to identify when an indicator is above or below a certain value (a threshold value). e.g. This solver would be used to identify the following conditions; when the CCI is above 100 or below -100. Or, when the Stochastics is above 80 or below 20, and vise-versa. Or, if the MACD is above zero or below zero. Those numeric values are threshold levels/values.
Tip: If you want to compare an indicator to another indicator, or bar prices, or some other non-fixed numeric value, then use the Comparison solver.

BloodHound v2

This section is specific to BloodHound v2. Scroll down for BloodHound v1.

Properties tab

This tab contains the essential settings for all nodes.

Modifiers

Click to see documentation

This section provides quick access to commonly used output modifiers.
Please note: The buttons only modify the individual instance of the selected node(s) on a Logic Board. Thus, the buttons are only visible when a node is selected on the Logic Board. The buttons are not available when a solver is selected in the Solvers panel, because the original solver's output can not be globally modified.
This behavior is consistent with adding a function node afterwards, so that the original solver's output remains unmodified elsewhere in the system. It is similar to having an SMA(50) on several charts. Changing the plot color on one chart does not modify the plot color on the other charts.

Enable button

Invert button

Swap button

Global Properties

These setting apply to all instances of the selected solver. Changing a setting in one of the instances changes the setting in all instances. If you only want to change one of the solvers, but not the others, you must make a copy of the solver. The copy will be a different solver, and a number will be added at the end of the name to differentiate it from the original.

Behavior

This option determines which outputs are calculated. Effectively turning off either the Long or Short output.

Evaluate

Input

This section determines what kind of data the solver will evaluate. Use the Type drop-down menu to select the data type.
The menu will change based on which data type is selected. Click on Type to see the various data types available and the corresponding documentation.

Click to see the Input » Type documentation

The Input section determines the data type to be evaluated. Use the 'Type' menu to select the data type. See below for detailed information on each data type.

Type: Fixed Value

Fixed Value

Fixed Value allows you to specify a single value that never changes, similar to the Threshold solver. This may be useful for instance when you require a fixed level or value to compare against, such as with the Comparison solver.
For example, detecting low volume bars. If 100 shares or less per bar is considered low volume, then 100 would be the fixed value to compare against Volume.
Note: Using Fixed Value does not make sense for solvers that only have one input. Such as the Inflection, Slope, Change In Slope, etc. solvers.

Type: Price

This option allows price data to be used in the solver. The price data can be shaped/manipulated by the various settings before it is used by the solver.

Custom Prices
SharkIndicators has custom system development prices that are very useful in reducing the number of solvers, and simplifying system logic, when evaluating the open and close (body) prices.
Body Top: Regardless of the bar direction, this returns the upper price of the candlestick body. Either the close or open price. Whichever is on top. e.g. For an up bar, the Body Top equals the close price.
Body Median: This returns the median price of the candlestick body. The formula is (Close + Open) ÷ 2.
Body Bottom: Regardless of the bar direction, this returns the lower price of the candlestick body. Either the close or open price. Whichever is the bottom. e.g. For an up bar, the Body Bottom equals the open price.

Price Type Long

Price Type Short

Lookback Period

Long Lookback Processing

Select how to process the price data in the Lookback Period. This process is used for calculating the Long output only.

Single Value: This returns the price from a historical bar referenced by the Lookback Period. No data processing is done. This makes the Lookback Period behave just like the Displacement parameter.

Average Values: This mode returns the average of the Price Type Long prices within the Lookback Period.

Max Values: This mode returns the highest selected price within the Lookback Period. It searches the Lookback Period historical bars and returns the highest value found. I.E. This is used when you want to use the highest High price of the previous 10 bars by setting Price Type Long = High and Lookback Period = 10.

Min Values: This mode returns the lowest selected price within the Lookback Period. It searches the Lookback Period historical bars and returns the lowest value found. I.E. This is used when you want to use the lowest High price of the previous 10 bars by setting Price Type Long = High and Lookback Period = 10.

Short Lookback Processing

Select how to process the price data in the Lookback Period. This process is used for calculating the short output only.

Average Values: This mode returns the average of the Price Type Short prices within the Lookback Period.

Max Values: This mode returns the highest selected price within the Lookback Period. It searches the Lookback Period historical bars and returns the highest value found. I.E. This is used when you want to use the highest Low price of the previous 10 bars by setting Price Type Short = Low and Lookback Period = 10.

Min Values: This mode returns the lowest selected price within the Lookback Period. It searches the Lookback Period historical bars and returns the lowest value found. I.E. This is used when you want to use the lowest Low price of the previous 10 bars by setting Price Type Short = Low and Lookback Period = 10.

Type: Volume

Volume will feed the current bar volume into the solver.
To evaluate the volume from a few bars back, use Indicator Value. Then select the VOL indicator, and use the Displacement setting.

Type: Indicator Value

Indicator

This is the default setting and most commonly used Input type. This allows the selection of any installed indicator that has one or more of the following output types; a Plot or Series<T>. A Plot is the plotted line or bar seen on the chart (e.g. the SMA line, RSI line, Volume bars, etc.), which always has a numeric value seen on the right or left edges of the chart. A Series<T> is not visible on a chart, but contains data like a plot and is readable by BloodHound. A Series<T> can contain the indicator’s main value, extra sub-data, or sometimes a signal of some kind. Note: Some NinjaTrader indicators are Not readable, per their design.
Tip: A signal value is typically +1, 0, or -1.
Technical note for programmers: BloodHound can read Series<T> data types Bool, Int, Double, and Long.

Select Indicator window

Clicking on the ellipsis button (to the right of the indicator) opens the Select Indicator window. This window is used to;
1) Select the indicator
2) Set the indicator properties
3) Select the Plot or Output you want the solver to analyze.

Search box: Typing the first three letters of an indicator will create a shorter filtered list of indicators to quickly find the indicator you are looking for.

Refresh Indicator List button: This button is only used once after a new indicator has been installed. It initiates rebuilding an updated list of indicators. If the newly installed indicator does not listed, please see the next section below. Also, a menu item to perform this same task has been added to the NinjaTrader Control Center » Help menu » SharkIndicators menu.

Add New button: Instead of double clicking on an indicator to add it, this button can be used.

Nest Input button: This provides a way to feed the output of one indicator in to the input of another indicator (e.g. Smoothing the CCI by nesting/feeding it in to the SMA indicator). Watch this video for more info on the process... How to Perform Indicator Nesting Within a Solver

Remove button: Use this to remove nested indicator from the list (in the bottom area).

Indicators Missing From the List
If an indicator is missing from the Select Indicator window, that means the indicator does not have a readable output or may have thrown an error when building the list. Or, the indicator may only draw information on the chart using custom graphics routines which are not readable.
In order for BloodHound to read 3rd party indicators the developer must follow NinjaTrader’s Best Practices for 3rd Party Developers > Expose Indicator States. This video will help you determine if your favorite 3rd party indicator has an output that is readable… How to analyze custom indicators in BloodHound.

Displacement

The Displacement parameter works the same way as the NinjaTrader Displacement parameter. It is used to analyze indicator values from previous bars back. On the chart, the Displacement moves the indicator plot forward X number of bars on to the current bar. Essentially, the same is occurring internally. the indicator value X number of bars back is moved forward on to the current bar. The Displacement operates on the chart/time-frame the solver is set to.
e.g. Setting Displacement to 5 will have the solver use the indicator value from the 5th bar back. In effect, this shifts the indicator plot from 5 bars back to the current bar.
i.e. Suppose you want to analyze an indicator’s value 5 bars ago. Set the Displacement value to 5 and the solver will always use indicator data that is 5 bars in the past.

Type: Swing Point Prices

Swing Point Prices uses the Swing Highs & Lows indicator plots for input price data. This is useful when you need to access the lowest low or highest high prices from the latest swing point.
You can add the bundled indicator Swing Highs & Lows to the chart to help you visualize the price data the solver is using.

Number of Swings

Swing type

Swing strength

Double Top/Bottom strength

Use Close prices

Long: Bottoms or Tops

Long: Widest or Tightest

Short: Bottoms or Tops

Short: Widest or Tightest

Type: Linear Regression Channel

Output Rules

Threshold Settings

Click to see the Rules window Introduction

The Rules window provides a more natural way to define charting conditions (a.k.a. trading system rules), and allows advanced combinations of multiple rules/conditions in a single solver, replacing BloodHound v1 need for multiple solvers to do the same complex task.

Rules Window Components

The individual rules can be expanded open or collapse close by clicking a rule's title bar. The title bar is the horizontal area with the word 'Fixed' or 'Range' within it.

Add Rule menu(top left): This menu adds the selected rule.
A Fixed rule is most commonly used. It provides a simple comparison of the Input to a numeric value or Input B.
A Range rule compares the Input to a lower and upper range limit of values. Or, a ranged distance between Input A and Input B.
Tip: To use a Range rule instead of a Fixed rule, delete the default Fixed rule, and then add a Range rule.

Output Calculation menu(top right): This defines how the output is determined when two or more rules exist and at least two of the rules are true.
First: The first rule that is true, starting at the top of the list and working down, that output value is used.
Largest: The largest output value from all the rules that are true becomes the output value. This is for fuzzy logic systems.
Smallest: The smallest output value from all the rules that are true becomes the output value. This is for fuzzy logic systems.
Average: The output values from all the rules that are true are averaged together and becomes the output value. This is for fuzzy logic systems.
Addition: The output values from all the rules that are true are added together and becomes the output value. This is for fuzzy logic systems.

Settings window: This applies to the Range rule. It provides more detailed control over the values being evaluated.
Outside - Inside: The evaluation can be either Inside the lower and upper range or Outside that range.
Exclude - Include: Include is equivalent to >= or <=. It include the values in the evaluation.
Exclude is equivalent to > or <, no equal. The lower and upper range values are not included in the evaluation.
Fixed Output - Ranged Output: The 'Fixed Output' outputs a simple digital type of output. Either the output is 0 (no output) or the value on the far right (usually 1).
Selecting the 'Ranged Output' enables the fuzzy logic capability of the solver. Two Output fields become visible. When the input value falls between the lower and upper range values, the solver's output is calculated by linearly interpolating between the two Output values.

Delete button: This deletes the rule. The Undo button can be used to restore it.

Threshold Rules widow

Fixed Rules:
A Fixed rule is used to identify when the Input is greater than a fixed value/level, lesser than a fixed value/level, or equal to a fixed value/level.
e.g. Typical examples would be, is the CCI greater than 150 or is it below -150. Or, is the MACD above the zero line or below the zero line.

This is the name of the Input. e.g. The indicator name or the bar price.

This menu selects the above(>), below(<), or equal(=) conditions to be identified.
= means 'equal to'. Input = Threshold value.
!= means 'Not equal to'. Input ≠ Threshold value.
> means greater than/above. Input > Threshold value.
< means less than/below. Input < Threshold value.
>= means equal or greater than/above. Input ≥ Threshold value.
<= means equal or less than/below. Input ≤ Threshold value.
e.g. CCI [ > ] [ 150 ] identifies when the CCI is greater then 150. See the Examples section listed below.

This is a numeric input field for the threshold value/level. When the Input is greater than, less than, or equal to this value, the condition is identified/true.

This sets the output value when the condition is identified/true. A value of 1 (meaning 100% confidence) is typical for most systems. Only advanced users, building a fuzzy logic system, should use a smaller value.

Range Rules:
A Range rule is used to identify when the Input is between two threshold value/level.

This is the name of the Input. e.g. The indicator name or the bar price.

This is a numeric input field for the first threshold value/level. For the Long calculation, this is typically the lower/smaller value of the range, and value ③ would be the upper/larger end of the range. For the Short calculation, this is typically the upper/larger value of the range, and value ③ would be the lower/smaller end of the range as seen in the image above. When the Input is between this value and value ③, the condition is identified/true.

This is a numeric input field for the second threshold value/level, and is the other end of the range, from value ②, to be identified.

Two Output fields, ④ and ⑤, appear when the Ranged Output option is enabled in the Settings menu. Only advanced users, building a fuzzy logic system, should enable this option.

This sets the output confidence value when the Input is at value ② of the range. If the value in ⑤ is 1 (meaning 100% confidence), then this value should be less. And, vice-versa if this value is 1.

This sets the output confidence value when the Input is at value ③ of the range. If the value in ④ is 1 (meaning 100% confidence), then this value should be less. And, vice-versa if this value is 1.

Options tab

This tab contains extra secondary settings common to all nodes.

Weight

Click to see the Weight documentation

This section is used to modify the node's final output value. It is only useful for designing fuzzy logic systems, or a fuzzy logic section of a system.
Please note: The Weight controls only modify the individual instance of the selected node(s) on a Logic Board. Thus, the Weight controls are only visible when a node is selected on the Logic Board. The controls are not available when a solver is selected in the Solvers panel, because the original solver's output can not be globally modified.
This behavior is different than BloodHound 1.0. In BH 1.0, the Weight controls modified the nodes output globally (every instance). This change in BH 2.0 adds more system design granularity.

Multiply Output: This takes the internal values (the Long and Short values determined via the Properties tab » Output Rules section) and multiplies them by this value. Note: the final output will not exceed a value of 1, as described in the Slider control below.

Slider control: The slider constrains the final output to a value of 0 to 1. The left side of the slider sets a minimum value that is output regardless if the solver condition is true or not. The right side sets a maximum value that is output. The output is capped.
e.g. Three indicator conditions are being checked, and thus three solvers are created. Only two out of the three indicator conditions are needed. An Additive logic node is used to add the solver's outputs together. Just two out of the three solvers need to add up to a value of 1. Therefore, the right slider (max output value) for all three solvers is set to 0.5. When two indicator/solver conditions are found, thus the outputs = 0.5, then the calculation, in the Additive node, is 0.5 + 0.5 + 0 = 1. A value of 1 means the two out of three condition is true.

Hedge Operation: This applies a mathematical formula to the internal value.

None: No modification is applied.
Very (square output): A squaring formula is applied. Output = value^² .
Somewhat (square root output): A square root formula is applied. Output = √value .

Description

This text area provides a place to write a full description of what the node is doing, used for, or what ever you want.
Note: The Description is global to all instances of the node. It is not applied individually to each instance as the Weight controls are.

Help tab

This tab displays the documentation page (from this web site) of the selected node.
Please note: NinjaTrader v8.0.26.0 or newer is required for the built in web viewer to work, and thus the documentation to be displayed.

Examples

Example 1A: The Basics

This demonstrates a basic function of the Threshold solver using the Stochastics » D plot. This Solver detects overbought, above 80, and oversold, below 20, conditions of the Stochastics » D.

Input Settings:

Set Indicator to Stochastics » D.

Rules window Settings:
Open the Rules window by clicking the ellipses button in section Output » Threshold Settings.

Set Thresholds values as shown.

The chart below shows a Long output when the Stochastics D is above 80, and a Short output when it is below 20.

Example 1B: The Basics with Volatility

This demonstrates a basic function of the Threshold solver to detect volatility. This Solver detects when the ATR is above 7. Some systems may need or take advantage of increased volatility to work well. The chart below shows when an increase in active trading picks up.

Input Settings:

Set Indicator to ATR.

Rules window Settings:
Open the Rules window by clicking the ellipses button in section Output » Threshold Settings.

Set Thresholds values as shown.
Note the Long & Short settings are the same.

The chart below shows a Long & Short output when the ATR goes above 7. The ATR is a non-trend directional indicator, which is why there is a Long & Short output at the same time.

Example 1C: A Basic MACD Filter

This demonstrates another basic function of the Threshold solver using the MACD line. This Solver will identify if the MACD is above the zero line for a long filter, or below the zero line for a short filter. The zero line threshold of the MACD switches the output between Long and Short.

Set Indicator to the MACD.

Rules window Settings:
No changes to the Rules window is necessary.

The chart below shows a Long output when the MACD is above zero, and a Short output when it is below zero. Quite a simple filter.

Example 2A: Using A Discretionary S/R Level

This demonstrates a simple way to use a discretionary support/resistance level to filter long only trading above a price level, or shorts only below that price level. The Solver identifies when the closing price is above or below the discretionary level.

Input Settings:

Set Input » Type to Price.

Rules window Settings:
Open the Rules window by clicking the ellipses button in section Output » Threshold Settings.

Set Thresholds values as shown.

The chart below shows a Long output when the close price is above 11,200.0, or a Short output when the close is below 11,200.

Example 2B: Two Discretionary S/R Levels

This demonstrates using two discretionary support/resistance levels. When the market (closing bar price) is above the lower S/R level a long only filter is applied. When the market is below the upper S/R level a short only filter is applied. When the market is between the two levels both long and short signals can occur.

Input Settings:

Set Indicator » Type to Price.

Rules window Settings:
Open the Rules window by clicking the ellipses button in section Output » Threshold Settings.

Set Thresholds values as shown.

The chart below shows a Long output when the close price is above 11,160.0, a Short output when the close is below 11,200, and of course a Long & Short output when the close is between the two levels.

BloodHound v1

This section is specific to BloodHound v1. Scroll up for BloodHound v2.

Indicator Threshold Solver

The Threshold solver compares the Input value against 5 possible threshold levels. This solver is primarily used to identify when an indicator is above or below a certain value (a threshold value). e.g. This solver would be used to identify the following conditions; when the CCI is above 100 or below -00. Or, when the Stochastics is above 80 or below 20. Or, if the MACD is above zero or below zero. Those numeric values are threshold levels/values.
When the input value falls between threshold levels the output value is calculated by linearly interpolating between the corresponding Output values.
Tip: If you want to compare an indicator to another indicator, or bar prices, or some other non-fixed numeric value, then use the Comparison solver.

Parameters

Display

Behavior

Input

Type

Thresholds

Please note, Threshold values must be entered in descending order.
I.E #1 A = 5, B = 4, C = 3, D = 2, E = 1
#2 A = 2, B = 1, C = 0, D = -1, E = -2

Long Output

Short Output

Sets the solver’s Short output values as described in Long Output above.

Video Tutorial

This video is from our weekly Workshop on July 10th, 2015.
For more benefit please watch in full screen mode, as this video is recorded in HD.

Examples

Example 1A: The Basics

This demonstrates the basic function of Indicator Threshold solver using the Stochastic D. This Solver will be used to detect over-bought, above 80, and over-sold, below 20, conditions of the Stochastic D. In this first example the 50 line of the Stochastic will be used to switch the output between Long and Short.

Add the Indicator Threshold solver
Set Indicator to Stochastic D
Set Thresholds values as shown
Set Long Output > Greater Than 80 to 1
Set Short Output > Less Than 20 to 1

The chart below shows a Long output when the Stochastic D is above the 50 line, and a Short output when it is below the 50 line.

Example 1B: The Basics of Going Digital

This demonstrates the basic function of Indicator Threshold solver using the Stochastic D. This time the Solver will be set to give a digital style output.

Add the Indicator Threshold solver
Set Indicator to Stochastic D
Set Thresholds values as shown
Set Long Output > Greater Than 80 to 1
Set Short Output > Less Than 20 to 1

Notice that At B & At C are set to 0, and thus no output values between 20 to 80. The chart below only shows values above 80 and below 20.

Example 1C: The Basics of a Smooth Transition

This demonstrates the basic function of Indicator Threshold solver using the Stochastic D. In this example the Solver’s output will transition between outputs always totaling 1. As the Long output increases the Short output decreases, and vise versa.

Add the Indicator Threshold solver
Set Indicator to Stochastic D
Set Thresholds values as shown
Set Long Output > Greater Than 80 & At 80 to 1
Set Short Output > At 20 & Less Than 20 to 1

The chart below shows a Long output when the Stochastic D is above 20, and a Short output when it is below 80.

Example 2: A Basic MACD Filter

This demonstrates the basic function of Indicator Threshold solver using the MACD line. This Solver will be used to detect if the MACD is above the zero line for a Long(buy) filter, or below the zero line for a Short(sell) filter. The zero line threshold of the MACD will be used to switch the output between Long and Short.

Add the Indicator Threshold solver
Set Indicator to the MACD
Leave the Thresholds values at zero
Set Long Output > Greater Than A to 1
Set Short Output > Less Than E to 1

The chart below shows a Long output when the MACD is above zero, and a Short output when it is below zero.

General

BloodHound

BlackBird

Other Indicators

Threshold Solver

BloodHound v2

Properties tab

Properties tab

Output Modifiers

Modifiers

Global Properties

Global Properties

Behavior

Input

Type: Fixed Value

Type: Price

Type: Volume

Type: Indicator Value

Type: Swing Point Prices

Type: Linear Regression Channel

Output Rules

Options tab

Options tab

Weight

Weight

Description

Description

Help tab

Help tab

Examples

Example 1A: The Basics

Example 1B: The Basics with Volatility

Example 1C: A Basic MACD Filter

Example 2A: Using A Discretionary S/R Level

Example 2B: Two Discretionary S/R Levels

General

BloodHound

BlackBird

Other Indicators

BloodHound v1

Indicator Threshold Solver

Parameters

Input

Thresholds

Long Output

Short Output

Video Tutorial

Examples

Example 1A: The Basics

Example 1B: The Basics of Going Digital

Example 1C: The Basics of a Smooth Transition

Example 2: A Basic MACD Filter

General

BloodHound

BlackBird

Other Indicators

Indicator Threshold Solver

Parameters

Input

Thresholds

Long Output

Short Output

Video Tutorial

Examples

Example 1A: The Basics

Example 1B: The Basics of Going Digital

Example 1C: The Basics of a Smooth Transition

Example 2: A Basic MACD Filter