Plot Type: Scatter Plot - ScottPlot 4.1 Cookbook

This page contains recipes for the Scatter Plot category.
Visit the Cookbook Home Page to view all cookbook recipes.
Generated by ScottPlot 4.1.67 on 8/13/2023

Scatter Plot Quickstart

Scatter plots are best for small numbers of paired X/Y data points. For evenly-spaced data points Signal is much faster.

var plt = new ScottPlot.Plot(600, 400);

// create sample X/Y data
int pointCount = 51;
double[] x = DataGen.Consecutive(pointCount);
double[] sin = DataGen.Sin(pointCount);
double[] cos = DataGen.Cos(pointCount);

// add scatter plots
plt.AddScatter(x, sin);
plt.AddScatter(x, cos);

plt.SaveFig("scatter_quickstart.png");

Custom markers

Markers can be customized using optional arguments and public fields.

var plt = new ScottPlot.Plot(600, 400);

int pointCount = 51;
double[] x = DataGen.Consecutive(pointCount);
double[] sin = DataGen.Sin(pointCount);
double[] cos = DataGen.Cos(pointCount);

// add scatter plots and customize markers
var sp1 = plt.AddScatter(x, sin, markerSize: 15);
sp1.MarkerShape = MarkerShape.openCircle;

var sp2 = plt.AddScatter(x, cos, markerSize: 7);
sp2.MarkerShape = MarkerShape.filledSquare;

plt.SaveFig("scatter_markers.png");

All marker shapes

Legend indicates names of all available marker shapes

var plt = new ScottPlot.Plot(600, 400);

int pointCount = 51;
double[] xs = DataGen.Consecutive(pointCount);

string[] markerShapeNames = Enum.GetNames(typeof(MarkerShape));
for (int i = 0; i < markerShapeNames.Length; i++)
{
    Enum.TryParse(markerShapeNames[i], out MarkerShape ms);
    double[] ys = DataGen.Sin(pointCount, 2, -i);
    var sp = plt.AddScatter(xs, ys);
    sp.LineWidth = 2;
    sp.LineColor = Color.FromArgb(50, sp.LineColor);
    sp.MarkerSize = 7;
    sp.MarkerShape = ms;
    sp.Label = ms.ToString();
}

plt.Grid(enable: false);
var legend = plt.Legend();
legend.FontSize = 10;

plt.SaveFig("scatter_markerShape.png");

Custom lines

Line color, size, and style can be customized. Setting markerSize to 0 prevents markers from being rendered.

var plt = new ScottPlot.Plot(600, 400);

int pointCount = 51;
double[] x = DataGen.Consecutive(pointCount);
double[] sin = DataGen.Sin(pointCount);
double[] cos = DataGen.Cos(pointCount);
double[] cos2 = DataGen.Cos(pointCount, mult: -1);

plt.AddScatter(x, sin, color: Color.Magenta, lineWidth: 0, markerSize: 10);
plt.AddScatter(x, cos, color: Color.Green, lineWidth: 5, markerSize: 0);
plt.AddScatter(x, cos2, color: Color.Blue, lineWidth: 3, markerSize: 0, lineStyle: LineStyle.DashDot);

var legend = plt.Legend();
legend.FixedLineWidth = false;

plt.SaveFig("scatter_lineStyle.png");

Random X/Y Points

X data for scatter plots does not have to be evenly spaced, making scatter plots are ideal for displaying random data like this.

var plt = new ScottPlot.Plot(600, 400);

Random rand = new Random(0);
int pointCount = 51;
double[] xs1 = DataGen.RandomNormal(rand, pointCount, 1);
double[] xs2 = DataGen.RandomNormal(rand, pointCount, 3);
double[] ys1 = DataGen.RandomNormal(rand, pointCount, 5);
double[] ys2 = DataGen.RandomNormal(rand, pointCount, 7);

plt.AddScatter(xs1, ys1, markerSize: 0, label: "lines only");
plt.AddScatter(xs2, ys2, lineWidth: 0, label: "markers only");
plt.Legend();

plt.SaveFig("scatter_random.png");

Scatter Plot with Errorbars

An array of values can be supplied for error bars and redering options can be customized as desired

var plt = new ScottPlot.Plot(600, 400);

int pointCount = 20;
Random rand = new Random(0);
double[] xs = DataGen.Consecutive(pointCount);
double[] ys = DataGen.RandomWalk(rand, pointCount);
double[] xErr = DataGen.RandomNormal(rand, pointCount, .2);
double[] yErr = DataGen.RandomNormal(rand, pointCount);

var sp = plt.AddScatter(xs, ys);
sp.XError = xErr;
sp.YError = yErr;
sp.ErrorCapSize = 3;
sp.ErrorLineWidth = 1;
sp.LineStyle = LineStyle.Dot;

plt.SaveFig("scatter_errorbar.png");

Scatter Plot with Shaded Error

A semitransparent polygon can be created and placed behind the scatter plot to represent standard deviation or standard error.

var plt = new ScottPlot.Plot(600, 400);

int pointCount = 20;
Random rand = new Random(0);
double[] xs = DataGen.Consecutive(pointCount);
double[] ys = DataGen.RandomWalk(rand, pointCount, 2.0);
double[] yErr = DataGen.Random(rand, pointCount, 1.0, 1.0);

plt.AddScatter(xs, ys, Color.Blue);
plt.AddFillError(xs, ys, yErr, Color.FromArgb(50, Color.Blue));

plt.SaveFig("scatter_shaded_error.png");

Lines Only

A shortcut method makes it easy to create a scatter plot with just lines (no markers)

var plt = new ScottPlot.Plot(600, 400);

double[] xs = DataGen.Consecutive(51);
double[] ys = DataGen.Sin(51);

plt.AddScatterLines(xs, ys, Color.Red, 3);

plt.SaveFig("scatter_lineplot.png");

Markers Only

A shortcut method makes it easy to create a scatter plot where markers are displayed at every point (without any connecting lines)

var plt = new ScottPlot.Plot(600, 400);

double[] xs = DataGen.Consecutive(51);
double[] ys = DataGen.Sin(51);

plt.AddScatterPoints(xs, ys, Color.Navy, 10, MarkerShape.filledDiamond);

plt.SaveFig("scatter_pointsplot.png");

Step Plot

A step plot is a special type of scatter plot where points are connected by right angles instead of straight lines.

var plt = new ScottPlot.Plot(600, 400);

double[] xs = DataGen.Consecutive(51);
double[] ys = DataGen.Sin(51);

plt.AddScatterStep(xs, ys);

plt.SaveFig("scatter_stepplot.png");

Add markers

Want to place a marker at a position in X/Y space? AddMarker() will create a scatter plot with a single point.

var plt = new ScottPlot.Plot(600, 400);

Random rand = new(0);
for (int i = 0; i < 100; i++)
{
    double x = rand.Next(100);
    double y = rand.Next(100);
    double fraction = rand.NextDouble();
    double size = (fraction + .1) * 30;
    var color = Drawing.Colormap.Turbo.GetColor(fraction, alpha: .8);
    var shape = Marker.Random(rand);
    plt.AddMarker(x, y, shape, size, color);
}

plt.SaveFig("scatter_AddMarker.png");

Want to modify the scatter points interactively? A ScatterPlotDraggable lets you move the points around with the mouse. As you move the points around, the values in the original arrays change to reflect their new positions.

var plt = new ScottPlot.Plot(600, 400);

double[] x = ScottPlot.DataGen.Consecutive(50);
double[] y = ScottPlot.DataGen.Cos(50);

var scatter = new ScottPlot.Plottable.ScatterPlotDraggable(x, y)
{
    DragCursor = Cursor.Crosshair,
    DragEnabled = true,
};

plt.Add(scatter);

plt.SaveFig("scatter_draggable_vertical.png");

Draggable Scatter Plot Vertical

You can restrict dragging to just X or Y directions.

var plt = new ScottPlot.Plot(600, 400);

double[] x = ScottPlot.DataGen.Consecutive(50);
double[] y = ScottPlot.DataGen.Cos(50);

var scatter = new ScottPlot.Plottable.ScatterPlotDraggable(x, y)
{
    DragCursor = Cursor.Crosshair,
    DragEnabled = true,   // controls whether anything can be dragged
    DragEnabledX = false, // controls whether points can be dragged horizontally 
    DragEnabledY = true,  // controls whether points can be dragged vertically
};

plt.Add(scatter);

plt.SaveFig("scatter_draggable.png");

Forest Plot

Scatter plots can be used to create forest plots, which are useful for showing the agreement between multiple estimates.

var plt = new ScottPlot.Plot(600, 400);

var plot1 = plt.AddScatter(new double[] { 2.5 }, new double[] { 5 }, label: "John Doe et al.");
plot1.XError = new double[] { 0.2 };

var plot2 = plt.AddScatter(new double[] { 2.7 }, new double[] { 4 }, label: "Jane Doe et al.");
plot2.XError = new double[] { 0.3 };

var plot3 = plt.AddScatter(new double[] { 2.3 }, new double[] { 3 }, label: "Jim Doe et al.");
plot3.XError = new double[] { 0.6 };

var plot4 = plt.AddScatter(new double[] { 2.8 }, new double[] { 2 }, label: "Joel Doe et al.");
plot4.XError = new double[] { 0.3 };

var plot5 = plt.AddScatter(new double[] { 2.5 }, new double[] { 1 }, label: "Jacqueline Doe et al.");
plot5.XError = new double[] { 0.2 };

plt.AddVerticalLine(2.6, style: LineStyle.Dash);

plt.SetAxisLimits(0, 5, 0, 6);
plt.Legend();

plt.SaveFig("scatter_forest.png");

Scatter Plot with Smooth Lines

Lines drawn between scatter plot points are typically connected with straight lines, but the Smooth property can be enabled to connect points with curves instead.

var plt = new ScottPlot.Plot(600, 400);

Random rand = new(1234);
double[] xs = DataGen.RandomWalk(rand, 20);
double[] ys = DataGen.RandomWalk(rand, 20);
plt.Palette = new ScottPlot.Palettes.ColorblindFriendly();

var sp1 = plt.AddScatter(xs, ys, label: "default");
sp1.Smooth = true;

var sp2 = plt.AddScatter(xs, ys, label: "high tension");
sp2.Smooth = true;
sp2.SmoothTension = 1.0f;

var sp3 = plt.AddScatter(xs, ys, label: "low tension");
sp3.Smooth = true;
sp3.SmoothTension = 0.2f;

plt.Legend();

plt.SaveFig("scatter_smooth.png");

NaN Values Ignored

When the OnNaN field is set to Ignore, points containing NaN X or Y values are skipped, and the scatter plot is drawn as one continuous line.

var plt = new ScottPlot.Plot(600, 400);

// create data that does NOT contain NaN
double[] xs = ScottPlot.DataGen.Consecutive(51);
double[] ys = ScottPlot.DataGen.Sin(51);

// plot it the traditional way
plt.AddScatter(xs, ys, Color.FromArgb(50, Color.Black));

// create new data that contains NaN
double[] ysWithNan = ScottPlot.DataGen.Sin(51);
static void FillWithNan(double[] values, int start, int count)
{
    for (int i = 0; i < count; i++)
        values[start + i] = double.NaN;
}
FillWithNan(ysWithNan, 5, 15);
FillWithNan(ysWithNan, 25, 1);
FillWithNan(ysWithNan, 30, 15);
ysWithNan[10] = ys[10];

// add a scatter plot and customize NaN behavior
var sp2 = plt.AddScatter(xs, ysWithNan, Color.Black);
sp2.OnNaN = ScottPlot.Plottable.ScatterPlot.NanBehavior.Ignore;
plt.Title($"OnNaN = {sp2.OnNaN}");

plt.SaveFig("scatter_nan_ignore.png");

NaN Values Break the Line

When the OnNaN field is set to Gap, points containing NaN X or Y values break the line. This results in a scatter plot appearing as multiple lines, with gaps representing missing data.

var plt = new ScottPlot.Plot(600, 400);

// create data that does NOT contain NaN
double[] xs = ScottPlot.DataGen.Consecutive(51);
double[] ys = ScottPlot.DataGen.Sin(51);

// plot it the traditional way
plt.AddScatter(xs, ys, Color.FromArgb(50, Color.Black));

// create new data that contains NaN
double[] ysWithNan = ScottPlot.DataGen.Sin(51);
static void FillWithNan(double[] values, int start, int count)
{
    for (int i = 0; i < count; i++)
        values[start + i] = double.NaN;
}
FillWithNan(ysWithNan, 5, 15);
FillWithNan(ysWithNan, 25, 1);
FillWithNan(ysWithNan, 30, 15);
ysWithNan[10] = ys[10];

// add a scatter plot and customize NaN behavior
var sp2 = plt.AddScatter(xs, ysWithNan, Color.Black);
sp2.OnNaN = ScottPlot.Plottable.ScatterPlot.NanBehavior.Gap;
plt.Title($"OnNaN = {sp2.OnNaN}");

plt.SaveFig("scatter_nan_gap.png");

Scatter Plot Labels

Individual points can be labeled.

var plt = new ScottPlot.Plot(600, 400);

double[] xs = { 18.5, 20.6, 22.3, 24.5, 26.6, 15, 15 };
double[] ys = { 1.43, 1.48, 1.6, 1.59, 1.53, 1.52, 1.6 };
string[] labels = { "A", "B", "C", "D", "E", "F", "G" };

var sp = plt.AddScatter(xs, ys);
sp.DataPointLabels = labels;
sp.DataPointLabelFont.Size = 24;

plt.SetAxisLimits(12, 28, 1.3, 1.7);

plt.SaveFig("scatter_labels.png");

Scatter List Quickstart

This plot type has add/remove/clear methods like typical lists.

var plt = new ScottPlot.Plot(600, 400);

double[] xs = { 1, 2, 3, 4 };
double[] ys = { 1, 4, 9, 16 };

var scatterList = plt.AddScatterList();
scatterList.AddRange(xs, ys);
scatterList.Add(5, 25);

plt.SaveFig("scatterList_quickstart.png");

Scatter List Generic

This plot type supports generics.

var plt = new ScottPlot.Plot(600, 400);

int[] xs = { 1, 2, 3, 4 };
int[] ys = { 1, 4, 9, 16 };

var scatterList = plt.AddScatterList<int>();
scatterList.AddRange(xs, ys);
scatterList.Add(5, 25);

plt.SaveFig("scatterList_generic.png");

Scatter List Draggable

There exists a Scatter Plot List with draggable points.

var plt = new ScottPlot.Plot(600, 400);

double[] xs = ScottPlot.DataGen.Consecutive(51);
double[] ys = ScottPlot.DataGen.Sin(51);

var scatter = new ScottPlot.Plottable.ScatterPlotListDraggable();
scatter.AddRange(xs, ys);

plt.Add(scatter);

plt.SaveFig("scatterList_draggable.png");

Scatter List Draggable Limits

A custom function can be used to limit the range of draggable points.

var plt = new ScottPlot.Plot(600, 400);

// plot sample data
double[] xs = ScottPlot.DataGen.Consecutive(20);
double[] ys = ScottPlot.DataGen.Sin(20);
var scatter = new ScottPlot.Plottable.ScatterPlotListDraggable();
scatter.AddRange(xs, ys);
scatter.MarkerSize = 5;
plt.Add(scatter);

// use a custom function to limit the movement of points
static Coordinate MoveBetweenAdjacent(List<double> xs, List<double> ys, int index, Coordinate requested)
{
    int leftIndex = Math.Max(index - 1, 0);
    int rightIndex = Math.Min(index + 1, xs.Count - 1);

    double newX = requested.X;
    newX = Math.Max(newX, xs[leftIndex]);
    newX = Math.Min(newX, xs[rightIndex]);

    return new Coordinate(newX, requested.Y);
}

scatter.MovePointFunc = MoveBetweenAdjacent;

plt.SaveFig("scatterList_draggableLimits.png");