How to Plot Time Series Data in Python: matplotlib, pandas, and plotly

Time series data—sensor readings, climate measurements, stock prices, patient vitals—needs special handling in matplotlib because the x-axis carries dates rather than plain numbers. Matplotlib understands Python datetime objects and pandas DatetimeIndex natively, but formatting the axis cleanly requires a few extra lines.

1. Basic time series plot

Plot the series with ax.plot(dates, values). Apply mdates.MonthLocator to place one tick per month and DateFormatter to control the label text.

import pandas as pd
import matplotlib.pyplot as plt
import matplotlib.dates as mdates

dates = pd.date_range("2024-01-01", periods=90, freq="D")
values = pd.Series(range(90), index=dates) + pd.Series(
    [i * 0.3 for i in range(90)], index=dates
)

fig, ax = plt.subplots(figsize=(9, 4))
ax.plot(dates, values, lw=1.5, color="#6b21a8")
ax.xaxis.set_major_locator(mdates.MonthLocator())
ax.xaxis.set_major_formatter(mdates.DateFormatter("%b %Y"))
fig.autofmt_xdate()
ax.set_ylabel("Measurement")
ax.set_title("Daily Time Series")
plt.tight_layout()

2. Rolling averages

Use Series.rolling(window).mean() from pandas. Plot the raw signal at low alpha and the rolling means at full opacity so the trend is readable.

import pandas as pd
import matplotlib.pyplot as plt

dates = pd.date_range("2024-01-01", periods=180, freq="D")
rng = pd.Series(range(180), index=dates).add(
    pd.Series([i ** 0.5 for i in range(180)], index=dates) * 0.5
)

rolling_7  = rng.rolling(7).mean()
rolling_30 = rng.rolling(30).mean()

fig, ax = plt.subplots(figsize=(9, 4))
ax.plot(dates, rng, alpha=0.25, color="#6b21a8", lw=1, label="Raw")
ax.plot(dates, rolling_7,  color="#f59e0b", lw=1.5, label="7-day mean")
ax.plot(dates, rolling_30, color="#10b981", lw=2,   label="30-day mean")
ax.legend(frameon=False)
ax.set_ylabel("Measurement")
fig.autofmt_xdate()
plt.tight_layout()

3. Dual y-axis (two variables, shared x)

Use a second axis sparingly — only when two datasets share a meaningful time relationship but have incompatible units. Colour-code both the lines and the axis labels to make the mapping unambiguous.

import pandas as pd
import matplotlib.pyplot as plt

dates  = pd.date_range("2024-01-01", periods=60, freq="D")
temp   = pd.Series([20 + i * 0.3 for i in range(60)], index=dates)
precip = pd.Series([5  - i * 0.05 for i in range(60)], index=dates)

fig, ax1 = plt.subplots(figsize=(9, 4))
ax2 = ax1.twinx()

ax1.plot(dates, temp,   color="#6b21a8", lw=2, label="Temperature (°C)")
ax2.bar(dates,  precip, color="#f59e0b", alpha=0.4, width=0.8, label="Precipitation (mm)")

ax1.set_ylabel("Temperature (°C)", color="#6b21a8")
ax2.set_ylabel("Precipitation (mm)", color="#f59e0b")
fig.autofmt_xdate()
plt.tight_layout()

Date tick locator reference

Common mistakes

• Overlapping tick labels — call fig.autofmt_xdate() or rotate manually with plt.xticks(rotation=45).
• Plotting integers instead of dates — ensure your x-axis is a DatetimeIndex or datetimearray, not a string column.
• Hiding raw data with only a smoothed line— show the underlying signal at low opacity alongside the rolling average.
• Dual axes without colour coding — readers cannot tell which line belongs to which axis without clear colour cues.