"One of the most common misunderstandings amongst climate deniers is the difference between short-term noise and long-term signal. This video shows how the same temperature data (green) that is used to determine the long-term global surface air warming trend of 0.18°C per decade (red) can be used inappropriately to "cherrypick" short time periods that show a cooling trend simply because the endpoints are carefully chosen and the trend is dominated by short-term noise in the data (blue steps). Isn't it strange how seven periods of cooling can add up to a clear warming trend over the last 4 decades? Several factors can have a large impact on short-term temperatures, such as oceanic cycles like the El Niño Southern Oscillation (ENSO) or the 11-year solar cycle. These short-term cycles don't have long-term effects on the Earth's temperature, unlike the continuing upward trend caused by global warming from human greenhouse gas emissions.
"The data (green) are NASA GISS monthly global surface temperature anomaly data from January 1970 through November 2016, with linear trends for the cherry picked time periods of Jan 1970–May 1977, May 1977–October 1979, October 1979–April 1988, April 1988–March 1997, March 1997–February 2002, February 2002–October 2009, and October 2009–April 2014 (blue), followed by the linear trend for the full time period (red)."
The same sort of thing could be plotted on a graph from each of the Hadley Centre's HadCRUT4 temperature data, Cowtan and Way's HadCRUT4-kriging, the Berkeley Earth data, and the NOAA NCDC data.
And a bonus video: confirmation of previous global warming models' predictions using the NASA GISS data for annual globally averaged mean temperatures. Teaser photo shows the famous Pause!