Category Archives: accuracy

accuracy, schmaccuracy.

Standard

it’s ground hog day once again, and poor phil. the famous groundhog

will get pulled out of a deep winter’s sleep.

many will gather to see what phil will predict.

will spring be just around the corner, or delayed for another few weeks?

how accurate are his predications, really?

At Gobbler’s Knob in Punxsutawney, Pennsylvania, weather-predicting groundhog Phil may see his shadow, which means he’ll call for six more weeks of winter-like conditions. Don’t get too bummed, though—Phil’s recent track record isn’t that great on the national level.

The tradition started in 1887, and since then, the groundhog chosen to represent Phil has seen his shadow 101 times. 

Phil has predicted 100% correctly in only five of the 26 years the National Climatic Data Center analyzed. Other than once, he only nailed the years in which he forecasted an early spring. Considering the contiguous United States just experienced its 2oth+ consecutive year with an above-average annual temperature, Phil may be wise to play the numbers and always predict an early spring.

Then again, perhaps we expect too much from a marmot with no access to nationwide climate data.

“well, the way we perceive accuracy and what accuracy is statistically

are really two different things.”

-nate silver

 

 

source credit: nick greene, mental floss