Histogram maker

A histogram sorts your numbers into ranges of equal width and draws how many landed in each. A bin covers everything from its left edge up to but not including its right edge, so a value of exactly 70 goes into the 70–80 bin rather than the 60–70 one; the last bin takes the maximum as well, so nothing is dropped. The table under the chart lists every bin with its count, which is what a frequency table is.

Left to itself the page uses Sturges' rule, the same starting point Excel uses: the number of bins grows with the logarithm of how many values you have. It is a starting point, not an answer.

The same numbers can look like one hill at six bins and like two at twenty. That is not a trick of the tool — it is the honest consequence of a choice that a histogram always makes, and why any serious write-up says how many bins were used. Too few and real structure is smoothed away; too many and every gap between values becomes a dip.

The useful habit is to try three or four bin counts and keep the one where the shape stays the same when you nudge it. If a second hump survives at seven, ten and fifteen bins, it is in the data. If it appears only at twenty-five, it is in the bins.

A histogram is for one column of measured numbers, and its bars touch because the axis underneath them is continuous. A bar graph is for named categories — apples, pears, cherries — and its bars stand apart because there is nothing between the categories. If your labels are words rather than numbers, the bar graph maker is the page you want.