Combined Rate Trap Calculator

Group A went from 90% to 95%. Group B went from 20% to 30%. Both improved.

The two of them together went from 55% to 35.9%.

Every one of those statements is exactly true of the same numbers. It is not a trick, an error, or a paradox in any useful sense — it is what a weighted average does when the weights move. Group A was half the total and is now a tenth of it, so the combined figure is a different mixture than it was, even though neither group got worse.

The mix is the entire mechanism. Whenever the group with the worse rate grows as a share of the total, the total is dragged towards it regardless of what either group did. Which means the first thing to look at when a total disagrees with its parts is not the rates. It is the sizes.

The page shows both groups, the combined figure, and the share each group holds, before and after.

Which figure is right depends on the question you are asking. If you want to know whether the thing you changed works, the groups answer that. If you want to know what happened to the business, the total does. They answer different questions and arithmetic does not settle which one you meant, so this page computes both and refuses to choose between them.

Two groups is the smallest example, not the limit. With more groups the effect is easier to produce rather than harder, and it hides better — which is why a headline rate with no breakdown underneath it is worth so little.

It will not tell you a channel is failing, a team is underperforming or a change did not work. Those are conclusions, and this page deals in the arithmetic that has to come before them.

Save as many versions as you like and put any two side by side, figure against figure, with the better one on each line marked and the difference in its own column. That screen is the one that decides things, and it is the part a free calculator cannot do at all. Two comparisons side by side is how you find out whether the reversal is a one-off or the shape of your reporting.

One HTML file. No account, no subscription, no internet needed. Opens on Windows, Mac and Android; the readme explains the extra step on iPhone and iPad, where Apple blocks JavaScript in downloaded files and every tool of this kind is affected.

Everything lives in your browser on your device. Nothing is uploaded, which is why it works offline and why nobody else can see your figures — and it also means clearing your browsing data deletes it. The tool does not pretend otherwise: the header shows how long since your last export at all times, turns amber after three days and red after seven, and export is one click that produces a plain file you keep.