I could not agree more.

99.9% of statistics is needed to think how to design an experiment and how to interpret the measured response in relation to the experimental factors. You don't need much expertise in statistics to actually analyze the data (that's done by software, anyway). And if you start to think how you might analyze your data (what method to use or what button to click) after the experiment, you already missed the most important part, and it is very likely that you have data that is insufficient or uncapable to answer your actual scientific questions.

I could not agree more.

99.9% of statistics is needed to think how to design an experiment and how to interpret the measured response in relation to the experimental factors. You don't need much expertise in statistics to actually analyze the data (that's done by software, anyway). And if you start to think how you might analyze your data (what method to use or what button to click) after the experiment, you already missed the most important part, and it is very likely that you have data that is insufficient or uncapable to answer your actual scientific questions.

Pithy, poignant, and pertinent.

Perhaps folks could submit their favorite statistical quotations. One of mine comes from George Box:

"All models are wrong, but some are useful."

Thanks to remember Giants.

We are standing on the shoulders of Giants.

To consult the statistician after an experiment is finished is often merely to ask him to conduct a post mortem examination. He can perhaps say what the experiment died of

and of course, learn not to repeat the same mistake again.