Well, I wanted to try it again, so I went for the last week and a half, almost 2 weeks, without eating grains. Got a bit more data. Plus I’ve been tracking this stuff non stop since the beginning of the year. So here’s what I came up with, comparing every day I’ve eaten grains with every day that I haven’t:

stomach with grains: [‘3.10’, ‘2.92’, ‘2.78’, ‘2.83’, ‘2.36’, ‘3.10’, ‘2.75’, ‘3.29’, ‘3.12’, ‘3.34’, ‘2.87’, ‘3.50’, ‘3.60’, ‘3.71’, ‘3.00’, ‘3.67’, ‘3.25’, ‘2.53’, ‘3.00’, ‘4.00’, ‘3.44’, ‘3.28’, ‘2.33’, ‘3.22’, ‘3.81’, ‘2.82’, ‘2.71’, ‘3.46’, ‘2.40’, ‘2.82’, ‘2.84’, ‘2.94’, ‘3.09’, ‘2.83’, ‘3.60’, ‘3.60’, ‘2.91’, ‘3.55’, ‘3.00’, ‘3.12’, ‘2.75’, ‘2.77’, ‘2.81’, ‘2.33’, ‘3.22’, ‘3.22’, ‘2.75’, ‘2.86’, ‘3.14’, ‘3.29’, ‘3.00’, ‘3.12’, ‘3.14’, ‘3.20’, ‘3.00’, ‘3.36’, ‘3.33’, ‘3.40’, ‘3.71’, ‘3.34’, ‘3.40’, ‘2.82’, ‘2.71’, ‘3.00’, ‘3.29’, ‘3.21’, ‘3.17’, ‘2.67’, ‘3.50’, ‘2.40’, ‘2.80’, ‘3.81’, ‘3.08’, ‘3.42’, ‘3.56’, ‘2.88’, ‘3.50’, ‘2.86’, ‘2.84’, ‘3.38’, ‘3.24’, ‘2.88’, ‘2.29’, ‘2.89’, ‘2.84’, ‘3.12’, ‘2.00’, ‘2.20’, ‘2.75’, ‘1.91’, ‘2.33’, ‘2.67’, ‘2.43’]
mean = 3.03
stomach without grains: [‘3.33’, ‘3.67’, ‘2.89’, ‘3.37’, ‘3.31’, ‘3.60’, ‘3.00’, ‘2.87’, ‘2.59’, ‘2.80’, ‘2.92’, ‘3.11’, ‘3.22’, ‘3.63’, ‘3.47’, ‘3.29’, ‘2.89’, ‘2.67’, ‘3.30’, ‘2.88’, ‘3.87’, ‘3.80’, ‘3.85’, ‘4.00’, ‘3.57’, ‘3.22’, ‘2.70’, ‘3.30’, ‘3.59’, ‘3.18’]
mean = 3.26

t-test on stomach before and after grains:

t = -2.67446580787, p = 0.00851982121181

So there we go! Going grain-free very probably helps my stomach feel better! If going grain-free didn’t help my stomach, there is only a 0.8% chance that the means would be this different.

However, there are complicating factors:

- travel days. On travel days stomachs are weirder.

- Google days. I used to be getting gourmet meals 3x/day, and now I have to fend for myself.

So let’s ignore the travel days:

t-test on stomach before and after grains:

t = -2.23123617479, p = 0.0277911206015

stomach with grains: mean = 3.09

stomach without grains: mean = 3.26

Still significant.

How about only comparing days after I left Google?

t-test on stomach before and after grains:

t = -1.06189037683, p = 0.294500544066

stomach with grains: (daily averages omitted)

mean = 3.13

stomach without grains: (daily averages omitted)

mean = 3.26

No significant effect.

Did Google just make my stomach worse? Let’s compare all the days while I was still at Google to all the days after I left.

t-test on stomach before and after Google:

t = 1.00310288377, p = 0.317813018172

stomach while I was still at Google: (daily averages omitted)

mean = 3.11

stomach post-Google: (daily averages omitted)

mean = 3.03

Nah, looks like the gourmet meals were nice after all. (or at least, probably no real effect either way.)

So I’m guessing that, all other things being equal, grain-free makes my stomach feel slightly better. It appears to have an effect, but it’s a small one. And I don’t have enough data from post-Google no-travel life to show statistical significance yet.