Going light on electrolytes worked. Then I started drinking half the water.

Three weeks ago I pulled electrolytes on every run to see if the headaches came back. They did, on the long ones. So I set a narrower rule and I’ve been running it since: mix only when I need it, plain water the rest of the time.

The rule turned out to be right. It also turned out to break my own math, which I did not expect.

The rule I settled on

Two triggers. Either one gets a dosed bottle:

• Distance: 16K (10 miles) or longer.
• Vert: more than about 2,000 ft of gain, whatever the distance.

The vert trigger is the new part. A 12K with 2,500 ft in it wrecks me in a way a flat 12K doesn’t, and I’d been treating them as the same run because the distance matched. They’re not the same. The climbing days cook me even when the mileage looks easy on paper.

Everything else, the flat evening 6Ks and 8Ks, gets water. That’s most of my week.

Three weeks in: no headaches. Not one. And the long runs themselves have felt better than they did in June, which I didn’t anticipate since I’m using less salt overall, not more. Fewer runs get the mix, so the mix lands on the runs that actually needed it and I’m not choking down salt water on a Tuesday for no reason.

That part’s a clean win. Here’s the part that isn’t.

My thirst fell off a cliff

Somewhere in these three weeks my body got more efficient, and I’m drinking about half the water I was drinking a month ago. Same distances, same climbs, same heat. Half the fluid.

I noticed it on refills. A 20K used to mean both vest bottles empty and a hit off the bladder. Now I come home with a bottle still mostly full and I never wanted it. I’m not rationing or toughing it out. I’m just not thirsty, and I feel fine.

Some of that is probably real heat adaptation. It’s been hot since May and this is my third summer of doing volume in it. Some of it might be that I’m running smarter, more even efforts on the climbs, and sweating less as a result. Honestly I don’t know the split, and I don’t have a good way to find out without redoing the sweat rate test a few times in the next couple of weeks. Which I should do.

Why that breaks the calculator

Here’s the mechanic. The calculator works out how much sodium you lose from your sweat rate and your duration, then spreads that sodium across the fluid it thinks you’ll drink. What you get back is a per-bottle recipe.

So the sodium is bolted to the water. Drink the bottles, get the salt.

Except I’m not drinking the bottles anymore. I’m drinking half of them, and the salt that was supposed to go in the other half is still sitting in the tub. My sodium need didn’t halve. My water intake did. The calculator has no idea, because it never asked me how much I actually drink. It just assumes I’ll finish what it prescribes.

That’s a real hole in the model, and it’s mine, so I get to own it. The fluid replacement target in there is set to 75% of sweat loss, which is a reasonable book number and is nowhere near what I’m doing right now. I’m probably closer to 40%.

What’s kept me out of trouble so far is that I’ve been drinking the dosed bottle first and the plain water second, or not at all. So I’ve been getting most of the salt by accident of ordering. If I’d been sipping evenly out of both bottles I’d have been running a real deficit for three weeks, and I’d probably have written a different post.

What I’m changing in the tool

The fix is to stop treating fluid intake and sodium intake as the same number. Sweat loss sets how much sodium you need. What you actually drink sets how concentrated it has to be. Those are two different inputs and the calculator currently only has one.

So I’m adding a way to tell it what you really drink per hour, separate from what you lose. Keep the total sodium fixed off sweat loss, then divide it into the fluid you’ll genuinely finish. Drink half as much, get twice the concentration, same salt in the body at the end.

Two things I have to be careful about before I ship it:

• Concentration has a ceiling. I sit in the salty-sweat bucket, so doubling my concentration doesn’t land somewhere polite. It lands in the range where the bottle tastes like warm broth and my stomach starts filing complaints on a climb. There’s a point where the honest answer is “you can’t fit this much sodium in this little water, drink more or take a capsule,” and the tool should say that instead of quietly handing you something undrinkable.
• Drinking half your sweat loss isn’t advice. It’s what I’m doing, and I feel good, and those are different claims. If the calculator makes low fluid intake easy to plug in, it shouldn’t read as an endorsement of low fluid intake.

I don’t have the interface figured out yet. Probably a second slider for actual intake that defaults to the current 75% assumption, so nothing changes for anyone who doesn’t touch it. Maybe a warning band when the concentration gets stupid. I’ll write it up when it’s in.

The short version of three weeks: the narrower rule works, the headaches are gone, and my body quietly changed the inputs on me while I was busy congratulating myself on the protocol. That’s the part I like about doing this in the open. I’d have never caught it if I weren’t measuring.

If your thirst dropped through the summer while your salt need didn’t, I’d like to hear about it.