Alright, so yes, I get high amount predictions every day (both in the Rachio app and the Apple Weather app). Every day. And every day, it does not happen (except yesterday, we had 1.59” of rain yesterday for the first time in almost three months). The prediction forces Rachio to skip watering, but the rain never comes. Repeat that a few days and you get wilting grass due to high sun. This is the predicate of the thread: Rachio should not allow the zone to end the day at 0 too many days in a row, even if predictions say it will rain. I start the day with end balance 110% every day but end up with end balance 0% every day. At some point, Rachio should think it should ignore the prediction and water regardless.
Last week had rain predicted every day, yet it never rained. Rachio never wanted to water because of the predictions. It’s an edge case, for sure. I just happened to stumble upon it, where predictions are far off many days in a row. Rachio currently plans to water next on June 20th 
How long have you had a Rachio?
How long have you had tempest?
Was this always an issue since day 1 or has something changed?
Rachio use for a couple months, Tempest user for over 6 months. This problem showed itself last week when rain started being predicted. Here are the current rain prediction for the next 7 days by Apple Weather (unrelated to Rachio/Tempest)
This false prediction never happened before to my knowledge. It’s been almost 10 days of this rain/no rain game.
In apple weather change the cloud drop down to precipitation. What is the hour by hour look like
To reiterate, this is possibly a once in a lifetime event - I have never ever seen so many consecutive rain predictions go south. It may or may not happen in the future, or in other places. The high predicted amount during the early hours when Rachio decides whether my before-sunrise-schedule should run or not, make Rachio skip the watering. Then later in the day, the prediction amount starts decreasing, eventually reaching zero, with no rain actually happening. Then repeat - in reality, the time rain was supposed to start kept being pushed back, as if the cloud slowed down. 10 days in a row… Every morning, rain expected at, say, 1pm, only to be pushed back to, say, 8pm throughout the day, then completely disappear and move to 1pm tomorrow. Hours are approximate, they weren’t the same each day.
I see two ways Rachio can improve this. The numbers are purely arbitrary, no math behind them, someone should math if either way is to be implemented 
a) the reactive way: include past end balance into the mix:
- if previous day does not end in 0% balance, use full predicted amount
- if only previous day ended in 0% balance, use 50% of the predicted amount
- if only last two days ended in 0% balance, use 25% of the predicted amount
- if three or more previous days ended in 0% balance, ignore the predicted amount, assume no rain, do your job, water that lawn
b) the proactive way: consider using the precipitation chance as well as the amount:
- if precipitation chance is less or equal to 30%, use 25% of the predicted amount
- if precipitation chance is between 30 and 50, use 50% of the predicted amount
- if precipitation chance is over 50, use full predicted amount
I think the reactive way is perhaps better as it has the least amount of impact for scenarios outside of this weird one.
I may be in the same situation as the OP. Rachio had my soil moisture in the 80’s yesterday with a 70% chance of rain and 0.74 in. We haven’t gotten any rain yet and Rachio says soil moisture is 110%. We could get some rain later today. Weatherbug has us at 40 - 55 chance today with 30 - 80% chance of storms/rain for the next 10 days.
Yeah this is a very corner case, where you need predictions that never come true and you need that to repeat several consecutive days. I doubt a well established lawn will die, a little stressed, but since I have newly laid sod, the lack of watering was much more visible immediately. It takes many factors to come together 
Dude,
Give up on the smart watering and just use a fixed schedule that you update as needed. You’ll go crazy otherwise. I did.
Maybe we should stop… No! We do not quit! Again!
https://getyarn.io/yarn-clip/a35c240e-74f9-432e-ba5a-04c55519fb73
Rachio’s weather for today had 70% chance with an amount of 1.05 in. We actually got 0.21 inches and it looks (by radar) like no more is coming today.
My next run is scheduled for the 19th.
4.52” so far today 
and Rachio just decided to go offline. Who’s going in the rain to troubleshoot wifi?
Next watering for me is Jun 24th, but I do have a solid 10 day rain forecast, so there’s that.
I live up north so probably not as hot as where you’re at, I’m looking on average every other day of watering.
This only happens if you get rain forecasted but it never rains, day after day. My next watering is June 24th because the forecast looks hellish with flooding rain - a low pressure tropical system formed in the gulf of Mexico, moving over South Florida this whole week. This time around, it WILL rain, 4.61” today and still going. Last week the forecast said rain but no rain ever came, so Rachio did its thing and kept skipping watering, as designed. The issue was the “no actual rain” part. By the time the rain forecast for “today” dissipated, Rachio had already skipped watering, and the story repeats the next day. Now I have too much water, anyone want some? 
1 Like
True I know what you’re saying. I still think there maybe a setting you can tweak to maybe balance it out.
I have a 40% chance of rain on Friday, showing .35” of total rainfall and Rachio is showing that it will irrigate that day.
It would be nice to get someone at Rachio’s take on this. @dane
Maybe just turn off weather intelligence and just use a physical rain sensor of how much water is collected.
I don’t expect this to be a problem in the long run, I have never experienced so many days of bad forecast, plus this lawn is not going to be new forever
To be honest what you were proposing , which is that Rachio should factor in the chance of precipitation percentage I could see that maybe work for you, but then may break all other customers.
As an example my forecast shows rain ever day with a 10% chance of rain, but it in fact rains every day, I wouldn’t want my irrigation to go off as that would be too much water.
Not sure if there is anything clever to come up with for this. Although I think maybe adding in a 2nd run time into the schedule may help. The 2nd run time will run at some time in the later part of the day and will only run if the 1st one was skipped due to forecast. Throughout the day forecast changed and rain didn’t occur, then 2nd start of on the schedule should kick off.
Even though watering late in the day or at night isn’t ideal this would only occur for those extreme forecasted weather abnormalities.