My system predicting Red Lake's ice-out

I've got a pretty good track record for picking the date for ice-out on Red Lake, usually pinpointing it within a few days. I thought others might be interested in how I do this.
First of all, I don't make my prediction until April. That is when spring melting actually begins and it is the weather in this month that accounts for 75 per cent of the timing of ice-out. The other 25 per cent comes from the weather in the first week of May.
I am emphasizing this to keep you from going down the rabbit hole of believing that last winter's conditions have anything to do with it. They simply do not.
I don't know how many times I have heard people say that because they had to put extensions on their ice augers in March that breakup would be exceptionally late and then the actual date was early. Or the opposite situation: there was less ice than normal so ice-out would be early -- except it wasn't.
Trust me on this; I've only seen it about 60 times.
Now, let's start with the fact that Red Lake's average ice-out date is May 8. That is just a fact gleaned from nearly a hundred years of data. Except for a handful of years, all the other ice-out dates are within one week either side of May 8. OK? So, you could say that normal ice-out is from May 1 to May 15. May 8 is right in the center.
When April comes around start paying attention to the highs and lows and how they compare to the averages for that date. The Weather Network conveniently does this for you in a graph on their 14-Day forecast.
If the temperatures are exactly normal all the time guess when ice-out will be? May 8! Do you see where this is going? Days below normal add to the time of ice-out and days above normal subtract.
I don't actually keep track of the days but rather the weeks. A week above normal moves the date to May 1. But if that is followed by a week below normal the date is back to May 8.
Finally, there are a couple of fudge factors. For instance how abnormal were the highs and lows for the weeks? If the deviation was only a few degrees, it will hardly make any difference, a day or two, for instance. If the temperatures are way out of whack, like 10 C, for most of the week, then it could move ice-out by an extra week.
A few other factors, like wind and rain, also play a marginal role. High winds, especially near May 8, will hasten the process. Rain any time does the same thing.
So, starting April 1 each year you can look ahead at the forecast and start to get an idea for your prediction then refine it as time goes on and you observe what actually happens weather-wise.
It's a fun thing to do while you anxiously wait for open water!
I'll just bring you up to date by telling you that the first two weeks of April were below normal but not critically so. That makes me move the ice-out date to about May 15.