This is really fascinating. I also like how the supreme intelligence points out how November-March does a great job of capturing the true PNW cold season. I do have a hard time believing 20-21' was the 5th coldest this Century.
Using the NCEI/NClimDiv statewide temperature dataset, and defining the winter as November through March—i.e., the 5-month period ending in March—the coldest winter of the 21st century so far appears to be 2022–23 (November 2022 through March 2023).
NCEI's Climate at a Glance system specifically supports multi-month time scales, including a 5-month period ending in March, and its Oregon statewide series begins in 1895. (NCEI)
Oregon's coldest November–March winters since 2000
Winter
Period
Relative cold
2022–23
Nov 2022–Mar 2023
Coldest
2010–11
Nov 2010–Mar 2011
Very cold
2008–09
Nov 2008–Mar 2009
Very cold
2005–06
Nov 2005–Mar 2006
Cold
2020–21
Nov 2020–Mar 2021
Cold
2016–17
Nov 2016–Mar 2017
Cold
The standout is 2022–23, and there's a good reason it can be easy to underestimate it if you only look at December–February. March 2023 was exceptionally cold statewide—NOAA ranked it the 3rd-coldest March in Oregon's entire 1895–2023 record. That very cold March pushed the five-month season substantially downward.
NClimDiv is the appropriate dataset for this calculation: NCEI describes it as a continuous monthly dataset from 1895 to present, with statewide temperature values derived from the underlying gridded climate-division data. (NCEI)
One important distinction: 2022–23 was not necessarily Oregon's coldest traditional meteorological winter (December–February). Your November–March definition captures the extended cold season much better for the Pacific Northwest, particularly because March 2023 was so cold.
If you want, I can also give you the exact NCEI statewide mean temperature for every November–March winter from 2000–01 through 2025–26, ranked from coldest to warmest.