Indeed, the EPS never bought into the -EPO. And I don't think it has been bad as the GEFS overall. But don't forget that the EPS has had enormous swings from cold to much warmer in just one run. Example: Here's last Thursday's (1/16) 0Z EPS, which had solidly BN over virtually the entire US in the 11-15 (see images below) and which looked cold at the end:
0Z 1/16 6-10: cool SE, AN much of N US
View attachment 31762
0Z 1/16 11-15: cold SE and much of US
View attachment 31760
The very next run, 12Z on 1/16, had a whopping 24 fewer HDDs! From another BB per a met there:
"Yeah, that 12z EE was incredibly warmer. 24 less HDD's which is a ton for an ensemble in just 1 run."
12Z 1/16 6-10: N SE and A to MA much of N US
View attachment 31763
12Z 1/16 11-15: N SE and much of US
View attachment 31761
Then the next two EPS runs, the ones on 1/17, each continued the sharp warming to the point that the 12Z 1/17 run had an astounding 50+ fewer HDD than the 1/16 0Z run!!
12Z 1/17 6-10: A to MA much of US including SE
View attachment 31764
12Z 1/17 11-15: SE is a little AN; N to AN most of US other than SW
View attachment 31765
And the latest EPS (12Z on 1/22), despite some modest cool changes and a better look, is still AN much of the US in the 6-15 as it warmed a good bit further over the long weekend:
12Z 1/22 6-10: A to MA most of US
View attachment 31767
12Z 1/22 11-15: AN most of US
View attachment 31768
So, I'm advising extreme caution in buying into a colder pattern based on how the EPS looks in week 2. Perhaps it will be different this time?? Who the heck knows?
Note that the 1st 6 maps are 850 anomalies whereas the last 2 are 2 meter. That's because I have old 850s saved but not old 2 meter. I would have preferred to display only 2 meter maps if I could. But the general point is still clear.