The Globe’s Frank Phillips has weighed in on the question of how Donald Trump’s presidency might impact the 2018 Massachusetts governor’s race.With the help of MassPoliticsProf Peter Ubertaccio, Governor Baker’s political advisor, Jim Conroy, Democratic pollster Lou DiNatale, and GOP strategist, Rob Gray, Philips reinforces my recent assessment of the potential for a serious anti-Trump backlash against Republicans on the ballot in 2018, including Charlie Baker.

The Phillips’ piece adds a couple of important wrinkles to the equation, namely the unprecedented fund raising prowess of the governor and the seemingly lackluster Democratic field, which conspicuously does not include Attorney General Maura Healy. These are both fair and reasonable rebuttal points, raised in the Philips’ piece by GOP operatives, to the notion that Baker is in trouble. I feel compelled, however, to sound some additional cautionary notes related specifically to the governor’s popularity and organizational strength, as well as the likely immeasurability of Trump’s potential for toxicity.

The governor’s very impressive popularity numbers almost certainly do not provide clear evidence that Bay State voters in 2018 will separate their feelings about Trump’s Washington from their perceptions of Baker’s Beacon Hill. Scott Brown illustrated this disconnect very well in 2012 by losing to Elizabeth Warren by 8 points despite his perch atop the polls as the “most popular politician in Massachusetts” at the time. Baker’s advisors are quite right, however, that Massachusetts voters have a history of separating national and statewide campaign narratives, having send a number of Republicans to the Corner Office despite sending none to Washington on the same ballot, and that Baker’s popularity and financial advantages will help him convince voters to do that again next year. All things being equal, I find Baker’s backers’ confidence reasonable enough. However, in a few months the dynamic of the 2018 national campaign narrative will begin to take shape and neither his popularity nor money will allow him to dictate the tenor of that narrative. As Professor Ubertaccio points out in the Philips piece, the “great big unknown” here is just how toxic Donald Trump and Washington Republicans will be next year. I would add an additional “unknown” here, namely, what will Elizabeth Warren’s role be in the 2018 narrative, and will that role suck up enough of the political oxygen to neutralize Baker’s conventional organizational and resource advantages? As he demonstrated in yesterday’s White House ceremony honoring the Navajo Code Talkers of World War II, Trump seems unable to resist raising Warren’s profile.

Were there a conventionally unpopular Republican in the Oval Office, I would be much more confident in Baker’s ability to establish a distinct campaign narrative in 2018, but because Trump has established a clear preference for narcissistic trolling over productive party leadership, I think conventional measures of Baker’s electoral strength and the potential for damaging “national narrative creep” in 2018 amount to a virtual crap shoot.

Finally, what say the Bay State’s small but loud band of Trumpers? Has Baker really built a Hillary-style firewall against GOP trouble makers? Are there really no publicity hungry wingnuts left in the MassGOP just itching to go “Full-Bernie” on the liberal Republican governor who just took a clear stand in defense of women’s reproductive rights and who will be increasingly on the lookout for high profile opportunities to separate himself from The Donald?

There are a lot more unknowns about the 2018 Massachusetts governor’s race today than there were a month ago. The pundits, pollsters, insiders, and blogging profs can be forgiven for attempting to apply conventional wisdom to this popular parlor game, but when it comes to the potential toxicity of Donald Trump in 2018, we may have to recognize a new species of unknowns that would probably even befuddle Donald Rumsfeld himself. In the Age of Trump we may have to get used to grappling with “virtually certain unknown unknowns.”