California's Top-Two Primary Is a Stress Test. pWAR Has Opinions

The mess that is California politics provides an interesting challenge for our analytics tool

California’s June 2 primary is now less than three weeks out, and the race to succeed Gavin Newsom has produced something genuinely unusual: two Republicans spent most of the spring consistently leading the polls in a state that hasn’t sent a Republican to statewide office since 2006. The reason is structural. California uses a top-two primary system, where all candidates from all parties appear on a single ballot and only the top two vote-getters advance to November, regardless of party. In a cycle where Democrats were badly fragmented across seven or eight serious candidates and Republicans were consolidated behind two, the arithmetic was uncomfortable enough that some Democratic strategists put the odds of an all-Republican November at 27%.

That scenario has receded. But the Democratic field is still fragmented, and the primary itself has been chaotic enough to warrant running the whole thing through pWAR, especially now that the dust from Eric Swalwell’s implosion¹ has begun to settle into something that looks like an actual race.

If you’ve been following this publication’s analytics work, you already know the framework. If not, the short version: pWAR (Political Wins Above Replacement) borrows from baseball’s WAR metric to measure a candidate’s value relative to a generic party-line placeholder. It asks not “did this candidate win?” but “did this candidate add anything beyond what any warm body with the right letter next to their name would have produced?” The formula combines Electoral Value (margin vs. party baseline), Financial Efficiency, a Difficulty Adjuster for hostile-environment performances, and Legislative Productivity and Constituent Service for incumbents, then subtracts penalties for primary losses, general election underperformance, and the dreaded Frequent Flyer Penalty for serial losers. California is not Maryland, so the replacement level baseline requires adjustment. In a state this blue, a Democratic replacement-level candidate is worth roughly 1.5 pWAR, and a Republican replacement-level candidate in a statewide race is worth roughly 0.0, given the party’s structural hole.

With that framing, let’s go through the field.

Xavier Becerra: +2.0 (revised upward from initial score of +0.8)

Becerra is now the polling frontrunner. The most recent Emerson poll (May 9-10) has him leading the field at 19%, having gained nine points since mid-April, with 31% support among Democratic primary voters. That matters for the pWAR score because the original +0.8 was an honest qualitative discount applied to a formula that produced +4.0 on paper. The discount reflected genuine uncertainty about whether his institutional record would translate into actual voter support. That uncertainty has resolved. It is translating.

The underlying formula still holds: long record as California AG and HHS Secretary, which generates real LP (+2.0) and CS (+1.0) credit. EV from his prior AG elections was always a soft input (+0.5) since those races were non-competitive. What’s changed is the FE_mod. His post-Swalwell donor surge has broadened his base meaningfully enough to move him from an average fundraiser (+0.5) toward grassroots-leaning (+1.0). He was pulling near-zero in fundraising before Swalwell’s exit opened a lane, then got a genuine small-dollar surge. No penalty history at the state level. DA_bonus = 0 for an environment he hasn’t faced competitively.

The formula now lands closer to +3.5. The qualitative adjustment comes down but doesn’t disappear: his debate debut drew criticism for lacking policy specifics and for grading Newsom’s homelessness record an ‘A’ on effort. That’s a real liability in a general election environment where homelessness is among voters’ top concerns.² The honest score is +2.0. Institutional Ace, and the numbers finally reflect it.

One additional context note worth flagging here: California Democrats are now actively urging voters to wait until the last minute to cast ballots and then support whoever is polling in front, a coordinated strategic voting effort designed to consolidate the Democratic field around a single candidate. If that cohort breaks for Becerra, and right now he is the likeliest beneficiary, the +2.0 may underestimate his actual June 2 result. pWAR cannot price in tactical voting behavior. The caveat belongs on record.

Steve Hilton: +1.5