I'm working through the estate distribution calculations. The total estate is approximately $2.9M, comprised of the house at $850k plus brokerage and bank assets and personal property. Under the current will, Sofia receives the house and brokerage totaling $1.55M, while Claire and Erik each get roughly $675k from the residue. If both Sofia's contingencies fail, Claire and Erik would split the entire estate evenly at around $1.45M each.

Now I'm mapping out the financial outcomes for Claire across different litigation scenarios. If she does nothing, she keeps her $675k. If she wins both contests, she gains $775k for a total of $1.45M. Winning just the codicil challenge nets her $425k additional, while winning only the account dispute adds $350k. If she loses with probable cause, she'd cover litigation fees of $150k-$250k, leaving her with roughly $425k-$525k net. The worst case is losing without probable cause, which triggers the no-contest clause in the codicil, forfeiting her entire $675k benefit plus fees—a net loss of around $200k—with the residue likely redistributing to Erik.

The key strategic insight is that challenging the account transfer alone appears to sidestep the no-contest clause since it targets a nonprobate transfer outside the will's language, creating a risk-free path where her worst case is $675k minus fees rather than total forfeiture. The exploitation claim under the statute is similarly positioned as independent of the probate contest, though there's a distinction: if she uses the exploitation claim to void the codicil itself, that might be recharacterized as a contest and trigger forfeiture. The safer approach is targeting the inter vivos transfers and POD accounts directly, where fee-shifting provisions under the exploitation statute would presumably favor the prevailing party.

Now I'm working through the undue influence analysis. The presumption clearly arises given the classic markers: confidential relationship as a live-in caregiver, active procurement (she found the lawyer, arranged the appointment, drove him, provided the asset list in her handwriting), isolation, and substantial benefit—the $850k house represents nearly 30% of the estate. The burden shifts to Sofia to rebut. Her defenses include the independent lawyer conference where she waited outside, the memo documenting that Erik was alert and adamant, the natural motive to reward three years of caregiving, and the fact that Claire's estrangement provides an alternative explanation for why she didn't receive the house. But Erik didn't entirely disinherit Claire—he only gave the house to Sofia, which complicates the narrative.

Weighing the evidence, I'm estimating undue influence on the codicil at around 55-65%, with capacity claims weaker at 25-30% given the five-week gap between the "clear and conversant" note and execution—though that memo is the strongest day-of evidence we have. Combined success on the codicil looks like 60-65%.

The convenience account theory is particularly strong here. The bank memo reads almost like direct evidence of non-donative intent—"so Sofia can handle my bills if I'm laid up"—and her presence at the branch supports an influence argument. This could potentially undermine the account transfer claim.

Now I'm thinking through the mechanics: the personal representative would normally bring the account recovery claim, but I need to figure out who that is and whether Claire has standing or needs to petition. If the account gets recovered, it flows back into the estate's residue and splits with Erik under the codicil, so Claire only nets half of any recovery. The litigation costs are substantial—probably $175k to $300k for a full trial, so I'm modeling around $200k for planning purposes.

For the account-only challenge without contesting the will, I'm calculating the expected value: if we win at 70% probability, we recover roughly $925k after accounting for fees, versus losing at 30% and netting $575k. That gives an expected value of about $820k, which beats the $675k baseline of doing nothing.

Now for the full contest strategy, I'm modeling all three claims together—the codicil, account, and exploitation allegations. Since they're correlated (same evidence influences all three), I'm estimating the joint probabilities: both succeed at 50%, codicil alone at 10%, account alone at 20%, and neither at 20%. The payoffs range from $1.225 million if both win down to a small recovery if both lose, since the presumption of undue influence is so strong that forfeiture risk is minimal even in the worst case.

Running the expected value calculation across all scenarios gives me roughly $937k for the full contest approach, which beats the account-only strategy at $820k. The key tactical insight is sequencing—I should file the account and exploitation claims first to build the factual record through discovery, then use that foundation to strengthen the codicil contest while staying within the statute of limitations.

There's also a clever strategic angle with Erik: since he gains the same $775k upside and has direct knowledge of the testator's statements, having him file as the named contestant while Claire supports from the sidelines could shield her from forfeiture risk if the contest succeeds. The forfeiture clause only applies to beneficiaries who actually contest, so if Erik leads and wins, both benefit without Claire bearing the downside—though I need to be careful about how much direct support she provides, since funding or orchestrating another's contest might cross into being deemed a contestant herself.

Now looking at the lawyer's memo through both lenses: it helps Sofia's position with the day-of capacity observations, the testator's own stated motivation to reward Sofia's care, and the fact that Sofia waited outside during the private conference with no opportunity to influence. But it also helps Claire's argument because Sofia was the one who located and engaged the lawyer, scheduled the appointment, and drove to the meeting—suggesting procurement and control over the process.

The memo's weaknesses for Sofia are significant: it shows no independent inquiry into undue influence, no capacity screening, no contact with family members, and no explanation for why the will deviated from prior arrangements. The asset list the lawyer reviewed came from Sofia's handwriting, which means she supplied the information and undermines the testator's knowledge of his own property—he was essentially relying on her version of his assets. The memo itself is brief and conclusory, and the lawyer's interest in defending his own work creates bias. Since Sofia referred the lawyer, he's not truly independent counsel for rebuttal purposes. The memo also doesn't document who did most of the talking or whether Sofia had already briefed the lawyer beforehand, so I need to target discovery at the lawyer's intake notes, phone records, billing entries, and communications about who initiated what.

The handwritten asset list cuts both ways: Sofia could claim she was merely transcribing at his dictation due to his poor handwriting or vision, which would be innocent, and the memo does say the client brought the list, suggesting he adopted it. But under the governing law, supplying information about assets constitutes active procurement regardless of motive, and it demonstrates he couldn't articulate his own holdings. On capacity, the diagnosis alone isn't enough; "good days and bad days" means fluctuating capacity, so whoever bears the burden of proving a lucid interval at execution becomes critical—typically the proponent must show capacity unless the contestant has to overcome a presumption.

The five-week-old "clear and conversant" note actually cuts against consistency and shows fluctuation rather than supporting it. The lawyer's memo from the day of execution is the strongest capacity evidence since it comes from a professional observer. The capacity standard itself is low: he knew his children (though the memo doesn't specify he named them—that's a gap; the doctor noted he recalled family details five weeks earlier), knew his disposition (giving the house to Sofia), and knew the nature and extent of his property (though he borrowed this from Sofia's list, which is the weakest link). So capacity becomes a close call.

On the rebuttal side, Sofia bears the burden of showing his act was free and voluntary, which means finding evidence of independent volition—prior statements to Erik or others about intending to leave her the house, statements to the doctor or neighbor, anything showing he decided this on his own. The critical discovery question is whether he told anyone besides Sofia about this plan; the absence of corroboration is nearly fatal. Her credibility matters too, along with whether she was actually paid wages for caregiving (if she was, the "rewarding care" narrative weakens considerably) and her lack of documentation.

On the brokerage account, the estate's burden is clear and convincing, but the bank memo works in our favor—it's contemporaneous and comes from a neutral source recording Henrik's own stated purpose of convenience rather than gift. The real problem is that POD designations don't help pay bills while he's alive; they only operate at death, so his stated purpose doesn't align with the instrument he chose, which suggests he either misunderstood what he was doing or someone else selected this mechanism. Sofia being present at the branch adds to the procurement concern and the lack of independent advice, though the banker's memo does show Henrik articulated a reason when they spoke, which complicates our narrative.

For discovery, I need to pull the bank forms to see whether Sofia filled them out or signed anything, get testimony from the branch employee, and determine if Henrik held a power of attorney—because if he wanted help paying bills, a POA would be the natural tool, not a POD, and that gap is telling. I should also examine account statements for any transfers to Sofia during his lifetime, which would show exploitation.

The no-contest clause in the codicil reaches "benefits under them" and could threaten her $675k residue share, but I can argue that a clause added by codicil shouldn't retroactively burden gifts made under the 2019 will—that's a construction argument worth raising.  The codicil republishes the will so the clause is probably enforceable, but here's the key: if Claire wins the codicil contest, the clause dies with it, so the forfeiture threat only matters if she loses. On probable cause, the facts are overwhelming—Sofia found the lawyer, scheduled the appointment, drove Henrik, wrote the asset list, and she's a live-in caregiver getting $1.55M, plus there's documented dementia and the bank memo. Claire's estrangement looks bad, but her reconciliation texts show good faith motive, so she should stick to the strong facts and avoid overreaching claims like alleging forgery. The no-contest clause doesn't mention nonprobate transfers, which is another angle to explore.

The strategy is to keep the codicil attack in the probate contest where probable cause gives her protection, and file the exploitation claim separately focused on the account and lifetime transfers rather than trying to unwind the house transfer through that route. For the exploitation claim itself, the vulnerable adult element is solid—she's 88 with vascular dementia and dependent on Sofia as her live-in caregiver—and Sofia clearly has a position of trust, so the focus shifts to proving she knowingly obtained or used Henrik's assets. I need to be careful about standing and who can bring the claim—the personal representative or Claire as an interested party. There's also the option of reporting to Adult Protective Services or making a criminal referral, though I have to be cautious not to use criminal prosecution as leverage in the civil case. If Sofia ends up as the PR or executor, that's a major conflict and I should push for her removal or appointment of a special administrator instead. For settlement, I'm thinking about bringing everyone together early—Sofia, Claire, Erik, and the PR—to negotiate a global resolution that covers the codicil, the account, any exploitation claims, lifetime transfers, and mutual non-disparagement provisions.

The key is structuring this so the no-contest clause doesn't block the settlement, which means getting court approval and buy-in from all beneficiaries. Sofia currently controls about $1.55M, which is roughly 53% of the estate, so a defensible settlement would have her keep somewhere between $300k and $550k—accounting for her three years of live-in care and the litigation risk—with the remainder flowing back to the residue. If Sofia takes around $400k in cash and the house gets sold or allocated separately, Claire and Erik could each end up with roughly $1.25M, which beats the expected value of litigation at $937k when you factor in legal fees, risk, and time. I'm considering whether Sofia keeps part of the POD account or takes a lump-sum cash payment funded from the house sale.

I also need to think through Erik's incentives carefully. He and Claire have identical economics, and he's a critical witness, but there's a structural conflict—if Claire drops out, Erik gains an extra $675k, which could tempt him to sit on the sidelines and free-ride. I should lock in a written cost-sharing and joint-prosecution agreement with Erik, ideally making him the lead petitioner and having him cover half the legal fees. The exploitation claim itself is a lever here, and I need to make sure the settlement addresses any wage or quantum meruit claims Sofia might raise.

On the preservation side, I'm looking at getting a TRO or lis pendens on the house, asking the court to freeze any brokerage POD proceeds, and tracing any funds already distributed. The key insight is that an exploitation claim reaches Sofia's personal assets—not just the estate—so it's not capped and survives even if the codicil is upheld. That asymmetry, combined with fee-shifting exposure and reputational risk, is what should push Sofia toward a global settlement. The strategy is to contest through the nonprobate and exploitation claims first, which sidestep the forfeiture risk entirely.

For immediate action: file a caveat and demand for notice in probate court, determine who the PR is, and issue preservation letters to freeze or trace the POD payout; if the house transfer is pending, record a lis pendens; if Sofia controls the estate, petition for a special administrator. Before pleading, I need the medical records from around the signing date, the drafting attorney's complete file including engagement letter and intake notes, and the bank and brokerage account documentation.

I should also secure Erik's declaration and statements from third parties who witnessed Henrik's isolation—physician, neighbors, housekeeper, mail carrier. Once I have the joint-prosecution agreement with Erik in place, send Sofia a pre-suit demand with a mediation proposal and probable-cause record; file both the exploitation and codicil contest claims within the statutory window if she doesn't settle. I'll need to preserve Claire's communications and prepare for the narrative that she's a disgruntled estranged daughter, while also evaluating whether her estrangement story holds up.

On the legal side, I need to document the probable cause before filing since that's what gets evaluated at the time of filing. I'm also going to work up a numbers table showing the break-even probability and fee threshold for the contest—comparing the expected value against the baseline $675k—and factor in the time cost: probate contests typically run 18 to 30 months, which means Claire's residue distribution gets held up entirely during litigation, adding another financial pressure to settling. I'll draft a memo with clear headers, bullets, and that numbers table, noting that all probabilities are judgment estimates. I should also flag the burden of proof issue—in most jurisdictions the proponent bears the initial burden, though that varies.