DECISION AND ORDER
LYNN ADELMAN, District Judge.Mohns, Inc. (“Mohns”) appeals an order of the bankruptcy court approving the trustee’s motion to compromise a claim of the estate under Federal Rule of Bankruptcy Procedure 9019(a). The claim involved the estate’s interest in a house owned by the debtors, John and Christine Wilson. Under the terms of the compromise, the Wilsons paid the trustee $10,000 in cash in exchange for the trustee’s relinquishing the estate’s interest in the house. Mohns is the debtors’ largest unsecured creditor. It has a claim in the amount of approximately $143,000, which represents the balance due to Mohns for constructing the house at issue. Mohns objected to the proposed compromise on the ground that if the trustee continued to litigate the remaining legal issues surrounding the es*909tate’s interest in the house and eventually-sold the house, he could obtain substantially more than $10,000 for the benefit of the estate. The bankruptcy court disagreed with Mohns and approved the compromise.
The present appeal is the second that I have had to decide in this case. In the first appeal, Mohns appealed an order of the bankruptcy court involving the debtors’ claim that $75,000 in equity in the house was protected by a homestead exemption under Wisconsin law. See Mohns, Inc. v. Wilson, 475 B.R. 674, 677 (E.D.Wis.2012). Mohns had argued that its claim against the debtors was one for the “purchase price” of the house and that therefore, under Wis. Stat. § 815.18(12),1 the debtors could not claim the house as exempt with respect to Mohns. The bankruptcy court rejected this argument and entered an order providing that “the claim of Mohns, Inc. is not a claim of a ‘purchase price’ within the meaning of § 815.18(12), Wis. Stats.[,] and the homestead exemption of the Debtors is not limited by reason of, or subject to, the Mohns, Inc. claim.” Mohns, Inc., 475 B.R. at 677. I disagreed with the bankruptcy court’s interpretation of the term “purchase price” and concluded that Mohns’ claim was in fact one for the purchase price of the house and therefore was not subject to the homestead exemption. Id. at 679-80.
My ruling in the prior appeal apparently created several issues that contributed to the trustee’s decision to enter into the compromise that is under review. In seeking the bankruptcy court’s approval of the compromise, the trustee indicated that those issues stemmed from the fact that, arguably, the debtors could continue to claim the house as exempt with respect to the claims of creditors other than Mohns. See Tr. of Feb. 14, 2013 Hearing at 75-76. If that were so, explained the trustee, then perhaps he would have to administer two separate estates — one for the benefit of Mohns and one for the benefit of the other creditors. However, during the hearing on the motion to compromise, Mohns conceded that it was not entitled to have a separate estate administered solely for its benefit and that the trustee could essentially stand in Mohns’ shoes and use the proceeds from the sale of the house for the benefit of all creditors. Tr. at 81, 101-02, 116. In light of this concession, the trustee no longer had to worry about the possibility of administering a separate estate. However, there were other issues surrounding the estate’s interest in the house. The trustee indicated that the debtors intended to take the position that they could still claim the homestead exemption with respect to the trustee, and that therefore the estate’s interest in the house was still subject to a $75,000 exemption. Tr. at 75. He also noted that the debtors intended to claim that, in the event the house were sold, they would be entitled to reimbursement for tax and insurance payments they made during the bankruptcy. Tr. at 76-77. The trustee stated that he wanted to accept $10,000 for relinquishing the estate’s interest in the house to avoid having to litigate these issues.
The trustee’s decision to compromise was not based solely on his desire to avoid having to litigate the remaining issues relating to the house. In addition, the trustee thought that the alternative to accepting the compromise — putting the house on *910the market — -would not yield more than $10,000 for the estate. The house was subject to two mortgages, and the trustee believed that it was probable that after selling the house, paying off the mortgages, and deducting brokerage fees and other sale expenses, the estate would be left with less than $10,000.
On appeal, the parties ask me to review the bankruptcy court’s decision under the legal standards that apply to a bankruptcy court’s decision to approve a compromise under Federal Rule of Bankruptcy Procedure 9019(a).2 A bankruptcy court may approve a compromise or settlement only if it is reasonable. See In re Fort Wayne Telsat, Inc., 665 F.3d 816, 820-21 (7th Cir.2011). Determining the reasonableness of a settlement requires comparing the amount of the settlement to the net expected gain of seeking a litigated judgment. The “expected gain” is the gain if the judgment is favorable, discounted (that is, multiplied) by the probability of a favorable judgment. The qualification “net” signals the need to subtract the cost of pressing ahead to judgment in order to estimate the value of litigating to judgment rather than of settling. Id. at 820.
A bankruptcy court’s approval of a settlement is reviewed for abuse of discretion. In re Holly Marine Towing, Inc., 669 F.3d 796, 799 (7th Cir.2012). However, the bankruptcy court may not simply accept the trustee’s word that the settlement is reasonable, nor may it merely “rubber-stamp” the trustee’s proposal. In re American Reserve Corp., 841 F.2d 159, 162 (7th Cir.1987). The bankruptcy court must make an “informed and independent judgment” about the settlement. Id.
In the present case, the bankruptcy judge did not issue a written opinion in which she explained the reasoning underlying her decision to grant the motion to compromise. Instead, she delivered an oral ruling at the conclusion of a lengthy hearing and colloquy with the lawyers. The judge’s final oral ruling is rather brief, but the judge did make extensive comments about the case in the course of her colloquy with the lawyers. Unfortunately, the judge never made any explicit findings of fact or conclusions of law, and it has proven difficult to identify the reasoning underlying the judge’s conclusion that the compromise was reasonable. To be sure, it is clear that the judge thought it was better for the estate to take the $10,000 that was on the table rather than risk obtaining less if the trustee proceeded to litigate the remaining legal issues surrounding the estate’s interest in the house and to eventually put the house on the market. Tr. at 123-24. What is unclear is how the bankruptcy court reached the conclusion that the net expected value of continuing to litigate and eventually putting the house on the market was less than $10,000. The judge did not make any findings as to the amount that the estate could expect to gain from a favorable judgment or a favorable sale of the property. She did not attempt to assign probabilities to obtaining a favorable judgment or a favorable sale price. Nor did she estimate what the estate’s litigation expenses would be. Instead, she stated in conclusory fashion that she thought the risk of netting *911less than $10,000 through further litigation and an eventual a sale was great enough to make settling for $10,000 a wise choice. Tr. at 121-25. Such a conclusory statement is insufficient; more detailed findings are required. See American Reserve, 841 F.2d at 162-63. Moreover, the comments that the judge did make raise some red flags. One red flag is that the judge seemed to be under the impression that she could not withhold approval of the compromise unless Mohns (or someone) could “guarantee” that if the trustee rejected the compromise he would net more than $10,000 for the estate. Tr. at 124-25. However, this is not the correct standard to apply when evaluating a compromise. If it were, no compromise could ever be disapproved, since no one could ever guarantee that rejecting a settlement and litigating to judgment will net more for the estate. Instead, the correct standard is whether it is probable that rejecting the settlement and litigating to judgment will net more for the estate. See, e.g., Fori Wayne Telsat, 665 F.3d at 820-21. Another red flag is that the judge seemed to defer to the trustee’s “business judgment.” Tr. at 94, 125. As I have already indicated, a bankruptcy judge may not simply accept the trustee’s word that a settlement is reasonable. American Reserve, 841 F.2d at 162.
If it were clear from the evidence presented at the hearing that the trustee could not expect to net more than $10,000 if he continued to litigate and if he eventually put the house on the market, then I could affirm the bankruptcy judge’s decision even though I cannot identify her precise reasons for finding the settlement adequate. See American Reserve, 841 F.2d at 162-63. However, it is not at all clear that the net expected value of continuing to litigate and putting the house on the market is less than $10,000. At the hearing, the bankruptcy court heard testimony from three real-estate brokers and also from other witnesses about the expected sale price of the house. All of the witnesses agreed that until the house was put on the market, no one would know for sure how much it would sell for or how long it would take to sell. However, the brokers opined about the price the house likely would sell for and about how long it would likely take to sell it. The low-end estimate was $430,000, Tr. at 5, and the high-end estimate was $460,000, Tr. at 51. The brokers seemed to agree that the time between listing the house and closing would be about three or four months. Tr. at 13-14, 49. If the house were sold, the estate would have to pay off the two mortgages on the house (which added up to $359,544, Tr. at 16), pay a commission to the broker (between 4.4% and 6% of the sale price, Tr. at 11, 51), and pay fees for transferring the title (three-tenths of a percent of the sale price, Tr. at 6) and for obtaining title insurance (half a percent, Tr. at 6). During the time that the house was on the market, the debtors would be expected to live in the house, continue making payments on the mortgages and for taxes and insurance, and cooperate with efforts to sell the house. The parties’ agreed that the debtors would be entitled to be reimbursed for doing these things, and the bankruptcy court seemed to conclude that, if it took three months to sell the house, the debtors would be entitled to reimbursement in the amount of $9,000 (although some of this would be principal on the mortgages, which the estate would recoup in the form of a lower mortgage pay-off at the time of sale). Tr. at 111.
Based on the above evidence, it seems that if the trustee put the house on the market, the likely outcome would be a gain to the estate of between $32,216 and $67,536. I calculate the low end of this range as follows. The lowest likely sale price is $430,000. From this, $359,544 would have to be used to pay off the *912mortgages. The brokerage commission would be 6% at most, and 6% of $430,000 is $25,800. The fees for transferring title and obtaining title insurance would be $1,290 and $2,150, respectively. Assuming that it will take three or four months to sell the house and that the estate will have to reimburse the debtors for their efforts during that time, another $9,000 would be subtracted from the sale price. Hence, the gain to the estate would be $32,216. As for the high end of the range, I use the highest likely sale price, which is $460,000. Again, I deduct $359,544 for the mortgages. Here, I use the 4.4% brokerage commission, and 4.4% of $460,000 is $20,240. Fees for transferring title and obtaining title insurance would be $1,380 and $2,300, respectively. Again, I assume that it would take three or four months to sell the house and that the estate would reimburse the debtors $9,000. Thus, the gain to the estate would be $67,536.
In estimating the expected gain to the estate, I also have to consider the probability that the trustee would lose any of the legal issues the debtors intend to raise. However, to the extent the record permits any reliable estimate of this probability, the probability would be quite low. Neither the trustee nor the debtors developed any cogent legal argument during the hearing which suggested that the debtors would be entitled to claim the property as exempt against the estate or that they would be entitled to reimbursement for past tax and insurance payments. Moreover, the bankruptcy court seemed to conclude that the trustee was almost certain to win any of the legal issues raised by the debtors. Tr. at 75-77. Thus, based on the record developed at the hearing, it appears that the probability of the trustee’s winning all of the legal issues raised by the debtors is quite high — let’s say 90%. Multiplying this probability by the expected gain from selling the house ($32,216 to $67,536) gives us an expected-gain range of $28,994 to $60,782. Obviously, this range is well above $10,000.
The final consideration is the trustee’s litigation costs, which he would incur in litigating the remaining legal issues surrounding the estate’s interest in the house. However, at the hearing, the trustee made no effort to quantify these costs. He did not say what his hourly rate was or estimate how many hours it would take to litigate the remaining issues or otherwise provide a basis for quantifying the cost of further litigation. Thus, although it is likely that the trustee’s litigation costs would not be zero, it is impossible to conclude based on the evidence presented at the hearing that they would be in the range of $19,000 to $50,000, which is the range they would need to be in for the value of the settlement to exceed the net expected gain of further litigation.
I also note that the bankruptcy judge expressed some concern over the fact that neither the debtors nor the trustee would sign a condition report for the property and that, because the property had a history of flooding in the basement, this might result in a lower sale price. Tr. at 84. However, there was no testimony that quantified the likely effect of the lack of a condition report and the history of flooding. One broker testified that the history of flooding might put downward pressure on the sale price, but he did not testify that it was likely to result in a sale price lower than $430,000. Tr. at 65-66. In fact, this broker thought the most likely sale price was $440,000. Tr. at 66. Thus, the record does not support the conclusion that the lack of a condition report and the history of flooding would have brought the net expected value of further litigation below $10,000.
In sum, I cannot say that the record supports the bankruptcy judge’s decision. Before I could affirm her decision, I would *913need her to explain how she arrived at the conclusion that the net expected value of continuing to litigate and selling the house was less than $10,000. Accordingly, I will vacate the order approving the compromise and remand so that she can make more detailed findings. On remand, the bankruptcy judge may conduct further proceedings and take further testimony as she deems necessary to make her findings. See American Reserve, 841 F.2d at 168.
For the reasons stated, IT IS ORDERED that the order of the bankruptcy court is VACATED and that this matter is REMANDED to the bankruptcy court for further proceedings consistent with this opinion.
. Wis. Stat. § 815.18(12) provides:
No property otherwise exempt may be claimed as exempt in any proceeding brought by any person to recover the whole or part of the purchase price of the property or against the claim or interest of a holder of a security interest, land contract, condominium or homeowners association assessment or maintenance lien or both, mortgage or any consensual or statutory lien.
. Mohns suggests that the standards governing a trustee’s decision to abandon property of the estate under 11 U.S.C. § 554 might apply to this case, but the trustee has not argued that the bankruptcy court's decision could be upheld under those standards. I also note that an argument could be made that the trustee’s decision to accept $10,000 for the estate’s interest in the house was a decision to sell estate property or an attempt to reduce estate property to money and therefore should be reviewed under either 11 U.S.C. § 363(b) or 11 U.S.C. § 704(a). However, no party has made such an argument.