BAILEY, J.
Robert challenges the trial court’s determination of his support obligations on two grounds. He argues first that the trial court should have used only one Child Support Worksheet to calculate his support obligations. He then argues that, in calculating his parenting time credit on the Child Support Worksheet, the court should have averaged the number of overnight visits per year between the two children.
The Guidelines anticipate that trial courts will use one Worksheet for all the children in a family. Robert takes issue with the trial court’s use of two Child Support Worksheets to determine the total amount of his child support obligation to be paid to Judy, noting that the use of two worksheets led to the trial court making two support calculations—one for each child—rather than using a single worksheet with a single support calculation, which inflated his basic child support obligation.
An examination of the Worksheets and calculation instructions in the Guidelines reveals the likely reason for the trial court’s separate calculations. The Child Support Worksheet combines parental income and then uses the Support Schedules to determine the weekly basic support obligation based on the total number of children. It then credits the non-custodial parent for the number of overnight stays the children have with that parent per year. This credit is applied against the non-custodial parent’s support obligation for all of the children. Thus, the Guidelines presume that each child will have the same number of overnight stays with the non-custodial parent.
The circumstances in the Blanfords’ case differ from the assumptions made by the Guidelines. The trial court found, after in camera interviews with the children, that M.B. would likely have no overnight parenting time with Robert, but that S.B. would have significant overnight parenting time with him—one worksheet presumes 182 days of overnight parenting time—and the court entered its order based on those findings. Given this difference in overnight parenting time between M.B. and S.B., the trial court could not have calculated Robert’s support obligations for both children using the Guidelines and Worksheet as they currently exist. See Hartley v. Hartley, 862 N.E.2d 274, 286 (Ind. Ct. App. 2007) (noting that “[t]he Child Support Guidelines offer no direction for calculating parenting time credit when a parent spends overnights with fewer than all of his children.”).
Given this difficulty, the trial court used separate Worksheets for each child. This led to calculations that treated each child as though he was an only child for the purposes of calculating Robert’s child support obligations. In challenging the trial court’s calculation of his basic support obligation, Robert directs our attention to the Commentary on the Indiana Child Support Guidelines. The Commentary notes that “intact households” spend more money per child “as family income increases” and that, assuming constant income levels, such spending drops “for each child as family size grows.” Child Supp. G. 1, Commentary. “[T]he amount of support required for two children is 1.5 times that required to support one child.” Child Supp. G. 4, Commentary. [Footnote omitted.]
We agree with Robert that, by using two worksheets in this manner, the trial court deviated from the Guidelines’ designated factor for calculating the basic support obligation for each additional child. The Worksheets used by the trial court have inflated the aggregate basic support obligation for M.B. and S.B. Under the Child Support Guidelines’ Support Schedules, the basic weekly support obligation for one child at Robert’s and Judy’s combined income level is $273.00; for two children, the basic weekly support obligation is $410.00. The trial court’s use of the separate worksheets doubled the underlying basic support obligation to an aggregate of $546.00—$273.00 for each of the two children—instead of fixing it at $410.00, which is 1.5 times $273.00, in keeping with the additional-child factor in the Guidelines and Support Schedules. This in turn inflated Robert’s support obligations for each child, since the rest of the support calculations were performed against that inflated basic support obligation figure.
Robert urges that the trial court should have used one Worksheet with the basic support obligation as expressed in the Support Schedules. He then goes on to argue that the trial court should have averaged the total overnight parenting time credit between the two children, allowing the calculation to be performed on one worksheet. Thus, in the example Robert posed in his brief, rather than receiving credit for 182 days of overnight credit against his obligations for S.B. and no overnight credit against his obligations for M.B., Robert would receive 91 days of overnight credit on one worksheet with one child support calculation accounting for two children.
This approach suffers from its own difficulties. In particular, the Parenting Time Worksheet used to determine the credit for overnight parenting time uses Table PT of the Guidelines to determine the multiplication factor used to calculate the amount of the credit given to the non-custodial parent. See Child Supp. G. 6, Commentary. Table PT includes a factor for duplicated expenses. By averaging the parenting time between the two children, Robert’s suggested 91 days of overnight parenting time credit might extend Robert too much or too little credit in calculating his support obligations.
Given this situation, the trial court could not have applied the Guidelines and used the Worksheets as written. The trial court nonetheless erred on two grounds when it ordered the child support using the separate Worksheets.
First, neither S.B.’s nor M.B.’s worksheet is correct because each child was treated as though he was an only child. The trial court used two Worksheets to determine the support obligations for the children, and the basic support obligations were calculated separately as if each were an only child. The use of a different method of calculating the basic support obligation is inherently a deviation from the Guidelines.
This deviation is itself the root of the second error. The Order deviates from the Guidelines without explaining its reasons for doing so. See Child Supp. R. 3; Child. Supp. G. 3(F), Commentary. Though the trial court made a finding that it did “not anticipat[e] that [M.B.] will choose to stay any overnights with the father,” the Order states in several places that its support obligation calculations come “Pursuant to the Indiana Child Support Guidelines,” even as the Order applies nonconforming basic support obligation calculations without further explanation. (Order 4.)
It may have been sufficient for the trial court to use the language from the Commentary to Guideline 3(F), “The court finds that the presumptive amount of support calculated under the Guidelines has been rebutted for the following reasons,” and then provide reasons for the deviation. Absent findings that explain the court’s reasoning, and given the unexplained approach to setting the basic support obligation, we cannot affirm the trial court because it has erred as a matter of law.
Given the errors in the trial court’s decision, we remand this issue to the trial court. Proper determination of the basic support obligation under the Schedules is crucial to the rest of the child support calculation. Thus, on remand, we direct the trial court to use one worksheet accounting for both children, rather than one worksheet for each child, and to use the Schedules’ basic support obligation for two children at the appropriate weekly income level for Robert and Judy. The trial court should then calculate the overnight parenting time credit using the basic support obligation for two children, and may adjust the number of days of overnight credit to reach what appears to be an appropriate result for setting Robert’s weekly support obligation. Because the Guidelines do not afford a basis on which to set the number of days of overnight credit, the trial court must explain the reasons for its use of the specific number of days of overnight credit in its order.
RILEY, J., and KIRSCH, J., concur.