Delores A. Laughlin (Laughlin) appeals from an order of the Court of Common Pleas of Beaver County (trial court), dated July 12, 1996, which directed the County Bureau of Elections (Bureau) to certify Joseph Zupsic (Zupsic) for the office of District Justice in Judicial District 36-3-03 (District) pursuant to the Pennsylvania Election Code (Code).1 We affirm.
After the general election on November 2, 1993, the ballot boxes from Beaver County, including twenty-two from the District, were unlocked. The votes were tabulated by automatic scanning machines and then the ballots were placed back into their respective ballot boxes which were reloeked. The seal numbers were unrecorded for three days after the election until the Return and Write-In Boards opened the ballot boxes to audit the results and count the write-in votes. The Bureau concluded that Zupsic received 3,783 votes and Laughlin received 3,747 votes making Zupsic the winner in the race for District Justice in the District by a margin of thirty-six (36) votes.
Laughlin filed a petition to open fourteen (14) of the twenty-two (22) ballot boxes from the District. The Recount Board met and recounted the contents of the fourteen (14) ballot boxes by hand. As a result of this hand recount, the Recount Board concluded that Laughlin was the winner by a margin of forty-six (46) votes. During the recount, Zupsic challenged the validity of sixty-two (62) ballots, while Laughlin challenged thirteen (13) other ballots.
Zupsic then filed petitions to open the remaining eight (8) ballot boxes alleging substantial fraud or error in the counting of the votes or in the marking of the ballots. The votes were recounted resulting in Laughlin receiving a total of 3,794 votes and Zupsic receiving a total of 3,748 votes; both candidates received one additional vote over the hand recount. During this second hand recount, Zupsic challenged an additional seven (7) ballots and Laughlin challenged an additional five (5) ballots as a result of the two hand recounts. Laughlin and Zupsic collectively challenged a total of eighty-seven (87) ballots.
The Bureau then conducted a second machine count of the votes which were close to the results of the two hand recounts; Laugh-lin received 3,792 votes and Zupsic received 3,750 votes. Zupsic then filed a petition to contest the general election nunc pro tunc with the trial court. After hearings on the petition, the trial court ordered that the challenged election be set aside.
In its opinion and order, dated April 8, 1994, the trial court found that “[i]n all probability, a sufficient number of ballots were altered ... so as to change the outcome of the election” (Court of Common Pleas of Beaver County, April 8,1994, Slip Opinion at 4 — 5, Finding No. 11) [hereinafter Slip Op. I ]. On May 5, 1994, the trial court filed its second opinion with findings of fact and explaining why it set aside the election [hereinafter Slip Op. II ]. Laughlin filed an appeal from this order to the Commonwealth Court, and Zupsic filed an appeal from the same order to our Supreme Court. By order dated May 2, 1994, the Supreme Court noted probable jurisdiction, and the Commonwealth Court transferred Laughlin’s appeal to the Supreme Court on May 4,1994.
In the instant case, our Supreme Court reversed the trial court’s order to set aside the election and concluded that “there was more than competent evidence to support the trial court’s finding of tampering,” which rendered those findings binding on appeal. In re Petition to Contest General Election for District Justice in Judicial District 36-8-03 Nunc Pro Tunc (Election for District Justice), 543 Pa. 216, 232, 670 A.2d 629, 637 (1996). The Court, however, also concluded that the trial court failed to make specific findings as to the degree of tampering or as *479to which of the ballots were altered. Id. at 227, 670 A.2d at 634. Consequently, the Supreme Court remanded the case to the trial court to make findings regarding the degree of tampering, to identify the ballots that were altered, and to explain its conclusions.
On remand, the trial court denied Laugh-lin’s request to admit additional evidence and to review ballots that did not belong to the group of eighty-seven (87) ballots which were challenged during the recount. In compliance with the Supreme Court’s order to designate which ballots were altered and how those ballots were altered, the trial court made specific supplemental findings concerning the altered ballots. (Court of Common Pleas of Beaver County, July 12, 1996, Slip Opinion at 10-19) [hereinafter Slip Op. Ill ]. As a result, the trial court concluded that Zupsic received more votes than Laughlin and ordered the Bureau to certify him as District Justice for the District.
Laughlin raises the following issues on appeal: (1) whether the trial court erred by refusing Laughlin’s request to consider additional evidence where the trial court was previously unable to determine which ballots were subject to tampering; (2) whether the trial court erroneously gave weight to inconsistent markings on the ballots; (3) whether the trial court stretched the value of inconsistent markings on the ballots; and (4) whether the trial court erroneously analyzed the ballots with a presumption of fraud based on the machine results. We will review each of these issues in seriatim,.2
I
Laughlin first argues that the trial court erred when it refused to accept additional evidence during the proceedings on remand from the Supreme Court. Such evidence, Laughlin contends, would have helped the trial court determine which ballots were subject to tampering because the trial court was unable to make that precise determination two years prior on the evidence available. Laughlin emphasizes that there is a manifest discrepancy between the trial court’s findings in Slip Op. II and Slip Op. III. The trial court in Slip Op. II indicates that a tamperer entered five (S) ballot boxes, (Slip Op. II at 2), and in Slip Op. Ill indicates that a tamperer entered seven (7) ballot boxes based on the same record. (Slip Op. Ill at 5). Laughlin argues that this discrepancy is the result of uncertainty and asserts that we should order the trial court to examine the unchallenged ballots in the districts in which it concluded there was tampering to provide a clear and convincing basis for determining whether there were alterations and whether other ballots were subject to tampering. We disagree.
The Supreme Court stated that the trial court “found ‘overwhelming’ evidence that more than five ballots were altered, noting that forty-five (45) of the eighty-seven (87) contested ballots had marks for Laughlin that were ‘substantially inconsistent’ with most other marks on the ballot. However, the court did not identify which forty-five ballots these were.” Id. at 227, 670 A.2d at 634. A reasonable reading of this language yields the conclusion that the Supreme Court was only concerned with directing the trial court to make specific findings as to which forty-five (45) ballots of the eighty-seven (87) challenged ballots were altered and why it deemed those forty-five (45) ballots to be altered and why the remaining forty-two (42) were unaltered. The only ballots that were before the trial court, or the Supreme Court, were the eighty-seven (87) contested ballots.
The Supreme Court’s remand order was for a limited purpose. The Supreme Court impliedly found that additional evidence was unnecessary when it stated that “there was more than competent evidence to support the trial court’s finding of tampering,” Id. at 232, 670 A2d at 637, and the trial *480court “should have identified the ballots that it was unable to classify as either altered or unaltered, and ... explain the reasons for its conclusions.” Id. at 235, 670 A,2d at 638. Absent direct instruction by the Supreme Court to admit additional evidence on remand, the admission or exclusion- of evidence is within the sound discretion of the trial court, and we may reverse only upon a showing that the trial court clearly abused its discretion. Commonwealth v. Spiewak, 533 Pa. 1, 7 n. 4, 617 A.2d 696, 699 n. 4 (1992).
II
Candidate Laughlin next argues that the trial court used a “bootstrapping” argument to justify its decision that forty-five (45) ballots were altered. Laughlin maintains that the only evidence that ballots were altered is that there were some extraneous graphite on the ballots. Laughlin contends that there is no reason to believe that, because a graphite mark for one candidate or on one spot of the ballot is different than another, the mark was made by a person other than the voter. Laughlin argues that independent clear and convincing evidence of a direct or of a substantial circumstantial nature must be adduced with respect to each and every ballot before fraud can be found for a given individual ballot. Laughlin contends that such evidence only existed with respect to five (5) ballots because the voters on those ballots gave direct clear and convincing testimony about their votes and the presence of fraud. We disagree.
Initially, we note that, although Laughlin claims that clear and convincing evidence existed for the alteration of only five (5) ballots, Laughlin previously conceded that fifteen (15) ballots were altered. Laughlin made these concessions in her brief, filed in the trial court on March 18, 1996, and orally represented to the trial court that she conceded the alteration of fifteen (15) ballots. Although Laughlin disingenuously argues here on appeal for the first time that this concession was for argument purposes only, our decision in In re Nomination Petition of Shuli (Petition of Shuli), 105 Pa.Cmwlth.462, 525 A.2d 6, 8 (1987), ajfd 514 Pa. 471, 526 A.2d 300 (1987), persuades us to deem Laughlin’s concession as an admission even though it was made in Laughlin’s brief and at oral argument rather than in a pleading. In Petition of Shuli, we stated the following:
It is true that in trials, especially in jury trials, it is necessary that the party who attempts to provide a fact crucial to his case by an admission in a pleading must do so by reading the averment and the admission into the record. We are of the opinion, however, that the same rule does not hold true in an election hearing where the pleadings are a necessary part of the evi-dentiary record in order to enable the hearing judge to determine what is at issue. This is especially so in election cases such as this one where there are many factual allegations for the judge to adjudicate. There is no necessity for counsel to call to the court’s attention what is already before it by virtue of the pleadings. We think it would be particularly unfair in such a setting for a party to disavow that which it has admitted in a pleading.
Id. (citations omitted).3 We, therefore, do not recognize Laughlin’s argument that only five (5) ballots were subject to tampering, and we will uphold the trial court’s finding that “[cjandidate Laughlin has conceded that fifteen ballots were fraudulently altered by unlawful tampering.” (Slip Op. Ill at 8-9).
After the Supreme Court concluded that there was substantial evidence to support the trial court’s findings that forty-five (45) ballots were fraudulently altered, it remanded the instant matter to the trial court and instructed the trial court to make specific findings regarding the degree of tampering. The Supreme Court cited In re Opening of Ballot Box of Greenwood Township, East, Election District, Columbia County (Greenwood), 344 Pa. 350, 25 A.2d 330 (1942), for the proposition that, on remand, the trial *481001111; had a duty to award the ballots to the candidate for whom the elector voted if such a determination could be made by clear and convincing evidence. Election for District Justice, 643 Pa. at 236, 670 A2d at 638-39.
In Greenwood, however, our Supreme Court observed that the trial court correctly made findings of:
fraudulent erasures and alterations, which findings [were] based not alone upon the testimony of the handwriting experts, but as well upon evidence indicating a probability of improper access to the boxes in question after the election officials had completed their computations, the physical condition of the boxes when produced, the testimony of the election officials, and the court’s own inspection of the ballots.
Id, at 354, 25 A2d at 332-33. With respect to this evidence of fraudulent tampering of ballot boxes and their contents after the official canvass of the votes, the Supreme Court stated that “[u]nder such circumstances, the [trial] court was fully justified in directing that the fraudulently altered ballots be counted for the ... candidate for whom the evidence shows they must necessarily have been counted by the election officials.” Id. at 355, 25 A2d at 333.
In relying on Greenwood, our Supreme Court, in the case sub judice, gave the following instruction to the trial court: “far from striking the altered ballots, the court of common pleas has the duty to enfranchise legitimate voters by awarding those ballots to the intended recipient if that recipient can be determined by clear and convincing evidence.” Election for District Justice, 543 Pa. at 236, 670 A2d at 639. Based on the reading of both Election for District Justice and Greenwood, the Supreme Court has clearly ruled that, upon a showing of clear and convincing evidence, fraudulently altered ballots must be counted in favor of the elector’s intended recipient, and that clear and convincing evidence of fraudulent tampering can be established by an accumulation of circumstantial evidence that defines a pattern that is consistent with tampering. Therefore, Laughlin’s argument, i.e., direct independent evidence of tampering for each and every ballot deemed altered is the only competent evidence that rises to the level of clear and convincing, is wholly untenable.
The same type of evidence that the Supreme Court found sufficient to establish that individual ballots were fraudulently altered in Greenwood, exists in the instant case. The overwhelming weight of the evidence supports the trial court’s conclusion that there was clear and convincing evidence that, of the eighty-seven individual ballots examined, forty-five (45) ballots were fraudulently altered. Of those forty-five (45) ballots, candidate Laughlin conceded that fifteen (15) ballots were altered from a vote for Zupsic to a vote for Laughlin. The trial court compared the remaining thirty (30) altered ballots to the fifteen (15) ballots that Laughlin conceded were altered. Collectively, these fifteen (15) ballots provided a standard or model with which other ballots could be compared. ■
In the process of examining the ballots, the trial court used the numbers assigned to each of the ballots by the Recount Board. The ballots challenged by Zupsic were la-belled as the “A” series of ballots, and the ballots challenged by Laughlin were labelled as the “B” series. Of the eighty-seven (87) challenged ballots, the trial court identified the forty-five (45) that were fraudulently altered. Of the forty-five (45) altered ballots, the trial court identified the fifteen (15) that Laughlin conceded were fraudulently altered and explained in detail its findings with respect to each of the remaining thirty (30) ballots that Laughlin did not admit were altered.
The trial court rendered specific findings regarding ballot numbers: A-l, A-5, A-6, A-7, A-8, A-9, A-10, A-ll, A-12, A13, A-22, A-27, A-28, A~29, A-30, A-31, A-32, A-34, A-37, A-38, A42, A-46, A-47, A-49, A-50, A-67, A-69, B-2, B-3, B^t. (Slip Op. Ill at 10-19) and found that the marks in the Laughlin oval are distinctly different from the marks in the other ovals in either size, shape, texture, shade, darkness, or color. These distinctions evidenced that someone other than the voter made the marks for Laughlin and, in most circumstances, evidenced the use of a *482different -writing instrument than that which was employed by the voter.
Additionally, the trial court compared each of the aforementioned ballots to the ballots of which Laughlin conceded an alteration. In this manner, the trial court found that, in many instances, the fraudulent alteration conformed to a pattern which was established by the fifteen (15) ballots that were admittedly altered in favor of Laughlin. Another contributing factor to the trial court’s findings was that many of the altered ballots were from ballot boxes with respect to which a pattern of fraudulent tampering had been established or in which concededly fraudulent ballots existed. Also, on many of the fraudulently altered ballots, the trial court found that, even though the oval for a straight democratic vote was filled in by the voter, which would include Zupsic as the democratic candidate for District Justice, there was a stray mark in the Laughlin oval, thereby contradicting the voter’s intent for a straight democratic vote. Some of the fraudulently altered ballots were voided and subtracted from Laughlin’s total.
The trial court also individually examined the remaining forty-two (42) challenged ballots. With respect to each of those ballots, the trial court made specific findings concerning the distinguishing marks on those ballots but concluded that either those marks did not conform to the pattern of fraudulent alterations or they were not otherwise remarkable enough to conclude that the voter did not make those marks. A review of the record as a whole reveals that the trial court’s findings are supported by substantial competent evidence.
Additionally, these findings supplemented the trial court’s original findings that despite the accuracy of the vote counting machines, there were substantial differences in the vote totals for each candidate and substantial differences in the under-vote and over-vote totals between the machine count on election night and the second machine count which favored Laughlin. (Slip Op. II at 1-2). The trial court also found that the Board of Elections had not adequately secured the ballot boxes and someone gained undetected access to the ballot boxes in question and altered the ballots. (Slip Op. II at 2). The numbered seals on the boxes were not observed and recorded until after the Return Board completed its audit and the Write-In Board completed its tabulation of write-in ballots.
This accumulation of competent evidence is the same type of evidence that our Supreme Court recognized as clear and convincing evidence of fraudulent alteration -with respect to specific ballots. In Greenwood, 344 Pa. at 354, 25 A.2d at 332-33, where the testimony of a handwriting expert was not the sine qua non for establishing clear and convincing evidence of fraudulent alterations, fraud was established by “evidence indicating a probability of improper access to the ballot boxes after election officials completed their tabulation of the votes and the court’s inspection of the ballots.” Id. We therefore, hold that the evidence in the instant case constituted clear and convincing evidence that the forty-five (45) designated ballots were fraudulently altered in favor of Laughlin where the intent of the elector was to east a vote in favor of Zupsic.
Ill
Candidate Laughlin next argues that the trial court stretched the objective value of the claimed inconsistent markings on forty (40) of the forty-five (45) ballots and that the court only had five (5) ballots, where there was direct evidence from five witnesses, which met the standard of clear and convincing evidence of tampering. Therefore, argues Laughlin, the trial court should have found that these were the only ballots subject to tampering. Laughlin argues that the clear and convincing evidence standard is met only where there is direct testimonial evidence from the elector, who can identify his vote, or from a handwriting expert, who can demonstrate that the elector did not make the claimed inconsistent markings. We disagree.
Although this issue is adequately addressed in Part II of this opinion, it is important to reiterate here that Laughlin conceded that fifteen (15) ballots were fraudulently altered. There was direct testimonial evidence from the voters on only five (5) of *483those fifteen (15) ballots. Therefore, Laugh-lin conceded that direct testimonial evidence is not necessarily required, as the sine qua non, in terms of meeting the clear and convincing standard of fraudulent tampering. Additionally, the accumulation of evidence established a clear and convincing pattern consistent with tampering, and forty-five (45) individual ballots fit this pattern. Moreover, the type of cumulative evidence and analysis relied upon by the trial court in concluding that forty-five (45) ballots were fraudulently altered in favor of Laughlin were approved by our Supreme Court in Greenwood, 344 Pa. 350, 25 A2d 330, as discussed in Part II of this opinion.
IV
Laughlin also argues that, until substantial evidence is brought forth, all marks on the ballot must be presumed to be those of the voter and not the wrong-doer. Laughlin maintains that the trial court erroneously approached its analysis with a presumption of fraud instead of a presumption of enfranchisement. Candidate Laughlin also contends that there is no specific evidence that the Bureau or anyone else tampered with the ballots. Laughlin contends that the trial court’s concern about preserving the public perception of the accuracy of the vote scanning machines caused the trial court to misapply the standard of clear and convincing evidence necessary to overcome the presumption of enfranchisement.
Laughlin specifically argues that the difference between the first machine count, which resulted in Zupsic receiving more votes than Laughlin, and the second machine count, which was corroborated by the hand recount and resulted in Laughlin receiving a winning number of votes, was not the product of fraud. The vote counting machines could have made an error during the first machine count. Three representatives from the vote counting machine manufacturer testified that human error or differences in marking the ballot sheets can result in the counting machine not reading the ballot if the voter made his markings too light for the machine to read. Laughlin contends that the vast difference in the voter’s markings on their own ballots is what could have caused the difference between the first machine count, which favored Zupsic, and the second machine count, which favored Laugh-lin and not, as the trial court concluded, because there were differences in the over-votes and under-votes as a result of tampering which transpired during the interim.4 We disagree.
Although the difference in the results of the two machine counts may not, by itself, be dispositive on the issue of whether tampering ensued during, the time between the two machine counts, the fact remains that the trial court had substantial competent evidence that there was indeed tampering at this time. As previously stated, five electors testified that their ballots were altered. Laughlin conceded that ten (10) other ballots were fraudulently altered without such direct testimonial evidence. The record as a whole reveals that there is substantial competent evidence of a clear and convincing nature to overcome the presumption of enfranchisement that forty-five (45) of the eighty-seven (87) challenged ballots were altered. The difference in the machine counts was merely one factor that contributed to the trial court’s conclusion, and it was corroborated by the court’s own analysis of the individual ballots. When the court examined the eighty-seven (87) challenged ballots and compared the altered ballots against the fifteen (15) ballots which both parties agreed were altered, evidence of fraudulent tampering became manifest, and the machine count became less significant in the accumulation of evidence of fraud.
Additionally, although Laughlin attempts to explain away the difference in the two machine counts as attributable to the inability of the vote scanning machines to read light voters’ markings, the trial court’s conclusion, i.e., the difference between the two machine counts demonstrates that tampering occurred between machine counts, is sup*484ported by the fact that the second machine count was almost exactly the same as the hand recount. Therefore, the machines were counting the votes as the human eye would count the votes and any claimed errors as a result of inconsistencies in voters’ markings did not cause inaccurate machine count results. Given the overwhelming evidence of fraudulent alterations of ballots in the instant case, Laughlin’s argument that the trial court erred in concluding that fraudulent alterations occurred as a result of the difference in the machine counts is wholly untenable.
Accordingly, the order of the trial court is hereby affirmed.
ORDER
AND NOW, this 4th day of June, 1997, the order of the Court of Common Pleas of Beaver County, dated July 12, 1996, is affirmed.
. Act of June 3, 1937, P.L. 1333, as amended, 25 P.S. §§ 2600-3591.
. The instant matter is both an election recount case, filed pursuant to Section, 1701 of the Code, 25 P.S. § 3261, and an election contest case, filed pursuant to Section 1756 of the Code, 25 P.S. § 3456. Therefore, our scope of review is limited to an examination of the record to determine whether the trial court committed errors of law and whether its findings are supported by adequate evidence. In the matter of the Recount of Ballots of Albany Township, 392 Pa. 602, 606-OS, 141 A.2d 389, 391-92 (1958).
. Although Laughlin's concession was made both in her brief and by way of oral representation before the trial court and not in the pleadings. Petition of Shuli emphatically pronounced the extension of this rule to election cases. A party cannot, on appeal, disavow a position for which that party repeatedly argued before the trial court and upon which the trial court relied in rendering its decision.
. An "over-vote” occurs when the voter votes for more candidates for a given office than are to be elected to that office. An "under-vote" occurs when the voter votes for fewer candidates for a given office than are to be elected to that office.