What 1,873 Investment-Property Loan Decisions Reveal About Financing Readiness | PAIUL™ Research
PAIUL™ Lending Decisions Research · Volume 1 · Data source: AAA Lendings

What 1,873 Investment-Property Loan Decisions Reveal About Financing Readiness

An analysis of AAA Lendings’ loan decisions, June 2024 – July 2026

Credit score and leverage are the first numbers most borrowers and brokers check. In 1,873 investment-property loan decisions made by AAA Lendings, approved and declined files looked much the same on both. Most declined files had strong credit, and loans outside the lender’s written program limits were declined about 4.6 times as often. This page sets out what the data shows, how it was analyzed and what it does not show.

Published October 28, 2026 · Aggregate figures only

1,873investment-property loan decisions
1,733approved or funded
140declined
$756M+total loan amount
45states

Key findings

Four findings, published with AAA Lendings’ written permission. All figures are aggregates, and none is based on fewer than 20 loans.

aThe dataset

1,873 decisions, more than $756 million, 45 states

1,873investment-property loans with a lender decision
Loan decisions in the datasetStacked bar of 1,873 investment-property loan decisions: 1,733 approved or funded and 140 declined.1,733 approved or funded140 declinedApproved or funded: 1,733 loansDeclined: 140 loansApproved or fundedDeclined

1,733 were approved or funded and 140 were declined. The loans were submitted between June 2024 and July 2026, span 45 states and total more than $756 million in loan amount.

Excluded: 1,478 files whose status showed neither an approval nor a denial (cancelled, withdrawn or still active), and all owner-occupied loans.

View as table
DecisionLoans
Approved or funded1,733
Declined140
Total1,873
bCredit and leverage

Approved and declined files looked alike on credit and leverage

765 vs 751average credit score, approved vs declined
Credit score and loan-to-value, approved vs declinedAverage credit score 765 approved vs 751 declined; median 770 vs 754; average loan-to-value 60.8% vs 63.4%.Approved or fundedDeclinedCredit score (300–850 scale)AverageAverage credit score, approved or funded: 765765Average credit score, declined: 751751MedianMedian credit score, approved or funded: 770770Median credit score, declined: 754754300500700850Loan-to-value (0–100% scale)AverageAverage loan-to-value, approved or funded: 60.8%60.8%Average loan-to-value, declined: 63.4%63.4%0%50%100%

The average credit score was 765 for approved loans and 751 for declined loans, and the medians were 770 and 754. Average loan-to-value (LTV) was 60.8% for approved loans and 63.4% for declined loans.

Valid credit scores only: six records with a credit score of 0 or 1 were removed.

View as table
MeasureApproved or fundedDeclined
Average credit score765751
Median credit score770754
Average loan-to-value60.8%63.4%
cDeclined files

Most declined files had strong credit

76%of declined files had a credit score of 720 or higher
Share of declined files with strong credit76% of declined files had a credit score of 720 or higher; 50% had 720+ credit and 70% LTV or less.Credit score 720 or higher76%Credit score 720 or higher: 76% of declined files720+ credit and 70% LTV or less50%720+ credit and 70% LTV or less: 50% of declined files0%100% of declined files

Half of the declined files had both a credit score of 720 or higher and an LTV of 70% or less. The highest-credit decline was a score of 816 at 70% LTV.

View as table
Declined filesShare
Credit score 720 or higher76%
720+ credit and 70% LTV or less50%
Highest-credit decline816 at 70% LTV
dProgram limits

Loans outside the program limits were declined about 4.6 times as often

99.0%of approved loans fell inside the encoded program limits (1,261 of 1,274)
Program limits: approvals inside the limits and decline rates99.0% of approved DSCR-program loans fell inside the encoded limits. Decline rate 35% outside the limits (7 of 20) vs 7.6% inside (104 of 1,365).Approved loans inside the encoded limits99.0%Approved loans inside the limits: 99.0% (1,261 of 1,274)1,261 of 1,274 approved DSCR-program loansDecline rateOutside the limits (20 loans)Outside the limits (20 loans): 35% declined (7 of 20)35% (7 of 20)Inside the limits (1,365 loans)Inside the limits (1,365 loans): 7.6% declined (104 of 1,365)7.6% (104 of 1,365)0%25%50%

PAIUL encoded the lender’s DSCR program credit-score and loan-to-value limits and checked the 1,385 DSCR-program loans that had valid values for both. The 20 loans outside the limits were declined 35% of the time (7 of 20), compared with 7.6% inside the limits (104 of 1,365): about 4.6 times as often (95% CI 2.5–8.6×; Fisher two-sided p ≈ 0.0006).

The limits were encoded from the lender’s written program guidelines, not fitted to these loan outcomes. The outcomes were used only to check the encoding, so this is a guideline-consistency test, not a measure of PAIUL Funding Score accuracy.

Method note: Five records with invalid credit scores (0 or 1) were excluded because they cannot be tested; the earlier draft AAA reviewed showed 98.7% (1,261 of 1,278), 25 loans outside and 32%.

View as table
DSCR-program loans tested (1,385)LoansDeclinedDecline rate
Outside the encoded limits20735%
Inside the encoded limits1,3651047.6%

Approved loans inside the limits: 1,261 of 1,274 (99.0%). Ratio of decline rates about 4.6 (95% CI 2.5–8.6); Fisher two-sided p ≈ 0.0006.

What the data is

An analysis of AAA Lendings’ loan decisions: who decided, which loans are included, and what was left out.

SourceAAA Lendings’ loan-system export of historical loan files, from which PAIUL used the investment-property loans with a lender decision. Published with AAA Lendings’ written permission (signed October 8, 2026), which names AAA Lendings as the data source.
Scope1,873 investment-property loans with a lender decision. Only loans with occupancy recorded as Investment or Investor were kept.
Period and reachLoans submitted June 2024 – July 2026, in 45 states, with more than $756 million in total loan amount.
ApprovedThe lender approved or funded the loan: 1,733 loans.
DeclinedThe lender declined the application: 140 loans.
Excluded1,478 files whose status showed neither an approval nor a denial (cancelled, withdrawn or still active), and all owner-occupied loans.
PrivacyBorrower names and street addresses were removed on receipt. Only aggregate figures are published, and no figure is based on fewer than 20 loans.

Method

Each step is simple enough to repeat. The figures were re-checked against the de-identified file before this draft was prepared.

  1. LabelingEach loan was labeled from its status in the lender’s system: a status that included a denial counted as declined; otherwise one that included an approval, funding or purchase counted as approved or funded. Files whose status showed neither were left out, though some have approval or funding dates in other fields. Two checks point the same way: leaving out the 36 files that also show a cancellation, and labeling every file by its funding and denial dates.
  2. Data checksSix records with a credit score of 0 or 1 are data-entry errors, not real scores. They were removed before computing the credit and loan-to-value figures and before the program-limit check; they are still counted among the 1,873 decisions.
  3. Credit and leverageAverages and medians were computed separately for approved and declined loans. Shares of declined files use the declined loans with a valid credit score and, for the combined test, a recorded LTV.
  4. Program limitsPAIUL encoded the credit-score and loan-to-value limits in the lender’s written DSCR program guidelines and checked each of the 1,385 DSCR-program loans with valid values for both. The check covered the program’s minimum credit score, its maximum loan-to-value for each credit-score band and loan purpose, and the states where the program is offered. Program limits are set on combined loan-to-value, so this check uses combined LTV; the averages in finding b use LTV. The gap in decline rates is reported as a ratio with a 95% confidence interval (log method) and a two-sided Fisher’s exact test.
  5. Correction to finding dFive records with invalid credit scores (0 or 1) were excluded because they cannot be tested; the earlier draft AAA reviewed showed 98.7% (1,261 of 1,278), 25 loans outside and 32%.

What this shows, and what it does not

What it shows

  • Credit score and loan-to-value alone did not separate approved from declined files in this data.
  • Most declined files had strong credit (76% at 720 or higher), and half had both strong credit and moderate leverage.
  • PAIUL’s encoding of the lender’s written program limits was consistent with the lender’s approvals, and loans outside those limits were declined more often.

What it does not show

  • It does not measure the accuracy of the PAIUL Funding Score or of any prediction. PAIUL did not predict these outcomes; it analyzed decisions that had already been made.
  • It describes one lender’s decisions over one period. Other lenders’ programs, overlays and outcomes differ.
  • It does not explain every decision. The data records each lender decision, not the reason behind it.
  • The program-limit check uses the lender’s current written guidelines. Loans made under earlier versions or approved as exceptions can fall outside today’s limits.
  • The comparison outside the limits rests on 20 loans, so the ratio is approximate (95% CI 2.5–8.6×). It shows an association, not a cause.
  • It does not indicate or guarantee how any future application will be decided, or on what terms.
  • It is not an endorsement. Publication permission does not constitute an endorsement of PAIUL by AAA Lendings.

What borrowers and brokers can check before submitting

General checks before you submit. Every lender’s rules differ, so confirm the specific lender’s requirements.

The property

  • Is it in lendable condition, with no health-and-safety issues, open code violations or major repairs outstanding?
  • If it is vacant, or has been converted (for example, a change in the number of units), does the program allow that, and was the work permitted?
  • Will it appraise, in its current condition, at the value the loan amount needs?

The lender’s own rules

  • For a cash-out refinance, has the property been owned long enough to meet this lender’s seasoning rule?
  • Is the property listed for sale, or was it listed recently?
  • Ask for the lender’s overlays in writing before you submit.

The program limits

  • Do the credit score, loan-to-value and loan purpose fit the program’s limits, and is the program offered where the property is located?

The file

  • Are income, asset and reserve documents complete and current?
  • Is any recent late payment or credit event explained and documented?
  • Is the file moving? Answer conditions quickly: files can be declined for inactivity.

The cash flow

  • Check the DSCR, then keep going: coverage is one check among several.

How PAIUL uses this

The research feeds two pre-submission checks. Neither is a prediction, and neither replaces underwriting.

Decline-risk flags before submission

PAIUL flags common decline risks before a deal is submitted: property condition and appraisal risk, lender overlays such as seasoning and listing status, documentation gaps and file activity. Each flag explains its reason, so the deal can be fixed, or matched to a better-fitting lender, before it is submitted.

Program-limit checks

PAIUL encodes lenders’ written program limits and checks a deal against them before submission. In this research, PAIUL’s encoding of AAA Lendings’ limits was consistent with 99.0% of the lender’s approvals.

Funding readiness, not an accuracy claim

PAIUL’s Funding Score is a rules-based funding-readiness assessment. Its funding-probability scores have not yet been calibrated against tracked loan outcomes, and this research is not an accuracy claim. PAIUL provides intelligence, not certainty, and is designed to support the judgment of borrowers, brokers and underwriters, not to replace it.

How to cite

Full citationPAIUL™ Lending Decisions Research, Volume 1 (2026). What 1,873 Investment-Property Loan Decisions Reveal About Financing Readiness: An analysis of AAA Lendings’ loan decisions, June 2024 – July 2026. Up to 100 Million Corp. https://paiul.net/research

Short creditSource: PAIUL™ analysis of 1,873 investment-property loan decisions by AAA Lendings, June 2024 – July 2026.

Please cite these figures as aggregate findings from AAA Lendings’ loan decisions, not as an endorsement or an accuracy rate.

Frequently asked questions

Why do strong investor files still get declined?

The main data records each lender decision, not the reason behind it, so it cannot say why a given file was declined. It does show that strong files are declined: 76% of declined files had a credit score of 720 or higher. Credit and leverage set where a file can go, and lenders also apply their program limits, their own overlays and a review of the property and the file.

Does a high credit score guarantee approval?

No. In 1,873 investment-property loan decisions made by AAA Lendings from June 2024 to July 2026, 76% of declined files had a credit score of 720 or higher, and half had both 720+ credit and a loan-to-value of 70% or less. The highest-credit decline was 816 at 70% loan-to-value. The average credit score was 765 for approved loans and 751 for declined loans.

Do credit score and loan-to-value decide whether an investment-property loan is approved?

Not on their own in this data. The average credit score was 765 for approved loans and 751 for declined loans, the medians were 770 and 754, and average loan-to-value was 60.8% and 63.4%. Approved and declined files looked much the same on both measures. They still set a program’s limits, and files outside those limits were declined far more often.

Are loans outside a lender’s program limits declined more often?

In this data, yes. PAIUL encoded the credit-score and loan-to-value limits in AAA Lendings’ written DSCR program guidelines and checked 1,385 DSCR-program loans. The 20 loans outside the limits were declined 35% of the time (7 of 20), against 7.6% inside (104 of 1,365), about 4.6 times as often. The comparison rests on 20 loans and shows an association, not a cause.

Did PAIUL predict these outcomes?

No. This is a descriptive analysis of loan decisions AAA Lendings had already made. PAIUL encoded the lender’s written program limits and used the outcomes only to check that encoding: 99.0% of approved loans (1,261 of 1,274) fell inside the limits. That is a guideline-consistency test. It is not a measure of the accuracy of the PAIUL Funding Score or of any prediction.

What is a lender overlay?

A lender overlay is a rule a specific lender adds on top of a loan program’s basic limits, such as how long a property must be owned before a cash-out refinance (seasoning), whether a recently listed property is eligible, or how long a file can stay inactive. Overlays vary by lender, so a deal one lender declines may fit another.

Where does the data come from?

From AAA Lendings’ historical investment-property loan decisions, shared with PAIUL and published as aggregate figures with AAA Lendings’ written permission. Borrower names and street addresses were removed on receipt, and no figure is based on fewer than 20 loans. Publication permission does not constitute an endorsement of PAIUL by AAA Lendings.

What is PAIUL?

PAIUL™ is a Predictive Funding Intelligence™ platform being developed to analyze real estate transactions using AI, structured data, underwriting logic, market intelligence and capital-provider criteria. It is designed to help users understand funding readiness, identify potential risks, explain the factors affecting a transaction, identify opportunities for improvement, and explore potential capital paths. PAIUL is the intelligence layer between a real estate deal and the capital that funds it. A product of Up to 100 Million Corp. PAIUL is not a lender, a mortgage broker, or a financial advisor. It provides decision-support intelligence and does not guarantee approval, pricing, terms, or funding.

Know your deal before capital does.

Run a free PAIUL QuickScan to see the funding-readiness flags on your own deal, from program fit to property and lender-rule checks, before you submit it.

Run your free QuickScan app.paiul.net/paiul-quickscan.php · Decision-support intelligence, not a lending decision.

PAIUL™ · Predictive Funding Intelligence™ · A Product of Up to 100 Million Corp.

Research source: AAA Lendings. Aggregate historical data only. Publication permission does not constitute an endorsement of PAIUL by AAA Lendings. Historical results do not guarantee future loan approval or funding.

PAIUL is not a lender, a mortgage broker, or a financial advisor. It provides decision-support intelligence and does not guarantee approval, pricing, terms, or funding.

About · Contact · Privacy Policy · Terms of Service

© 2026 Up to 100 Million Corp. All Rights Reserved.