About Frank — No Dealer Ties. No Brand Loyalty. Just the Data.

Frank is a used car expert and consumer advocate who has spent 20 years buying, inspecting, and evaluating used cars — so you don’t have to learn the hard way. This site exists because most used car advice is either AI-generated filler or dealer marketing dressed up as guidance.

Frank evaluates every model year using NHTSA complaint data, IIHS safety ratings, RepairPal repair costs, and KBB depreciation curves. He has personally bought or helped buy over 50 cars across 15 brands. His methodology, experience, and core beliefs are documented below.

Who Is Frank?

Frank has spent 20 years buying, inspecting, and reselling used cars across 15 brands and hundreds of model years. He is not a dealer. Not a mechanic. Not an automotive journalist.

Frank is an independent buyer and consumer advocate who helps real people avoid bad used car purchases.

Dealers profit from information asymmetry. They know which model years have transmission problems. They know the 2013 RAV4 has 847 NHTSA complaints while the 2019 has 127. They price accordingly — but they don’t tell you why.

Frank eliminates that gap. Every recommendation on this site comes with the data behind it: NHTSA complaint counts, IIHS crash ratings, RepairPal repair cost estimates, and KBB fair market values.

No dealer ties. No brand loyalty. Just the data and 20 years of buying used.

The site covers Toyota, Honda, Ford, Chevy, Nissan, Subaru, Mazda, Jeep, Hyundai, Kia, Dodge, Lexus, Acura, BMW, and GMC. Each brand has a dedicated hub page with model-by-model reliability breakdowns and specific years to avoid.

Why Does This Site Exist?

Most used car advice online is either AI-generated filler or dealer marketing dressed up as guidance. Generic articles repeat “check the CarFax” and “get a mechanic inspection” without telling you which specific model years have catastrophic engine failures or how many NHTSA complaints a car actually has.

NHTSA complaint counts vary 10x between model years of the same car. The 2013 Toyota RAV4 has 847 complaints. The 2019 RAV4 has 127. Same name. Same badge. Completely different reliability story. That kind of data changes a buying decision — but most sites never show it to you.

Frank built this site because every recommendation deserves a data trail. Every “avoid” label links back to NHTSA complaint counts, IIHS safety ratings, KBB depreciation data, and RepairPal repair costs. No gut feelings. No brand allegiance.

People don’t want “this car is generally reliable.” They want specific complaint counts and real repair cost ranges. That’s what Frank delivers.

How Does Frank Evaluate Used Cars?

Frank evaluates every used car model year using a data-first methodology that cross-references four independent sources before making any recommendation. The process always follows the same order: data first, then 20 years of personal experience, then verdict. Never the reverse.

What makes this different from other used car sites is the math. Frank calculates complaint density — complaints per 1,000 units sold — to normalize raw complaint counts by sales volume. He tracks repair cost vs purchase price ratios and year-over-year complaint trends. No other used car site does these calculations.

What Data Sources Does Frank Use?

Frank cross-references four primary data sources for every model year evaluation.

Secondary sources include Consumer Reports for reliability scores, Edmunds for pricing cross-checks, and J.D. Power for dependability rankings.

These are Tier 2 — reference material that supplements the primary data.

Reddit and car forums fall into Tier 3 — anecdotal owner experiences. Frank reads them but never bases a recommendation solely on forum posts. Anecdotal evidence confirms patterns in the data. It doesn’t replace the data.

What Makes Frank’s Approach Different from Other Sites?

Most used car sites tell you a car is “reliable” or “unreliable” without showing you the data behind that label. They say “avoid the 2013 RAV4” and move on. Frank says something different.

Frank says the 2013 RAV4 has 847 NHTSA complaints versus 127 for the 2019. The transmission accounts for 38% of those complaints. At $18,000 the risk is too high. At $8,000 with $2,000 budgeted for potential transmission work, the risk-adjusted math might still work.

That’s the difference. Frank calculates complaint density ratios, repair cost vs purchase price ratios, year-over-year complaint trends, and “at what price does it make sense” thresholds. No other site normalizes complaint counts by sales volume. No other site tells you the price where a “bad year” becomes a reasonable gamble.

“Avoid” doesn’t mean “never buy.” It means “know the risks and negotiate accordingly.” Every other site just says “avoid” and moves on. Frank shows you the math.

What Is Frank’s Experience with Used Cars?

Frank has personally bought, inspected, or helped friends and family purchase over 50 used cars across 20 years. The experience started at auctions in the early 2000s. Between 2005 and 2010, Frank helped 30 or more friends and family members buy used.

In 2010, Frank discovered the NHTSA complaint database — and the methodology shifted from gut instincts to data. Systematic tracking with spreadsheets started in 2018. frankusedcar.com launched in 2024.

A few experiences shaped how Frank evaluates cars today:

The 2007 Camry oil burn. Helped a neighbor buy a 2007 Toyota Camry with 85K miles — “it’s a Toyota, it’ll be fine.” Three months later, burning a quart of oil every 1,000 miles. Blue haze at every red light.

Even Toyotas have bad years. The 2AZ-FE piston ring defect taught Frank that brand reputation alone means nothing without year-specific data.

The $800 Nissan CVT surprise. Found a 2014 Nissan Altima at auction — $6,200, clean body, 72K miles. CVT started shuddering at 78K. Dealer quoted $3,400 for replacement. That $6,200 “deal” became a $9,600 mistake. That grinding, rubber-band-stretching feel when the CVT slips at highway speed — Frank will never forget it.

The PPI that saved $4,000. A client was about to buy a 2016 Jeep Grand Cherokee — beautiful, low miles, great price. Frank insisted on an independent mechanic inspection.

It found an oil leak from the rear main seal, warped rotors, and a leaking transfer case. The dark stain spreading across the concrete under the Jeep told the whole story. Never trust a clean CarFax alone.

The Tacoma that wouldn’t die. Bought a 2006 Toyota Tacoma with 152,000 miles for $9,800. Everyone said it was crazy. Changed the timing belt, flushed the coolant, new brakes. That Tacoma hit 240K before Frank sold it — for $7,500. The satisfying thunk of a solid Toyota door close at 200K miles. Still tight. No rattles.

The Theta II engine seizure. Neighbor’s 2015 Hyundai Sonata engine seized at 67K miles — out of warranty by 2K miles. The knocking sound that started softly and grew louder every day. Metal-on-metal, like a slow countdown. That engine is the biggest recall in US automotive history for a reason.

The Mazda that changed everything. Frank had never recommended Mazda until a client insisted on a 2017 CX-5. Skyactiv engine, traditional automatic, stellar crash ratings. The precision of that steering — firm, responsive, nothing like the vague floating feel of most crossovers. Mazda is the most underrated used car brand. The Mazda CX-5 years to avoid guide explains why Frank now considers it a top pick.

Frank maintains a 200K-mile club — every car he has personally seen cross 200,000 miles. Toyota: 14 entries. Honda: 9. Everyone else: 4 combined. The Toyota RAV4 years to avoid guide shows which specific model years earn that kind of longevity.

What Does Frank Believe?

Frank’s recommendations follow five core principles that never bend to brand loyalty or advertiser pressure.

1. Data over opinions. Every “avoid” label has a data trail. NHTSA complaint counts, IIHS ratings, RepairPal costs. If Frank can’t show the numbers, Frank doesn’t make the recommendation.

2. No brand loyalty. Toyota has bad years. Honda has bad years. The data decides, not the badge. Frank has criticized Toyota (2007-2009 Camry oil consumption) and praised Nissan (Frontier with manual transmission) in the same article.

3. Every car has a price where it makes sense. A “year to avoid” at $15,000 might be a reasonable buy at $5,000 if the buyer knows the risks. Context is everything. Frank always provides the price threshold where a “bad year” becomes worth considering.

4. Transparency about uncertainty. When data is limited — low complaint counts, no IIHS test, discontinued model — Frank says so. “Not enough data to recommend or warn” is a valid position. Frank never fills gaps with guesses.

5. Consumer protection above all. Frank exists to prevent bad purchases. Dealers profit from information asymmetry. Frank eliminates it. Every article serves one purpose: help the buyer make a smarter decision.

Frank’s specific stances on CVT transmissions, problem engines, safety ratings, and brand-level reliability are documented in every article on this site.

How to Use This Site

frankusedcar.com is organized around one question: which model years should you buy, and which should you avoid?

The site covers 15 brands, each with a dedicated hub page. Hub pages give you the full reliability landscape for a brand — best models, worst model years, common engine problems, and Frank’s brand-level verdict.

From each hub, you can drill into model-specific “years to avoid” guides with NHTSA data per model year, or “best and worst years” pages that combine both recommendations and warnings.

Shopping for a specific brand? Start with the brand hub:

• Used Toyota buying guide
• Used Honda buying guide
• Used Mazda buying guide

Know the model but want year-by-year guidance? Go directly to the model’s “years to avoid” page from the brand hub.

First-time used car buyer? Start with the used car buying guide — Frank’s complete process for finding a reliable car without getting burned.

Every page on this site links to the data sources behind its recommendations. Every “avoid” label shows the complaint count. Every “solid pick” shows the repair costs. The numbers don’t lie.