How to buy a car using data part III: The cheap-ass car version

Guest post by Dev Nambi

Remember when Dev introduced us to buying a car using data part I and II? He’s back with part III, and this time it’s all about buying cheap-ass cars.

Screen Shot 2013-05-11 at 11.43.27 AMMy sister called me from her trusted car repair shop. Her ’95 Ford Escort had been troublesome for months, and was now truly dead. This means the car’s demise had left my sister, her husband, and their two-year-old without transport. Worse, they had a 20+ mile commute to work, didn’t have time off, and would be fired if they couldn’t get to work at a moment’s notice.

I had less than 72 hours to find a replacement car. So I broke out my data nerd skills and got to work.

My first step was to find out about what features my sister cared about the most in a car:

  • Space for a child seat and groceries
  • Reliable
  • Less than $5,000
  • Low operating cost: the cost to run the car each year, repairs, insurance, and gas.

I had researched how to buy a car using data. Sleuthing on Craigslist and AutoTrader revealed that vehicles this cheap are 9+ years old and have 100K+ miles. Many seemed of dubious reliability.

There was no way to know the reliability of a car from its description. That suggested there were both ripoffs and deals in the listings. This was an information asymmetry problem. The seller had perfect knowledge and the buyer had little.

Where to start: Make and Model

Internet sleuthing led to FleetBusiness, which reported how long different brands last before they die (are junked). I also found TrueDelta, which had reports from car owners about repairs, mileage and cost. Here’s what I found in the FleetBusiness data:


The car brand that died off quickest was Suzuki. The brands that died off slowest were Toyota, Honda and Subaru. The die-off rate was not a straight line… it was an S-shape, like the continuous normal distribution. Looking at the scrap rate per year, I saw a roughly normal distribution:


Most cars died after 10-20 years. The cars I was looking at were the worst possible age. The odds were good the car I purchased would die within the next 5-10 years. However, the cars I was looking at were 10-13 years old. Any cars that died before then weren’t for sale so I could exclude that percentage.


The most reliable brands to buy at 10 years’ age were Honda and Toyota, followed by Chrysler. I picked 6 reliable models:

  • Toyota Corolla
  • Honda Fit
  • Honda Civic
  • Toyota Camry
  • Hyundai Sonata
  • Hyundai Elantra

I added two Hyundai models, the Elantra and Sonata, because I heard their later-generation models were well-built. This was not data-driven and foolish.

What’s on sale?

I collected 117 car listings. My goal was to have enough listings that there were a few good deals.

The biggest cost of owning a car is depreciation: the difference between your purchase price and what you sell it for. Buying a cheap car that lasts a long time seemed the best way to reduce that cost.

I didn’t care about car mileage or age. I wanted a car with as many miles remaining as possible. I needed to find out how long each car model would last. If a car has 125K miles already there’s a big difference between a car that lasts 200K miles vs. 150K miles. The 200K car will get you 3X farther.

I guessed mileage was roughly 5X more important than age. Maintenance costs would increase exponentially as mileage and age increased. I puzzled out an equation to compute a “quality score for each car.

Score = fnNormalize ( Age^1.2 ) * 20% + fnNormalize ( Mileage^1.4 ) * 80%


The ratio of this score to the price is the “value score.” Higher value scores were better deals:


Roughly, better-quality cars were more expensive. However, there isn’t a straight line. There were ripoffs (in the upper left, with smaller dots) and potential deals (in the lower right, with larger dots).

Let’s go shopping!

Now I had a shopping list: the five cars with the highest value scores.

  • The first car had sold, in under an hour.
  • We went to see the #3 car at a nearby dealership. The test drive was illuminating: the car was junk. The brakes barely worked, the fan belt made a whistling sound, and the lowest gear didn’t work… in an automatic. We left in a hurry.
  • For car #2, I wasn’t hopeful after that first test drive, but was surprised when this car handled well. The engine, brakes, and steering all worked perfectly. A roller-coaster route through West Seattle found no issues. We made plans for my trusted mechanic to look over the car.

Open your bonnet and say “vrooom”

The car and seller were legitimate. A check of the vehicle’s VIN number found no thefts or accidents.

The mechanic confirmed car #2 was in good working condition except the it burned some oil when accelerating. Some hasty Internet searches suggested this was not unusual for old Toyota Corollas and didn’t mean the engine was toast. We quickly bought the car. Success!


  • Work quickly. Good deals sell fast, in a day or two.
  • Hundreds of cars in Seattle were listed on Craigslist and AutoTrader each day.
  • The dealer car we tried was worse and more expensive than the private seller. A NADA report shows that used-car dealerships’ profit margins were 12% for used cars. A $5,000 dealer car would cost $4465 on by a private seller.

Comments on How to buy a car using data part III: The cheap-ass car version

  1. How did you calculate how many miles a car had left, since that was a big part of your value analysis? All of your previous figures just looked at age.

    • Hi Rowany,

      I couldn’t find the find the mileage-at-death statistics for different models/brands of car, so I assumed that the mileage was age * 12.5K per year, which is the average number of miles driven per year ( That’s a risky assumption, because it assumes each make/model of car is driven the same amount, but it’s the best I could do with the data.

      So, the formula would be (Expected Lifetime of Car In Years * 12,500) – Current Mileage. I’d love to find a better formula, though.

  2. This maybe isn’t part of your study or experience, but I bought a car with a diesel engine simply because it is expected to last for much longer than a gasoline engine. A VW with a diesel engine could have 250k miles and still be in great condition and at least another 100k before it’s needs serious repairs. People don’t really like them because they are noisy etc, but I get like 46 mpg, and they last and last….

    • Hi Jessi,

      I was curious about that, too. Unfortunately I couldn’t get data about VW that separated gas from diesel engines. Also, I assume a car can die from many different reasons (engine, transmission, electrical issues, rust, etc). I *really* wish I had this info, because I’m a big fan of diesel engines.

  3. Nice data!

    Also, thank you for adding in the snippet about taking it to your trusted mechanic to look it over. My hubby is a mechanic, and actually likes it when people take cars to his shop to look over prior to purchasing it. It’s saved some families from buying complete lemons.

    As a side note, when looking over the car check for these things: rust, paint variances, clean title and body damage. Remember, not all accidents are reported to Carfax. Unfortunately some people fail to report accidents to insurances, so it leaves the Carfax with a clean record, when it really shouldn’t be.

    If there’s a lot of rust located underneath the car, that could mean that the car was in a flooded area at one time, which comes with it’s own problems. Usually a lot of rust means a no go for me, namely because I don’t want to deal with those problems. To check this, lay down on the ground and look at the underbelly of the car. A little dirty, but worth it.

    When looking for body damage, look around the car for large bumps, cracks or variances in paint. Most of the time, the large bumps or cracks can indicate what type of collision it was in (bumps/cracks in the front = front end collision = bad). In some cars, the placement of these large bumps can actually indicate if the car was totaled (for example: hitting a Ford Mustang in the rear quarter panel usually means the car is totaled. Also, if it was a strong enough collision for the very expensive airbags to go off in the front, usually means totaled as well).

    As for looking for variances in paint, it’s pretty easy to do. Look at the reflection of your arm/hand in the paint of the car. If it’s wavy, or anything but a perfect mirror image, chances are that section of the car was hit, repaired and (poorly) repainted.

    Lastly, make sure that the title is clean/valid! My hubby and I bought a 88 notchback mustang thinking the title was clear, only to later realize that it was deemed as scrap/parts only. Granted, it’s a racetrack only car but still… sucks that it’s not street legal due to the title.

    Sorry for the long post, just that my hubby and I buy project cars on Craigslist a lot. Tis a small hobby of ours. Hopefully the stuff I posted helps everyone along with your data. :hugs:

    • Hi Chrissy,

      I didn’t put it into the blog post (for sake of brevity), but I had a big checklist of things to look over when inspecting a car, including looking under the car for signs of flood damage.

      I can’t believe I didn’t put the note in about clean title…that’s hugely important, especially with disasters like Hurricane Sandy putting lots of flooded cars on the market.

      • It’s alright. That’s one of the main things that I love so much about the offbeat community: we help each other out. So, if you forget something, no biggie. Someone on here is bound to mention it to help the others.

        Wow, we kinda sound like an online commune. LoL

  4. I just bought a “cheap” car after my other one was totaled, trying to get one under the price of what my insurance would pay. I got a really good deal on a car, because I took it to a mechanic. The car was “drivable” but would need some minor repairs costing me about $700. I used that report to get the seller to agree to a lower price, since I would have to make repairs. It also gave me piece of mind that the low price wasn’t masking some huge problem that was going to blow up in my face in a few months. I ALWAYS recommend paying the extra $50 for a mechanic to look over a car before you buy!

  5. I recently sold my 22.5 year old Honda Civic, while it only had 65,000 miles on it and ran well, I knew that any future problems were going to cost more than it was worth. And then there was the odd fluctuation in the electrical system . . . I figured getting 1500 for it with 500 back on insurance, I will take the bus for awhile before getting another car.
    The next one will probably be a Civic as well.
    Your sister is very lucky to have such a good researcher helping her find a new-ish car.

  6. I have a 2002 Honda Accord ready to turn 257000 miles. No major problems. Main thing is keep up the routine maintenance. It has more than paid for itself. It is a 4 cyl and gets an steady 30 mpg and on a long highway trip I can get up to 32. No complaints!!! Body is still in great shape. I am not a great car wash and waxer, so that is saying a lot. Interior is in near mint condition. Rugs need a good cleaning. Some brands just do a better job of building a car to last.

  7. This incident reminded me one that happened with me. Then my friend suggested me to buy a used car online ( rather than spending money on a new one. That car which got broke down was my favourite and I got almost like that on my online car shopping deal. You just have to list your car requirement details and most matched is offered to you. This was easy process and convenient too.

Join the Conversation