How to analyze rental properties with hedge fund algorithms

The OC Housing News analyses every for sale property on the MLS for its potential as a cashflow investment based on advanced hedge fund algorithms.

small_investorsBetween late 2010 and early 2012, I purchased 53 homes in Las Vegas — sight unseen. I used a service that provided pictures of the properties, but since these were all auctions, sometimes the inside views were not available, so I had no idea what I would find if I won at auction.

Does that sound scary or crazy?

It was a manageable risk. Usually, whenever I bid on a property where I couldn’t see the inside, I simply bid less to give myself an allowance in case I had to replace everything, and sometimes I still won the property at auction. I purchased one property only to discover after we opened the door that the former owner removed the entire kitchen, cabinets, counters, and even the expensive appliances from unclutterer. However, I bought that property with such a large discount that I didn’t spend near the savings on replacing the entire kitchen.

If you know your after-renovation resale price, and if you know construction prices in the area, buying houses sight unseen is not as risky as you might imagine.

Automating the process

Las_Vegas_foreclosureThe most time consuming part of bidding on properties was analyzing each deal. I manually pulled rental and resale comps on over 1,500 properties in order to obtain the 53 I purchased. I had to copy and paste this information into a spreadsheet that crunched the numbers to generate my bids. I knew there had to be a more efficient way, but I couldn’t justify the expense of a system that would fully automate the process — I envisioned such a system, and I hoped some day to implement it.

In the spring of 2012, I was approached by an OC real estate developer who was asked by his capital partner to put together a proposal to buy rental properties at auction similar to the REO-to-rental companies that sprang up during that period. With my experience in Las Vegas and my understanding of residential rental acquisitions, I was responsible for developing the systems we would use to analyze deals.

Although these REO-to-rental companies like to come across as being immensely sophisticated, none of it is rocket science. The basic math is actually quite simple. As I worked to bring the venture to fruition, I also developed the algorithms necessary to assess rental and resale values automatically. When the deal fell apart, I was left with a lot of disappointment but also a fully developed property analysis system — the system implemented on this site today.


Racing to Buy Homes Sight Unseen

Residential real estate to be the next frontier for speed-based investing

By Timothy W. Martin, April 13, 2015 11:53 a.m. ETcash_cow

ATLANTA—It took Akuansa Graham seven minutes on a recent morning to craft a $124,000 bid for a three-bedroom Buford, Ga., home he had never seen.

The Starwood Waypoint Residential Trust executive went to public auctions in the years after the financial crisis looking to buy homes lost to foreclosure. Now the 38-year-old crouches over a computer and relies on algorithms that evaluate home values, proximity to schools and crime rates to outrace rivals for any remaining bargains offered by real-estate agents.

This is exactly what I did each morning. It’s far easier now with a fully automated system to weed out the poor candidates. If I were doing similar work today, I would use the automated system to generate a list of potential targets, then I would analyze comps to verify the accuracy of the projections. The system eliminates most of the wasted time that otherwise would be spent evaluating unsuitable properties.

With the low-hanging fruit from the housing bust mostly picked, Wall Street-backed buyers of real estate are increasingly turning to quantitative data analysis as a way of accelerating their search for a dwindling supply of available homes that can be transformed into rental properties. Math-driven models powered by historical patterns can size up homes sight unseen and calculate future income in minutes, allowing private-equity giant Blackstone Group LP, the Alaska Permanent Fund Corp. and other bulk purchasers to skirt neighborhoods with softer rental demand or properties that need costly repairs.

Advances in how companies use technology to evaluate mountains of data has quickened everything from stock trading to student test-performance evaluations to patient care. Behind the new speed in real estate is a change in how big buyers find most of their properties.

And this same system is available to everyone here on this site.cashflow_problem

Oakland, Calif.-based Starwood Waypoint, one of the six, said it has cut the time it takes to calculate a first bid on a house to eight minutes. “We encourage our guys to make an offer before they see the house,” said Ali Nazar, Starwood Waypoint’s chief experience officer. “I don’t want to wait for anyone else. Our competitors are also fast.”

Since their bids always contain a number of contingencies allowing them to get out of the contract, bidding quickly, sight unseen, is not a big risk. Since these aren’t auctions, they will see it before they buy it.

Operating from the firm’s largest branch office at a former meatpacking plant in Atlanta, Starwood Waypoint’s acquisitions team evaluates potential purchases with a data map that ranks the “livability” of local neighborhoods according to information provided by local employees of the firm who frequently visit the area. Factors include proximity to retailers and how noisy the neighborhood is.

Employees ultimately weigh about 15 variables when calculating a bid price, the monthly rent and renovation costs. Employee bonuses are based on the accuracy of their home bids and rental estimates.

While this additional data is not necessary to evaluate financial performance, it’s useful for small investors who plan to hold properties long term. I also provide community information on crime, schools, neighborhood amenities, and so on.

For more details, look at the investor calculations on today’s featured property in Victorville. If you click on the address, you will be taken to the property details page. There are tabs for comparable resales and comparable rentals to verify pricing and rental income. On the comparable rentals tab, you can examine both active and closed sales. This property shows that it could easily rent for the $1,000 price generated by the automated system. If someone purchased this as a rental, it would likely perform as the investor calculations show. This particular property has a cap rate of 5.4% with a cash-on-cash return of 8%. In today’s market, that’s not bad.


[dfads params=’groups=23&limit=1&orderby=random’]
[listing mls=”SB15078698″]