Opendoor just reported one of the ugliest earnings in recent memory, and the stock rallied 16%. Let me show you what actually happened, because this is a masterclass in financial sleight-of-hand.
The headline numbers: Revenue down 15%. Homes sold down 13%. Net loss of $1.3 billion—more than three times worse than last year. Shareholders diluted by 33% because the company couldn't pay its debts in cash, so it printed shares instead. Oh, and contribution margin hit 1%, meaning they made just $4,000 per home after all costs.
Four. Thousand. Dollars.
You could flip a couch on Facebook Marketplace and make better margins.
So why did the stock rally? Because the new CEO built the entire bull case on 150 homes bought in October, called them "the most profitable October cohort in company history," and refused to tell you the actual margin number.
Let me walk you through the con.
The October Cohort Trick: Out of 1,978 homes sold in Q4, only about 118 came from the new model. That's 6% of the total. The other 94% were older inventory that got dumped at 1% margins. But the entire earnings call focused on those 118 homes.
Why? Because when you cherry-pick a tiny sample, you can make anything look good. They said the October cohort margin was "above our target range of 5-7%," but wouldn't give the actual number. If it was 12%, they'd shout it from the rooftops. The fact that they won't say tells you everything.
Also, "best October ever" is meaningless. October is one of the weakest months in housing. This is like Applebee's bragging about their best Tuesday lunch service in February. Congratulations, you beat a low bar.
The small volume problem: They only bought 1,169 homes in Q3, the lowest on record. When you're that selective, of course margins look good—you're only taking the slam dunks. The question is whether this holds when they scale to 6,000 homes per quarter, their stated target. Zero evidence it will.
And here's the kicker: the CFO literally said, "You should not expect every quarter to look like October on a margin basis." Translation: margins will compress the moment we try to grow. We've seen this movie before. It was called Zillow Offers, and it ended with a $500 million writedown.
The $933 million "noncash" loss: Everyone is waving away the GAAP loss as "noncash," but let me explain what actually happened. Opendoor couldn't pay their debts, so they handed shares to creditors. Shares outstanding went from 720 million to 957 million. Every existing shareholder's slice of the company got 25% smaller overnight.
That's not a fake loss. That's real dilution. You owned a percentage of the company, and now you own less. The accounting system records that as a loss because it is one.
And they still have $193 million in convertible notes due within 12 months. If those convert, more dilution is coming.
What investors should watch for: Can they scale? The path to profitability requires 3.5x growth in volume over 10 months while maintaining 5%+ margins. That's a huge ask in a softening housing market. One hiccup—rising rates, falling home prices, economic slowdown—and the whole thing falls apart.
Also, pay attention to how many of those October homes have actually sold. They reported at 50% sell-through, which means the first half—always the best homes—are gone. The back half is where margins go to die. Every prior cohort has degraded as sell-through increased.
Red flags to watch: If contribution margins don't improve next quarter, the October story was a fluke. If acquisition volumes stay low, they can't scale. If they have to raise more capital, expect more dilution. And if housing demand weakens, the whole business model is in trouble.
The stock might keep rallying on meme energy and retail hype. Momentum is a real force, and sometimes bad companies have good charts. But the fundamentals here are brutal, and pretending otherwise doesn't change the math.
If they won't tell you the actual October cohort margin, there's a reason.
