Investors in NVIDIA (NVDA 1.75%) have had a rough time over the past year. Shares of the specialist chipmaker, whose wares are found in everything from cars to video game equipment to data centers, are down more than 35% over the trailing 12 months.

NVDA Chart

NVDA data by YCharts

NVIDIA's high-octane chips can be used for tough processing jobs, including data science and artificial intelligence. Nevertheless, demand has slowed since last fall, resulting in a free-falling stock price. Will the losses continue to pile up when NVIDIA reports earnings on May 16? Not if the company sells enough graphics processing units (GPUs) to big customers. Here's what to look for.

When expectations go low, (some) stocks go high

Analysts are calling for at least $2.197 billion in revenue and $0.81 a share in profits after accounting for one-time and non-cash items in NVIDIA's fiscal first quarter. That's about in line with what NVIDIA reported in Q4 -- $2.205 billion and $0.80, respectively -- but down big from last year's Q1, which brought in $3.207 billion in revenue and $2.05 a share in adjusted earnings. Talk about a downgrade!

We'll know Thursday whether followers of the stock are being too pessimistic. Should revenue come in meaningfully above $2.2 billion, we'll know that NVIDIA is back on track.

In the meantime, it's worth noting that NVIDIA executives have emphasized the size of the addressable market the company is attacking. In data centers alone, the company says, the opportunity is nearly 5 times the entire company's current revenue run rate.

Comparing NVIDIA's addressable market in two charts, with the growth going from $37 billion to $50 billion

NVIDIA sees a big opportunity ahead, but can the company cash in? Data source: NVIDIA investor presentation.

CEO and co-founder Jensen Huang put NVIDIA's thesis in perspective in his stage presentation for Investor Day, declaring: "Every single industry will benefit from data science. That's why there are so many people ... because the amount of data that you're working on is so gigantic."

Translation: Be patient; the profits are coming.

Clouds rolling in, and storms could follow

Despite the hyperbole, analysts remain cautious with their estimates, which seems fair to me. Even after the sell-off in the stock, NVIDIA trades for a premium to virtually all of its peers on the basis of price to sales (8.3, versus 5.5 for the industry median) and price to estimated earnings (31.26, versus 17.19 for the industry median).

Nor is there a guarantee that NVIDIA will crush its depressed target. The scientific computing market in which Huang places so much faith is already being served by the "hyperscale" providers he's counting on to be continual buyers. Yet fellow data center supplier Arista Networks (ANET -0.96%) saw its stock take a hit on slowing demand from the likes of Facebook and Alphabet. NVIDIA could just as easily report similar slowing this week.

The Foolish bottom line

I'm not a buyer of NVIDIA stock, yet. The story is solid, and the opportunity is most certainly real. But you don't have to buy NVIDIA shares to get meaningful exposure to the data science or artificial intelligence trends. Far cheaper stocks with serious growth potential are enjoying similar tailwinds. Stitch Fix (SFIX -2.82%), for example, is one of the biggest internal users of data science to help foster deep relationships with its clients, a strategy that appears to be taking hold.

So don't rush into NVIDIA stock. Watch it, wait, and keep some cash handy. Odds are this one is going to remain volatile for a while.