Three months ago Powell Industries delivered the kind of quarter that usually resets a stock higher: a record $934 million in new orders, a 3.0 times book-to-bill ratio and a backlog that jumped to $2.4 billion. Since then the shares have fallen about 31%, trailing the S&P 500 by roughly 27 percentage points. When the switchgear maker reports fiscal fourth-quarter results after the close on November 16, the question is whether the business is still tracking the expansion story management described, or whether the market has spotted cracks the order book alone cannot paper over.

The Street is looking for revenue of about $328.4 million, up roughly 10% from $298.0 million a year ago and about 5% above last quarter's $311.7 million. That is solid growth, not a step change. Consensus EPS of $1.52 implies a 64% year-over-year decline, but that headline figure looks misleading. Management presented last quarter's $1.42 on a split-adjusted basis, and the year-ago $4.22 appears to reflect the pre-split share count. A fairer read is that analysts expect earnings to edge up from last quarter's $1.42, consistent with modest revenue growth and stable to improving margins. Powell offers no formal numeric guidance, so the consensus is anchored mainly to the backlog and to management's increasingly assertive tone.

That tone is what this report has to back up. The big-ticket wins last quarter, including a phase-one behind-the-meter data center order of more than $400 million plus $60 million LNG and $75 million petrochemical awards on the Gulf Coast, are long-dated. Management acknowledged that only about 54% of backlog is expected to convert within 12 months, down from the low 60s a quarter earlier, and revenue slightly missed expectations last time. So the order surge is not expected to show up in revenue right away. What investors should want is evidence that conversion is stabilizing rather than slipping further, plus another quarter of orders healthy enough to suggest the $934 million figure was a peak in a rising trend rather than a one-time spike. A sharp drop in bookings would be expected after a mega order, but commentary on the next phases of the data center campus and on the pipeline of smaller awards will matter.

Margins are the second test. Gross margin rose to 30.6% from 29.6% in the prior quarter, but management also flagged moderate inflation in copper, aluminum and steel, SG&A costs from standing up new facilities, and tightening craft labor heading into fiscal 2027 and 2028. The Jacinto Port expansion, which management said would finish within a month or two of the last call, should now be complete, and any update on its ramp toward more than $100 million of incremental annual revenue is important. So is a decision on the $70 million to $100 million greenfield plant. Holding gross margin near 30% while absorbing startup costs would validate the operating leverage thesis; a dip back toward the high 20s would feed the bear case that capacity spending is getting ahead of profitable output.

Sentiment has shifted from mildly bullish before the last report to mildly bearish now, and the stock tells the same story. At $133.12, shares sit far below the 200-day moving average of $213.08 and only about 10% above the post-earnings low of $121.13, a dramatic comedown from last quarter's range of $184.61 to $328.00. Expectations have clearly reset lower, which gives Powell a less demanding audience than it had in the summer.

The core issue is simple. The backlog proves demand; this report has to show Powell can turn that demand into steadily rising revenue at sustained margins while its petrochemical business, down 49% last quarter, moves toward the recovery management hinted at. Do that, and the selloff looks like a misread of a longer runway. Show slower conversion and margin slippage, and the market's skepticism will look earned.