In the world of investment research, the pressure to present “high-conviction” ideas is pervasive, despite concerns about the relationship between confidence and accuracy. This dynamic was recently highlighted by a note from JPMorgan’s oil research team, which for the first time since the onset of the Iran conflict admitted it lacked a clear baseline outlook on crude oil prices. The analysts stated that modeling the conflict’s “endgame” was presently impossible and questioned the assumption that global oil supply disruptions were merely temporary.

This candid admission broke a long-standing norm in the financial industry, where analysts typically feel compelled to express definitive views, regardless of underlying uncertainty. The JPMorgan note drew mixed reactions, including criticism on social media for breaking from the tradition of projecting certainty. However, some observers suggest the industry might benefit from such humility, arguing that acknowledging uncertainty could enhance the credibility of analysts in a field where overconfidence often leads to significant investment losses.

For years, investors and analysts alike have placed premium value on conviction calls, even if those calls are built on tenuous premises. Experienced research professionals recall being instructed to produce a set number of “high-conviction” recommendations regularly, a practice that sometimes led to recycling mediocre ideas dressed with exaggerated confidence. These so-called conviction lists often included outdated or poorly performing stock picks boosted by inflated price targets to mask underperformance.

Critics of this approach highlight that excessive certainty may be a greater risk factor than the absence of a clear view. Markets involve complex human behavior and unpredictable variables, making rigid attachment to a thesis susceptible to costly errors. The ability to manage cognitive dissonance and admit uncertainty is considered an underappreciated skill in investment management.

The emergence of artificial intelligence and big data analytics has complicated the landscape further. Hedge funds and asset managers once believed that proprietary data—ranging from satellite imagery to web-scraped pricing—could consistently generate alpha, or above-market returns. While these methods provided some edge, many quantamental funds have since tempered expectations, noting that such data often increases subjective confidence rather than improving decision quality substantially.

This increased certainty can slow the willingness to cut losses on poor investments, a factor detrimental to overall performance. Market participants are reminded of the importance of optionality—the ability to refrain from committing capital when prospects are unclear, akin to a gambler choosing not to bet on every race despite the odds.

Some industry voices argue that analysts’ willingness to admit “I don’t know” should be appreciated rather than ridiculed, potentially marking a shift toward more measured, less dogmatic research practices. In this context, the long-cited observation from poet William Butler Yeats—that “the best lack all conviction, while the worst are full of passionate intensity”—resonates as an apt reflection on the nature of investment insight. Those who succeed may not be those who are most certain, but those who balance confidence with an awareness of their own fallibility.