Episode summary
Top Decile CIO Mark Steed on Sports investing, AI and tracking high convictions for pension teams
Summary
Mark Steed says Arizona Public Safety Personnel Retirement System’s growth from $10 billion in 2019 to $25.5 billion came from two forces working together: stronger investment performance and higher contributions from employers. He says the fund simplified its asset allocation, improved trust with constituents, and used a probabilistic framework to overweight US equities after COVID, which helped returns while low rates encouraged cities and towns to borrow and contribute more to reduce unfunded liabilities.
He frames his investment style as a blend of intuition and quantitative analysis, saying both matter but each must be checked against logic, assumptions and data quality. That same mindset led him to early work on AI and machine learning. Mark Steed says the fund is now focused on practical uses such as streamlining due diligence, extracting data from reports, and testing forecasts on paper before relying on AI for investment decisions. He is cautious about black-box models and says guardrails are essential, even if they reduce explanatory power.
On sports investing, Mark Steed says the fund liked the resilience of sports assets and found soccer attractive because of growth in US popularity and undervaluation. He now prefers adjacent service businesses over the teams themselves. He also argues that in a more uncertain economy, conviction should come down, portfolios should stay closer to target weights, and liquidity providers may be best positioned to benefit.
Key takeaways
- Mark Steed says Arizona Public Safety Personnel Retirement System grew from $10 billion to $25.5 billion through a mix of stronger returns and higher employer contributions.
- He says the fund’s AI work is still in R&D, with a focus on routine tasks first and predictive use only after enough sample forecasts are collected.
- Mark Steed argues that black-box AI needs guardrails because models can hallucinate and be hard to explain to boards or constituents.
- He says sports-team investing can work, but he prefers the service businesses around teams because they offer more diversification and less league-level control risk.
- In today’s uncertain market, Mark Steed says convictions should come down and liquidity providers may have the best opportunities.
Notable quotes
“We simplified the asset allocation to create fewer buckets so that, you know, there was more competition for deals”
“the third phase will be using AI to help make predictions and inform decision-m”
“we're very clear about if you're going to use quantitative data for analytics or anything else, we're clear about whether you have a large enough sample size”
How the episode opens
Hi, I'm Christine Jordano. I'm here with Mark Steed. Mark Steed is the chief investment officer of the Arizona Public Safety Personnel Retirement System.
Now, in 2007, he started at the fund as an intern. He was already a smart guy having gone to school and starting his degree in predictive analytics from Northwestern. But in 2019 he became the CIO of the fund that was $10 billion.
It's now $25.5 billion which is making this fund one of the fastest growing in the United States if not the world. He has a fascinating process by which he gets convictions from his investment team before making his investments and we'll go into all of that. Welcome Mark.
Thanks for having me. — Real pleasure. Let's start with your growth story. Now, as you were noted as one of the fastest growing funds in the United States, can you explain all of that?
I'll start out start out by saying I can't take credit for all of it. in uh 2019 which was the first year that I started in earnest as the CIO. At that time we were as you said 10 billion our performance was in the autumn cortile across all the major time horizons.
Now there were some reasons for that. There was some structural features in the benefit side that forced us to have a pretty defensive portfolio but long story short those nuances I think were lost on a lot of the constituents. So they just didn't have a lot of trust in the investment program.
So I knew that I had to, you know, put up some points pretty fast. And luckily, we were able to change the investment policy. We simplified the asset allocation to create fewer buckets so that, you know, there was more competition for deals.
From the episode’s captions: roughly the first two minutes, lightly cleaned. Keep watching on YouTube →
Topics
- sports investing
- ai in investing
- high conviction process
- pension fund investing
- due diligence
Watch the episode
Summary, takeaways and quote selection by Octagon. Quotes are short excerpts; the full conversation is in the episode.
More from Markets Group
-
The €3bn German Allocator Who Said No to Alternatives -
RE West 2026: Valerie Red Horse Mohl of the California Endowment Spoke to Courtney McQuade -
RE West 2026: Steve Biggs of The Mather Group Spoke to Courtney McQuade -
RE West 2026: Martha Delgado of Saint Francis Foundation Spoke to Courtney McQuade -
Two Stints as Interim CIO: Howard Hodel on Risk, Rebalancing and Reassuring the Board -
Stefan Beiner on Repo, Resilience, and Why Governance Beats Prediction in Pension Investing
Markets Group, this episode and its artwork belong to their owners. Octagon is not affiliated with the show; this page is an independent summary.
