Episode summary
Daniel Mahr – Glass Box Quant at MDT Advisers (EP.472)
Capital Allocators with Ted Seides
· · Hosts: Ted Seides · Guests: Daniel Mahr
Summary
Daniel Mahr argues that MDT’s edge comes from combining machine learning with a transparent, repeatable stock-selection process rather than chasing secret information. He traces his path from flipping IPOs in the dotcom bubble to joining MDT in 2002, where he saw firsthand that a disciplined system could outperform instinct, especially when humans struggle to buy “bad story” stocks. MDT evolved from simple factor tilts to a decision-tree framework that asks a sequence of context-dependent questions about companies and then aggregates many shallow trees into a “forest” used for portfolio construction.
Mahr says the firm’s approach is a “glass box,” not a black box: the team can see which factors drive each trade, test whether signals still add value, and remove factors like book-to-price when they stop working. He emphasizes overfitting, underfitting, liquidity, trading costs, and risk controls as central constraints. MDT also avoids relying on alternative-data arms races, preferring long historical datasets and analytical edge over informational edge.
Looking ahead, Mahr says GPT-style large language models are not yet used in the stock-picking process because of backtest and in-sample concerns, though AI tools may help with software productivity. He remains optimistic that more data, faster computing, and better algorithms will keep creating opportunities.
Key takeaways
- Mahr says MDT’s durable advantage is analytical, not informational, and comes from decision trees and machine learning applied to long historical datasets.
- He says the firm uses a “glass box” model so traders can see why the portfolio is making each decision and remove factors that lose explanatory power.
- Mahr says book-to-price and other older signals have faded over time as the economy has become more intangible.
- He says the model can surface surprising effects, including strong reversals in stocks down 70% to 80%, even when traders are emotionally resistant.
- Mahr says MDT is not using chatGPT in stock selection because commercial large-language-model backtests are not trustworthy out of sample.
Notable quotes
“We like to position our investment strategies as being a glass box. There's a lot of machinery on the inside, but we can see into it”
“Our view in the machine learning space is that transparency is exceedingly important to understand precisely how these models are working”
“we have a 24 year head start on someone who is new to the game”
“the model seems to be making less use of this over time. What if we made zero use of it”
“The good news is we don't need to know what's driving it. The important thing is having strategies that are active and that are able to take advantage of inefficiencies when they present themselves”
How the episode opens
Our head of trading would come to me every day we bought a stock that was down 70 or 80% and say, "Dan, we need to override this trade. The company's CEO just resigned in disgrace. Or, "Dan, this is a one product biotech company and they've just missed their target.
It looks like the whole company's got nothing. We have to override these trades. We can't do them.
That's when the light bulb goes off. This is precisely why this strategy works is because even quantitative investors who are intentionally trying to buy these stocks find it hard to overcome the human emotions involved with buying a bad story. I'm Ted Sides and this is Capital Allocators.
My guest on today's show is Daniel Maher, head of MDT, the $26 billion quantitative equity investing group at Federated Hermes that oversees a suite of actively managed mutual funds, ETFs, collective investment trusts, and separately managed accounts. Dan joined the firm in 2002 as a junior analyst and took over leadership of the team 6 years later, guiding its evolution through vast changes in data, computing power, and investment methodology. Our conversation traces Dan's path from flipping IPOs as a college student to running machine learning models across global equity markets.
We discussed the development of MDT's decision tree framework, a glassbox approach to stock selection that blends transparency with sophistication, and how the team balances analytical rigor with human judgment. Dan explains lessons from two decades of modeling markets, including the challenges of overfitting and underfitting data and MDT's steadfast focus on analytical edge rather thanformational edge. Before we get going, it's that time of year when we turn to traditions, like the tradition of Thanksgiving, gathering family and friends to share what we're thankful for.
One of which is the ability to eat enough turkey and tryptophen to fall into one of the best slumbers all year.
From the episode’s captions: roughly the first two minutes, lightly cleaned. Keep watching on YouTube →
Topics
- quantitative investing
- machine learning
- stock selection
- overfitting underfitting
- active vs informational edge
Companies and tickers mentioned
- MDT Advisers
- Federated Hermes
Watch the episode
Watch “Daniel Mahr – Glass Box Quant at MDT Advisers (EP.472)” on YouTube
Summary, takeaways and quote selection by Octagon. Quotes are short excerpts; the full conversation is in the episode.
More from Capital Allocators with Ted Seides
-
Aydin Senkut: Chasing Demand, Not Hype, at Felicis -
Scott Ramsower: Continuation Vehicles from the LP Seat -
Vik Uppal: Real Estate Special Situations With a Flexible Mandate -
Zack McGuire: Dynamic Allocation and Concentrated Conviction at Children's Healthcare of Atlanta -
Ben Harrison: AI Tools for the Investment Firm -
Tracy Britt Cool: The Berkshire Model for $5–50M Businesses
Capital Allocators with Ted Seides, this episode and its artwork belong to Ted Seides – Allocator and Asset Management Expert. Octagon is not affiliated with the show; this page is an independent summary.
