Skip to abstract

Research note · ECPND · v0.3 · Updated 8 October 2026

ECPND Large Cap 50: an S&P 500 selection with the earnings call participation network.

Abstract

Earnings Call Participation Network and Dynamics (ECPND) is a daily score for US stocks, derived from who takes part in earnings calls. This note applies it where most equity capital is invested: among the largest companies of the S&P 500. Each quarter we take the 200 largest companies of an S&P 500-based universe, keep the 50 with the highest ECPND score, with every sector of the index represented by at least one company, and hold them at market-capitalisation weights, scaled so that every sector weighs what it weighs in the index, with no stock above 15%. The portfolio therefore looks like the index, holds the same kind of companies at the same sector weights, and differs from it only in which companies the score keeps within each sector. In a backtest from January 2022 to September 2026, after 10 bp trading costs, it returned +19.0% per year. SPY, bought once and held with dividends reinvested, returned +12.2% per year. $100 grew to $226 in the selection and to $172 in SPY.

Unlike our volatile-stock selection, this portfolio kept its risk at the level of the index. Its largest interim loss was -24.8% against -22.3% for SPY, and its volatility 17.8% against 16.6%; the Sharpe ratio was 1.06 against 0.77. The lead was never large in the wrong direction. The selection finished ahead of SPY in four of five calendar years and less than a point behind it in 2023; measured over all rolling one-year windows, it was ahead in 99% of them, and its largest shortfall against SPY from any earlier relative high was 6%. These are the results of a historical simulation and do not guarantee future performance.

Author · SGG ResearchUniverse · US large capsPrice cut-off · 18 September 2026

01Introduction

Who takes part in an earnings call is information in itself. External analysts decide which companies they follow and question, and because the same analysts recur across companies and quarters, their participation links companies into a network of professional attention. Earnings Call Participation Network and Dynamics (ECPND) measures the structure of this network and the way its connections form, persist and shift over time, and expresses the result as a daily score for US stocks. Our published study examines the score as an overlay on stock rankings; a second note lets it select stocks on its own among the most volatile fifth of the market.

That volatile selection earns its lead over the index with a portfolio that looks nothing like the index: fifty mid-sized, fast-moving stocks at equal weight, a market sensitivity well above one and a tracking error of around 15% per year. An investor who holds the S&P 500 and asks whether the score can improve on it needs a different construction. The portfolio has to hold the companies that make up the index, at weights close to the index, and it may deviate only where the score gives a reason. This note builds that portfolio and compares it with SPY.

02Construction

The experiment follows a single account from 10 January 2022 to 18 September 2026. Every quarter it repeats the same three steps, 19 times in total. The universe, the prices and the account rules are those of the published study; the share counts used for market capitalisation are a new input.

  1. Take the largest companies. We start from an S&P 500-based universe of 477 stocks on average. For each stock we compute its market capitalisation on the decision day from the share count last reported before that day and the opening price, and keep the 200 largest companies. These make up about nine tenths of the value of the whole universe.
  2. Apply the score. Within the 200, the stocks are ordered by their ECPND score as it was available on that day. No price forecast, no model and no other factor enters the decision. The 50 stocks with the highest score are selected for the coming quarter; stocks without a score cannot be selected. If a sector of the 200 has no company among the 50, its highest-scored company takes the place of the weakest selection, so that every sector of the index is represented.
  3. Weight and trade. The selected stocks are weighted by market capitalisation, and the weights are then scaled sector by sector so that each sector carries the weight it has among the 200 largest companies. The score decides which companies are held; the index decides how much technology, energy or health care the account holds. No stock may exceed 15% of the account; any excess is spread over the other holdings in proportion to their size. At the next market open the account sells what is no longer selected, buys what is new and adjusts the rest. Until the next quarter nothing is traded.

The account starts with $1 million, keeps 95% invested and 5% in cash, and pays 10 bp on the value of every purchase and every sale. Its value is recorded at each daily close. The comparison is SPY, bought once at the open of the first day and held to the end with the same invested share and the same costs. All prices are adjusted for splits and dividends, so both accounts include reinvested dividends. Share counts are taken from company filings as reported by Yahoo Finance and adjusted for the same splits as the prices.

03Results

Figure 1$100 in ECPND Large Cap 50 and in SPY
Daily value of $100 invested in SPY and held, and of $100 in the 50 stocks with the highest ECPND score among the 200 largest companies, capitalisation weighted, from January 2022 to September 2026.
ECPND Large Cap 50 rebalanced quarterly · SPY bought and heldECPND Large Cap 50 · $226SPY, buy & hold · $172

Figure 1 follows $100 in each account. The selection was ahead of SPY on 84% of all trading days since the start. The lead opened in 2022, when the selection fell less than the index, and widened in 2024 and again in 2026. In between it paused: measured against its own best relative level, the selection fell at most 6% behind SPY, in August 2022. Over the first half of the period the selection returned +10.5% per year against +6.4% for SPY; over the second half +27.7% against +18.1%.

The holdings moved with the score, the sector weights did not. In October 2022 only 4 of the 50 names were technology companies, so the sector's index weight rested on those few, next to the largest holdings of that quarter such as T, LOW, CB, WMT; by July 2026 the selection contained 27 technology names, led by NVDA, AAPL, AVGO, MU. The largest position reached the 15% cap at 7 of 19 rebalancings, usually Apple or NVIDIA. The score therefore still followed where the participation network registered growing professional attention, but the sector scaling kept the account from turning that into a sector bet.

Table 1ECPND Large Cap 50 and SPY, January 2022 to September 2026
PortfolioReturn p.a.Total returnLargest lossVolatilitySharpe
ECPND Large Cap 50+19.0%+126%-24.8%17.8%1.06
SPY, buy & hold+12.2%+72%-22.3%16.6%0.77

The selection returned +19.0% per year against +12.2% for SPY, with a largest loss of -24.8% against -22.3% and volatility of 17.8% against 16.6%.

Table 2Return by calendar year
Portfolio20222023202420252026 to Sep
ECPND Large Cap 50-11.2%+24.4%+33.8%+26.6%+21.1%
SPY, buy & hold-15.3%+24.7%+23.7%+17.1%+12.2%

The selection finished ahead of SPY in 2022, 2024, 2025 and 2026 and behind it in 2023, by less than a point.

Table 3Controls and variants, return per year
PortfolioReturn p.a.Largest lossSharpe
200 largest, capitalisation weighted+13.4%-22.9%0.82
ECPND Large Cap 50, capitalisation weights without sector scaling+20.0%-24.0%0.94
200 largest, 50 weakest ECPND+4.8%-21.9%0.37
ECPND Large Cap 50, equal weight+17.4%-24.9%0.97
ECPND Large Cap 50 without NVDA+17.3%-23.4%1.04

The 200 largest companies at capitalisation weights stay close to SPY (+13.4% against +12.2%); the difference is the larger members outpacing the smaller ones in this period. Without the sector scaling the same selection returned +20.0% with a largest loss of -24.0%. Selecting the 50 weakest ECPND scores instead of the strongest returned +4.8%, so the score separates the two halves by 14 points per year. Equal weights lose 2 points; leaving NVDA out of the large-cap set loses 2 points.

Table 3 separates the parts of the result. Holding the 200 largest companies at capitalisation weights stays within about a point of SPY per year, which confirms the share counts and prices and shows how little the largest members alone gained over the whole index. The sector scaling is what brings the volatility down to the index: the same selection at plain capitalisation weights returned +20.0% per year with volatility of 21.6%, the scaled account +19.0% with 17.8%. Selecting the 50 weakest scores instead of the strongest returned +4.8% per year, below the index; the score orders the large caps in the right direction, and the spread between its two ends is 14 points per year. Capitalisation weighting matters: the same 50 stocks at equal weight returned +17.4% and were ahead of SPY in far fewer one-year windows. The result also does not rest on one stock. With NVIDIA removed from the large-cap set before selection, the portfolio still returned +17.3% per year, 5 points above SPY.

04Limits

The lead over SPY is not evenly spread. In rolling one-year windows the selection was ahead in 99% of cases and at worst -2 points behind. The year 2023 shows what a weak stretch looks like: the account held technology at the index weight, but the index's return that year came from a handful of the largest technology companies, and the score held at most two of those seven at any rebalancing in 2023; the account finished less than a point behind the index. An investor who adopted the selection at the start of 2023 would have waited about 8 months to draw level with SPY. The tracking error against SPY was 7% per year, about half of that of the volatile-stock selection (15%), but far from an index fund.

The construction was chosen in hindsight. The number of large companies, the number of holdings and the cap were fixed after reviewing variants on the same history: with the 150 largest companies instead of 200 the return falls by 2 points per year, with the 100 largest by 6 points, with 40 holdings it is +21.9% and with 60 holdings +15.2%, while caps between 10% and 20% change it by at most 0.4 points. These are degrees of freedom that a single history cannot settle, and the sensitivity to the size of the large-cap set is the most important of them. The requirement that every sector be represented was added in this edition, after the first live portfolio built from the rule held no financial, consumer staples, utility or real estate company; on the same history it costs 1.8 points of return per year and deepens the largest loss from -19.9% to -24.8%, but it makes the lead over SPY more even: the share of rolling one-year windows ahead of SPY rises to 99%. We prefer the rule that matches its description.

The period covers one market cycle, in which large technology companies led the market for most of the time. The sector scaling removes the choice of sectors from the result, but within technology the outcome still depends on whether the score held the right few names, and from 2024 on it did. The accounts assume execution at the next open, costs of 10 bp and no market impact; at the size of these companies capacity is not a constraint, but taxes are not modelled. Delisted and acquired companies are missing from the universe, which flatters all portfolios including the index replica, although the comparison with SPY uses the fund's actual prices. This note is a historical simulation and not a forward test. A prospective record of this construction has been kept since 7 October 2026: quarterly selections are committed before the opening and valued at every close, and the record is available to subscribers.