Expected Body Performance Highlights: 5/15/25

TL/DR: Exposure weightings based upon the ratio of Vector Model based “Expected Body” metrics to corresponding Sigma implied “Expected Body” metrics, have generated positive alpha over the highly volatile prior ~90 days spanning 2/18/25 thru 5/14/25, as they have for our overall 3+ year out of sample period, before accounting for transaction costs, financing, and using simplifying assumptions about execution price.

First, a refresher (or quick intro) on Expected Body metrics and why you should care:

Expected Up Body (EUB) and Expected Down Body (EDB) are the probability-weighted average prices between the model date price and the 95th probability percentile price upward and downward, respectively. We offer these metrics because “expected value”, inclusive of tail outcomes in a world of often fat tails, can easily clash with our common sense understanding of the term “expected”.

EUB and EDB are likely of interest to (1) investors seeking “base case” return scenarios to the upside and downside, respectively, such as those running portfolio optimization analytics, (2) investors involved in call or put spreads, though we would also point such investors to our Option Fair Value Estimate performance report, (3) alpha seeking investors, for, as we detail below, the ratio of Vector Model EUB and EDB relative to Sigma based EUB and EDB tends to generate alpha when used to guide position weightings.

Important Expected Body Performance Definitions and Disclosures

In this blog we will highlight excerpts from the Expected Body Performance Report on vecviz.com related to Return on EUB (and EDB) Based Capital (ROEUB and ROEDB, respectively)¹. ROEUB and ROEDB indicate how return seeking and risk avoidant strategies, respectively, perform when using the support and resistance / machine learning-based Vector Model to guide return expectations and associated ticker exposures, instead of bell curve-based Sigma based Expected Body metrics².

More specifically, Sigma ROEUB and Sigma ROEDB are ascribed the underlying ticker return, and Vector Model ROEUB (ROEDB) is ascribed a multiple of the underlying return based on the ratio of Vector Model EUB (Sigma EDB) to Sigma EUB (Vector Model EDB), subject to a cap and floor of 3.00x and 0.333x.

Since EUB and EDB are influenced heavily by the distance to the 95%tile, and the 95%tile for the Vector Model is very frequently different than the 95%tile for Sigma, both to the upside and downside, we point your attention to the alpha of Vector Model ROEUB and ROEDB to the underlying ticker returns we ascribe to Sigma ROEUB and ROEDB. Systematic differences between Vector Model and Sigma Expected Body metrics, be they driven by systematic differences in 95%tile or otherwise, get reflected in beta, leaving the alpha a purer reflection of value add from ticker differentiation and/or timing³.

Before we begin, please note that (1) what we refer to as “Sigma” is also commonly referred to as Exponentially Weighted Moving Average (EWMA) Sigma – it weights more recent return observations more heavily in the context of a rolling 2 year lookback window, (2) aggregate measures of ROEUB and ROEDB reflect simple averages of ticker level EUB and EDB analytics for the Vector Model and Sigma, and associated actual forward returns, (3) results for forward horizons > 1d reflect overlapping periods, for both the Vector Model and Sigma, unless noted otherwise, (4) ROEUB and ROEDB do not include any adjustments for transaction costs, position financing costs (or credits), and assume execution occurs at end of day prices – a shortcoming of the analysis that we partially mitigate by focusing upon 10d forward performance in our commentary. With that said, let’s get into it.

Fig1: EUB and EDB implied forward 10d returns across tickers, by day:

VecViz’s entire ~150 ticker coverage universe, for both the Vector Model (blue) and Sigma (red), are reflected above, for each day in VecViz’s entire 3+ year out of sample performance period.

You can see that over the prior 3 months (small margin right of “2005-01”, onward till the end), Vector Model EUB was lower than Sigma’s, suggesting ROEUB will reflect more conservative positioning. However, Vector Model EDB was less adverse than Sigma EDB, suggesting that ROEDB will reflect more aggressive positioning.

Fig2: Average 10d ROEUB and ROEDB across tickers, by day:

Average ROEUB vs Sigma and associated Alpha and Beta

Vector Model (bold bars) and Sigma (pastel bars) depict the average ROEUB (right chart) and ROEDB (left chart) for the 90 days spanning 2/19/25 thru 5/14/25, for the 1d forward, 10d forward, and 21d forward time horizons.

We see that on an outright basis the Vector Model tended to underperform Sigma on ROEDB but outperform on ROEUB, consistent with the relative positioning implied by the expectations depicted in Fig1, given the context of negative average returns for the ticker coverage universe. However, both ROEUB and ROEDB delivered positive alpha during this time frame, for each forward time horizon. Note that 10d and 21d ROEDB alpha was statistically significant at the 1% threshold, and 10d ROEUB was significant at the 25% threshold.

Tickers driving ROEUB and ROEDB performance over the last 90 days

Here we look at top and bottom 10 tickers driving ROEUB (left) and ROEDB (right) performance for the Vector Model (blue) and Sigma (red) over the 2/18/25 thru 5/14/25 period. These were extracted from lists of 30 such tickers in the full report.

A broader view of Vector Model “Expected Body” performance vs. Sigma

On the left we see the % of criteria listed on the right that were met by each horizon (ranging from 1d to 252d) for each applicable lookback window (30d thru the entire 3+ year out of sample period, “AllD”). On the right we see that alpha has been greater than zero for most look back windows and applicable horizons, for both ROEDB and ROEUB, so the performance of the last 90 days is not atypical.

Related reports and reference info:

See the Reports page of our site for the full Expected Body Performance Report, and similar reports on Value at Risk (VaR), Opportunity at Risk (OaR), Option Fair Value Estimates, and our V-Score. There is also a report that aggregates the report cards of all the metrics.

See our Methodology oriented blogs, including those relating to Vector Strength and Support and Resistance, to learn more about the Vector Model.

The report also addresses mean absolute average error (MAE), the metric we use when evaluating accuracy. For example, when evaluating EUB, we take the average absolute value of the difference between the actual forward horizon price and the model date EUB price for the given model, for ticker model dates that are in the given model’s “evaluation set”. ↩︎
As explained in the full Expected Body performance report, Sigma EUB and EDB for a ticker are +/-0.657 standard deviations from the Model Date price ↩︎
That said, in the Expected Body Performance “Report Card” we also consider ROEUB and ROEDB on a “distance to 95%tile Vector Model -Sigma differential” adjusted basis, and we encourage you to review the report to learn more about how we do that. ↩︎