I have a question about optimizing my portfolio based on the alpha and beta factors using the Treynor Black model. It seems that I can analyze an optimized portfolio, but can't adjust a previously created portfolio based on the template. Do I have to go through trial and error to lower the beta of my portfolio, or is there a way to calculate it separately in Excel? Is there a better way to manage this process within the platform?

Treynor Black model

We just made it easier to find better ETFs and mutual funds in the same category. Instead of opening a dozen tabs and running screeners, you can now see all the relevant alternatives ranked by the things that actually matter — like expense ratio, risk-adjusted returns, and drawdown — right on the fund's page.

### What we added

Every ETF and mutual fund on PortfoliosLab now has an **Alternatives** page. Open any fund — say, [VOO](/symbol/VOO/alternatives) — and you'll see a breakdown of similar funds from the same category, ranked by the things that actually matter when choosing between them.

<img height="412" width="696" src="https://portfolioslab.com/static/user-content/1/f0ahvv66ifgewz9r8k7e5bx6.png" />

The page is split into five sections, each answering a different question:

**Cheapest alternatives** — same category, sorted by expense ratio. If you're paying 0.50% and there's a nearly identical fund at 0.03%, you should know about it.

**Best risk-adjusted alternatives** — ranked by PortfoliosLab's combined score that accounts for volatility, drawdown, and return consistency over one year. Not just "what went up the most" but "what went up without falling off a cliff."

**Top performing (YTD)** — straightforward year-to-date returns. Useful for spotting momentum, less useful on its own.

**Lowest volatility** and **lowest drawdown** — for when your goal is sleeping through a correction, not chasing returns.

Below the ranked tables, you'll also see funds from the same issuer and a list of what other investors commonly compare with the fund you're looking at — based on real usage data, not editorial picks.



<img height="302" width="692" src="https://portfolioslab.com/static/user-content/1/q4gdr87fspmly06qqijbz9wf.png" />

The alternatives page starts from the fund you're already researching and surfaces what you'd find if you spent an hour digging through screeners — ranked, filtered to the same category, and ready to compare side by side. Each row links directly to a full head-to-head comparison if you want to go deeper.

### A few ways to use it

You hold a broad-market ETF and want to check whether a cheaper option exists in the same category. Open its alternatives page and look at the expense ratio table.

You're evaluating a fund someone recommended and want to see how it stacks up on risk. Check the risk-adjusted and drawdown tables to see where it falls relative to peers.

You want to understand what else is out there before committing to a position. Scan the full page, then compare directly with anything that catches your eye.

We just made it easier to find better ETFs and mutual funds in the same category with a dedicated alternatives comparison page.

Find Better ETFs and Mutual Funds Without the Guesswork

When you're looking at a portfolio, knowing it returned 12% last year isn't enough. You want to know how it behaves relative to the market. Was your strategy just riding the market or did it have a real edge? The block shows five metrics at a glance: Alpha, Beta, R², Upside Capture, and Downside Capture.

<img height="254" width="706" src="https://portfolioslab.com/static/user-content/1/s8z19ssdg4dcuxaedeod2m9i.png" />

To make it more useful, we also added a few insights alongside numbers. Raw numbers alone can mislead. A beta of 0.5 looks defensive, but only means something if R² is high enough. A portfolio capturing 200% of upside and 106% of downside is technically "favorable" but that's not the whole story. The block catches these cases and calls them out directly.

When R² is too low for the metrics to be meaningful, it tells you to pick a better benchmark rather than letting you draw the wrong conclusions.

Available on all portfolio pages where a benchmark is selected and in the [Portfolio Analysis tool](https://portfolioslab.com/tools/portfolio-analysis).

We added a Benchmark Metrics block to all portfolio pages and to the portfolio analysis tool results.

New Portfolio Metrics: Alpha, Beta, R², Upside and Downside Capture

Great stuff Dmitry, as you know I working a lot with PRT for the actual investing. Therefore as you wanted to you could refine the walk-forward approach by implementing anchored and non-anchored and calculating WFE. (the following is quoted from the public guide, [https://www.prorealtime.com/en/pdf/probacktest.pdf)](https://www.prorealtime.com/en/pdf/probacktest.pdf\)) 

Walk Forward analysis is an essential part of developing a trading strategy. It allows you to optimize a trading system and to validate its robustness and stability over time. First, the walk forward analysis optimizes a set of variables on an initial period called "in-sample data" or the "optimization period", then it tests the best parameters on the following period called "out of sample data" or the "test period", and repeats the process by shifting forward the time windows. This automated process can be repeated as many times as necessary. The optimization periods (in-sample) may have a single starting point (Anchored mode), or different starting points (Non-anchored mode). NON-ANCHORED MODE ANCHORED MODE By comparing the results of these 2 periods, the method allows you to: Check that the performance of the strategy on the optimization periods (in-sample data: shown in blue) are consistent with the test periods (out-of-sample data: shown in grey). This avoids the risk of "overfitting" by projecting the strategy on a period of time not used itself in the optimization. Check how the strategy behaves in changing market conditions Test the strength of the strategy on past data To determine if a strategy is robust or not, we use the WFE ratio (walk forward efficiency ratio). The walk forward efficiency ratio is a qualitative indicator of the optimization process. It compares the annualized gain of the test period to the annualized gain of the optimization period. This measure of robustness is a critical part of walk forward analysis. WFE Ratio \= Annualized gain of the test period Annualized gain of the optimization period Annualized gains are the gains realized over a given period, expressed in annual terms. They let us compare the gains of different periods on a common basis. Typically, the optimization period (in-sample/blue in the image above) represents 70% of the total analyzed period, and the test period (out of sample/grey) 30%. To measure the reliability of the analysis, the results of the two periods need to be evaluated with a common basis. Studies of this method indicate that if at least 3/5 test periods display a WFE ratio greater than 50%- 60%, the strategy can be considered to be robust. After the walk forward method, it is interesting to analyze the results of several points. It is necessary to examine the regularity of gains from post-optimization periods. If the results show that the strategy shows a risk of being over-optimized (\= the gains from test periods are much lower than those of optimization periods), the platform lets you adjust different parameters (variables, stop and target levels, trading periods) and launch a walk forward analysis again, as many times as needed, to obtain a robust strategy

How strategy backtests usually work. You input your portfolio positions and ask the optimizer to find an allocation with best performance. The optimizer then looks at the historical data, picks a few stocks that did well, drops the rest and presents you with a portfolio that would have crushed it over the last years. When comparing with the original allocation the backtest looks great. Of course it does. The optimizer already knew what happened.

This is called lookahead bias, and it makes backtests almost useless for predicting real-world performance. So we've added a way to test your portfolio strategy without it.

### What we built

In [Portfolio Optimization](/tools/portfolio-optimization), there's now a **Reoptimize Frequency** setting with three modes: Once, Quarterly, and Yearly.

Set it to Quarterly or Yearly, and the tool will re-run the optimization at the start of each period — using only the data available at that point in time — then measure how the resulting portfolio actually performed over the next period. This repeats across the entire timeline, giving you a realistic picture of how the strategy would have played out if you'd been running it live.

This is different from the **Optimization Date** setting, which works in the default "Once" mode. That one splits your data at a specific point: the optimizer trains on everything before that date, then you see how the resulting weights performed afterward. Useful, but it's still a single optimization — one set of weights tested against one future period.

Walk-forward takes it further by re-optimizing repeatedly, so every period gets weights that were calculated without any knowledge of what came next.

<img height="74" width="637" src="https://portfolioslab.com/static/user-content/1/ffzakrfu7msavorg3oe3ew43.png" />

### Why it matters

A single-point optimization tells you what worked in the past. Walk-forward optimization tells you whether your strategy is robust enough to keep working as the market changes.

If the out-of-sample results hold up across multiple rebalancing periods — through rate hikes, drawdowns, recoveries — you have something. If they fall apart, better to know that here than after you've deployed capital.

<img height="402" width="637" src="https://portfolioslab.com/static/user-content/1/ryec7uikdlm7acfynu12vmvd.png" />

*Harry Browne's Permanent Portfolio with a periodic optimization vs the original static allocation.*

### A few ways to use it

Set Reoptimize Frequency to Yearly and compare the walk-forward result against your original allocation. If the optimized weights consistently outperform through different market regimes, that's a meaningful signal about the strategy's durability.

Try different training windows alongside it. A 1-year window will react faster to recent market shifts; a 3-year window produces more stable weights. Walk-forward backtesting helps you see which one actually generalizes instead of just fitting the past.

### Availability

Walk-forward backtesting is available on Plus and Pro plans. Single-point optimization (the default "Once" setting) works the same as before for all users.

Portfolio Optimization now supports walk-forward backtesting. Re-optimize your portfolio quarterly or yearly using only past data.

Platform Updates

Discussions by tag treynor black model