Portfolio Turnover Explained: Reading the Number Right

Two fund fact sheets sit open on my screen. One reports portfolio turnover of 28 percent for the year. The other reports 165 percent. The instinct is to read the first as patient and the second as frantic, then move on. That read is usually wrong, because neither number means much until I know how each fund defined it and over what window. Portfolio turnover is one of those figures that looks precise and comparable, and it’s often neither.

This is an accounting measure rather than a verdict. It tells you how much a book changed over a stated period, and it earns its keep only when you pair it with the method behind it. Get that pairing right and turnover becomes a fast way to read a strategy’s trading behavior. Get it wrong and you’re comparing two numbers that were never on the same scale.

What portfolio turnover actually measures

Turnover captures how much of a portfolio was bought and sold over a period, scaled by the size of the portfolio, and expressed as a percentage. The convention most mutual funds report takes the lesser of total purchases or total sales over the period and divides it by average assets. On that definition, 100 percent turnover means the fund traded the equivalent of its entire book once during the window. Fifty percent means half. Three hundred percent means the whole portfolio was replaced, on average, three times over.

Two details in that formula do quiet work. The denominator is average assets across the period, not the starting or ending balance, so a fund that grew sharply mid-year gets measured against a fairer base. And turnover is a rate, not a trade count: ten small trims can print less turnover than one wholesale switch of a single large position. The percentage answers how much of the book moved, never how many times someone hit the button.

The lesser-of-purchases-or-sales choice is deliberate too. It strips out the buying and selling that comes purely from investors adding or withdrawing money, so the figure reflects the manager’s own trading instead of fund flows. That’s the standard, but it isn’t the only convention in circulation, and that’s exactly where comparisons start to break.

Why the convention and the window change the number

Some reports use the lesser-of method. Others sum gross trading activity, counting both the buys and the sells. Others summarize turnover position by position. These aren’t small differences. Run the same trades through them and you can get headline numbers that differ by a factor of two.

Consider a case I keep in mind. Picture a five-stock book, equally weighted at 20 percent each. Over a quarter the names don’t change at all, but the weights shift to 30, 25, 20, 15 and 10 percent. Adding to the first two positions takes 15 percent of the portfolio; trimming the last two frees the same 15 percent. The lesser-of method reports 15 percent turnover. A gross-activity method adds the buys and the sells together and reports 30 percent. Same trades, same quarter, two defensible answers, and they disagree by double.

The window matters just as much. A figure of 30 percent over three months isn’t the 30 percent an annual report shows. Annualized, that quarterly number is closer to 120 percent. So before you compare two funds, you name the convention and the period for both. Skip that step and the comparison is noise dressed up as a metric. It’s the same discipline that protects you when a strategy adds or drops names through scheduled index reconstitution, which mechanically prints turnover that has nothing to do with a manager’s conviction.

Where turnover actually comes from

A portfolio changes for many reasons, and only some of them involve swapping the list of holdings. Scheduled rebalancing back to target weights creates turnover. So do changing signals, additions and removals from the investable universe, corporate actions like mergers and spinoffs, cash flowing in and out, risk controls that cut an oversized position, and any deliberate change to target weights. Each one moves shares, and every share moved counts.

Cash flows deserve their own mention, because they generate turnover with no change of view at all. A 15 percent inflow that has to be put to work buys across existing positions, and a redemption forces sales to raise cash, both leaving a footprint in the gross figure even though the lesser-of convention is built to net most of it out. Corporate actions do something similar: a merger can retire a holding you never chose to sell, and a spinoff hands you a new line you never chose to buy. The book changed, but the manager’s conviction never moved.

The trap is assuming turnover means the names changed. It often doesn’t. My five-stock example prints turnover with the same five tickers held from start to finish, purely because the weights moved. An equal-weight or risk-parity book can generate steady turnover at every rebalance while holding an almost identical roster, simply because it keeps resizing positions back to plan. Read turnover as a proxy for how many holdings got replaced and you’ll misjudge a book that’s really just rebalancing. Weight discipline shows up as turnover, and that’s a feature of the method, visible in any serious position sizing method that resets exposure on a schedule.

The frictions turnover hints at

Turnover matters because every unit of it drags the portfolio across real-world costs. Spreads come first. Every time you trade you pay part of the bid-ask spread, so a book running 200 percent turnover crosses that spread far more often than one running 20 percent. Then come commissions where they still apply, market impact when an order is large relative to available liquidity, taxes on realized gains, timing differences between the signal and the fill, and the plain operational load of placing and settling more trades.

Notice what turnover doesn’t tell you here. It flags that friction exists and roughly how often you’ll meet it. It doesn’t tell you the size of the bill. A fund at 100 percent turnover trading deeply liquid large caps at a one-cent spread pays a trivial cost. Another fund at the same 100 percent turnover, trading thin small caps, can bleed far more through market impact alone. The number points at the door. It won’t tell you what’s behind it.

Reading turnover against the rest of the picture

On its own, a turnover figure is half a sentence. I read it next to holding period first, because the two are sides of one coin: 100 percent annual turnover implies an average holding period around a year, while 400 percent implies roughly three months. Then liquidity, because the same turnover is cheap in large caps and expensive in micro caps. Then assets under management, because a nimble strategy at 50 million dollars can run a turnover that quietly destroys itself at 5 billion, once its own orders start moving the market. Then the trading schedule, and finally whether performance is reported net or gross of those trading costs.

This is also where turnover connects to a manager’s whole philosophy. A long-horizon trend follower like Ed Seykota runs naturally low turnover, letting winners extend for months, while a short-horizon mean-reversion system churns positions weekly by design. Neither turnover level is better than the other. Each is consistent with a different edge. And when you read a strategy’s reported track record, low turnover paired with a clean backtest still needs the same scrutiny for survivorship bias as any other claim, because a tidy turnover figure says nothing about which names quietly dropped out of the sample.

What a single turnover number can’t tell you

The limit is worth stating plainly. One turnover statistic can’t reveal trade quality, the exact costs paid, the capacity of the strategy, or the reason the portfolio changed. Two funds can both report 80 percent turnover while one earned its trades through disciplined risk cuts and the other churned on noise. The figure treats them identically. It also floats free of return: a portfolio that holds the same names all year prints near-zero turnover, and another that replaces several holdings can report the very same return before costs. Turnover describes the activity, not the outcome.

Cross-strategy comparison is where this bites hardest. When two managers define turnover differently, or measure it over different windows, ranking them by turnover is a category error. I’ve seen a gross-activity 120 percent sit right beside a lesser-of 60 percent that actually represented more trading, and a reader who took the headlines at face value drew the backwards conclusion. Before turnover tells you anything useful about a strategy, you have to make the two numbers speak the same language.

Reading turnover like an accountant, not a scorecard

Turnover is one of the most useful quick reads on a strategy and one of the easiest to misuse. Treated as a scorecard, it invites lazy verdicts about who’s patient and who’s reckless. Treated as an accounting link between a portfolio’s stated method and the frictions it will meet, it does honest work: it tells you how often the book trades, roughly how long it holds, and where to look next for the real costs. Name the convention, name the window, then read the figure against holding period, liquidity, size, and net-versus-gross reporting. That’s the whole discipline, and it’s enough to keep a single percentage from telling you a story it was never built to tell.

Learn the pattern. Ride the trend. Keep the gains.

Educational content only. Not investment advice. Trading involves risk. You are responsible for your decisions.