Rebalance timing luck for a monthly Russell 3000 momentum strategy (20 top-quality stocks from the top 100 momentum names among the 1,000 most liquid) is a 3.5% CAGR spread between luckiest and unluckiest reformation days (1991-2024; daily data; assumed Interactive Brokers costs).
How material is the rebalance timing luck (RTL) induced by picking a trading day to reform a monthly stock momentum strategy? Is there a way to manage the risk of bad luck? In their November 2025 paper entitled "The Tranching Dilemma. A Cost-Aware Approach to Mitigate Rebalance Timing Luck in Factor Portfolios", Carlo Zarattini and Alberto Pagani investigate monthly momentum portfolio tranching, holding multiple portfolios with the same strategy but with different reformation cycles, as a way to manage RTL. Their test strategy each month:
Identifies the 1,000 most liquid components of the Russell 3000 Index.
Finds the 100 stocks with highest total returns from 12 months ago to one month ago.
Reforms an equal-weighted portfolio of the 20 out of these 100 stocks with the highest momentum quality based on the percentage of days with positive and negative returns during the ranking interval.
They run this strategy with reformation cycles from the close on trading day one of each month to the close on trading day 20 (or last) of each month. They then consider effects on RTL of holding 2, 4, 5, 10 and 20 tranches across these cycles. They assume portfolio reformation frictions as standard Interactive Brokers fee of $0.0035 per share with minimum $0.35 per trade (doubled for sell transactions to account for SEC clearing fees). Using daily data for Russell 3000 Index components during 1991 through 2024, they find that:
U.S. Stock Market End-of-Quarter Effect PremiumS&P 500 quarter-end (QEND) 8-day windows showed modestly higher average returns than turn-of-the-month windows, with Q3 consistently weak across subperiods (daily closes, Jan 1928–mid-Aug…
Turn-of-the-Month Effect for Currencies? PremiumTurn-of-the-month (eight trading days centered on month-end) average abnormal returns for UUP, FXE and FXY versus all eight-day intervals were very small, with no…
Does the Turn-of-the-Month Effect Work for Asset Classes? PremiumAverage abnormal turn-of-the-month returns across 10 asset class ETFs (vs all eight-trading-day intervals) were 0.4% in first sample halves versus 0.1% in second halves…
Does the Turn-of-the-Month Effect Work for Sectors? PremiumAverage abnormal turn-of-the-month returns across nine Select Sector SPDR ETFs declined from 0.6% (first half) to 0.1% (second half) and 0.0% (last fourth) (Dec…
Stock Market Behavior Around Mid-year and 4th of JulyFreeS&P 500 shows modest positive bias around mid-year through July 4th (0.087% vs 0.036% average daily return; 1950-2024; 10-day windows), but inconsistencies across subperiods…
