Six Years of Buying on the 15th, and the Second Tab That Broke the Ritual

At 9:05 on a Sunday mornin last November, with the kettle arguing on the stove behind me, I sat at the kitchen table with my laptop open to the second tab of a spreadsheet I had been avoiding for months, a printed stack of six years of statements at my left elbow. The tab had 72 rows. Each row was one month of my investing life from March 2019 thru February 2025, and row 41 was about to end an argument I had been having with myself since my first payroll deposit.

The argument was about $43,200. That figure is $600 a month for seventy-two months into FXAIX, an S&P 500 index fund, plus a parallel $200 a month into QQQM, an ETF logging the Nasdaq-100, snagged on the same day every month because my stomach had decided the 15th was the cheap day. My stomach had opinions. It had never once been asked to show its work.

Row 41 was May 2022, and next to it in column F sat the price I paid on the 15th beside the price on the 2nd of the same month, a gap of 4.1% that no superstition of mine had ever accounted for. The question the whole tab existed to answer was simple and slightly humiliating: had six years of ritual buying actually snagged me anything? The answer lived in row 73. I sat there a moment before scrolling down.

the birthday of a superstition, March 2019

The superstition had a birthday: March 15, 2019, my first automated purchase, three days after my first real paycheck from the school district, and the fund promptly rose 1.8% that fortnight, which my twenty-nine-year-old brain filed as evidence instead of coincidence. Payday was the 15th and the 30th. The money went in the day it arrived, half from each check, and for six years every dip that happened to follow a mid-month purchase reinforced the feeling that the calendar was on my side.

Confirmation bias is patient. It will take six years of Tuesdays to build a shrine outta two lucky Fridays, and it never once volunteers the months where the 15th was the most expensive day on the calendar, like April 2021, when I snagged within 0.3% of that year's high and felt clever doing it. The spreadsheet doesn't take sides. That is what I built it for.

the payday that moved to the 26th

The ritual got stress-tested in September 2024, when I changed districts and payday moved to the 26th, and I discovered I had genuine feelings about this — a flash of anxiety at the idea of buying on "the wrong day," as if the S&P kept a ledger with my name on it. I kept buying on the 15th anyway, moving money manually, outta what I can only call superstition with a routing number. The experiment had to happen after that.

four weekends of clicking through statements

Building the second tab took four weekends of clicking through six years of statements, pulling closing prices for the 2nd, the 8th, the 15th, the 22nd, and the 28th of every month, and simulating what each $600 would have snagged instead. Where a target date fell on a weekend, I used the Friday close, the same rule for every column so the comparison stayed honest. Seventy-two rows. Five columns of prices. One column of quiet shame.

$61,780 versus $62,043

Here is row 73. Buying strictly on the 15th for seventy-two months turned $43,200 of contributions into $61,780 by February 2025. Buying on the 2nd of the same months, identical dollars, identical fund, turned the same $43,200 into $62,043. The 28th came in at $61,905, the 8th at $61,660, and the 22nd, which I had privately nominated as my new lucky day, at $62,010. The spread from best day to worst day, across every day I tested, was $383 on a $62,000 account — six tenths of one percent, six years of ritual, priced at about $64 a year.

the four months that did all the damage

I figured out, after sorting column F from largest to slight, that nearly the whole gap came from four months: May 2022, October 2022, April 2024, and September 2024, when mid-month purchases happened to land within days of local bottoms or tops, and row 41 alone — May 2022's 4.1% — accounted for more than a third of the total. Randomness dressed up as signal. The 15th didnt outperform or underperform over any stretch longer than a quarter; it wore whatev the month handed it, and I spent six years reading the outfit as intention.

the 26th, minus the superstition

The new rule took one sentence and no luck: money gets invested the day payroll lands, on the 26th, and any cash waiting between paychecks sits in the high-yield savings account earning 4.25% instead of idling in checking where it whispers about meme stocks. The automation does the buying. Two months in, I found I dont even look at the execution price anymore, and that is the actual dividend the experiment paid — not basis points, attention reclaimed.

the ritual, retired with honors

The second tab is still in the spreadsheet, and on the first Sunday of this July I opened it again at 9:05 in the morning with the kettle going, purely to visit, the way you reread a letter that once changed your mind. Row 73 still says $263. I checked the 26th too, out of curiosity; it comes in at $61,998, statistically indistinguishable from the 15th and from everythin else, which is exactly the point and exactly the relief. The 15th isn't lucky. It never was. The buying was the whole trick, and it works on any day the market is open.