The Two Times I Bought All-Time Highs and the 746 Days Between Them

At 3:48 in the afternoon on Monday, January 3, 2022, I sat at the nurses' station on the fourth floor with a browser window open to my brokerage, an order for $18,000 of VTI filled in and waiting, and the S&P 500 ticker reading 4,796 — a number the index had never closed above. My hands were shaking. Not metaphorically. The desk phone rang twice while I sat there, some family member with a question about room 412, and I let it ring because I was busy doing the exact thing a hundred forum threads had warned me against: buying an all-time high with my entire annual bonus.

The $18,000 was a retention bonus, taxed and wired, equal to eleven weeks of take-home pay, and it was earmarked for nothing in particular, which made it the most dangerous money I owned. Three years a charge nurse. The plan was to put the whole bonus into a broad-market ETF and let it sit til 2047, and the only obstacle between me and that plan was that the market had just spent thirteen months climbing almost without breathing.

I clicked submit at 3:51, and the fill came back three cents above the price I had been staring at, which felt like an omen and was in fact just spread. What happened over the following 746 days is the reason I am writing this, cuz the second all-time-high purchase — the one I made on January 19, 2024, at a record close of 4,839 — cost me nothin in nerves the market had not already billed me for once. Same desk. Same ticker. Different hands.

the $3,840 paper loss by October

The first month was fine. Then 2022 arrived in earnest: rate hikes, a war, a June inflation print that read 9.1%, and by late September my $18,000 of VTI showed a balance of $14,160 — a paper loss of $3,840 that I recalculated on the bus home with increasing precision and decreasing composure. The index's drawdown bottomed near 25% in October. Mine sat at 21%, slightly cushioned by the dividend, a cushion I clung to like a raft.

the October 2022 conversion on the way down

I did not sell. I want that on the record in a boring sentence, cuz the interesting sentences belong to what I did instead: I kept the $500 automatic monthly buy running through the whole decline, added a second $1,000 purchase in October 2022 at prices 24% below my first fill, and ran a Roth IRA conversion of $28,000 the same month, deliberately, on the way down, because converting at depressed values meant paying tax on a modest number. My accountant and I hashed out the conversion on a Tuesday call. He figured I was gloomy. I was greedy in a specific, spreadsheet-flavored way.

746 days to get even

The index needed 746 days to close above 4,796.56 again. From January 3, 2022, to January 19, 2024, my original $18,000 spent approximately two full years underwater, and for most of 2023 it hovered between 8% and 12% below my fill — close enough to breakeven to be taunting, far enough to sting on red days. The monthly $500 buys, meanwhile, filled a notebook of lots at prices from $238 down to $181. Water finds its level. So does patience, eventually, at a compounding rate.

Getting even is an underrated genre of victory. On January 19, 2024, when the S&P closed at 4,839.81, my January 2022 lot was back above water by 0.9%, a margin so thin I could have lost it to a bad week, and I remember feeling not triumph but vertigo — the realization that the market had simply resumed, as if two years of my attention meant nothing to it. The market doesnt know weve met. I have found that comforting ever since.

the second ticket at 4,839.81

That afternoon I snagged again: $12,000, moved deliberately outta the high-yield savings account where it had been earning 4.6% while I waited for a dip that never arrived, into the same ETF, at a new all-time high, from the same fourth-floor desk, with the same order screen and a different set of hands doin the clicking — steadier, or at least steadier by half. The record close was 4,839.81. I know the number by heart. I typed the order without rereading it.

the dip that arrived two years late

The dip I had been waiting for since October 2022 ultimately showed up in April 2025, seventeen months after the second purchase, when tariffs took better than 10% outta the index in a week, and my hands, for the record, stayed perfectly still while I added that month's $500 on schedule. The cash that waited in savings from 2022 to 2024 earned 4.6% while the ETF returned 21% over the same stretch. The waiting cost about $1,970. I did that math once, in a weak moment, and filed it under tuition as well.

746 days of tuition

The second purchase is up 29% as I write this in March 2026, and the first — the one that spent 746 days underwater — is up 33% including dividends, which means the "worst possible entry" of January 2022 quietly beat the 4.6% savings account it competed against in my head the entire time. I keep both lots in a note on my phone labeled DO NOT SPLIT. Mostly it reminds me what splitting costs. The math forgave me first.

The second click was easier because the first one had paid the tuition. I knew, sitting at that desk in January 2024, exactly what a 20% drawdown felt like in my hands, my sleep, and my bus-ride arithmetic, and I knew the monthly buys had turned the 2022 hole from catastrophe into rounding error by 2025, and that knowledge is the only edge I have possessed in fourteen years of owning the same boring fund. Not timing. Not nerve. An accurate memory of surviving.

the phone at 3:48 PM

Last Monday in March, at 3:48 in the afternoon, the desk phone at the nurses' station rang while I was charting, and I picked it up on the second ring — a family member with a question about room 412's visiting hours, as it happens, cuz some things are permanent. Afterward I glanced at the ticker out of habit: 4,796 is long gone, buried under two years of gains, and the shaking went with it. I snagged records twice. The records didnt care. That was the good news both times.