Volatility Harvesting for Palantir
The Market Wave Algorithm for Equities against Palantir in three windows. Buy and hold and DCA beaten in all three.
Palantir: the AI software stock that doubled to $207, nearly halved to $106, and climbed back to $190.
Why this case study, and why now
Palantir moves. It rose 24% in a day in February 2025, doubled to $207 by November, fell to $106 by June 2026, and rose 29.5% in a day this August. Holders felt every swing. Buy and hold turned none of them into cash.
This study asks whether those swings can be harvested. Each window uses a recipe of the Market Wave Algorithm for Equities built to buy and sell large portions of the position.
What did a year that went nearly nowhere pay?
Strategy+112.6%
- Buy and hold
- +7.7%
- DCA
- +26.8%
What if you bought the gap?
Strategy+277.3%
- Buy and hold
- +84.2%
- DCA
- +36.4%
What if the recipe reads smaller waves?
Strategy+312%
- Buy and hold
- +84.9%
- DCA
- +35.7%
Every run starts from the same hypothetical $100,000 in cash and zero shares.
Backtested results. Past performance does not guarantee future results. For educational purposes only.
The subject
Palantir builds data and AI software for governments and large companies. Its AI Platform turned it into one of the fastest growing software companies in the market, and one of the most debated.
- Feb 4, 2025Fourth quarter results beat. The stock rises 24% in a day to a record near $104.
- Feb to Apr 2025It slides from about $125 to below $70 in the April selloff.
- Nov 3, 2025All time high of $207.52, roughly double the February gap.
- Feb 2026Revenue up 70%, guidance far above consensus. An 11% pop, then a slide with the software selloff.
- Jun 25, 2026A low of $106.37, down about 40% for the year. Its worst month on record.
- Aug 4, 2026Revenue up 93%. The stock jumps 29.5%, from $125.65 to $162.66.
- Sep 23, 2026Closes at $191.79, its highest in nearly a year. Near $189 today.
Here is the catch that matters for this study. The business kept delivering and the stock kept swinging. Holders lived through every move and were paid for none.
The investor's dilemma
Palantir rewards the story and punishes the ride. It sets up three failure modes that human investors handle badly.
Rose in a day this August
Chasing the gap
Buying the day after a 24% jump feels like buying the top. Within ten weeks it was down more than 30% from there. The natural responses were to sell in April or to freeze.
The paper gain trap
$100,000 at the February 2025 gap was worth nearly $200,000 at the November peak. By June 2026 it was worth about $102,000. Almost a year of gains, gone, and none of it banked.
Movement as stress
Palantir moved 24% in a day in 2025 and 29.5% in a day in 2026. To a holder, that is something to endure. To a rules-based seller, it is inventory.
DCA deserves credit. In 2026 it kept buying through the slide to $106 and finished ahead of buy and hold. But it buys blind, never sells, and never realizes a dollar. Every swing passed straight through it.
A rules-based algorithm turns the swings themselves into the return.
The algorithm
The Market Wave Algorithm for Equities does three things a human holder struggles to do at the same time.
Ride confirmed moves
It buys when price breaks from the wave with a buffer. It does not need to know why the stock is moving.
Exit on rules, not nerves
It sells a set share of the position when a target is hit. It sold into the November peak and the August surge.
Step aside in chop
When the signal is unclear, it waits in cash instead of guessing.
Every recipe here moves most of the position. The two runs from the February 2025 gap buy with 93% to 98% of cash and sell 99% to 100% on an exit. The 2026 run buys with 99%, sells 70% at a time, and only trades below and above the wave.
Macro recipe, buy below the wave, sell above it.
Mid scope recipe, near full swings.
Micro recipe, near full swings.
Backtested results. Past performance does not guarantee future results. For educational purposes only.
“The 2026 harvest”
January 1, 2026 to October 1, 2026 · Hourly chart · Macro recipe, buy below the wave, sell above it
Buy and hold earned +7.7%. DCA, buying monthly through the decline, earned +26.8%.
The algorithm bought each deep dip below the wave, sold each rebound above it, and sold out into the August rally.
+112.6%
- Buy and hold +7.7%, outperformance +104.9 percentage points
- DCA +26.8%, outperformance +85.8 percentage points
- Ending value
- $212,642
- Realized profit
- $112,573
- Cash
- $212,454
- Shares held
- 1
Backtested results on the hourly chart, valued October 1, 2026. Past performance does not guarantee future results. For educational purposes only.
“The gap buyer”
February 4, 2025 to October 1, 2026 · Daily chart · Mid scope recipe, near full swings
Suppose you bought the day Palantir gapped up 24%, at about $104: the entry most investors regret. You then rode a slide to below $70, a run to $207, a fall to $106, and the August surge. Buy and hold earned +84.2%. DCA earned +36.4%.
The algorithm bought the spring 2025 lows, sold into the summer and November highs, and rebuilt in the 2026 decline.
+277.3%
- Buy and hold +84.2%, outperformance +193.1 percentage points
- DCA +36.4%, outperformance +240.9 percentage points
- Ending value
- $377,319
- Realized profit
- $244,603
- Unrealized profit
- $32,715
- Shares held
- 1,937 at $174.33 avg cost
- Cash
- $6,926
Backtested results on the daily chart, valued October 1, 2026. Past performance does not guarantee future results. For educational purposes only.
“The gap buyer, finer wave”
February 4, 2025 to October 1, 2026 · Six hour chart · Micro recipe, near full swings
Same entry, same $100,000, same twenty months. The change: a faster chart and a micro scope of 1, so the algorithm reads smaller waves and trades more of them. It still commits 93% on a buy and sells 99% on an exit. Buy and hold earned +84.9%. DCA earned +35.7%.
+312%
- Buy and hold +84.9%, outperformance +227.1 percentage points
- DCA +35.7%, outperformance +276.3 percentage points
- Ending value
- $412,055
- Realized profit
- $279,782
- Unrealized profit
- $32,272
- Shares held
- 2,028 at $174.27 avg cost
- Cash
- $26,370
Backtested results on the six hour chart, valued October 1, 2026. Past performance does not guarantee future results. For educational purposes only.
Side by side
Read the realized profit row. Buy and hold and DCA never sell, so they banked nothing.
| The 2026 harvest | The gap buyer | The gap buyer, finer wave | |
|---|---|---|---|
| Window | Jan 1 to Oct 1, 2026 | Feb 4, 2025 to Oct 1, 2026 | Feb 4, 2025 to Oct 1, 2026 |
| Chart | Hourly | Daily | Six hour |
| Recipe | Macro scope 16, buy below, sell above | Scope 8, near full swings | Micro scope 1, near full swings |
| Strategy return | +112.6% | +277.3% | +312% |
| Buy and hold | +7.7% | +84.2% | +84.9% |
| DCA | +26.8% | +36.4% | +35.7% |
| vs buy and hold | +104.9 pts | +193.1 pts | +227.1 pts |
| vs DCA | +85.8 pts | +240.9 pts | +276.3 pts |
| Ending value | $212,642 | $377,319 | $412,055 |
| Realized profit | $112,573 | $244,603 | $279,782 |
| Shares held | 1 | 1,937 | 2,028 |
| Cash | $212,454 | $6,926 | $26,370 |
All three from $100,000 and zero shares.
Backtested results. Past performance does not guarantee future results. For educational purposes only.
Why the results differ
The three runs are the same algorithm answering different questions.
The 2026 harvest asks: can a near-flat year still pay?
Answer: The macro recipe buys only below the wave and sells 70% above it. $112,573 in cash.
The gap buyer asks: what if I bought at the worst-feeling moment?
Answer: The daily recipe swings nearly the whole position in and out, banking each leg.
The finer wave asks: what if the algorithm reads smaller swings?
Answer: More of Palantir's movement becomes realized profit.
Buy and hold made one decision and lived with it.
The edge is not prediction. It is getting paid on the swing itself.
Backtested results. Past performance does not guarantee future results. For educational purposes only.
The real lesson
- 01
Volatility is inventory
A stock that moves 30% in a day is hard to hold and easy to sell into. Rules turn the movement holders endure into realized profit.
- 02
A bad entry is not a bad outcome
Both twenty month runs started the day after a 24% gap, then fell more than 30%. Rules made far more of that entry than conviction did.
- 03
Cash is a position
The 2026 run ends with $212,454 and one share. If Palantir falls again, it has the cash to buy. If it rises, the year is already paid.
The client decision is how you want to be paid: swings harvested into cash, or harvested into a larger position. The Market Wave Algorithm for Equities has a recipe for each.
Backtested results. Past performance does not guarantee future results. For educational purposes only.
A path forward
Arch Public runs the Market Wave Algorithm for Equities in your own brokerage account.
Zero custody. Full control.
Rules, not reactions.
What do you want your account doing the next time Palantir moves 30% in a day?
Buying Palantir's February 2025 gap and holding returned
+84.9%
The same $100,000 under the six hour recipe returned
+312%
with $279,782 realized.
In every run the algorithm banked more than its starting capital in profit: $112,573 in nine months, $244,603 and $279,782 over twenty. The swings holders endured are the swings the algorithm sold.