In July the Morgan Stanley TMT momentum pair suffered its largest five-week drawdown since the series began in 2000. Larger than the dot-com unwind, larger than the GFC, larger than COVID. The “biggest winners minus biggest losers” trade, which had been printing money for three years, gave back a year of gains in a few weeks. Leverage, narrow leadership and crowded positioning all broke at once. Extraordinary stuff..
Three months on, the market has done what it usually does after a fright. It has climbed back. The ASX 200 had its own moment in March, briefly dipping more than 10% from its high before bouncing, and the April 2025 tariff sell-off (-15.8% peak to trough) was fully recovered within weeks. The pattern is familiar enough that it should no longer surprise anyone, and yet every time it happens the same two mistakes get made. Investors sell into the hole, or they sit in cash waiting for a bottom that is only visible in the rear-view mirror.
This note is about the third option: having a plan for the sharp pullback before it arrives, and using it to ride the cycle back up.
Pullbacks are the rhythm, not the exception
Ben Carlson at Ritholtz has done the simplest and most useful work on this. Using S&P 500 data back to 1950, a 10% correction arrives roughly every 1.8 years and a bear market roughly every five. Just over half of all 10% falls went on to become 15% falls. Capital Group’s long-run data says the average bear market since 1950 has lasted about 12 months and cost 33%, while the average bull has run 67 months and returned 265%. The arithmetic of that asymmetry is the whole argument for staying in the game.
What happens if you buy into the hole? Carlson’s March note measured forward returns from every month-end where the index sat 10%, 20% or 30% below its high. Buying at a 10% drawdown averaged roughly +15% one year later, +42% after three and +72% after five. From a 30% drawdown the numbers were +21%, +48% and +88%. Dimensional, using the Fama-French total market series, gets the same shape: five years after a 10%, 20% or 30% decline, average cumulative returns all exceed 50%. Losses over those horizons were uncommon. Not impossible, uncommon.
I tend to feel that the first job in any sharp pullback is simply to remember those base rates. The second job is to distinguish which kind of pullback you are in.
The best days hide inside the worst weeks
The reason “step aside and come back when it’s calmer” fails is mechanical, not psychological. JPMorgan Asset Management’s data over the last twenty years shows seven of the ten best days in the market occurred within two weeks of the ten worst. Invesco counts 24 of the 30 best days of the past 30 years falling inside the tech wreck, the GFC and the pandemic. Miss the ten best days between 2005 and 2024 and a $10,000 investment in the S&P 500 ends at $32,871 instead of $71,750. The best days are not scattered randomly through bull markets. They are the violent snapbacks that happen while everyone is still terrified.
Schwab ran the cleanest version of this experiment. Five hypothetical investors, twenty years, the same annual cheque. The one with perfect timing finished with $186,007. The one who invested the day the money arrived, every year, with no view at all, finished with $170,555. The one who sat in cash waiting for the right moment finished with $47,357. Perfect timing was worth about 9% over two decades. Bad timing cost you three quarters of your outcome. That is the trade-off the “I’ll wait for the bottom” crowd is actually making.
Nobody rings a bell at the bottom
Warren Buffett wrote “Buy American. I Am.” in the New York Times on 16 October 2008 with the S&P 500 already down 40% from its peak. He said plainly that he had no idea whether stocks would be higher or lower in a month or a year. The index then fell a further 28% before bottoming on 9 March 2009. He was five months and 28% early, and it was still one of the great calls of the era, because the question was never “is this the low” but “is this price cheap against what these businesses will earn over the next decade”.
David Tepper went on CNBC on 23 March 2020 and said he was buying “a little” tech and healthcare, while still expecting the market to fall another 10% to 15%. That was the day of the low. He did not know it. He was scaling in.
Howard Marks wrote “The Limits to Negativism” on 15 October 2008, in what he called the greatest panic he had ever seen. His point was that scepticism has to cut both ways. We are trained to be sceptical of euphoria, but almost nobody is sceptical of pessimism when it is the consensus, and at some point the price already discounts every bad outcome being discussed. In March 2020 he put the same idea in practical terms: since the bottom cannot be timed, you buy as prices fall, in steps, moving the portfolio from defence to offence as the discount widens. Not all at once, and not after the all-clear.
The hardest part of any sharp fall is working out what kind of selling you are looking at. A share price that has halved can be telling you two completely different stories. Sometimes the business has changed: the capex budget has blown out, the contract has been lost, the earnings are genuinely lower. That deserves a lower price. More often than people think, the selling is mechanical: a placement’s new shares still digesting, a sector rotation, an index rebalance going the wrong way, a margin call ripping through a crowded register. None of that says anything about what the business will earn, and mechanical sellers eventually run out of stock. Same chart, two different answers. That distinction is the entire skill. A fully funded business trading near book value because a rebalance went against it is a pullback you buy. A business whose earnings have changed is not, whatever the chart looks like.
What disciplined money actually does
Having watched a lot of these, the people who consistently come out of sharp pullbacks richer do a handful of unglamorous things.
1. They decide the rule before the fall.
Vanguard’s rebalancing research recommends a simple 5% drift threshold. The point is not the number. The point is that a rule forces you to buy what has fallen when your instincts are screaming the opposite. Morgan Stanley’s list of the biggest mistakes in volatile markets has, at the top, failing to rebalance back into equities after the drop. People rebalance out of winners happily and never rebalance into losers.
2. They buy in tranches, not in one go.
Buffett’s October 2008 and Tepper’s March 2020 were both partial. Marks calls it calibrating. A third at the first 10%, a third at 20%, a third held back. You will never buy the low with all of it, and you will never miss the recovery with none of it.
3. They treat volatility as a regime, not a trigger.
StoneX’s work on the VIX shows that when the index has been above 30, the following twelve months returned +22.3% on average and were positive 87.6% of the time. The short-horizon numbers after a spike above 50 can still be ugly, so it is not a buy signal in the sense of “buy today”. It tells you that you are in the part of the cycle where the forward returns are earned, and that cash on the sidelines is now expensive.
4. They expect the junk to lead the first leg.
Daniel and Moskowitz’s “Momentum Crashes” paper explains why the most beaten-down names rip hardest off a bottom: after a panic, the losers behave like a leveraged call option on the market. That is why the first fortnight of every recovery looks like a rubbish rally and why the quality names only take over later. Knowing this stops you selling the good businesses in disgust in week one.
5. They keep a watchlist and dry powder, and they know where the dry powder comes from.
The investors who did best out of March 2020 were not the ones who predicted it. They were the ones with a list of businesses they already wanted at prices they had already decided on, and the capacity to act when the phone rang. For a founder or a substantial holder, that capacity often does not mean cash sitting idle. It means knowing in advance what you can borrow against the holding you already have, on what terms, before the lender is nervous too. We spend a great deal of time at Gleneagle on exactly that conversation, and it is a far better one to have in the calm than in the panic.
The honest caveat
Nick Maggiulli at Of Dollars and Data ran the ultimate version of “buy the dip”. Give an investor perfect foresight of every bottom between 1920 and 1979 and let him hold cash until each one arrives. Over 40-year windows, that God-like dip buyer still lost to a dull monthly investor more than 70% of the time. Coutts found the same thing on more recent data: investing on 1 January every year from 1990 to 2024 returned 12.1% a year, while waiting each year for a 10% fall returned 6.6%. Exploitable dips are rarer than they feel, and cash waiting for them is a cost every single day.
So the edge is not in waiting. The edge is in being invested, having the rule and the list and the capacity ready, and then acting when the sharp pullback arrives rather than freezing.
The local twist
One more thing before anyone treats every dip as April 2025. Market Index counts fifteen ASX 200 corrections since 2000 and the one-year forward returns range from +30% (June 2012) to -41% (January 2008). Just over half of our 10% falls deepened further, and the ones that did were the ones that ran into a recession or a rates shock. With the RBA’s Deputy Governor telling the Citi conference last week that the inflation outlook is “materially worse” than August, a fifth hike this year now better than a coin toss for Melbourne Cup day, and a cash rate heading to its highest since 2008, the next Australian pullback may be a rates story rather than a positioning story. Those take longer to resolve and the first leg down is rarely the last.
Which is the point of the whole note. A sharp pullback is only an opportunity for the investor who has already worked out which businesses they want, what price makes sense, what the rule is, and where the capital comes from. Everyone else is just watching the screen.
Let’s see what comes.
