DCA +523% vs Averaging Down +2,713%

What Is Averaging Down?

Averaging down is a strategy where you buy additional shares of a stock after its price has fallen, thereby lowering your average cost basis. The opposite (buying more as the price rises to increase your average) is called averaging up.

The core principle is simple. If you bought 100 shares at $100 and the price drops to $80, buying another 100 shares at $80 gives you an average cost of $90. You break even when the price recovers to just $90. But behind this simple math lies the need for unlimited capital, endless patience, and the conviction that "this stock will definitely come back."

The Highest Win-Rate Strategy

In my years of stock investing, I've often wondered: purely in terms of win rate, what strategy wins most often? It's also the most common strategy among beginners.

Averaging down. The strategy of averaging down and waiting indefinitely until the price recovers has the highest win rate. However, it completely ignores time and opportunity cost. And the critical requirement is that your capital must be virtually unlimited.

No matter how good a stock is, if the market turns against it, the price just keeps falling. If you average down by doubling your investment every time the position drops 20%, trying to keep your loss at just -5%, you can quickly calculate how much capital that requires.

Then you just wait until it goes back up. Most stocks have a major rally at least once every 5–10 years. If you keep averaging down and holding until then, you do make significant profits.

This isn't just me, it's the approach of many wealthy individual investors. These people look at monthly and yearly charts.

Averaging Down Must Be Done in Multiples

The key to averaging down is that the additional buy amount must be large enough relative to your existing position for the average cost to drop meaningfully. Fixed small-amount buys become almost useless as the cumulative investment grows.

The math makes this clear.

Setup: Principal ₩10M, purchase price P, stock drops -50% → current price 0.5P

New avg cost = Total invested / Total shares

Original shares = 10M / P
Additional shares = Additional amount / 0.5P  (half price = 2x shares per won)

Bigger Multiples Don't Eliminate Your Loss

Increasing the averaging-down multiple does lower the breakeven rise, but the curve only approaches 0% (it never reaches it. As the multiple grows, each additional unit of capital produces less and less benefit) a classic case of diminishing returns. The "optimal multiple" is the point on the chart where the curve starts to flatten.

Formula summary:

Additional buy = Principal × N

New avg cost = (1 + N) / (1 + 2N) × P

Post-averaging return = (current price / new avg cost) - 1
                      = (0.5 × (1+2N) / (1+N)) - 1

Rise needed to break even = (new avg cost / current price) - 1
                           = ((1+N) / (1+2N)) / 0.5 - 1

The pattern is clear:

• 0.1x (₩1M): Average cost drops from P → 0.917P, only 8%. Return improves from -50% → -45.5%, still needs +83% to break even. This is the limit of fixed-amount averaging.
• 1x (₩10M): Average cost drops to 0.667P, down 33%. Return halves from -50% → -25%, and only +33% is needed to break even.
• 2x (₩20M): Average cost hits 0.6P, only +20% needed to break even. The effect becomes substantial.
• 3x (₩30M): Average cost at 0.571P, only +14.3% to break even. But you need ₩30M in additional capital.

In short, averaging down only works meaningfully at 1x or more, ideally, you need to invest at least as much as your current position. This is the mathematical basis for "you need unlimited capital." If you average down 1x at -50%, your total invested doubles. If it drops another -50%, you need another 1x, making it 4x, 8x, 16x... growing exponentially.

The simulations below show how this principle plays out with real data.

Simulation: Doosan Enerbility (A034020) 2009–2026

The chart below shows Doosan Enerbility's monthly candlesticks from January 2009 to February 2026, along with the average cost basis trend lines for three simulation strategies. When the price (candle) is above the cost basis (line), you're in profit; below means a loss. The markers (▲) indicate Simulation 2's buy points.

Simulation 1. DCA Long-Term (Fixed ₩1M/Month)

Simulation 2. Averaging Down (Buy 1x Cumulative When Avg Cost -10%)

Buy Event Summary (30 total)

Simulation 3. Improved Averaging Down (Avg Cost -10% + Green Candle Only)

Buy Event Summary (20 total)

Months Where Sim 2 Bought but Sim 3 Skipped (red candle = skip)

Three-Strategy Comparison

Simulation 3 Key Takeaways:

• A single green-candle filter reduced buys from 30 → 20, cutting capital from ₩536.9T → ₩524.3B (approx. 1/1,024)
• Skipping 10 buys produced a 2^10 = 1,024x capital savings (exponential structure effect)
• Returns dropped only slightly from 2,713% → 2,456%, but required capital went from astronomical to large-corporation level (₩524.3B)
• By not buying during red candles, you avoid "catching a falling knife" and only enter after a confirmed bounce

Drawdown Period Analysis

Simulation 1 (DCA): 134 months in loss / 206 total (65.0%)

Simulation 1: Periods below -50%: a staggering 44 months (3 yrs 8 mo)

The reason is that for the fixed monthly amount to meaningfully lower the average cost, the new investment needs to be a significant multiple of what's already invested. By this point, so much capital had accumulated that ₩1M/month barely moved the needle, the averaging effect became negligible as time went on. Ultimately, the profit came not from averaging down but from having accumulated a large position by the time the price finally surged.

Simulation 2 (Averaging Down): 67 months in loss / 206 total (32.5%)

Simulation 2 aggressively lowers the average cost with each buy by investing 1x the total accumulated capital. As a result, the return never reached -50%. The worst was -18.4% in March 2020, and months below -10% totaled just 5. There's virtually no psychologically devastating period. The trade-off: ₩536.9 trillion in required capital, completely unrealistic.

Simulation 3 uses a green-candle filter to reduce buys, resulting in a higher average cost than Sim 2 (₩4,070 vs ₩3,711) and longer loss periods (96 vs 67 months). However, the worst return of -47.4% never breached -50%, and the required capital of ₩524.3B is 1/1,024th of Sim 2.

Comprehensive Summary

1. Prerequisites

• The stock must be in a confirmed sideways consolidation after bottoming or in a long-term uptrend. Averaging down on stocks headed for delisting or permanent decline leads to total loss.
• Simulation 2 (buy 1x cumulative at -10%) keeps drawdowns manageable: 67 months in loss (32.5%), worst at -18.4%. But it requires ₩536.9 trillion over 30 buys.
• Simulation 1 (fixed ₩1M DCA) is realistic capital-wise, but demands enduring 134 months (65%) in loss, the longest streak of 8 years 8 months, and a worst return of -82.2%.
• Simulation 3 (green-candle filter) reduces required capital to ₩524.3B while achieving a 2,456% return.

2. Constraints

• Fixed-amount averaging loses effectiveness as total invested grows. Adding ₩1M to a ₩200M position barely changes the average cost.
• 1x-cumulative averaging is effective but requires capital to double with each buy: ₩1M → ₩2M → ₩4M → ... → ₩268.4T. This is what "you need unlimited capital" actually means.
• Adding a green-candle filter skips 10 buys for a 2^10 = 1,024x capital savings, but you miss low-price entries and end up with a higher average cost.
• All three strategies require the stock to survive. Doosan Enerbility surged after 2020, but not every stock does.

3. Opportunity Cost

• In the DCA simulation, from March 2012 to October 2020: 8 years and 8 months of continuous losses. During this same period, the S&P 500 returned approximately 200%.
• In the averaging-down simulation, ₩536.9T was locked in a single stock. Even a fraction of that diversified into rising stocks would have produced far better risk-adjusted returns.
• Both strategies use time as a weapon. For time to be a weapon, you must sacrifice potential returns elsewhere (opportunity cost).
• Ultimately, averaging down requires the conviction that "this stock will definitely go back up," and if that conviction proves wrong, you lose both time and capital.

4. Did DCA Actually "Work" in Simulation 1?

Simulation 1 involves monthly ₩1M "averaging down," but the profit didn't come from the averaging effect. The proof: 8 years 8 months of consecutive losses from March 2012 to October 2020. Fixed monthly contributions barely move the average cost once the position grows large. Adding ₩1M to a ₩100M position changes the average by less than 1%.

The real reason Simulation 1 achieved +523.5% returns is that Doosan Enerbility surged 34x from ₩3,028 to ₩104,400 after 2020. Without that surge, the position would have remained deeply underwater.

In other words, Simulation 1 is not a case where averaging down "worked"; it's a case where a massive price surge rescued a losing position. The averaging down didn't create the profit; the large accumulated position simply converted into profit when the stock finally surged. Had Doosan Enerbility continued trading in the ₩10,000–20,000 range, it would still be at -50%+ losses after 17 years.

5. When Does Averaging Down Actually Work?

For a stock like the NASDAQ that trends consistently upward on monthly charts, no strategy is needed, just buy and hold. Averaging down is inherently a "decline-then-recovery" strategy, so it doesn't even trigger on uptrending stocks.

Averaging down can be effective in two types of chart patterns:

A. Crash → Sideways → Rally (Doosan Enerbility type)

    ╲
     ╲___________╱╱
     Crash  Sideways  Rally

• This is where averaging down shines. You lower your cost during sideways trading and reap explosive profits on the rally.
• Risk: Extreme opportunity cost and capital freeze. As the Doosan simulation shows, sideways periods can last 8–10 years. Capital is locked up with no guarantee of when (or if) the rally comes.
• Worst case: the sideways phase never ends, or leads to further decline. "I thought it was the bottom, but there was a basement" happens frequently in practice.

B. Surge → Sideways → ? (High-consolidation type)

         ╱‾‾‾‾‾‾‾╲?
    ╱╱╱           ╲?
    Surge  Sideways  Crash?

• When a stock that's already rallied hard consolidates at highs, you average down on small dips while hoping for further upside.
• At least conclusions come quickly. If it rallies further after consolidation, profit; if it drops, you cut losses quickly.
• Risk: Recovery from a crash is virtually impossible. Since you averaged down from highs, your average cost is elevated. A crash can push you to -80% like Simulation 1. A stock down 50% from its peak needs to rally 100% just to return to break-even.

C. When Averaging Down Doesn't Work

6. Strategy Diversification: Shortening the Buy Cycle

The simulations above are all long-term strategies based on monthly candles with buys at month-end. However, the same averaging-down logic can be adapted to a shorter buy cycle for a medium-term approach.

For example, if your investment horizon is one year, you could apply the same strategy (buy 1x cumulative when avg cost -10% and green candle) on a weekly basis every Friday. With 4 decision points per month instead of 1, you get finer timing and faster responses during sideways periods.

The catch is that capital requirements remain significant. In fact, since the buy cycle is shorter, running this strategy over a long period causes required capital to grow even faster than in the simulations above. Therefore, weekly-based averaging down should be operated as a take-profit-and-reset cycle, closer to medium-term rotational trading than long-term holding.

Conclusion: Averaging down is only valid for stocks you are genuinely certain will recover. That conviction must be grounded in business fundamentals, industry outlook, or broad indices (KOSPI, S&P 500), essentially betting on an economy, not blind faith in a single stock. Averaging down with vague conviction about individual stocks can lead to the worst possible outcome: losing both time and capital.

That said, when conditions are met, it produces the highest returns (that's also a fact. Simulation 2's +2,713% and Simulation 3's +2,456% are 4–5x the return of DCA (+523%) over the same period. If the stock's fundamentals are intact, you have sufficient capital reserves, and there's clear evidence supporting a recovery) averaging down isn't just "holding through losses"; it can be the most aggressive yet logical cost-basis management strategy available.