The Real Risk of One Stock

Concentration risk is the danger that too much of your net worth rides on a single stock. One name can move several times as much as the broad market over a year, and it can fall much further from its own high. This page measures both: the annualized volatility and the peak-to-trough drawdown of six well-known stocks set against the S&P 500 index, using the last year of minute-level data.

How much more does one stock swing than the index?

Volatility is the standard measure of how much a price bounces around. Annualized realized volatility takes the day-to-day percentage moves over a year and scales them to a yearly figure: a higher number means wider swings in both directions. The index blends roughly 500 companies into one line, and their individual jumps partly cancel, so the blended line travels less ground than any of its parts.

SELECT ticker,
       round(stddevSamp(ret) * sqrt(252) * 100, 1) AS annual_vol_pct
FROM (
    SELECT ticker, dt,
           c / lagInFrame(c) OVER (PARTITION BY ticker ORDER BY dt) - 1 AS ret
    FROM (
        SELECT ticker,
               toDate(toTimeZone(window_start, 'America/New_York')) AS dt,
               argMax(toFloat64(close), toTimeZone(window_start, 'America/New_York')) AS c
        FROM global_markets.delayed_stocks_minute_aggs
        WHERE ticker IN ('SPY', 'KO', 'JNJ', 'PG', 'NVDA', 'TSLA', 'PLTR')
          AND window_start >= now() - INTERVAL 400 DAY
          AND (toHour(toTimeZone(window_start, 'America/New_York')) * 60
               + toMinute(toTimeZone(window_start, 'America/New_York'))) BETWEEN 570 AND 959
        GROUP BY ticker, dt
    )
)
WHERE ret IS NOT NULL AND dt >= today() - 370
GROUP BY ticker
ORDER BY annual_vol_pct

Over the trailing year the SPY ETF posted the lowest reading in the group at 12.5%, and PLTR sat at the top with 51.5%. Every single stock in the 7-name comparison swung wider than the index. Even the steadiest blue chips, the consumer-staples and healthcare names people picture as sleepy, carried more year-long volatility than the diversified basket. The gap widens sharply for the high-growth names at the right of the chart, where a single earnings night can move the stock more than the index moves in a month.

How far can a single name fall?

Volatility counts moves in both directions. Drawdown counts only the painful one. Maximum drawdown is the largest peak-to-trough drop over a window: it answers "if I had bought at the worst possible moment, how far underwater would I go before the stock set a new high?" For money you might need on a fixed timeline, drawdown is often the figure that matters more than volatility.

SELECT ticker,
       round(abs(min(c / peak - 1)) * 100, 1) AS max_drawdown_pct
FROM (
    SELECT ticker, dt, c,
           max(c) OVER (PARTITION BY ticker ORDER BY dt
                        ROWS BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW) AS peak
    FROM (
        SELECT ticker,
               toDate(toTimeZone(window_start, 'America/New_York')) AS dt,
               argMax(toFloat64(close), toTimeZone(window_start, 'America/New_York')) AS c
        FROM global_markets.delayed_stocks_minute_aggs
        WHERE ticker IN ('SPY', 'KO', 'JNJ', 'PG', 'NVDA', 'TSLA', 'PLTR')
          AND window_start >= now() - INTERVAL 400 DAY
          AND (toHour(toTimeZone(window_start, 'America/New_York')) * 60
               + toMinute(toTimeZone(window_start, 'America/New_York'))) BETWEEN 570 AND 959
        GROUP BY ticker, dt
    )
    WHERE dt >= today() - 370
)
GROUP BY ticker
ORDER BY max_drawdown_pct

The shallowest drawdown in the group was 8.5% and the deepest was 48.3%, on PLTR. The index and the blue-chip names sit toward the shallow end of the chart; the high-growth names stretch across the deep end. A drawdown near 48.3% means a portfolio built entirely on that one stock would have been cut nearly in half at its low, and would then need to almost double to get back to even. The index's own worst drop over the same window lands down among the calmest single names, not out with the deepest.

What one stock's path looks like next to the index

Numbers on a bar chart flatten the experience of holding a position. This next panel indexes both a single high-flyer and the S&P 500 to 100 at the start of the window and tracks them month by month, and the two lines share a starting point so you can watch them separate.

SELECT formatDateTime(mo, '%b %Y') AS month,
       round(anyIf(idx, ticker = 'SPY'), 1) AS spy_indexed,
       round(anyIf(idx, ticker = 'TSLA'), 1) AS tsla_indexed
FROM (
    SELECT ticker, mo,
           mclose / first_value(mclose) OVER (PARTITION BY ticker ORDER BY mo
               ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) * 100 AS idx
    FROM (
        SELECT ticker, mo, argMax(c, dt) AS mclose
        FROM (
            SELECT ticker,
                   toStartOfMonth(toDate(toTimeZone(window_start, 'America/New_York'))) AS mo,
                   toDate(toTimeZone(window_start, 'America/New_York')) AS dt,
                   argMax(toFloat64(close), toTimeZone(window_start, 'America/New_York')) AS c
            FROM global_markets.delayed_stocks_minute_aggs
            WHERE ticker IN ('SPY', 'TSLA')
              AND window_start >= now() - INTERVAL 400 DAY
              AND (toHour(toTimeZone(window_start, 'America/New_York')) * 60
                   + toMinute(toTimeZone(window_start, 'America/New_York'))) BETWEEN 570 AND 959
            GROUP BY ticker, mo, dt
        )
        GROUP BY ticker, mo
    )
)
GROUP BY mo
ORDER BY mo

Both lines begin at 100. Twelve-plus months later the index finished near 122.1 and the single stock near 124.3, close to each other at the finish. The paths in between were nothing alike. The index climbed a fairly smooth staircase. The single name spiked well above the index line, gave most of it back, and swung through a range several times wider before the two nearly met again. An investor who checked the account only on the first day and the last saw similar results. Anyone who watched month to month, or who needed to sell during one of the deep dips, lived a very different year.

Diversification is the counterweight

The diversified index sits at the low end of both volatility and drawdown in every panel here, and it gets there through breadth: hundreds of independent outcomes averaged into one line, where the individual jumps partly cancel. Spreading a portfolio across holdings whose ups and downs do not move in lockstep pulls the combined swing below that of any single name inside it. A portfolio built out of one stock, or several stocks in the same industry that rise and fall together, carries the wide volatility and deep drawdown shown above rather than the index's narrow ones.

When a single position is too large to sell outright, its swing can still be capped rather than removed. A protective put sets a floor under the holding, and a collar brackets it between a floor and a ceiling. The broad-market line in the panels above is the benchmark both the raw position and any hedge on it are measured against.

FAQ

What is concentration risk?

Concentration risk is the exposure that comes from holding too much of your wealth in a single stock, sector, or asset. If that one holding drops, nothing else in the portfolio cushions it. The single-stock volatility and drawdown figures above quantify how much larger those swings run than a diversified index's.

How much more volatile is a single stock than the S&P 500?

It varies by the stock. Over the last year the S&P 500 ETF's annualized volatility was 12.5%, while the single names in the same window ranged up to 51.5% for PLTR. Every stock measured swung wider than the index.

Does diversification reduce risk?

Spreading money across many holdings whose ups and downs do not move in lockstep lowers the volatility and the drawdown of the combined portfolio versus any single holding. The index in the panels above is the clearest example: it posted the lowest volatility of every line measured.

Is my own company stock a concentration risk?

If a large share of your net worth, salary, or equity compensation depends on one company, yes. That single bet exists before you invest a dollar elsewhere. Building the rest of the portfolio out of the same or closely related stocks deepens the exposure rather than balancing it.

Every panel above ships with the SQL behind it. Expand any one to audit the numbers, or measure your own holdings against the index on the Strasmore terminal. For the volatility concept in depth see implied volatility, and for a real market-wide drawdown see the COVID crash of March 2020.