Strasmore Research
市场回顾 Matt Connor作者： Matt Connor · data as of July 22, 2026 · refreshed weekly

美股2026年涨幅榜与跌幅榜（年初至今）

2026年迄今美股表现最好和最差的个股，均按真实流动性从交易所成交记录中排名，剔除拆股和杠杆基金的干扰。

2026年迄今表现最好的股票，目前由 SNDK 领跑，从年初首个交易日到最近一个已完成交易日累计上涨 469.1%。本页在真正有成交的股票中，对2026年美股涨幅榜和跌幅榜进行排名，数据直接取自交易所成交记录，而不是数据商的整理摘要，并会随着这一年逐步推进持续更新。在排名之前，拆股和杠杆基金已经被剔除——这正是多数“年初至今”榜单在不知不觉中出错的地方。

2026年美股涨幅榜

下表中的每一只股票在排名之前都先通过了流动性门槛：过去一个月内，每个已完成交易日的平均成交金额至少达到 1 亿美元。一只无人问津的股票即便涨幅巨大，那也只是一个报价，而不是一个真实的结果。

查询2026年美股涨幅榜：交投活跃股票的年初至今前十名
每个数字背后的完整 SQL
WITH complete AS (
    SELECT toDate(toTimeZone(window_start, 'America/New_York')) AS d
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE ticker = 'SPY'
      AND window_start >= toDateTime('2026-01-01 00:00:00')
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
    GROUP BY d
    HAVING count() >= 380
),
universe AS (
    SELECT ticker,
           sum(toFloat64(close) * toFloat64(volume)) / uniqExact(toDate(toTimeZone(window_start, 'America/New_York'))) AS adv
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE window_start >= now() - INTERVAL 21 DAY
      AND toDate(toTimeZone(window_start, 'America/New_York')) >= today() - 20
      AND toDate(toTimeZone(window_start, 'America/New_York')) IN (SELECT d FROM complete)
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
    GROUP BY ticker
    HAVING adv >= 100000000
),
edges AS (
    SELECT ticker,
        argMinIf(toFloat64(open), toTimeZone(window_start, 'America/New_York'), toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT min(d) FROM complete)) AS year_open,
        argMaxIf(toFloat64(close), toTimeZone(window_start, 'America/New_York'), toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT max(d) FROM complete)) AS latest_close,
        countIf(toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT min(d) FROM complete)) AS bars_open,
        countIf(toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT max(d) FROM complete)) AS bars_close
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE ((window_start >= toDateTime('2026-01-01 00:00:00') AND window_start < toDateTime('2026-01-10 00:00:00'))
        OR (window_start >= now() - INTERVAL 8 DAY))
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
      AND ticker NOT IN ('SPCX')
      AND ticker NOT IN ('KORU','SOXL','SOXS','SOXY','TQQQ','SQQQ','QQQU','SPXL','SPXS','UPRO','SPXU','SPYU','LABU','LABD','FAS','FAZ','TNA','TZA','YINN','YANG','UDOW','SDOW','BOIL','KOLD','UCO','SCO','USD','SSO','SDS','QLD','QID','ERX','ERY','DRN','DRV','CURE','MUU','SNXX','UVXY','SVXY','UVIX','SVIX','BULZ','WEBL','WEBS','DPST','DRIP','GUSH','AGQ','ZSL','GDXU','GDXD','FNGU','FNGD','DUST','JNUG','JDST','NUGT','BITX','BITU','SBIT','ETHU','ETHT','NVDL','NVDS','NVD','NVDX','NVDU','NVDD','NVDQ','TSLL','TSLQ','TSLZ','TSLR','TSLT','TSLS','TSDD','AAPU','AAPD','MSFU','MSFD','GGLL','GGLS','AMZU','AMZD','METU','METD','PLTU','PLTD','SMCX','SMCZ','CONL','CONI','MSTX','MSTU','MSTZ','BRKU','AMDL','AMUU','AMDD','ELIL','ELIS','HOOX','AVGX','AVGU','TSMX','TSMZ','MULL')
      AND ticker NOT IN (SELECT ticker FROM global_markets.stocks_splits
                         WHERE execution_date BETWEEN toDate('2026-01-01') AND today())
    GROUP BY ticker
    HAVING bars_open >= 100 AND bars_close >= 100
)
SELECT e.ticker AS ticker,
    round((e.latest_close / e.year_open - 1) * 100, 1) AS ytd_return_pct,
    round(u.adv / 1e6, 0) AS avg_daily_dollar_m
FROM edges AS e
INNER JOIN universe AS u ON e.ticker = u.ticker
WHERE e.year_open >= 10
ORDER BY (e.latest_close / e.year_open) DESC
LIMIT 10

年内领跑的是 SNDK，涨幅 469.1%；其次是 MXL，涨幅 321.1%；第三名是 AEHR，涨幅 268.3%。排在第十位的 AXTI 仍然上涨了 183.9%，可见整张榜单的涨幅都远远甩开了任何一个宽基指数。

每一次涨跌旁边的成交金额一列，信息量不亚于百分比本身。AXTI 的日均成交金额约为 408 百万美元；排在榜首的 SNDK 日均成交金额则是 19525 百万美元。榜上每只股票都跨过了每日1亿美元的门槛，但门槛之上的跨度依然很宽，宽到同样一个涨幅百分比，放在区间两端可能代表着两种完全不同的含义：一只交投清淡的股票更容易被推高，也更难在屏幕上显示的价格上真正退出。

2026年美股跌幅榜

统计方法完全相同，只是方向相反：在通过同一流动性门槛的股票中，找出年初至今跌幅最深的那些。

查询2026年美股跌幅榜：交投活跃股票的年初至今后十名
每个数字背后的完整 SQL
WITH complete AS (
    SELECT toDate(toTimeZone(window_start, 'America/New_York')) AS d
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE ticker = 'SPY'
      AND window_start >= toDateTime('2026-01-01 00:00:00')
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
    GROUP BY d
    HAVING count() >= 380
),
universe AS (
    SELECT ticker,
           sum(toFloat64(close) * toFloat64(volume)) / uniqExact(toDate(toTimeZone(window_start, 'America/New_York'))) AS adv
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE window_start >= now() - INTERVAL 21 DAY
      AND toDate(toTimeZone(window_start, 'America/New_York')) >= today() - 20
      AND toDate(toTimeZone(window_start, 'America/New_York')) IN (SELECT d FROM complete)
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
    GROUP BY ticker
    HAVING adv >= 100000000
),
edges AS (
    SELECT ticker,
        argMinIf(toFloat64(open), toTimeZone(window_start, 'America/New_York'), toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT min(d) FROM complete)) AS year_open,
        argMaxIf(toFloat64(close), toTimeZone(window_start, 'America/New_York'), toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT max(d) FROM complete)) AS latest_close,
        countIf(toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT min(d) FROM complete)) AS bars_open,
        countIf(toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT max(d) FROM complete)) AS bars_close
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE ((window_start >= toDateTime('2026-01-01 00:00:00') AND window_start < toDateTime('2026-01-10 00:00:00'))
        OR (window_start >= now() - INTERVAL 8 DAY))
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
      AND ticker NOT IN ('SPCX')
      AND ticker NOT IN ('KORU','SOXL','SOXS','SOXY','TQQQ','SQQQ','QQQU','SPXL','SPXS','UPRO','SPXU','SPYU','LABU','LABD','FAS','FAZ','TNA','TZA','YINN','YANG','UDOW','SDOW','BOIL','KOLD','UCO','SCO','USD','SSO','SDS','QLD','QID','ERX','ERY','DRN','DRV','CURE','MUU','SNXX','UVXY','SVXY','UVIX','SVIX','BULZ','WEBL','WEBS','DPST','DRIP','GUSH','AGQ','ZSL','GDXU','GDXD','FNGU','FNGD','DUST','JNUG','JDST','NUGT','BITX','BITU','SBIT','ETHU','ETHT','NVDL','NVDS','NVD','NVDX','NVDU','NVDD','NVDQ','TSLL','TSLQ','TSLZ','TSLR','TSLT','TSLS','TSDD','AAPU','AAPD','MSFU','MSFD','GGLL','GGLS','AMZU','AMZD','METU','METD','PLTU','PLTD','SMCX','SMCZ','CONL','CONI','MSTX','MSTU','MSTZ','BRKU','AMDL','AMUU','AMDD','ELIL','ELIS','HOOX','AVGX','AVGU','TSMX','TSMZ','MULL')
      AND ticker NOT IN (SELECT ticker FROM global_markets.stocks_splits
                         WHERE execution_date BETWEEN toDate('2026-01-01') AND today())
    GROUP BY ticker
    HAVING bars_open >= 100 AND bars_close >= 100
)
SELECT e.ticker AS ticker,
    round((e.latest_close / e.year_open - 1) * 100, 1) AS ytd_return_pct,
    round(u.adv / 1e6, 0) AS avg_daily_dollar_m
FROM edges AS e
INNER JOIN universe AS u ON e.ticker = u.ticker
WHERE e.year_open >= 10
ORDER BY (e.latest_close / e.year_open) ASC
LIMIT 10

年初至今跌幅最深的是 EOSE，跌幅 -65.3%，其次是 EPAM，跌幅 -56.3%。排在第十位的 SMR 跌幅为 -45.3%。这份榜单上有好几只都是被广泛持有的大公司，而不是投机性的小盘股，这正是跌幅榜最出人意料的一点：充足的流动性和分析师覆盖，并不能给股价托底。任何把其中一只股票作为投资组合过大比重持有的人，都要承受这一整段跌幅，集中持股的风险一文用波动率和最大回撤数字对此做了测量。

2026年市场整体涨跌几何

榜单只展示分布的两端，市场广度展示的是分布的主体：包括指数基金在内的整个筛选样本，按涨跌方向和涨跌幅度分入不同区间后是什么样子。

查询年初至今市场广度：筛选样本按涨跌区间的分布
每个数字背后的完整 SQL
WITH complete AS (
    SELECT toDate(toTimeZone(window_start, 'America/New_York')) AS d
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE ticker = 'SPY'
      AND window_start >= toDateTime('2026-01-01 00:00:00')
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
    GROUP BY d
    HAVING count() >= 380
),
universe AS (
    SELECT ticker,
           sum(toFloat64(close) * toFloat64(volume)) / uniqExact(toDate(toTimeZone(window_start, 'America/New_York'))) AS adv
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE window_start >= now() - INTERVAL 21 DAY
      AND toDate(toTimeZone(window_start, 'America/New_York')) >= today() - 20
      AND toDate(toTimeZone(window_start, 'America/New_York')) IN (SELECT d FROM complete)
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
    GROUP BY ticker
    HAVING adv >= 100000000
),
edges AS (
    SELECT ticker,
        argMinIf(toFloat64(open), toTimeZone(window_start, 'America/New_York'), toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT min(d) FROM complete)) AS year_open,
        argMaxIf(toFloat64(close), toTimeZone(window_start, 'America/New_York'), toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT max(d) FROM complete)) AS latest_close,
        countIf(toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT min(d) FROM complete)) AS bars_open,
        countIf(toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT max(d) FROM complete)) AS bars_close
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE ((window_start >= toDateTime('2026-01-01 00:00:00') AND window_start < toDateTime('2026-01-10 00:00:00'))
        OR (window_start >= now() - INTERVAL 8 DAY))
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
      AND ticker NOT IN ('SPCX')
      AND ticker NOT IN ('KORU','SOXL','SOXS','SOXY','TQQQ','SQQQ','QQQU','SPXL','SPXS','UPRO','SPXU','SPYU','LABU','LABD','FAS','FAZ','TNA','TZA','YINN','YANG','UDOW','SDOW','BOIL','KOLD','UCO','SCO','USD','SSO','SDS','QLD','QID','ERX','ERY','DRN','DRV','CURE','MUU','SNXX','UVXY','SVXY','UVIX','SVIX','BULZ','WEBL','WEBS','DPST','DRIP','GUSH','AGQ','ZSL','GDXU','GDXD','FNGU','FNGD','DUST','JNUG','JDST','NUGT','BITX','BITU','SBIT','ETHU','ETHT','NVDL','NVDS','NVD','NVDX','NVDU','NVDD','NVDQ','TSLL','TSLQ','TSLZ','TSLR','TSLT','TSLS','TSDD','AAPU','AAPD','MSFU','MSFD','GGLL','GGLS','AMZU','AMZD','METU','METD','PLTU','PLTD','SMCX','SMCZ','CONL','CONI','MSTX','MSTU','MSTZ','BRKU','AMDL','AMUU','AMDD','ELIL','ELIS','HOOX','AVGX','AVGU','TSMX','TSMZ','MULL')
      AND ticker NOT IN (SELECT ticker FROM global_markets.stocks_splits
                         WHERE execution_date BETWEEN toDate('2026-01-01') AND today())
    GROUP BY ticker
    HAVING bars_open >= 100 AND bars_close >= 100
),
rets AS (
    SELECT e.ticker AS ticker, (e.latest_close / e.year_open - 1) * 100 AS ret
    FROM edges AS e
    INNER JOIN universe AS u ON e.ticker = u.ticker
    WHERE e.year_open >= 10
)
SELECT bucket,
       names,
       round(100 * sum(names) OVER (PARTITION BY side) / sum(names) OVER (), 1) AS side_share_pct
FROM (
    SELECT tup.1 AS bucket, tup.2 AS names, tup.3 AS ord, tup.4 AS side
    FROM (
        SELECT arrayJoin([
            ('Down 40%+',   countIf(ret < -40),               1, 'down'),
            ('Down 20-40%', countIf(ret >= -40 AND ret < -20), 2, 'down'),
            ('Down 10-20%', countIf(ret >= -20 AND ret < -10), 3, 'down'),
            ('Down 0-10%',  countIf(ret >= -10 AND ret < 0),   4, 'down'),
            ('Up 0-10%',    countIf(ret >= 0 AND ret < 10),    5, 'up'),
            ('Up 10-25%',   countIf(ret >= 10 AND ret < 25),   6, 'up'),
            ('Up 25-50%',   countIf(ret >= 25 AND ret < 50),   7, 'up'),
            ('Up 50%+',     countIf(ret >= 50),                8, 'up')
        ]) AS tup
        FROM rets
    )
)
ORDER BY ord

在通过筛选的股票中，62.3% 年内上涨，37.7% 年内下跌。这条分布曲线，正是前十名榜单所掩盖的部分：一端有 26 只股票落在“Down 40%+”区间，另一端有 123 只落在“Up 50%+”区间；而紧贴“持平”两侧的“Down 0-10%”区间有 143 只，“Up 0-10%”区间有 194 只。在任何一年里，市场的大多数成员做的都是些平淡无奇的事，上面两张榜单上的名字，是从这些平均数中被挑出来的例外。

大盘自身的表现如何

把一只个股的这一年放到大盘自身的这一年旁边，含义会大不相同。下面是四大指数 ETF 在完全相同窗口内的表现：

查询四大指数ETF年初至今的表现（同一测量窗口）
每个数字背后的完整 SQL
WITH complete AS (
    SELECT toDate(toTimeZone(window_start, 'America/New_York')) AS d
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE ticker = 'SPY'
      AND window_start >= toDateTime('2026-01-01 00:00:00')
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
    GROUP BY d
    HAVING count() >= 380
),
edges AS (
    SELECT ticker,
        argMinIf(toFloat64(open), toTimeZone(window_start, 'America/New_York'), toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT min(d) FROM complete)) AS year_open,
        argMaxIf(toFloat64(close), toTimeZone(window_start, 'America/New_York'), toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT max(d) FROM complete)) AS latest_close,
        countIf(toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT min(d) FROM complete)) AS bars_open,
        countIf(toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT max(d) FROM complete)) AS bars_close
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE ((window_start >= toDateTime('2026-01-01 00:00:00') AND window_start < toDateTime('2026-01-10 00:00:00'))
        OR (window_start >= now() - INTERVAL 8 DAY))
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
      AND ticker IN ('SPY', 'QQQ', 'DIA', 'IWM')
    GROUP BY ticker
    HAVING bars_open >= 100 AND bars_close >= 100
)
SELECT multiIf(ticker = 'SPY', 'S&P 500 (SPY)', ticker = 'QQQ', 'Nasdaq 100 (QQQ)',
               ticker = 'DIA', 'Dow (DIA)', ticker = 'IWM', 'Russell 2000 (IWM)', ticker) AS label,
    round((latest_close / year_open - 1) * 100, 1) AS ytd_return_pct,
    (SELECT toString(min(d)) FROM complete) AS measured_from_date,
    (SELECT toString(max(d)) FROM complete) AS measured_through_date
FROM edges
ORDER BY ytd_return_pct DESC

上面的两张榜单和这一面板，统计的是同一个窗口：从2026年第一个完整交易日（2026-01-02）到最近一个已完成交易日（2026-07-20）。在此期间，Russell 2000 (IWM) 的涨跌幅为 18%，Dow (DIA) 的涨跌幅为 7.5%。涨幅榜上的每一只股票，涨幅都远远超过这四条指数线——这正是指数运作的算术原理：数百只成分股被平均成一条线，个股之间的涨跌在这条线里部分相互抵消了。这条线之下的离散程度，远比这条线本身宽得多，上半年板块排行榜一文把全部十一个板块两两排名对照。

拆股会让年初至今涨跌幅失真

拆股会改变股数和每股价格，但不会改变持仓本身的价值。在这么长的窗口内，会有不少公司发生拆股，如果不做复权处理直接计算年初至今涨跌幅，就会把每股价格的变化误读成业绩表现。下表中的每一只股票，2026年都发生过足以让价格变动至少五分之一的拆股或股票股利，且都通过了同样的流动性门槛——也就是说，在被拆股筛选剔除之前，它们本来都有资格出现在上面的榜单里。

查询被剔除出榜单：2026年幅度大到足以伪造涨跌幅的拆股
每个数字背后的完整 SQL
WITH complete AS (
    SELECT toDate(toTimeZone(window_start, 'America/New_York')) AS d
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE ticker = 'SPY'
      AND window_start >= toDateTime('2026-01-01 00:00:00')
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
    GROUP BY d
    HAVING count() >= 380
),
universe AS (
    SELECT ticker,
           sum(toFloat64(close) * toFloat64(volume)) / uniqExact(toDate(toTimeZone(window_start, 'America/New_York'))) AS adv
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE window_start >= now() - INTERVAL 21 DAY
      AND toDate(toTimeZone(window_start, 'America/New_York')) >= today() - 20
      AND toDate(toTimeZone(window_start, 'America/New_York')) IN (SELECT d FROM complete)
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
    GROUP BY ticker
    HAVING adv >= 100000000
),
split_terms AS (
    SELECT ticker,
        argMax(split_to, execution_date) AS new_shares,
        argMax(split_from, execution_date) AS old_shares,
        round((old_shares / new_shares - 1) * 100, 1) AS split_price_effect_pct,
        replaceAll(argMax(adjustment_type, execution_date), '_', ' ') AS split_type,
        toString(max(execution_date)) AS split_date
    FROM global_markets.stocks_splits
    WHERE execution_date BETWEEN toDate('2026-01-01') AND today()
    GROUP BY ticker
    HAVING new_shares > 0 AND old_shares > 0
       AND greatest(new_shares / old_shares, old_shares / new_shares) >= 1.25
),
edges AS (
    SELECT ticker,
        argMinIf(toFloat64(open), toTimeZone(window_start, 'America/New_York'), toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT min(d) FROM complete)) AS year_open,
        argMaxIf(toFloat64(close), toTimeZone(window_start, 'America/New_York'), toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT max(d) FROM complete)) AS latest_close,
        countIf(toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT min(d) FROM complete)) AS bars_open,
        countIf(toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT max(d) FROM complete)) AS bars_close
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE ((window_start >= toDateTime('2026-01-01 00:00:00') AND window_start < toDateTime('2026-01-10 00:00:00'))
        OR (window_start >= now() - INTERVAL 8 DAY))
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
      AND ticker NOT IN ('SPCX')
      AND ticker NOT IN ('KORU','SOXL','SOXS','SOXY','TQQQ','SQQQ','QQQU','SPXL','SPXS','UPRO','SPXU','SPYU','LABU','LABD','FAS','FAZ','TNA','TZA','YINN','YANG','UDOW','SDOW','BOIL','KOLD','UCO','SCO','USD','SSO','SDS','QLD','QID','ERX','ERY','DRN','DRV','CURE','MUU','SNXX','UVXY','SVXY','UVIX','SVIX','BULZ','WEBL','WEBS','DPST','DRIP','GUSH','AGQ','ZSL','GDXU','GDXD','FNGU','FNGD','DUST','JNUG','JDST','NUGT','BITX','BITU','SBIT','ETHU','ETHT','NVDL','NVDS','NVD','NVDX','NVDU','NVDD','NVDQ','TSLL','TSLQ','TSLZ','TSLR','TSLT','TSLS','TSDD','AAPU','AAPD','MSFU','MSFD','GGLL','GGLS','AMZU','AMZD','METU','METD','PLTU','PLTD','SMCX','SMCZ','CONL','CONI','MSTX','MSTU','MSTZ','BRKU','AMDL','AMUU','AMDD','ELIL','ELIS','HOOX','AVGX','AVGU','TSMX','TSMZ','MULL')
      AND ticker IN (SELECT ticker FROM split_terms)
    GROUP BY ticker
    HAVING bars_open >= 100 AND bars_close >= 100
),
ranked AS (
    SELECT e.ticker AS ticker,
        round((e.latest_close / e.year_open - 1) * 100, 1) AS unadjusted_ytd_pct,
        s.split_price_effect_pct AS split_price_effect_pct,
        s.split_type AS split_type,
        s.split_date AS split_date
    FROM edges AS e
    INNER JOIN split_terms AS s ON e.ticker = s.ticker
    WHERE e.year_open >= 10 AND e.ticker IN (SELECT ticker FROM universe)
)
SELECT ticker, unadjusted_ytd_pct, split_price_effect_pct, split_type, split_date
FROM (
    SELECT ticker, unadjusted_ytd_pct, split_price_effect_pct, split_type, split_date,
           row_number() OVER (ORDER BY unadjusted_ytd_pct ASC) AS rn_low,
           row_number() OVER (ORDER BY unadjusted_ytd_pct DESC) AS rn_high
    FROM ranked
)
WHERE rn_low <= 14 OR rn_high = 1
ORDER BY unadjusted_ytd_pct ASC

BKNG 经历了一次forward split，把每股价格拉动了 -96%，若不做复权处理，未调整的筛选结果会显示为 -96.7%；VGT 经历了一次forward split，把价格拉动了 -87.5%，未调整的结果会显示为 -85.1%。这两只股票的持有人在拆股生效当天都没有损失任何价值：每股价格按同一个比例被除低，股数按同一个比例被乘高。“拆股价格影响”一列，单独隔离出了纯粹由拆股造成的价格变化，而这正是未经调整的筛选器会误读为业绩表现的部分。

面板最后一行方向相反：DD 经历了一次reverse split，把每股价格推高了 200%（不是拉低，而是推高），未调整的结果显示为 235.9%。任何跳过这一步的榜单，都会在榜单顶部和底部同时发布虚构的数字——这正是本筛选器把发生拆股的代码整体剔除、而不是尝试修补其价格的原因。基金和公司一样都可能出现在这个面板里：指数基金也会拆分份额，本筛选器对两者一视同仁。

统计方法说明

筛选的每一个阶段，以及每个阶段结束后仍然留在样本中的股票数量：

查询筛选流程逐级说明：每一阶段留存的股票数量
每个数字背后的完整 SQL
WITH complete AS (
    SELECT toDate(toTimeZone(window_start, 'America/New_York')) AS d
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE ticker = 'SPY'
      AND window_start >= toDateTime('2026-01-01 00:00:00')
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
    GROUP BY d
    HAVING count() >= 380
),
universe AS (
    SELECT ticker
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE window_start >= now() - INTERVAL 21 DAY
      AND toDate(toTimeZone(window_start, 'America/New_York')) >= today() - 20
      AND toDate(toTimeZone(window_start, 'America/New_York')) IN (SELECT d FROM complete)
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960
    GROUP BY ticker
    HAVING sum(toFloat64(close) * toFloat64(volume)) / uniqExact(toDate(toTimeZone(window_start, 'America/New_York'))) >= 100000000
),
edges AS (
    SELECT ticker,
        argMinIf(toFloat64(open), toTimeZone(window_start, 'America/New_York'), toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT min(d) FROM complete)) AS year_open,
        argMaxIf(toFloat64(close), toTimeZone(window_start, 'America/New_York'), toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT max(d) FROM complete)) AS latest_close,
        countIf(toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT min(d) FROM complete)) AS bars_open,
        countIf(toDate(toTimeZone(window_start, 'America/New_York')) = (SELECT max(d) FROM complete)) AS bars_close
    FROM global_markets.delayed_stocks_minute_aggs
    WHERE ((window_start >= toDateTime('2026-01-01 00:00:00') AND window_start < toDateTime('2026-01-10 00:00:00'))
        OR (window_start >= now() - INTERVAL 8 DAY))
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) >= 570
      AND toHour(toTimeZone(window_start, 'America/New_York')) * 60 + toMinute(toTimeZone(window_start, 'America/New_York')) < 960

    GROUP BY ticker
    HAVING bars_open >= 100 AND bars_close >= 100
),
flagged AS (
    SELECT ticker,
        year_open,
        ticker IN (SELECT ticker FROM universe) AS liquid,
        ticker NOT IN ('SPCX','KORU','SOXL','SOXS','SOXY','TQQQ','SQQQ','QQQU','SPXL','SPXS','UPRO','SPXU','SPYU','LABU','LABD','FAS','FAZ','TNA','TZA','YINN','YANG','UDOW','SDOW','BOIL','KOLD','UCO','SCO','USD','SSO','SDS','QLD','QID','ERX','ERY','DRN','DRV','CURE','MUU','SNXX','UVXY','SVXY','UVIX','SVIX','BULZ','WEBL','WEBS','DPST','DRIP','GUSH','AGQ','ZSL','GDXU','GDXD','FNGU','FNGD','DUST','JNUG','JDST','NUGT','BITX','BITU','SBIT','ETHU','ETHT','NVDL','NVDS','NVD','NVDX','NVDU','NVDD','NVDQ','TSLL','TSLQ','TSLZ','TSLR','TSLT','TSLS','TSDD','AAPU','AAPD','MSFU','MSFD','GGLL','GGLS','AMZU','AMZD','METU','METD','PLTU','PLTD','SMCX','SMCZ','CONL','CONI','MSTX','MSTU','MSTZ','BRKU','AMDL','AMUU','AMDD','ELIL','ELIS','HOOX','AVGX','AVGU','TSMX','TSMZ','MULL') AS not_geared,
        ticker NOT IN (SELECT ticker FROM global_markets.stocks_splits
                       WHERE execution_date BETWEEN toDate('2026-01-01') AND today()) AS no_split
    FROM edges
)
SELECT tup.1 AS screen_step, tup.2 AS names
FROM (
    SELECT arrayJoin([
        ('Traded the first and the last full session', count()),
        ('Averaging $100m+ a day over the past month', countIf(liquid)),
        ('Opened 2026 above $10 a share', countIf(liquid AND year_open >= 10)),
        ('Not on the excluded-fund list', countIf(liquid AND year_open >= 10 AND not_geared)),
        ('No split or stock dividend in 2026', countIf(liquid AND year_open >= 10 AND not_geared AND no_split))
    ]) AS tup
    FROM flagged
)

完整的方法说明如下：

  • 窗口范围。 从2026年第一个完整交易日，到成交记录上最近一个完整交易日。两端的日期都从分钟K线中读取，只有常规交易时段K线数量完整的交易日才会被计入，因此一个仍在进行中的交易日永远不会被当作已完成处理。
  • 涨跌幅的计算。 用年内首个常规交易时段的开盘价，对比最后一个常规交易时段的收盘价，数据来自分钟K线，盘前和盘后的成交都被忽略。这是价格涨跌幅，股息不会被加回，因此高股息股票在这里的读数，会比按总回报计算略微逊色。
  • 流动性门槛。 过去 20 个日历日内，只统计完整交易的交易日，常规交易时段的日均成交金额至少达到 1 亿美元。如果算入未走完全程的交易日，就会用一整天去除一部分成交金额，从而低估每只股票的实际成交规模。仅这一道筛选，就把样本从 4476 个代码缩减到 1056 个。
  • 价格门槛。 股票年初开盘价须高于 10 美元。10 美元以下的股票，很小的价格变动就能产生很大的百分比。
  • 杠杆和反向基金通过一份人工维护的名单逐一剔除，完整名单打印在每张表格下方的 SQL 中，涵盖了跟踪指数的杠杆基金，以及放大单只公司单日涨跌幅的单股杠杆产品。一只杠杆基金完全可能在没有任何公司层面消息的情况下登上未经筛选的涨跌榜榜首。这份名单靠人工维护，这也是它的弱点：本月才上市的基金，只有在有人把它加入名单之后才会被剔除。
  • 普通基金不会被过滤掉。 只要通过了流动性和价格门槛，未加杠杆的指数或板块 ETF 会和运营公司一起出现在榜单上，也会被计入广度统计。基金本身就是一个由同一批股票组成的篮子，因此这些行反映的是这个篮子的表现，而不是某一家公司这一年的表现。
  • 拆股。 任何在 2026 年内生效过拆股或股票股利的代码都会被整体剔除，无论比例大小。这是一个偏保守的选择：会让极少数真正表现出色的股票失去上榜机会，但绝不会把股数的变化当作涨跌幅发布出来。上面的面板展示的正是那些拆股幅度大到足以伪造涨跌幅的被剔除股票。
  • 晚于年初上市的公司。 在年内第一个交易日之后才上市的公司没有1月份的价格，因此不会出现在这些表格里——在新股上市密集的年份，这是一个真实存在的缺口。
  • 最终结果。 共有 966 只股票参与排名，其中涨幅前十和跌幅前十，就是本页的两张榜单。

同一个问题，三种时间窗口

年初至今只是这架阶梯上的一级，同一次涨跌在不同的时间跨度下看起来并不相同。周维度是本周美股涨幅榜与跌幅榜，月维度是本月美股涨跌榜，本页则是年度维度。各级的门槛并不相同：本页按过去一个月的日均成交金额 1 亿美元筛选，而周榜和月榜是按各自窗口内的成交总额筛选，因此一只股票完全可能通过其中一张榜单的筛选，却在另一张上落选。一只因单次财报反应登上周榜榜首的股票，放到全年来看往往只排在中游。想了解这一年更宏观的叙事，上半年市场回顾涵盖了指数、利率和板块层面的细节，杠杆ETF的运作原理一文解释了本页为何要把加杠杆的基金排除在外。

FAQ

2026年表现最好的股票是哪一只？

在日均成交金额至少 1 亿美元、年初开盘价高于 10 美元、2026 年内未发生拆股、且不在杠杆基金名单上的股票中，SNDK 是目前2026年表现最好的股票，较年初首个交易日累计上涨 469.1%。MXL 排名第二，涨幅 321.1%。

2026年跌幅最大的股票有哪些？

在本页的筛选样本中，EOSE 是2026年表现最差的股票，跌幅 -65.3%，其次是 EPAM，跌幅 -56.3%。完整的2026年美股跌幅榜列在上方第二张表格中，并随这一年的推进持续更新。

年初至今涨跌幅是如何计算的？

年初至今涨跌幅，比较的是最新价格与日历年年初价格，并以百分比表示涨跌。本页具体计算方式，是用2026年首个常规交易时段的开盘价，对比最近一个已完成交易时段的收盘价。年内发生拆股的股票会被整体剔除，而不是做复权调整，因此读者看到的数字和成交记录之间，不隔着任何调整系数。

2026年涨幅榜里包含杠杆ETF吗？

不包含。杠杆或反向基金会把某个指数或某只个股的单日涨跌幅成倍放大，因此经常在没有任何公司层面消息的情况下登上未经筛选的涨跌榜榜首。这些基金在排名之前就会按名单逐一剔除，完整名单可以在每张表格下方的 SQL 中查看。普通的指数和板块 ETF 则另当别论：只要通过与个股相同的门槛，就会被纳入排名，因此基金完全可能出现在任意一张榜单上。

本页每张表格都是一条附带 SQL 的已存储查询。展开任意面板即可核查筛选逻辑，也可以在 Strasmore 终端上换成自己的门槛重新构建。

#market movers#gainers and losers#year to date#stock splits#market breadth