14

I bought 16 years of historical hourly stock options data for my personal investment study to test how particular strategies may have played out over a long period of time.

I'm looking at the data for a DISH 1/21/2012 5/2.5c credit spread (yes I know, an absurd position, but due to this weird scenario, my software flagged it) and this is the weirdest activity I've ever seen unless it's an error on the data supplier's part. Maybe someone who knows the ins and outs of the market better than me can explain what happened here.

Summary at market close:

  • the options both saw an hour of 52,000 volume
  • 553 bids were made on the $2.5c (open interest)
  • 553 asks were made on the $2.5c (open interest)
  • 550 bids were made on the $5c (open interest)
  • 550 asks were made on the $5c (open interest)
  • both options were deeply in-the-money

At market open the next day, both options died:

  • both options were still deep ITM, with DISH experiencing a $2 (8%) drop from the previous market close and then holding stable for the following 2 months.
  • both options went to zero open interest
  • both options went to zero volume
  • both options went to zero bid / ask
  • they stayed that way until expiry roughly 2 months later, with no further trades or bid / ask movement

On 11/14/2011 at market close:

  • 1/21/2012 DISH $2.5c had the following state:

      open: 20.5,
      high: 20.5,
      low: 20.5,
      close: 20.5,
      bid: 20,
      ask: 21.1,
      underlying: "DISH",
      strike: 5,
      last_volume: 52000,
      bidsize: 553,
      bid_date: "2011-11-14 16:00:00",
      asksize: 553,
      expiration_date: "2012-01-21",
      option_type: "call",
      root_symbol: "DISH",
      underlying_bid: 25.54,
      underlying_ask: 25.55
    
  • 1/21/2012 DISH $5c had the following state:

      open: 23,
      high: 23,
      low: 23,
      close: 23,
      bid: 22.5,
      ask: 23.6,
      underlying: "DISH",
      strike: 2.5,
      last_volume: 52000,
      bidsize: 550,
      bid_date: "2011-11-14 16:00:00",
      asksize: 550,
      expiration_date: "2012-01-21",
      option_type: "call",
      root_symbol: "DISH",
      underlying_bid: 25.54,
      underlying_ask: 25.55,
    

The next morning, on 11/15/2011 at market open:

  • 1/21/2012 DISH $2.5c had the following state:

      open: 0,
      high: 0,
      low: 0,
      close: 0,
      bid: 0,
      ask: 0,
      underlying: "DISH",
      strike: 2.5,
      last_volume: 0,
      bidsize: 0,
      bid_date: "2011-11-15 11:30:00",
      asksize: 0,
      expiration_date: "2012-01-21",
      option_type: "call",
      root_symbol: "DISH",
      underlying_bid: 23.74,
      underlying_ask: 23.75
    
  • 1/21/2012 DISH $5c had the following state:

      open: 0,
      high: 0,
      low: 0,
      close: 0,
      bid: 0,
      ask: 0,
      underlying: "DISH",
      strike: 5,
      last_volume: 0,
      bidsize: 0,
      bid_date: "2011-11-15 11:30:00",
      asksize: 0,
      expiration_date: "2012-01-21",
      option_type: "call",
      root_symbol: "DISH",
      underlying_bid: 23.74,
      underlying_ask: 23.75
    

Can anyone explain how this could plausibly happen? It seems like a highly unusual anomaly in options trading and I haven't seen anything like it before, since with an ITM option with so much previous volume should have at least some holders selling or at least making asks, even if an algo-trader bought up most of the volume. Should I consider this a probable flaw in the historical data I purchased, or rather is there a reasonable explanation?

33

If you look at DISH's dividend history, you can see that on 20111101 DISH declared a special $2/share dividend payable on 20111201. The ex-date for that 8% dividend was... 20111115. The $2/share drop you saw from the 14th to the 15th was the stock going ex-dividend.

So the stock options (which are American options) were deep in-the-money with a large dividend looming. This is one of the rare cases when early exercise is optimal. If you exercised the options on the 14th, you got that $2 special dividend while if you held the options you lost out on the dividend. The options holders all realized this and exercised.

| improve this answer | |
  • 2
    News are an important part of Equities trading: dividends, share split/consolidation, results announcements, forecast announcements, acquisitions/sales, impactful regulations, supplies' price going up/down, ... and none of this is featured in historical market data. – Matthieu M. Aug 8 at 9:34
  • 2
    @MatthieuM. Would records like these not be available in long format, such as SEC filings? It's not that the data isn't published, but it's not raw trading data so it doesn't feature along side the transactional data of bid/ask. – rbrtl Aug 9 at 19:21
  • 2
    @rbrtl - You are correct. These filings should be in EDGAR and some sources make them available as a dataset. So while these events do not appear in trade and quote data, they are often available historically from market data vendors. – kurtosis Aug 9 at 19:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.