Theoretical and practical problems with linear minimum tick sizes, relative spreads

Discussion in 'Strategy Development' started by stephencrowley, Feb 1, 2006.

  1. I was just thinking about this, and thought I'd throw it out.

    Sub-penny pricing has been eliminated for all stocks >$1, but is still allowed for anything less.

    Why is $1 so magical? If a stock moves beyond the $1 range then the minimum tick size change introduces a structural break into the systems.. spiking volatility, causing havoc, etc.

    Seems pretty silly to me.. here is an example. Say a stock is trading at 99 cents.

    (1.00-0.999)/0.999= 0.0010010

    So the minimum tick size is (0.1%)

    As soon as it reaches a dollar the minimum relative tick size is

    (1.00-0.99)/0.99 = ans = 0.010101

    From 0.1% to 1%, the relative spread changes by an order of magnitude!

    This is a huge change in trading costs/risks/spread etc for no other reason than someone decided $1 was special.

    Now, I don't trade penny stocks so I don't really care that much, but the same problem still applies for high priced stocks.. say, google. Trading at $400 and the relative spread is as small as 0.0025%, but SIRI $5 has a minimum relative spread of 0.2%.

    Has anyone ever thought of the practical and economic implications of minimum sizes, scaling, stock splits, etc?

    This is why logarithmic charts and transformations are so popular, but it doesnt make the problem go away.
  2. You're getting bogged down in mathematical minutiae. Sure, a one-penny spread is more valuable on a $2 stock compared to a $100 stock. In floor trader parlance, the "edge" is more valuable on a less volatile contract(Eurodollars) than a more volatile contract(EuroCurrency). Focus more on limiting losses and implementing a system that you feel comfortable trading. The thing you're trading is irrelevant.
  3. I'm not getting bogged down. I thought about this for about 5 minutes when I was analyzing relative spreads of high priced stocks vs low priced stocks. My commission structure makes it much cheaper to trade high priced rather than low.

  4. Then pull out your urination hose and let fly on the people who aren't as sagacious, wise, knowledgable, experienced, well-rounded, well-versed, topflight, or as expert as yourself. And: make some money in the process. That'll show em.
  5. Haha. I'm not sure if you were being serious or facetious, but funny none-the-less. And that is exactly what I'm doing. :p

  6. has occurred to me as well, particularly with the gold spike, inflation concerns and branching out into us traded foreign equity where the exchange aspect becomes important

    what about pricing securities in %s, reindexed every day based on prev day close. wouldn't be perfect, but it could level the cost and precision concerns associated with granularity

    not sure what sort of system would be required to maintain consistent position value overnight. granularity does seem to lend seemingly arbitrary personality to specific assets.

    or log pricing, not sure