Clearly, the tax advantages of long-term gains are superior. But I feel that algorithmic trading, by its very nature, likely doesn't work well for long-term trades. Are these traders happy paying the ~32-40% tax on short-term gains? Why not write algorithms that let you forecast more long-term prices?

closed as off-topic by Chris W. Rea, littleadv, Ganesh Sittampalam, Victor, Noah Apr 10 '15 at 14:51

  • This question does not appear to be about Personal Finance within the scope defined in the help center.
If this question can be reworded to fit the rules in the help center, please edit the question.

  • Who says what the traders are paying for taxes? What if the funds are a pension fund or mutual fund that may not pay taxes directly itself? – JB King Apr 1 '15 at 22:00
  • 4
    I'm voting to close this question as off-topic because it isn't about personal finance. – Chris W. Rea Apr 1 '15 at 22:12
  • Can you recommend what category this might fit under, then? I believe this is personal because I'm planning on trading algorithmically, for personal gains. I just wanted to understand how a large number of trades (what algorithmic traders do) would affect my tax situation, and whether or not they trade in long or short term. – alichaudry Apr 1 '15 at 23:07
  • 1
    If you're going to make this about your tax situation, then say so and please disclose your country. There's no such distinction as "long term gains" in certain places. – Chris W. Rea Apr 2 '15 at 2:32
  • Assuming you're talking about the US, I think you're missing a fair bit of knowledge about the tax code. For one thing, a lot of algorithmic trading is done in the futures markets, and income from futures trading is taxed in a totally different way from normal investment income. Definitely read up on the tax implications before deciding to pursue trading as an avocation. – dg99 Apr 2 '15 at 15:03

Algorithmic trading essentially banks on the fact that a price will fluctuate in tiny amounts over short periods of time, meaning the volatility is high in that given time frame. As the time frame increases the efficiency of algorithmic trading decreases and proper investment strategies such as due diligence, stock screening, and technical analysis become the more efficient methods.

Algorithms become less effective as the time frame increases due to the smoothing effect of volatility over time. Writing an algorithm that could predict future long-term prices would be an impossible feat because as the time frame is scaled up there are far less price fluctuations and trends (volatility smooths out) and so there is little to no benchmark for the formulas.

An algorithm simply wouldn't make sense for a long-term position. A computer can't predict, say, the next quarter, an ousted CEO, a buyout, or anything else that could effect the price of the security, never mind the psychology behind it all. Vice versa, researching a company's fundamentals just to bank on a 0.25% daily swing would not be efficient.

Tax advantages or not, it is the most efficient methods that are preferred for a given time-scale of trading.

  • 2
    It is true that long-term strategies are typically less profitable than short-term strategies, but it's because of reduced certainty over time, not reduced volatility over time. No long-term strategy cares about volatility; volatility is only important to trading because it provides many opportunities to capitalize on mean reversion. But reversion to the mean does not really hold in the long term, and even if it did, a strategy would not be able to profit meaningfully from it. – dg99 Apr 2 '15 at 15:19
  • Great answers/comments! So the gist of it is that I should use algorithms to develop short-term strategies, and that the volatility in the short-term market should be high enough that even after short-term capital gains taxes are accounted for, there is a profit. – alichaudry Apr 3 '15 at 0:25

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