5 Ways You Can Combat High Frequency And Algorithmic Trading

Alright, I admit it.  For a long time, a number of traders who are much smarter than me told me tales of stops being run and seeing mysterious outlier volume prints during the trading day.  For the most part I dismissed these claims, likening them to being the stuff of conspiracy theories.  You know the type.

Man never landed on the moon.  Okay, sure buddy.

JFK was killed by the mob.  Uh huh…..right, right.

My 45-year-old cousin who has never been married and owns a vintage dress shop called “Out of the Closet” isn’t gay.  Yeah whatever….!

But about four or five years ago I slowly started to change my tune as I realized that manipulation by HFT’s and algorithmic trading was in fact true.

The seeds of this manipulation were sown when the exchanges switched to decimalization back in 2000.  This meant that instead of 1/16th of a point (6.25 cents) being the smallest price change that a stock could move, they could now move in pennies.

The change was supposed to benefit the retail trader, giving them tighter spreads and ideally more liquidity, but technology began to bastardized the concept, slowly turning it into a disadvantage for the average trader.

The problem was only exasperated when the exchanges were allowed to go public and the overriding idea was more profit which meant more transactions.  It has now gotten to the point where the computers for HFT firms sit next to the exchange’s servers and rivals compete to see who can have the shortest CAT 5 cable connecting them.

In addition to seeing this problem from the trading side, over the last few years as I ventured over to the brokerage side, I began to get an even clearer picture of the shenanigans that go on.  Now I regularly see fill reports from the exchanges that go six places to the right of the decimal (as seen in this example of a trade in $KBH from yesterday).


What can you do as a retail trader to combat the companies, traders, and bots that have almost unlimited funds and resources and perpetuate this problem?  You have to think “out of the box,” and here are five suggestions on how to do that.

Change the way you trade patterns –  It used to be that trading a breakout from a traditional chart pattern was the money.  But then everybody in town traded the breakout, and HFT’s took advantage of that and trading the false breakout became the money.  But now, trading the breakout in the original direction of the pattern after the false breakout is the money.

Point is, you have to give patterns more leeway, even to the point where traditionally they would be considered “distorted” or “broken.”  Ha, ha, ha, laugh all you want, but if you trade a pattern in the traditional way these days you might scare off a nice baby who’s ready to party.  Sorry, I watched “Swingers” last night.

Embrace the “tail” –  I spent twenty years trading in the opposite direction from tails; the long thin extremities most apparent on candlestick bar charts.  The idea being that price had penetrated those levels, was rejected, and then should reverse away from the tail.

Back in the good ole’ days, meaning 2007, I started to see Trader X talk about the ends of tails as triggers to get into a trade.  For example, using the low of a hanging man candle at support as a trigger to get short, illustrated in the chart below of $FB.


Because I used to be the “world’s smartest trader” and didn’t listen to anyone else, I brushed aside the idea.  But now due to the tendency of HFT’s to “probe” past support/resistance levels momentarily, tails are more prevalent and less significant than they used to be.  Try using levels created where multiple tails line up as the true support/resistance levels to trigger your entries instead of the more traditional levels.

Re-think your ideas on stops –  In the past I have written about three different techniques for placing stops that can help thwart having them run by HFT’s.  The concepts revolve around being more dynamic in the way you place them and using a chess like mentality to think a number of steps ahead.  You can read the specifics of these ideas in my post, “How To Place More Effective Stops.”

Stick to high volume stocks – As of this writing it is still pretty hard for HFT’s to manipulate the movements of highly liquid stocks like $AAPL or $MSFT as well as many widely followed ETF’s.  Keep in mind that the higher the volume and tighter the spread, the less susceptible a stock is to mechanical machinations.

Take a longer time frame – Once again, as of this writing HFT’s can only affect price in the short-term.  As you go towards longer time frames, like in swing trading, you can almost completely eliminate their influence.  The one codicil to that is that when you are tying to find an initial entry for that swing position you may have to scale in or give the position a little bit more room so as not to get stopped out.

BONUS TIP: (don’t say I’m not a giver)

Switch asset classes – HFT’s are most prevalent in equities and somewhat in futures. One area that they are not as involved in, or if they are don’t seem to make as much of an impact, is in options (especially weekly options) or forex.

Trading successfully always involves adaptation.  It’s the nature of the game and always will be.  The HFT’s and algorithmic bots that are currently running amok in the markets can cause difficulty, but with some creativity and strategy their effects can be muted to a great extent.

If you want to learn more about trading and the markets, all while helping to kick pediatric cancer’s ass, check out my book, Trading: The Best Of The Best – Top Trading Tips For Our Times.  It includes over 200 market tips from 60 active traders and investors, and all proceeds go to the McKenna Claire Foundation to help support research for a cure.  Just click the banner below or go directly to Amazon by clicking here.

Updated Amazon Banner

Why not subscribe to bclund.com for free  Via E-mail or Via RSS and follow me on StockTwits and Twitter?

What bclund is, is the intersection of markets, trading, and life (with some punk rock, pop culture, and off-beat humor mixed in). Check out “The Best Of bclund” to get started.

Click here to “Like” the bclund Facebook page.

Brilliant stuff like this rains down like..well, rain, on my stream during the week. If you want to get wet, follow me on Twitter and StockTwits. You can also pick up my book Trading – The Best of the Best: Top Trading Tips For Our Times by clicking here.

  • Saurabh

    I’m just guessing on this, but it would seem that HFT makes very short time frame trading for eighths and quarters much more difficult but doesn’t affect swing or position trading as much unless your positions are huge and those eighths and quarters matter. The flash crash seems to be (so far) a special case. I think longer term price moves are still governed by longer term players (funds) accumulating or dumping enormous positions over time. TA still shows that happening and HFT can’t stop that from happening. HFT increases the volume numbers but if they are net flat at the end of the day, they haven’t really changed the supply/demand balance as the funds do. But they do (usually) increase liquidity, which is good. I will say however that I don’t think it’s right that some players get faster access to price information even if they pay for the privilege. Markets are based on trust and special arrangements erode that.

  • Pingback: Is Day Trading Really A Sucker's Game? | The bclund BlogThe bclund Blog

  • Pingback: Top clicks this week on Abnormal Returns | Abnormal Returns

  • scrilla_gorilla

    Logically, tails may possibly have significance only when a bar represents a “natural” time period (like a day, where the bar represents the open to the close). If a bar represents an arbitrary time period like 1hr or 5m (and to a lesser extent 1 month) then there is no reason why tails would have significance.

    For example, consider this string of prices:
    0:00: 50.00
    0:30: 49.00
    1:00: 50.00
    1:30: 49.00
    2:00: 50.00
    2:30: 49.00
    3:00: 50.00

    It is obvious the price is just oscillating in a range of 49 and 50. But if viewed on an hourly chart, all three bars will have prominent bottom tails.

    Now consider the exact same string of prices but shifted 30m.
    ….
    0:30: 50.00
    1:00: 49.00
    1:30: 50.00
    2:00: 49.00
    2:30: 50.00

    The price action is exactly the same, but now all three bars will have prominent top tails. This may have some significance when you are talking about days (a unit of time that represents a full “natural” period in the real world), but because hours is an arbitrary time period the tails here have no significance. The price actions in example #1 and #2 are essentially identical, but lead to two completely different conclusions based on the tails.

    I think the concept of tails is one that may have made some sense in certain contexts, but like many things in technical analysis is more often applied in nonsensical ways.

    There is one obvious strategy for human traders to take advantage of HFT that you didn’t mention: Wait and do nothing but keep vigilant watch for dislocations caused by runaway algos, and deploy capital to take advantage when the algos go off the rails.

    I can understand when people complain about predatory algos that front-run orders and lift liquidity before your order can fill, but I CANNOT understand when people complain about erratic market behavior caused by algos doing stupid stuff. When algos do stupid stuff, that only provides other participants with more opportunity to profit.

    • http://bclund.com/ bclund

      You sir need to start your own blog (seriously). Good stuff. Thanks for reading.

    • goldmar

      Nice ideas but the problem is: HFT algos can’t fail. By definition. Because when they do something REALLY stupid (like the flash crash), their stupidity gets reversed. For the retail investor this is a lose-lose game.

      • http://bclund.com/ bclund

        Josh Brown had a great quote on this at Stocktoberfest

        “The way you beat high frequency trading is with low frequency trading.”

  • Maisie

    i agree abt the tails. i’ve started to think of patterns as having “cores” and “skins”, the cores formed by the bodies and the skins, the shadows/tails. so far current thinking is, best off waiting for a break of the “skin”, setting stop just 2 or 3 (5m) bars back and taking first half profits *very early*. then if u ultimately get stopped out at least u’ve most likely banked some, plus which u’ve reduced yr risk to 1/2.
    it’s only a matter of time before the hft snake eats its own tail, don’t u think? it’s just an accident waiting to happen. some total disaster will happen at exactly the wrong moment, congress will get into an uproar about it (not least because all their positions will be a mess, lol) and wham, hft/algo will be banned? in the meantime, just have to keep thinking creatively as you do, and make sense of what they do. great thing about the market is, the charts make it all visible and comprehensible if you think enough.

    • http://bclund.com/ bclund

      I like that. I may rip off your “cores” and “skins” analogy…. :)

  • Pingback: Pseudo Random News and Comment | Mortality Sucks

  • Pingback: 5 Ways You Can Combat High Frequency And Algorithmic Trading – bclundbclund « insipidbanality

  • Pingback: Thursday links: stacked decks | Abnormal Returns

  • http://kiddynamitesworld.com Kid Dynamite

    most important tip: HIT THE BID or TAKE THE OFFER…. pay that 1c spread – smallest in history, and be done. bingo bango bongo. Often times, as you even noted in your post, you’ll do 1/2 penny better due to your order being internalized and your net cost will be 1/2c…