12

Whenever I buy a security, I always use a Limit Order. Just to be safe, I input a price just below the Bid Price, e.g. 21.80. Why? In case the spot price dips suddenly! See below.

This cheap trick matters more for the Options Contract Multiplier! A penny (→ dollar for options) saved is a penny (→ a dollar for options) earned!

If my order doesn't fill at 21.8, then I notch up the limit order price by one penny to 21.81. If my order doesn't fill, then I notch up the limit order price by one penny to 21.82. I repeat until I fill my order.

  1. What's the official term for this algorithm? I spoke to an MFin CFA and MFin MBA CFA, but I had to explain for ten minutes before they understood.

  2. What do you reckon of this algorithm? Is it beneficial? Can it be improved? Is there a better method to save on the bid-ask spread?

7
  • 1
    what's the maximum price you would be willing to pay? why not just put that as the order price?
    – user253751
    Apr 29 at 8:34
  • 23
    You make it sound like using a limit order gives you and edge over other market participants and that's not true. All you are doing is selecting a buy price below the market and hoping that if price drops, you'll get a fill there. Apr 29 at 11:38
  • I may be ignorant, but if you put in a buy order at 21.80, but there are some available at 21.75, won't the purchase be made at the lower point? What are you gaining by bumping up by a penny instead of, say, a nickel? Or, as @user253751 says, just by putting in the most you are willing to pay? Apr 29 at 16:37
  • 1
    @Michael Richardson - The OP is placing a limit order below the bid ($21.80) not above it. Therefore, it is just a limit order. Apr 29 at 16:58
  • 1
    Interactive Brokers (as per your screenshot) has a bot algorithm somewhat like this that they call "Adaptive Algo". From the KB: An adaptive limit order goes in near the bid price (for a buy) or the ask price (for a sell), and waits to fill. If it doesn't fill, it is resubmitted at a slightly more aggressive price. It continues this pattern of "wait and move" within the spread until it fills, is canceled, or hits the limit price cap/floor. Apr 29 at 22:26

9 Answers 9

47

"If the price dips .01 before I hit enter, my method saves me .01!"

And what happens if the price bumps up .01 before you hit enter? And then you raise it a smidgen, and the price bumps up .01 again? And again? Will you chase the price up until the order fills?

I would say that splitting hairs on the sub-penny cost of an order is a sign that you are either:

(a) A high frequency trading algorithm run by a large financial institution; OR (b) A novice trader who is overly confident in the precision of their methodology.

Be careful that you don't trick yourself into getting involved in financial dealings outside your abilities. Day trading is risky business.

1
  • Frequent trading can also have tax implications (classification as income from work). Different type of risk.
    – MSalters
    May 2 at 12:52
23

I don't see how your approach achieves anything.

If the current price is 21.9, then offering to buy at a limit of 21.8 will always fail. Even 21.89 will fail. You'll end up increasing the limit bit by bit until you hit 21.9 and then you will succeed.

The simpler answer is to decide how much you're willing to pay. If that's 21.95, then offer 21.95. If the current price is 21.9, then you'll still pay 21.9, but will have saved a lot of pointless effort in the process.

9
  • 3
    +1. Keen advice: "decide how much you're willing to pay". That's the foundation of all negotiation, which trading is just a specialized form.
    – bishop
    Apr 29 at 15:54
  • 1
    what is the evidence that placing an order below the bid will fail? Apr 29 at 18:13
  • 4
    Placing a buy order below the bid is going to fail unless price drops. You also have a VERY good chance that you get no fill and you miss out on a better price. Apr 29 at 19:21
  • 2
    @RalphWinters What you just be said makes no sense to me. Apr 30 at 18:12
  • 1
    What do you mean by "current price" - best bid, best ask, or last traded?
    – Nayuki
    Apr 30 at 19:19
17

You're basically emulating what a market order will do. Thought at a much slower timescale.

You can claim that you're saving pennies here and there, but you have to remember that every time you change the price, your order will then have the lowest priority at the new level.

For instance, if you have a buy order, and increment the price from $4 to $5, fundamentally, it will only trade when after all other $5 buy orders are filled.

Assuming that all the previous $5 buy orders are filled, you will be the next cab off the rank. But rather than be a $4 buy order, you have a $5 buy order. You've just potentially wasted yourself $1.

Your algorithms success basically is dependent on what the market will do in that time. If you're looking to buy, and the price is going to go up, a market order that fills quickly will be better than your algorithm which may be chasing orders to fill quite slowly, and your final trade may be at a much higher price than you would have gotten initially.

Conversely, your algorithm will "do well" if the market goes down because that initial order will fill at once price step below spot price as opposed to potentially multiple levels above spot price.

In summary:

If there is money to be made with such "algorithms" the high frequency traders are already doing it. And they will do it better than you.

1
  • +1 for losing priority (which may cause you to pay more than you would have otherwise as you increment into higher prices), and another +1 for "the HFT guys are ahead of you." If the OP's strategy were trading against the simple market-making system I wrote way back when (which was not particularly high frequency, actually), my system would have been making money from the OP.
    – cjs
    May 2 at 6:00
6

I do not see how your mechanism is beneficial for a couple of reasons.

First, when you place a limit order, it is an upper (or lower) bound. If the market maker can improve upon it, they are obligated to do so. It is also in their interest to do so because you are allowing them to substitute your liquidity for theirs. You are adding liquidity to the market and it is in their interest to reward you for doing that.

I think part of the problem is that you do not seem to have an actual reservation price. If you had a reservation price of 22, then an ask of 21.90 puts you ahead. If your true reservation price is 21.80, then you are worse off by raising the price to 21.81.

It sounds like you are wanting it to fill, but you want to feel like you won something extra. I think you are overestimating your role in the transaction.

As already mentioned, if the price continues to climb, you will be chasing the price increase and exceed the 21.90 that you could have had. However, if it falls, it isn't likely due at all to your actions. Indeed, you may be working against your own interest.

If the price falls, it is because someone else is selling shares, maybe a block of shares.

If the dealer wants to buy inventory, they will use your price as a floor. When you raise the price, they will raise the floor. If they get a large order, then they will invade the floor and take your order. If the dealer needs to unload inventory, they may take your price at the first opportunity if there are no intervening orders. You may get your 21.80 if you are simply patient. It may also never reach that point.

This discussion gets a bit more complex if you can see the order book because there are a number of shenanigans played in order books by other participants such as iceberg orders and so on. Indeed, you seem to be trying to play shenanigans but I don't believe you are having any impact.

I think you are seeing this as if you are approaching a cliff edge, trying to touch the edge without falling over. The part you may be missing is that the cliff edge is moving. At 21.80, if you could see the future, you may find that you are way over the cliff edge already and that by altering your order to 21.81 you lost a penny. The difficulty is that you cannot see the counterfactual case where you did not alter your order. Your system may well be losing you money, but as an opportunity cost so that you cannot actually see it happen.

I don't think any of us can recommend a strategy to you because we do not understand or know your goals. Your goals should drive your strategy. The rules that you use to accumulate a large position will differ from those attempting to engage in day trading. The other problem is that this forum does not support mathematical notation, so potential strategies to conserve cash flows cannot really be discussed here. You could try reading on the Quant SE.

This strategy doesn't have a name because its function appears to be emotional and not financial. I don't think you could systematically improve on the bid-ask spread with this method, even if you were a mega trader.

4

For normal trading where you're looking for price improvement on a wide bid/ask spread, if your broker offers it, you can use a Pegged-to-Primary order.

You select a maximum amount that you are willing to pay as well as an offset amount which is the increment. For buying, as the bid rises. you buy order is increased by the amount of the increment until the maximum amount is reached. If share price increases past your maximum amount, your order stays on the book as a limit order at your maximum price.

0

What you are doing is "trading in the spread", meaning that you are trying to avoid having your order filled at the bid or at the ask.

In reality you are probing the price of market makers and some advanced, invisible market orders. Not all orders are visible, so not all prices are visible either. By changing your order a price epsilon each time, you are trying to touch one of these invisible prices, if any.

It is beneficial on the options market, as you already noted, but is because the options markets are very liquid and they have a mathematical model behind. This will be less effective in markets that are slow.

As it can be improved, no, at least not manually. Is here where algorithmic trading shines. These market makers and advanced orders mentioned above? They are basically placing or taking offers from the market that are somewhat beneficial to them. Exactly what you are trying to do manually.

0

To answer your first question, the closest term I have found to what you are doing is called a "Pegged to Primary" order. Here is a description from the Interactive Brokers Website:

https://www.interactivebrokers.com/en/index.php?f=613

"Relative (a.k.a. Pegged-to-Primary) orders provide a means for traders to seek a more aggressive price than the National Best Bid and Offer (NBBO). By acting as liquidity providers and placing more aggressive bids and offers than the current best bids and offers, traders increase their odds of filling their orders. Quotes are automatically adjusted as the markets move, to remain aggressive."

This is controlled by an algorithm in which you could put in a Buy order at the current Bid (or Ask) and then add an offset, e.g +1 or -1 ticks, and the your order will change as the bid/ask goes up and down. That is why it is also referred to as a Relative Order. But not all brokers implement this type of order. I am just referring you to a description of this so that you can discuss this terminology with your broker and have them comment on this. I am not commenting on whether it works or not.

0

This algorithm makes sense in the context of delayed market information, e.g. Interactive Brokers provides delayed information about market data for free but you have to pay extra for real time market data (https://ibkr.info/article/2966). On Interactive Brokers you will get notified immediately if your orders fills or not after you have placed it. So this playing around with order price gives you some information about the current bid or ask without having to pay for real time quotes.

Example

Suppose you want to buy a stock trading on Tokyo Stock Exchange trough Interactive Brokers. In that case you only see market data with a delay of 20 minutes. The ask price you see in the interface is 2000 JPY per share. You place a limit buy order at 2002 JPY per share. It doesn't fill. You now know that the current ask is above 2002 JPY per share.

1
  • 1
    Yes, but if your trading strategy depends on current information then it doesn't matter anyway because you will be completely outperformed by other traders who are paying their fees for current information and who are manifestly better equipped to exploit that information. All real opportunity windows will still be closed before this strategy can successfully make you money (for reasons other than luck).
    – J...
    May 2 at 12:40
0

With wide spreads and trading in a range you'll probably get better fills by sitting on the bid while bumping it up. The reason is securities wiggle around a lot and chances that the price will come to you are pretty good. But if that's what you're hoping for you might as well sit on the bid or even below that and you'll still have good chances to get filled. Where you'll lose out is when the security is moving against you, and you'll have to chase it.

Some people mentioned REL orders, in my experience they don't work very well with volatility. I believe HFTs have ways to work them all the way up to ASK and fill you there, and then you'll see the bid immediately drop to where it was previously.

Adaptive orders - maybe, but here again I have noticed that you're either getting filled on ask in urgent mode, or just sitting for long periods if normal or patient mode, and running the risk of the market moving against you.

Remedies - in my opinion it is best to place limit orders in the middle of the spread, or market orders if you must get filled.

New contributor
pyproper is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.

Your Answer

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