What level of slippage to expect with stocks?

[AMSH]

Hey all,

I've run into a wall with my systems development as I've got no idea of what level of slippage to build in (as I've done little live trading). As far as I can see this is something that's determined through experience so I'm looking for people's opinions on the level of slippage that is encountered when trading stocks and on the questions below:

1. What is the minimum price tick on stocks (ASX)? I've been told both .5c and .1c. Does the minimum tick change with SP - i.e. does a $20 stock trade in half cents and a 10c stock trade in tenths of a cent?

2. Does the price info provided by comsec (for example for last trade at 10.905) represent the same value that that person taking the trade would have paid/received? In other words, does comsec fill orders at prices to 3 decimal places, or could that trade actually have been filled at 10.9048?

3. What level of slippage have you experienced from live trading? This might be slightly too broad a question, but any opionions would be helpful.

I presume that the level will largely depend on the share price of the stock (or possibly the average range relative to the SP) so perhaps information classified by SP would be appropriate?

Any input would be greatly appreciated.

Cheers,

AMSH

 

[tech/a]

1. What is the minimum price tick on stocks (ASX)? I've been told both .5c and .1c. Does the minimum tick change with SP - i.e. does a $20 stock trade in half cents and a 10c stock trade in tenths of a cent?

Under 1c in .001 of a cent
1c-10c in .01
10c to around 50c in .5c
50c + 1c

2. Does the price info provided by comsec (for example for last trade at 10.905) represent the same value that that person taking the trade would have paid/received? In other words, does comsec fill orders at prices to 3 decimal places, or could that trade actually have been filled at 10.9048?

Probably at the opening or closing auction.

3. What level of slippage have you experienced from live trading? This might be slightly too broad a question, but any opionions would be helpful.

Logical stop points like pivots or support resistance will give greater slippage as thats where most place their stops I place mine above these levels (And so do more and more I notice) can be nasty!

I presume that the level will largely depend on the share price of the stock (or possibly the average range relative to the SP) so perhaps information classified by SP would be appropriate?

Illiquid stocks can be murder just dont trade them.

 

[MS+Tradesim]

1) Slippage depends on the stock (and the trader). In penny stocks you often have no slippage or can get the bid/ask (some people might consider paying the spread to be slippage but I don't so you could get some different opinions there). In higher price liquid stocks you may find it a bit more "jumpy" but comparitively (on a % basis) slippage can be managed with sensibly placed limit orders...unless you are chasing a momentum driven stock and then all bets are off.

2) Tick size. Up to 10c a tick is 0.001. Up to $2 a tick is 0.005 and then above is 0.01.

3) Orders are filled according to the tick value. If an order is executed over several price levels, the average price may go beyond 3 decimal places but no actual transactions occur at less than 3 dec. places.

4) I don't get a lot of slippage but that's probably just experience. IMO, slippage is only a real issue for scalpers who are shooting for very small moves. Maybe go 2%. But that seems high to me. Dunno. Maybe someone else can help with that one.

 

[AMSH]

Thanks for the replies guys, that's exactly what I was after. If anybody's interested, I've written some code for MS to add in slippage by price tick. Let me know if you want it. Unfortunately, three ticks on the buy signal just about killed my short term system. Back to the drawing board.

 

[howardbandy]

Greetings --

I do not know if this is helpful to you or not, but I have looked into the bid-ask spread and slippage on the US exchanges.

If the issue being traded has average daily liquidity (that is, price times volume) of $100,000,000 or more, the bid-asked spread is usually 1 or 2 cents -- sometimes a little higher during the periods of low activity during the middle of the trading day. There are about 500 issues, including 300 of the 500 stocks in the S&P 500 and over 100 ETFs, that meet those criteria. The average share price on US exchanges is higher, say around $25 for the stocks in the S&P 500, than on Australian exchanges.

Commissions are so low that they can be ignored. Interactive Brokers charges $0.005 per share for the entry and $0.005 again for the exit -- a total of one cent per share per round trip. A speaker representing the brokerage industry whom I heard recently described the trend of commissions for retail traders going "toward zero". A few brokers already charge no commission when the account balance is above a minimum (which can be as low as $5000).

Total commission plus slippage on liquid issues is about $0.03 per share per round trip, or 0.12% of $25. If a trading system gains one percent (1.00%) per trade on a $25 stock, the gain is $.25. Compare with $0.03 for frictional costs.

Thanks,
Howard

 

[AMSH]

Thanks for that Howard. The US stuff gives me a general idea, but I'd be interested on your opinion on whether trading on US exchanges is (1) feasible and (2) worth it for a beginner system trader (with very limited capital). Is the main advantage of trading on a US exchange the general liquidity and the range of stocks or are there others - (I notice many reasonable systems that are killed by restricting the trade to 5% of volume traded and it'd be good to have a larger stock base to play with)?

On a related note, would you guys think that restricting trades to 5% of the traded volume (and I do understand the problems with this measure as volume isn't known till the close) too restrictive or too loose? In a penny stock would trying to open a position at 5% of the volume traded that day be enough to cause a large amount of slippage?

Sorry about the nonstop questions. I just want to have some truly realistic backtesting results before I start putting my money behind a system.

Cheers,

AMSH

 

[MS+Tradesim]

AMSH,

5% on a penny stock can cause slippage - depending on the volume for the day. On short term systems I restrict it to 2%. I want to be in and out - I don't want to be responsible for shifting the price, and I don't want to have to take time filling a position that I may only hold for a few minutes to a day.

I have one exception. If I'm taking a position on or near to close, afternoon sell-offs often mean I can get bigger and cheaper fills.

 

[howardbandy]

Hi AMSH --

Right now, February 2009, the US is on standard time and Australia is on Daylight time. The US NYSE and all US floor trading opens at 9:30 AM in New York and closes at 4:00 PM. In Sydney or Melbourne those times are 1:30 AM and 8:00 AM, respectively. In Brisbane (no daylight time) -- 12:30 Midnight and 7:00 AM. In Adelaide -- 1:00 AM and 7:30 AM. In Perth -- 11:30 PM and 6:00 AM.

The volume is highest at the open and at the close. So you can trade the US markets if you either stay up late or get up early. If you are taking your signals at the close and trading at the close, the Australian morning corresponds with the US close.

If you have a trading system that uses end-of-day data and computes its signals after the close, there is a very large reward for taking the position on the close of the daily bar that gives the signal rather than waiting for the next day's open. You have to anticipate what the close will be, but there are many ways of doing that. For example, use real-time quotes and trade two minutes before the close; or compute the signal price in advance and place limit-on-close orders (providing your broker accepts them).

Almost all of my personal trading is done with Exchange Traded Funds that follow broad market indexes or sector indexes and that have average daily liquidity of US$100 million or more. I often use the leveraged ETFs. My holding periods usually range from intra-day to a few days, and sometimes as long as a few weeks. I also trade highly liquid options on broad market indexes. My positions are so small relative to the volume of the issue I am trading that they never move the market. You can do this from Australia.

The bid-ask spread on the broad market and sector ETFs is usually one cent on ETFS that range in price from $10 to $80. The bid-ask spread to purchase or sell the option is five cents or less (it can be as low as one cent if your broker uses "penny pricing" for options, otherwise five cents) on options priced at around $3.00 to $5.00.

Stocks that are very low in price cause any bid-ask spread, slippage, or commission to be a larger percentage of the trade than for a higher priced issue. But very low priced issues move a higher percentage more easily than higher priced issues do, so you might make that up easily. It is the liquidity that I would worry about more than the share price. Personally, I want to see $100 million per day, or close enough to it that the bid-ask spread is one or two cents. There are only a few stocks listed on the ASX that trade that much -- BHP, etc. I do not follow Australian stocks intra-day. You might want to do some studies using intra-day prices and bid-ask spreads. Let us all know what you find.

Thanks,
Howard

 

[AMSH]

Hey Howard,
Thanks for the concise and in-depth reply. The info re the US is interesting but I'm not ready for it yet. It's crazy how much profit there seems to be in the specy stocks on the ASX and frustrating that at least 90% of it is off limits due to volume / activity constraints. Looks like I'm going to have to look at altering my approach to trading as the volume constraints are currently killing everything I try, but I don't think leveraged instruments are the way to go yet. I'd like to be able to make a consistent profit (even a small one) with ordinary shares first; then I'll look into ETF's and options.

With regards to the slippage, I'm starting to think that the amount involved with a system is probably too heavily based on the type of system and instruments traded to allow blanket statements about what level I'll encounter. I think it's probably going to be a case of starting out expecting a lot and then adjusting my expectations (and system criteria) based on at least some level of active trading. Thanks for all the input. This stuff really is more complex than one would expect isn't it? Damn.