I am getting ready to use a system that I have been developing to trade the emini (S&P) for a 1 hour hold period. I am trying to get a better grasp on my expected costs, and I am very surprised to learn that it could be as high as 20% of my expected average profit (+ 20% of my expected average loss). With costs this high, how is it even possible to make a profit trading the emini over a 60 minute period? Here is my logic: Assume the average 1 hour range is around 3 points. My risk control of 5% of capital per trade on $100,000 account means that I can trade 33 mini S&P contracts. If I expect to give up .25 points on the entry and .25 points on the exit as slippage, that would total ($12.50 * 2) * 33 contracts = $825 slippage. I need to add $4.50 per contract for fees and commissions ($4.50 x 33) = $148.50 for a total average cost per trade of $825 + $148.50 = $973.50 If I can expect to earn an average of 3 points per winning trade (3*$50)*33 contracts = $4950, my $973 cost works out to a whopping 20% of my profit potential! That seems like a pretty high cost to me - am I calculating this correctly?? (and this doesn't even include that cost against my loosing trades). To compare, I figure it would be cheaper to bet on red or black on a roulette wheel in Vegas. There is a 1 / 27 chance that my capital goes to the house (landing on green). So, my costs to play roulette are slightly less than 4%.
You can earn the spread. You don't have to enter at market and pay slippage. Spreads are not a cost unless you don't know what you're doing.
Even if you assume slippage on both entry and exit, your slippage is only 1 tick (0.25) not 2 x 0.25. As an example, imagine you enter a trade long at 1860.00 with bid at 1859.75 and offer at 1860.00. Now you immediately want to close your position without the market having moved at all. You can sell the bid at 1859.75. You've lost commission and 1 tick, not two. Of course, if you are trading 33 contracts and using market orders, you probably won't receive entire fills at the current bid and ask. Think of it this way, you have to make 1 tick ($5) on a trade to cover the $4 commission per contract if you are a retail trader. The larger your profit targets the less this will impact your profitability.
Thanks pspr. I am trying to estimate the difference between my back-testing results where I have used the published settlement price, and what I might expect when actually trading. I don't understand why I shouldn't consider 1 tick slippage ($25) for both the entry AND the exit? ie: if the open price is 1100.00, then I would buy (get filled) at 1100.25 (cost = $12.50) if the close price is 1110.00, then I would sell (get filled) at 1109.75 (cost = $12.50) In total, there is $25 LESS profit than what the back-testing results had shown. Right? Also - I only use market orders, as the risk of not getting filled on a limit order negatively effects the profitability of this system.
You are assuming the spread is 2 ticks. It usually is just one. There are only bid and ask prices. Not a price then an ask above that and a bid below it. There is only a bid and ask. I don't use MOC orders so I don't know what to tell you to expect there vs. settlement. Look at the trading ranges vs 1 tick during your average holding period to estimate your potential percentage costs per trade if your system is viable.
Not necessarily. Let's say you always go in the direction of the market (which is the least favorable for slippage.) Last trade of 20 contracts hits the ask at 1100.25. However, there were 500 contracts offered at 1100.25, so now there are 480. Your buy stop triggers at 1100.25, and your 33 contracts only take out part of the remaining 480 contracts. No slippage. However, your stop could be behind 5 large stop orders that trigger at the same time, and ES is offered at 1100.50 by the time you hit it. 0.25 slippage. Of course, everything changes so quickly, you can only make a rough guess as to your chances of slippage.
According to the 1 tick slippage method of estimating slippage, + commissions and fees, I calculate that it's WAY cheaper to trade the SP pit traded futures contract than the SP emini (# contracts adjusted for equal $ profit). It's interesting that when the exchanges move away from the floor traded contracts to the e contracts, they are shifting the profits from the floor trader (who earns the bid/ask) to the exchange and the broker (who earns the fees and commissions). contract #contracts tickvalue fees total emini 33 $12.50 $4.50 $561 sp full 6 $25 $4.50 $177 This seems way out of line. The costs for trading a full S&P are 1/3 rd of trading a mini?? I must be making a mistake - the slippage for a full must be typically more than 1 tick.
Gregk, /ES futures have cheaper tax rates then $SPY, but $SPY has cheaper commissions and spread. Heres the breakdown: Every 4 ticks (or 1 point) on the ES = 10 Cents on the SPY. So 1 tick on the ES = 2 ½ cents on the SPY. This means every time you make a trade with the ES, the market has to move an extra 1 ½ cents to cover the spread (before you can even think about making a profit on the trade). Then theres commissions, trading the SPY, its like a stock, you can have a fixed commission with unlimited shares per trade. With futures you have to pay a seprate commisson for each contract (usually about $4 roundtrip). Heres an example of a trade with the SPY: You buy 5000 Shares at $110.50, the price has to move to $110.71 for you to make 2 points ($1000) on your trade. Your commission is $14 roundtrip ($7 + $7) so your total profit is $986. Heres an example of the same trade with the ES: You buy 10 contracts at 1100.50, the price has to move (9 ticks) to 1102.75 (thats $110.50 to $110.73 on the SPY) for you to make 2 points ($1000) on your trade. Your commission is $40 roundtrip (10 X $4) so your total profit is $960. Botton line, if your going to be scalping for 4 - 6 ticks (0.10 - 0.15), I think $SPY will be the way to go.
That's an interesting perspective. I've been trading the big S&P contract for years (many years prior to the e mini). I just hadn't considered that trading a SPY - like stock fund would be economical. The only problem is that there is not the same kind of leverage opportunities when trading the actual index (or SPY). What kind of typical slippage is there when trading SPY? Greg