On 1/27/07 C2 reports a return after commission of $475K and after real-life slippage of $464K. Dividing the difference by 4590K shares the system traded since its inception, slippage is assumed to be $0.0024/share. My average slippage is ~20 times that figure!!! (with 100-200 shares per trade)
Question to stock traders [I trade only futures]. this extreme-os review states very high slippage. Has anyone found that trying limit orders with C2 stock systems reduces slippage? Or are the stocks too thinly traded for this approach?
Has anyone found that trying limit orders with C2 stock systems reduces slippage? Or are the stocks too thinly traded for this approach?
Ross, my expereince is that slippage is directly related to instrument volatility and speed of trading. The faster one trades any instrument, the more slippage there will be. In fact the reason most would give for faster trading speed is to take advantage of volatility as well as leverage.
The faster one trades stocks – especially more volatile ones – the more slippage will be an issue. Sure some stocks are thinly traded, but not many from my perspective, and I avoid those.
Its really very simple… the longer you can wait for the price you want, the more likely you are to get it. That’s why I perfer weekly stock trading on equities with sufficient trading volume.
I forgot to mention that I trade only limit orders precisely to avoid slippage.
>Sure some stocks are thinly traded, but not many…
I would say there are several thousand thinly traded, low
volume stocks between the NASDAQ, NYSE, and AMEX.
> I would say there are several thousand thinly traded stocks…
I define this as frequent gaps on say a 5 minute chart…IOW, there
are often no trades at all for 5 minutes at a time. Moreover,
there are frequent quotes at minimal volume (say 100 shares) for
the entire 5+ minute time period.
I think the question is too general. It depends on the system, the liquidity of the stocks that it trades, the number of subscribers, and their volumes. I trade extreme-os with about the same volumes as that reviewer. The entry orders are limit orders about 10 cent above the C2 price. Usually they are filled and then my slippage at this side is 10 cent or less. If the limit order is not filled then it will be converted to a market order after some time and then the slippage can be more; say 30 cent. The exits are market orders and I think that this slippage is much more often larger than 10 cent. One possible explanation is that this is caused by a large number of subscribers sending a market order at the same time. If this explanation is true, then a limit order at the exit could reduce the slippage. Another explanation is that the (discretionary) timing of his exits is better than the (mechanical) timing of the entries. If this explanation is true, then a limit order at the exit can increase the slippage if many subscribers are not filled. I think it will be hard to determine what the net effect is without a randomized experiment. And even then it will only apply to this system.
I would say there are several thousand thinly traded, low
volume stocks between the NASDAQ, NYSE, and AMEX.
True, but…
1) Volume criteria differ. If you require average daily volume (ADV) greater than 100,000 shares, then yes, you have eliminated just under 50% of about 8k US stocks. If you require 50,000 shares ADV, then you have eliminated over 1/3. If you require 10,000 shares ADV, then you have eliminated 1/4. 2) Remember the law of supply and demand. Lower volume is not always bad, and higher volume is not always good.
3) The faster you trade and the larger your trading account, the more important liquidity is. Personally, trading weekly, I have not had any significant problem getting fills in larger equity accounts even trading stocks with less than 100,000 shares ADV.
3) Bottom line is whatever your ADV requirement, there are thousands of stocks to choose from.
My experience with stock systems is this: If a system has high slippage from market orders, forget about trying to match the C2 hypothetical fills with limit orders, unless you’re a very skilled trader and use the entries and exits more as “indicators” for your own trading rather than hard entry and exit points.
What happens is that as soon as the signal is issued for a popular system, all subscribers are trying to get in or out. This means that before you would be able to enter your limit order, many times the stock is already trading above (long entry) or below (long close) the hypothetical C2 fill price. If you try to match this price with a subsequent limit order, you essentially hold on an hope (how funny…) that the price bounces back to the level of the hypothetical C2 fill. Nine out of ten times this might actually happen, but the 10th time it might not and you’re forced to get in or out at a much worse price and give up all your previous gains.
In addition, it is a common misconception that high slippage is always a problem. It all depends on slippage relative to the average profit per trade. E.g. even if a system trades very thinly traded stocks with $0.30 slippage per share, slippage shouldn’t be a real issue if the average profit per share is $3.
This senario is much more true for day trading than longer term.
System popularity not withstanding, it is very rare for any stock to not retrace its position a few days or even a week or so later. Why is it unwise to chase an entry? Because stocks so often retrace. IMO, waiting for your price is wiser and therefore often times trading slower is better.
WRT system popularity creating excess demand at long entries or supply at long exists, etc, I was just reading on another message board, a thread about Gorilla Trades, a very highly promoted system. Many subscribers or former subscribers told of their experiences, and slippage was not a major issue primarily because the trading time frame is longer and the number of positions held is large. Therefore meeting exact entry and exit price is not crucial to success. Sometimes you might do better, sometimes worse. It averages out. This is my experience as well.
Again, if you are taking about day trading, that’s a different story.
I agree that my observations might apply less to longer timeframes.
But in your scenario, where the probability of a retracing is high, the vendor could have used limit order signals himself in the first place; no need to use market orders in that case.
True. I said earlier that I always use limit orders. However, I have found that quite often a few days or even weeks later I may have a better price than I got with my original limit order (meaning that my trade went into draw down mode.) IMO the market is not very predictable and 3 of 4 stocks follow the market direction, so it is not possible to avoid this senario on every trade. Just get the best price you can at the time, and know your exit point at the time of entry.
> E.g. even if a system trades very thinly traded stocks with $0.30 slippage per share, slippage shouldn’t be a real issue if the average profit per share is $3.
Don’t agree with that. If you mean .30 per side that’s 20%, or 10% if you
mean per RT. Take 10%-20% out of EVERY trade (win, lose, or draw)
it more often than not IS a problem.
Often those kinds of costs will turn a system with 75% winners into
a 50/50 or less system. Test it (TS can add any kind of slippage you want to a system test) and see. Nearly all the small marginal or breakeven trades will push into the loss column. Maybe the system will still eek out a profit, but the real life psychology of trading the system would be dramatically altered.
Perhaps I’m confused because I will have dinner in 2 minutes, but if the average profit per share is $3 and there is .30 cents slippage per share on each side, then the average profit will become $2.40, which is still positive.
> if the average profit per share is $3 and there is .30 cents slippage per share on each side, then the average profit will become $2.40, which is still positive.
The key word here is the big IF. The reality is even the very best stock
trading systems on C2 average about .60 per trade. Many very good ones are around .40. Can you see where .30 slippage per share on each side might be a problem? Anyway, my point wasn’t even that these systems would become unprofitable, but that a high percentage of small winning trades would become small losing trades. I looked at some of the very best stock systems, and even assuming 10% additional slippage (.04-.06) per RT trade the winning percentages dropped from
75-85% to 50-65%. For the more marginal systems the story is much, much worse. There will also be a dramatic effect on stats like consecutive winners and losers, etc. While these stats may not have that big an influence on the total profit, they have a dramatic influence on
how trade-able and sell-able a system may be. Again, anyone who has ever fiddled with TradeStation or similar software will have seen this before.
Sam, as I’ve said elsewhere, what you are describing applies mostly to faster ‘day trading’ stock systems. Those that hold positions for days or weeks are not nearly as vulnerable to slippage.
>what you are describing applies mostly to faster ‘day trading’ stock systems.
I was responding to the specific post that said 30 cents slippage wasn’t a problem. Relatively large slippage can be a problem for long term
systems too. The specific examples I gave were all from longer term (not day trading) stock systems. Once again the very best stock systems I found had average P/L’s per unit around 60 cents. I suspect yours is around 40 cents, but that is just a guess because there are no closed trades.
> Those that hold positions for days or weeks are not nearly as vulnerable to slippage.
True to an extent, but there are systems here on C2 that trade size
in thin markets. Whether you take it out in slippage, the RF, or whatever, it is something to consider. Doesn’t matter if your time frame is weeks or minutes if the market is only at your price for a few seconds.
I see your point. Perhaps the clearest example is what I read about ‘extreme-os’ in subscriber reviews. Looking at its stats, I noticed the small $ difference between average winners and losers. To me that’s a hint that slippage could be a problem.
But I continue to postulate that if your trading plan is to hold well past the first bounce (or drop when shorting) that slippage will be far less of a problem.
I already had understood all the issues, but I was wondering whether anyone instead tried limits, even when the vendor uses market orders…
It SEEMS that using limit orders (at least, for entries) could take out the slippage, even if you missed some trades. But on the other hand, this might causse you to miss the best trades, if the market strongly moves towards a profitable trade/exit. Thus it is partly a tradeoff between avoiding slippage and missing good trades…
I agree with the person that the market often retraces back once you are in the trade. And perhaps waiting til it retraces or diong limit orders, you are competing less with OTHER people following the same system who are trying to get in…
I was responding to the specific post that said 30 cents slippage wasn't a problem.
Not quite. As you do quite often, you changed the question/statement to fit your answer so that you can be right. The post didn't say that 30 cents slippage wasn't a problem. The post said 30 cents slippage is not a problem if Average Profit is $3. You conveniently ignored the second part, because it doesn't fit your answer.
The statement was:
"E.g. even if a system trades very thinly traded stocks with $0.30 slippage per share, slippage shouldn't be a real issue if the average profit per share is $3. "
You said you disagree with that and continue to use an example of a system with $0.4 average profit. With $0.4 average profit, $0.30 slippage is a problem, but this is not the point which was made. The point which was made is that with large average profits, slippage is not such a big deal. You are talking about systems with small average profits and although what you said is true, it's not relevant to the point which was made and you are talking about something different.
