I have just glanced at the proof and not studied it long enough to refute it. However, it seems everyone is talking about something different. So I will point out what I was talking about. I was referring to the fact that the max draw down INCREASED when applying a stop. SO this means that if you take the biggest peak to valley on the PL chart its bigger than had you not used the stops. Yes it may be true that for each trade it lowers the max drawdown, but when applied to a system of trades it can increase the drawdown. The reason for this is pretty simple. Typical price action may have the trade going into the red -$500 before going positive $500... well if you had a stop smaller than -$500 you get stopped out. Well you can argue that this reduced your max draw down. Which is true for the trade, because perhaps the trade would have continued down to -$2000 or something, but instead was stopped at -$500. So for a specific trade this is valid. Now what happens in a system of trades? Well it has to do with the probability of an event happening. What if very often the trade goes -$500 before going positive.. or the trade doesn't even have to go positive after, it just improves to say -$200.... So assuming with a high probability a trade goes negative 500, then everytime it happens your getting stopped out.. so over a series of trades what if you got stopped out and realized the full $500 loss. So imagine a lot of trades getting stopped out that would have gone for a profit or improved from that point had you let it run its designed course. This only becomes the correct course of action if its RARE for the strategy to reach this stop value, and if its RARE for it to rebound after reaching this area. Sure this stop may protect you from a few -$1000 plus type days.... but is saving that extra 500 a VERY small percentage of the time worth possibly realizing that 500 loss very often? Anyway, this is just late night ramblings and I probably did not properly illustrate what I am trying to say. Maybe someone else can chime in and build on what i'm saying.... In the end proof or no proof I know from lots of testing that normally stops produce smaller profits and bigger draw downs than without.
agreed i dont think anyone quite understood what my original question was addressing. and i agree with you that a stop can increase the drawdown when applied to a system, for a specific case (specific chain of events). like with yours going from -500 to positive, but what you see in testing, or what you expect, doesn't speak to the 'maximum drawdown' of your (or any) specific system. but if you take all the possible outcomes of market movement, ie. getting worse after hitting -500$pl, the system will always have a greater 'max possible drawdown' without the stop, since the PL can always get worse than -500$p etcl. probability of events shouldn't factor in when you are choosing to look at the 'worst case scenario', since you dont care how infrequent it happens. take this sequence of events, every trade goes to -500, then gets worse, never better. which is entirely possible, that series of trades will give you your maximum drawdown of your system. average or expected drawdown is a whole different story, which is probably what you are thinking more about, since it has more of an bearing on overall performance, max drawdown has more to do with risk. not that im saying you should use stops or focus more on risk control. was just surprised about your results when testing. i wouldn't have been surprised if you saw a change in avg drawdown when adding stops, but that you saw an increase in the maximum on your test data. i guess that speaks to the probability of hitting that true maximum drawdown on a finite test set. which unfortunately for systems without stops, since the maximum may be greater than the account value on many systems. my apologizes for my hard to read midnight ramblings too.
this convo just gave me an epiphany about why some systems blow out accounts even though all their backtesting shows very manageable draw downs. i guess in the worst case every system, regardless of stops or not, can blow out an account. the difference is with stops, you can define your maximum loss per trade, so you know how many losing trades must happen before your account is gone. without stops a single trade can wipe the account, regardless of what backtesting has shown, since the chances of seeing the max drawdown is very small. its important to understand the difference between max observed daily drawdown and max possible daily drawdown.
the discussion is kinda dancing around the point i was trying to make earlier... any strat can have a hard/ emergency stop. you simply find the value that does not hinder the performance. it may be a wide stop, or not. i doubt any developer would say, "ok , i need a stop", and arbitrarily apply a xxx$ stop. they would test.. if i tested my system..and it performed best with an excessively wide stop (subjective?).. i might have 2nd thoughts.. but thats just me
Hmmm ... I have been experimenting with a reversion to mean type system applied to a (screened) basket of stocks eg DOW 30. It trades intraday - always flat at EOD. I have not been able to find any stop loss strategy that improves any measure of performance - net profit, maximum drawdown, profit factor, Sharpe ratio etc. You name it - they are all worse if stops are applied. The simple measure of closing all open positions with MOC order beats anything else for positions that do not hit their target during the day. As I said in a previous post, diversification can be the best way of dealing with unacceptable risk.
I guess we are talking about two different things - your proof seems to refer to max (unrealized) drawdown of a trade while the quote that I cited and refuted is referring to the realized loss or profit of a trade: Since that statement said "loss" not "drawdown" I stand by my assertion that it is false. Adding a stop can most definitely result in a greater loss on a specific trade.
Its quite possible to take a system that consistently unprofitable, reverse the buy/sell signals and end up with a system that is consistently unprofitable.
Its telling that folks that actually have done systematic backtesting have one position regarding stops while it seems those in the other camp advoacting stops do not seem to have any real experience in this area. Its definitely not intuitite - I must admit I was surprised the first time I applied a stop to strategy I had and saw how much worse the results became. Then I analyzed it, trade by trade and saw what is happening. Its very rare for any trade to be immediately profitable upon entry. The majority of trades spend some amount of time underwater. If your stop is too tight then you turn many winning trades into losing trades. If your stop is too wide then it really doesnt help because you still take heavy losses on those stopped out trades and you still miss out on a few trades that would have turned around and ended up positive. In a trend following system you really shouldnt need a stop. Assuming that the system trade exit criteria is watching that the trend has reversed then trades should naturally exit on their own anyway (if you are long and the trend turns down and stays down then the system should want to exit on its own).
No, me too. Whatever the inner workings of the (automated) system are, for it to, after opening a trade, to accept a contary move as large as the avarge profit/trade sounds awkward to me. Assuming that the system doesn't use some kind of 'magic' and only tries to find high-probability entry-points based on passed price/volume/time action, isn't the fact that it moves to to the wrong side by that much enough evidence to make the entry invalid? Sure, in backtesting it might end up ok, because it does so 50% of the times, but surely this is not based on the system method. Ursa..