What's a Winner & What's a Loser?

[timsk]

This might seem an odd question coming from a 'T2W Advisor' and longstanding member - but there you go! Actually, to be more precise, it's what constitutes a break even trade that the thread is really all about. Consider the three trades below:

TRADE A
Gross P/L: +3
Commission: -2
Net P/L: 1

TRADE B
Gross P/L: +2
Commission: -2
Net P/L: 0

TRADE C
Gross P/L: +1
Commission: -2
Net P/L: -1

A large percentage of my trades oscillate around the break even point and, to my way of thinking, all of them are failed trades, including trade A. Even though it has a nominal profit, I don't really want to include it as a winning trade, as it wreaks havoc with the profit ratio. The success ratio can be affected too, if the results are evaluated literally. So, what do you do - does break even literally mean any trade with a Net or Gross P/L of 0? Or can a break even trade have a Net or Gross P/L of plus or minus 1 point, 2 points, 3 points or more? Do you evaluate your results before or after taking commissions into account? If you include comm's, do you also include an additional sum per trade to cover the other trading expenses such as data and ISP subscriptions etc?
Tim.

(Note to mod's: I'm not sure this thread is in the best place - feel free to move it if there's a more appropriate home for it.)

 

[TheBramble]

timsk it is indeed a strange question.

A Winner is a trade which increases your P/L - a Loser is one that reduces your P/L.

Don't get hung up on what element of the trade's P/L constitutes commissions, slippage etc - it's the final whistle when you close the trade that matters. If you're running your trading as a business, sure you can take infrastruture, consumables and incidentals into account, but that's an formal corporate accounting issue and really nothing to do with the expectancy of your systems.

Profit ratios and all that good stuff are a distraction, unless you're selling your system of course!

If "A large percentage of my trades oscillate around the break even point" is because you are scalping then the ONLY way this is going to work long-term is if you're trading sufficient size to make the spread/commissions a very small part of your expenses.

If you're pulling 2-3 pips per scalp, you're not going to make a living paying 2-3 pips commission/spread on each of those trades...

If the majority of your trades are hanging around B/E for other reasons, then I think you know that no type of accounting, however creative, will help.

 

[nine]

I don't count genuine break even (0 to +1 net) trades in my score. FWIW.

On the other hand I will only go be if I scratch the trade because I either entered incorrectly or I realised that the trade has gone for too long without delivering its potential. For me a be+ is usually a limit order not a stop. I don't know if that helps.

For me trade C is a loser (it must be 0+ including commission to actually be be+).

 

[Captain Currency]

Step back your too close....

Percentage of winners for show, bottom line P&L for dough!


Kind Regads,

 

[Crap Buddist]

hmm well all 3 are winners in my book, but your expenses as a % to generate those revenues are huge. Which results in zero net gain for the 3 trade period. (less any additional expenses, which will tilt it into a loss making period) net loss on operations

So ways to increase gross profits, look at increase in revenue gains per trade and/or decrease costs.

 

[Crap Buddist]

then I think you know that no type of accounting, however creative, will help.

hmm, maybe if you could get a pip rebate and charge that to the period (discounts received, income). A bit like what Wickes did a few years back but they hadn't actually earned the rebate at the time ! Then the stock tanked !

 

[Splitlink]

This might seem an odd question coming from a 'T2W Advisor' and longstanding member - but there you go! Actually, to be more precise, it's what constitutes a break even trade that the thread is really all about. Consider the three trades below:

TRADE A
Gross P/L: +3
Commission: -2
Net P/L: 1

TRADE B
Gross P/L: +2
Commission: -2
Net P/L: 0

TRADE C
Gross P/L: +1
Commission: -2
Net P/L: -1

A large percentage of my trades oscillate around the break even point and, to my way of thinking, all of them are failed trades, including trade A. Even though it has a nominal profit, I don't really want to include it as a winning trade, as it wreaks havoc with the profit ratio. The success ratio can be affected too, if the results are evaluated literally. So, what do you do - does break even literally mean any trade with a Net or Gross P/L of 0? Or can a break even trade have a Net or Gross P/L of plus or minus 1 point, 2 points, 3 points or more? Do you evaluate your results before or after taking commissions into account? If you include comm's, do you also include an additional sum per trade to cover the other trading expenses such as data and ISP subscriptions etc?
Tim.

(Note to mod's: I'm not sure this thread is in the best place - feel free to move it if there's a more appropriate home for it.)

I think that a winning strategy is more important.

Those trades that you closed for a breakeven or a loss. How many went up afterwards?

If you closed all your trades, in a given number, using the same procedure with all, and you ended with a poor result and the trades went profitable after you closed then your stops are too close. Whether they are, individually, winners or loses does not come into it.

If I don't make an acceptable return on my trade but I get a couple of points out of it, then I am thankful but it is still, for me, a loser and am disappointed.

You can't say the trade was a success if the broker's fees makes it negative, stop kidding yourself!

 

[timsk]

Thanks for the replies so far.
To put flesh on the bones of the thread, here are my trades from last Friday. Clearly (I think) HAL and WM are losers and SLB and BRCM are winners. The other 5 trades are all oscillating around the b/e mark and it's these that prompt the thread.
Tim.

 

[Crap Buddist]

You can't say the trade was a success if the broker's fees makes it negative, stop kidding yourself!

Well at the end of the period it tallies to zero, but if one can reduce costs, to say 1 but run the biz then same then all the trades are still winners and theres a profit of 3 units ?

Running the numbers can communicate to the end user area's which can be worked on.

So can we get cheaper comms ? can we crank up the revenue.

Or one might get volume rebates which might tilt then end of the period into profit.

 

[Splitlink]

Well at the end of the period it tallies to zero, but if one can reduce costs, to say 1 but run the biz then same then all the trades are still winners and theres a profit of 3 units ?

Running the numbers can communicate to the end user area's which can be worked on.

So can we get cheaper comms ? can we crank up the revenue.

Or one might get volume rebates which might tilt then end of the period into profit.

If you are looking to cut corners to get a profit of 1, I would say we are in the wrong business!

 

[Splitlink]

Thanks for the replies so far.
To put flesh on the bones of the thread, here are my trades from last Friday. Clearly (I think) HAL and WM are losers and SLB and BRCM are winners. The other 5 trades are all oscillating around the b/e mark and it's these that prompt the thread.
Tim.

You've got two good winners, there. If you can do that on a regular basis you don't need much help from me. :whistling

Are you asking this question for tax purposes i.e. tax losses?

Split

 

[Crap Buddist]

If you are looking to cut corners to get a profit of 1, I would say we are in the wrong business!

not cut corners , cut costs. then if the trader can flip that formula over in the market as often as he can end of the period after a couple thousand flips and if one unit= /10/100 /1000, then its just a numbers game. It might be viable if improvements can be made.


Often when companies are taken over bods will look at maintaining/increasing sales and slash costs, sack half the workers if they can and deliver the same standard might turn the co into profit.


The theory stands ,whether it pans out that way ???

 

[timsk]

Are you asking this question for tax purposes i.e. tax losses?

Hi Split',
No, in answer to your question - I'm just interested to know how everyone else defines a break even trade.
Of the nine trades I took on Friday . . .
4 were winners with an average profit of $46.81
3 were losers with an average loss of $21.33
2 were closed at break even $0.00
Therefore, based on this tiny sample . . .
the Profit Ratio is 2.19 : 1 and the Success Ratio is 1.33 : 1
However, had the -$1.00 loss on MER been a +$1.00 win, then the Success Ratio would rise to 2.5 : 1!!!Okay, I accept fully that this is way too small a sample to be statistically meaningful. Nonetheless, defining what a winning, losing and break even trade is will have a big impact on the Success Ratio. As this (crude) example illustrates, the ratio can easily be skewed to produce a very false impression that I'm doing much better than in reality I really am. This is what concerns me. I'd rather be conservative and air on the side of caution than the other way around. I hope all that makes sense!
Tim.

 

[MrGecko]

Whether a trade is a "winner" or a "loser" depends on the risk;

A "win" is a trade that returns excess reward given the risk taken, including costs and everything else is a loser. To determine the performance of a trade (or string of trades); use

(Reward - Risk - Costs) / Risk

which should be >1. All the "losing" trades that return a profit, but didn't make up for their risk, can be considered a contributions to exchange fees etc... Yes it improved your bottom line, but it underperformed against its benchmark (the risk). I mean, it seems crazy to me that a +10 profit is considered a "win" if it carried a -50 SL! Any trader that pulls out a string of trades like this is just waiting for it all to go t!ts up.

This philosophy isn't for everyone, but its JMHO

 

[TheBramble]

You increased your bottom line by £123.25 on Friday. That's the single most important piece of data. That you have a W:L of 4:3 is probably not indicative of your normal trading day I'd guess, and Fridays are not the best day of the week.

The thing is, you rarely ever get a dead B/E trade - you either make a wee bit or lose a wee bit. I think what you’re asking is how do you handle these ‘outliers’ if you want to calculate a reasonable profile of how you’re performing.

You could set an arbitrary level of P/L (+/- £x) at which trades do not included in your profitability calculations, or you could set a dynamic cut-out where any trade which yields less than +/- x% of your average trade P/L, do not get included.

Personally the ‘almost rans’, plus and minus tend to cancel each other out over time and I’m not that bothered. I realise they do make a mess of your Aw:Al, but so what. Unless your comparing your stats with someone else or preparing them for someone else, there’s not a lot of point.

I'm sorry I can't be more enthusiastic about this timsk, but I remember getting into all this good stuff in a big way in my early trading days and it really didn't help me. I found myself grabbing a small profit just to click up another WIN rather than leaving the trade to do what it should have been doing which is to clock up a lot more $ for my bottom line than I allowed it to, just in case it became a LOSS! This is quite independent of any market related information which would have under normal circumstances been the only thing that governed that decision. Ridiculous.

I don't want my head turned by anything when I’m in the market – all I’m looking at is the magic of the moment when I'm trading. It would be very different if I were running a fund and trading others bods' money, I'm sure, I'd need to pay attention to these things.

Your not thinking of starting up Timsk International Fund Management are you????

 

[TheBramble]

Whether a trade is a "winner" or a "loser" depends on the risk;

A "win" is a trade that returns excess reward given the risk taken, including costs and everything else is a loser. To determine the performance of a trade (or string of trades); use

(Reward - Risk - Costs) / Risk

which should be >1. All the "losing" trades that return a profit, but didn't make up for their risk, can be considered a contributions to exchange fees etc... Yes it improved your bottom line, but it underperformed against its benchmark (the risk). I mean, it seems crazy to me that a +10 profit is considered a "win" if it carried a -50 SL! Any trader that pulls out a string of trades like this is just waiting for it all to go t!ts up.

This philosophy isn't for everyone, but its JMHO

Understand what you're saying Mr. Gecko, but R:R is data more properly considered in the area of expectancy, and only ever very indirectly in the area of P/L and W:L.

You could quite conceivably run a system with a Risk:Reward of 50:10 and providing your W:L was greater than 5:1, you’d have a winning system (forget the deep pockets for potential drawdown for now....)

 

[MrGecko]

Yes Bramble, of course what you say is true.

R:R is always done before a trade is entered, so I consider my trades "winners" or "losers" depending on ther actual performance compared to expected performance - it's just the method that makes the most sense to me.

And as for 5:1 R:R trades, yes it possible to profit from them consistently; the whole R:R discussion is almost useless, as it treats the outcomes of winning or losing as equally likely to occur - and of course we make trades on the judgement that one set of events is more likely than the other (i.e. writing options; take out the probabilities, and the R:R is awful).

Perhaps the two points that I wanted to make are moot, but

1) I consider my trades W/L relative to their benchmark, it's just the way I like to do it. No better nor worse than others.

2) the whole returns a reward given the risk taken statement. W:L ratio's are just one of the many ways to monitor performance. I use this method, as if it is >1 it means I am returning a reward greater than I should have expected, given the risks I took. The result of this is often a sobering reminder of the difference between how good I am and how good I think I am

But as you said, all of this is pointless because it treats each outcome as equally likely, which of course it shouldn't be. And anyway, the bottom line is the bottom line; if you have an increasing equity curve, who cares? "If it ain't broke, don't fix it!"

 

[timsk]

I'm sorry I can't be more enthusiastic about this timsk, but I remember getting into all this good stuff in a big way in my early trading days and it really didn't help me. I found myself grabbing a small profit just to click up another WIN rather than leaving the trade to do what it should have been doing which is to clock up a lot more $ for my bottom line than I allowed it to, just in case it became a LOSS! This is quite independent of any market related information which would have under normal circumstances been the only thing that governed that decision. Ridiculous.

I don't want my head turned by anything when I’m in the market – all I’m looking at is the magic of the moment when I'm trading. It would be very different if I were running a fund and trading others bods' money, I'm sure, I'd need to pay attention to these things.

Your not thinking of starting up Timsk International Fund Management are you????

Hi Tony,
Thank you for your comments.
I quite take your point about grabbing a small profit just to be able to chalk up another 'win'. The large number of trades at break even or +/- a diddly amount are a natural consequence of a highly risk averse strategy. I'm an intra day trend trader and I move my stop loss to B/E at the earliest opportunity - probably too early for most. If they go as expected, then I trail my stop at a respectful distance but, at the start of the trade and before it's into reasonable profit - all I think about is not losing money. Or, if I do lose, ensuring the loss is kept to an absolute minimum. (A lot more work still to be done on this front!)
No plans just yet to start Timsk International Fund Management' but watch this space!
Tim.

 

[Captain Currency]

all I think about is not losing money.

Hello Tim,

Im on the tee box for the third hole, there is a lake in front of the green. I say to myself "dont hit your ball into the water" guess what happenns.........? Ball ends up in the drink! I think it all about our minds NOT been able to process a negative thought or image. I bet all my mind heard was ball and water!


You are correct from your post 13 that if one of your trades from Friday had a slightly greater or lesser value (even as low as 1 pound), this would throw your winning/losing ratio or percentages out by a noticable number - however your bottom line P&L would only be either up or down by 1 pound.
I was going to break your results down but you bet me too it.

From my own experience, moving your stop-loss to break even when the trade is up X amount of points can look like it is helping in the short run, but this practise could be stopping you from going forward and profiting in the longer run.

Just one of the many dilemmas that we as traders have to go through

Kind Regards,

 

[nine]

Tim,

I like what Captain Currency said and it made me think about something else. How about changing your measurements to something more meaningful.

This is just a thought but perhaps you primary measurement should be a points expectancy and a secondary might be a win ratio. So expectancy would include all trades, winners, be+, be- and losers because its a true measure of financial outcome. One way to measure it would be:

E = (total gained today - total lost today) / (Number of trades * Amount risked including slippage and commissions)

Another one might take into account the achieved "real" risk for the day:

E = (total gained today - total lost today) / (Number of trades * (Total Lost Today/Number of Losing trades))

But if you were measuring win ratio you would exclude trades in the BE zone if you wanted to encourage yourself to BE+ a trade rather than holding to your system stops. If you wanted to encourage yourself to hold trades until your strategy says "its a loser" rather than scratching difficult trades then you might include the BE's in your win ratio statistics BUT to force holding you could consider a BE+ that you exited before your system exit to be a loss.

There's an old saying that you manage what you measure - so it's important to make sure that your measurements reward you for doing what you need to do to be successful. IMHO.