Aug 022011

I started a new blog post (which I’ll post tomorrow) and while writing I was watching ES test the prior day low.  We’re in the Globex session and it’s normal for this low to be tested.  I saw strong selling going into this low and the trend is down, so this would be a fade trade.  I was going to pass on it as I don’t like fade trades, but I saw that it was holding.

I calculated my risk and where I’d put my stop.  I have a support level at 69.25 but that would make a stop over 2 pts wide.  Then I thought, why let it go that far.  I could put my stop 2 ticks under yday low and it holds or it doesn’t.  So I took the trade and my stop was just 3 ticks away.  I’d say the probability would be greater than 50% that I’d stop out.  However, my target was the Globex VPOC at 76.  So if I’m right, I’d make 20 ticks.   So I risk 3 to make 20.  Even if the win probability was 20% I’d still make money (4*-3 + 1*20 = +8 expectancy).

So I entered the trade and sellers came in just 1 pt above my entry.  I scaled half at +4 ticks and that put me in position to win 1 tick.  I have no risk.  However there is a small catch, that alters my expectancy calculation.

Now I’m risking 6 ticks to make 24.  With 20% win rate, my expectancy is now 0.  So I would have to win more than 20% of the time to make this work.  This shows how scaling out can hurt expectancy.  I’m not sure I understand how all that works really because the expectancy is actually changing as the trade unfolds.  Seeing strong selling right after my entry lowers my probability of a winning trade.  Taking half off puts me at no risk and makes my probability of a winning trade 100%.  It gets complicated.  My current thinking is that a planned, forced scale at +x ticks can hurt expectancy.  Instead I prefer to go for something bigger like 4 pts and if the trade starts looking bad or that it won’t hit my target, I scale one off.

This trade side-tracked my discussion other post a little but since it came while writing this I thought it would be worthwhile to include it in this post.  I started my other post because I didn’t plan on taking any trades here, since it’s against the trend.  But then the calculations made sense to me and I thought I’d give it a shot.

It’s like calling with pocket 2’s in a big multi-way pot.  Your chance of winning is small, but if you flop a 2 you can win a big pot.  And this is an example of how studying poker has helped improve my trading.  I’m constantly calculating my outs & odds while playing poker.  Shouldn’t I be doing the same while trading?

Cutting winners is another problem of mine.  That’s like folding a decent hand in a huge pot.  Notice here if I cut my winner short and say take profit at 75, then I’ve completely changed my expectancy calculations.  The reward would be 16 ticks instead of 20.  That requires a higher win rate, and unfortunately win rates the most difficult thing to predict as the market can do anything.  Those who trade the same quantifiable setup over and over have a much better idea on their win rate, I’m not there yet.

So I let this go.  It may pull back a couple points but I’m not cutting it short unless I see massive selling & absorption just before my target.  Actually I just noticed VWAP dropped down to 76.00 so I have VWAP & VPOC there and I should front run by 2 ticks so I moved my target to 75.50 and will leave it a lone.

It may come back and stop me out.  But if it does I’ll know I made the right play.  Because if I cut the winners short each time, over time I’ll actually end up with less profits.



 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>