I realized that this chipping away approach was what I should be doing, not putting myself at big risk, trying to collect a ton of dough - Tony Saliba.
(Disclaimer: Just to re-iterate that I am NOT giving recommendations to buy or sell any of the securities mentioned in this blog.)
The first purchases were made this morning. We need to remind ourselves that this system is not supposed to work overnight. It's a long term weekly system, designed to make consistent money, over the long term. In fact, even in backtesting, the first year is often not profitable. So I would expect a single digit gain or loss on closed equity in the first year. That said, we should expect some nice open profits at this stage (after 12 months). Professional traders don't consider open profits as their own.
Start up is hard for long term systems because the trades that are closed out first are often the losers, and we need to give winners time to run. From testing, the average holding time for a winner was about 200 days, compared to 60 days for a loss. And the really big winners (which are the trends we are trying to catch) often run for much longer than a year.
The Total Trading Capital column in the spreadsheet is the money allocated to existing trades minus brokerage fees plus money in the bank. This is what TradeSim uses to calculate position sizing so I will do the same. MIN I paid double in brokerage because I made a mistake with the position sizing and bought less than I should have the first time around.
9 comments:
Nizar,
First of all, your blog is a fantastic learning tool for those of us just starting out such as myself so thanks for posting it. Congrats on the commencement of trading i wish you every success and nothing less.
Couple of questions... I recall having a conversation with you re the amount of capital invested (i think its called position sizing) and you mentioned the superior performance of using 10% of remaining equity rather than 10% of total equity. Is there any reason you have not followed this strategy?
Also, in the spreadsheet the Average Price = Current Value for all the stocks except LRF. I checked the mkt price and and it was still around $1.64.... perhaps an error in the sheet? or am i missing something?
Congrats again, and i look forward to yout reply.
Jeton
Hi Jeton,
Thanks for dropping by and for your kind words.
I have learnt alot from the online trading community so hopefully this blog will give a little bit back.
I'm not sure what you mean regarding the position sizing. I probably explained it wrong because I didn't have a correct understanding of the way TradeSim works out position sizes until only a few days ago. An option in TradeSim that I was exploring previously was to use 10% of total trading capital for each position.
At the time, I thought that total trading capital was either defined by money in the bank + open trades OR just money in the bank. Turns out it's neither. TradeSim defines total trading capital as money in the bank + total cash that has been allocated to existing trades.
I'll give you an example.
Say you have a system and to date have taken 2 trades, and you put $4k in one trade and $5k in another trade, assuming you have $50k left in the bank and brokerage of $30 each way, your next position will be 10% of (50k+4k+5k) = 59k minus $90 (3*$30, including brokerage for this trade).
So thats 10% of $58,910 so $5,891 will be allocated for the third trade. The current value of the 2 open positions is not taken into account.
Then when trades are closed (either in profit or loss) the net proceeds (or losses) are added to (or subtracted from) the bank.
I currently use a fixed percent risk position sizing model that risks 1.5% of total trading capital per trade.
There was no real difference in results between this method or the 10% method described above. It was just personal choice. Results for both were very similar.
What's important is that both methods allow you to risk more when you are winning and less when you are losing.. This works to compound the account quicker when you are winning, and at the same time, decrease risk of ruin when you are losing.
Hope that helps.
Also regarding LRF, thanks for pointing it out to me. I was relieved to discover it's just an Excel error on my part. That specific cell was somehow set to "Show no decimal places" (while the rest of the column showed 2) so it rounded the value up to $2. But all the calculations were worked out from $1.64. This has now been corrected.
Nizar.
Just a correction.
"Then when trades are closed (either in profit or loss) the net proceeds (or losses) are added to (or subtracted from) the bank"
Net proceeds (whether the trade was closed in profit or loss) are obviously always only added to the bank!
Nizar.
Nizar,
Were all of these positions opened on the 9th? Or are they paper purchases?
Thanks
Hi,
They were purchased today on-market.
Nizar.
I am surprised you got a signal for all at the same time. Is that correct? If so did your system give more signals that were ignored due to lack of capital etc?
Yes.
I got 19 signals.
It seems correct.
Even in backtesting on the first day I was getting alot more signals than I could buy.
Just like in backtesting, I could only afford 10 stocks.
Why are you suprised?
Nizar.
I just double checked.
When backtesting on what was originally my out-of-sample, with start date 01-01-2004, on the first day, I got 24 signals.
I took 10.
From the 19 signals that I got for this week I also checked them through the chart, to see how it looks and to to make sure they all meet the required criteria (white candle, ROC, breakout, liquidity).
MIN only just made the liquidity filter cut off which is $1million turnover per week.
Nizar.
Thanks Nizar you answered my query. I hope it all goes well.
Post a Comment