The attached spreadsheet shows my tracking of JP's trades for June, so far. As I am not yet able to trade at JP's size, I have been wondering what the effects would be for smaller trade size. Knowing that JP weights his partials so that the majority is out by approx. 5 pts, I have assumed 20 contracts (cell B1) and a weighting as shown in cells D1 to N1. I suspect that JP trades much heavier than this but it gives a starting point for a partial profile. The maximum number of partials was 11 and a few trades were "front loaded" for reasons such as imminent reports. By changing the partial profile in cells D1 to N1, it is possible to see the effects of trading any number of contracts in any profile.

Excellent spreadsheet uktrader - many thanks for sharing it with us.

Your comments about size and how to duplicate JP's success are extremely important. I tried to address this in an article: Size of Position but not sure if I did it justice.

In your spreadsheet, I would change cell B1 to =SUM(D1:N1)
Gives you a bit more flexibility over looking at the final results with different partials at each level. I like the structure of your sheet though and the way that you've collected and collated the data. It's fairly difficult to keep track of JP's trades because you need to be in the room 100% of the time and hear everything that he does. Also, how does one classify a trade that meets 2 or more criteria? For example, if the VAH is also a Single Print, then is the short that you take there a VAH short or a Single Print short.

One way of addressing this is to have a column dedicated to each trade type and put in a 1 if that trade type is relevant. That allows you to track 1 trade over multiple strategies. Using excel's filter function you can then look at each trade type which would also allow you to see if the clustering of trade areas made the trade a better, worse or indifferent "bet."

These comments are also obviously addressed at brokebackbernie who is doing a very similar thing.

I think that this sort of work is extremely important because it will help isolate areas that work better and allow package scaling. This will also help the smaller traders who are not able to scale 10 or more times.

Another thing that is important when looking at JP's strategies is how to handle MacGyvers. Remember that JP doubles his position at the MacGyver level so if he gets stopped out then his loss is 2 points on 20 contracts and 1 point on 40 contracts. In other words, if you average that, he loses 2 points on a total of 40 contracts so you need to take that into account in the stop loss column.

The spreadsheet treats a mcgyvered position as a new trade (with a tighter stop) so automatically assumes that size has been doubled. I have changed B1 to sum(B1:N1). Thanks for the suggestion.

One of the problems with treating a MacGyver as a separate trade is when you are summing up and collating the results to see what works and what doesn't and what needs a later entry and what needs an early entry.

A trade with a MacGyver is still part of the original strategy and shouldn't really be treated separately. A long at VAL with a MacGyver is still the same trade and once you have collected data on 100 or more trades and want to analyse them then you want to know how many VAL trades needed MacGyvers. By adding this in as a column with the trade types and putting a 1 in there it is easy for Excel to sum it up using the reporting feature and either the Count or Sum on that column. The cell doesn't have to have a 1 in it because the Count function in the reporting will treat it as a 1 and so you can put more information into that MacGyver column to add more info to the output.

...Excuse me for just rambling what's in my head at the moment but you may get some ideas from this.

So, if you had a bunch of columns to the right of the current cells then you would have for example: VALL (Long at VAL), VALS (Short at VAL), SP (Single Print), MG (MacGyver) Plus many more - and you can always add new strategies as you go along.
So if JP also MacGyvered on a VALL then you'd put a 1 or TRUE in both these columns and leave the rest blank.

Now at some point in the future when you are looking at all the VALL trades ONLY you suddenly notice that they all ALSO had a MacGyver. This is telling you a LOT of information.

If collected in a well structured manner it becomes extremely easy to extract very good information out of the spreadsheet.

Please don't think that I'm criticising what you or brokebackbernie are doing because I think that you're doing amazing stuff. I'm just vomiting some of my thoughts and experience onto the keyboard.

Thanks for the comments - feedback always welcome

I don't fully understand what you are saying regarding the mcgyver trades. By treating them as separate trades, I thought that I was making it easier to see how macgyvered trades compared with non-macgyvered trades. It also makes it easy to account for the trade size increase. My excel skills are fairly limited so I may be missing a valuable suggestion here. If you could post a simple example then I might be able to better understand and incorporate your suggestions.

I've put this together fairly quickly in response to the posting mentioned above and written some notes here.
I'll also post these notes in the forum.

The main purpose of what I have done here is to collect enough information for further analysis at a later date.
As such, the analysis formulae and techniques are not in the spreadsheet nor are they documented here.

By treating each trade as 1 strategy from (multiple) entries to multiple exits we can record the success or
failure of the whole strategy concept and work out probabilities of it working. By including the MacGyver
"flag" in the same trade we also include a means for filtering (not discussed here) results at a later date
and finding what type of trade (if any) is more prone to a MacGyver style entry and perhaps finess the
trades that we pick or the size that we trade them.

More columns are needed for the exits and this spreadsheet gives you the ability to "go in light" on certain
positions by entering (say) 10 contracts on the entry.
By placing the number of contracts against the entry type you can use a count() function or the reporting
function to determine how often a strategy signaled an entry. If multiple are applicable then split the contracts
across all the are applicable.