TPS (Time, Price, Scale-in) Revisited

Trading Systems
101

Another lack lustre day to be sure.

Took profit on the NASDAQ and EUSTOXX50.
Took partial profit on Italy.
Could not scale-in further on the Hang Seng.
Am not scaling in further on Gold (not happy with where the close today is compared to the previous lowest close while in the trade).

Given that tomorrow is already Wednesday: the results for this week will not even remotely resemble those of last week unfortunately. It is what it is.

The above refers to the TPS system only.

I was looking around for an article that I found some years ago and which probably is one of the reasons I started trading this system in blind faith some years ago. I found it and much to my delight the article has been updated i.e. system thoroughly and correctly backtested professionally using TradeStation on the S&P 500 SPY ETF from 2000 to 2016. What’s more the same site contains some interesting insights as to position sizing and risk management for this system. Will post link tomorrow i.e. working on my iPad right now and I never get it right to copy and paste links.

102

Morning.

For the skeptics and the cautious…

Take a look at the information on the links provided below. As noted previously: their initial tests were completed at the end of 2013 and, arguably, it’s the main reason why I decided to delve seriously into this trading system and trade it in good (blind???) faith. The updated testing was done at the end of 2016. I have no reason to question the results presented for any reason whatsoever especially given that the work is done independently and, furthermore, the tests are done using TradeStation. It is worthy to note that in spite of one bad year for this trading system (2011) it has right up until the end of 2016 shown an overall profit and a stellar one at that. In the absence of anything else that has ever generated consistent profits for me (in my case over the last five year period) I shall trade this system even if it ends up being my core trading system while I find something with more action that may be profitable (as anybody that knows reading this thread I’ve an unfortunate urgency).

Now one other thing that needs clarification (only thought about this last night when I managed to track down their new site). The abbreviation “TPS” is ACTUALLY describing an ENTRY METHODOLOGY. In other words: TPS is not a trading system in and of itself. I note this because you will note in the test results that different criteria are being used to enter and exit the trades. Different from both my criteria as well as Connors’ original criteria (although their criteria used is not too dissimilar to one or two of the other trading systems detailed in the e-book). This easily explains why I get far more trades from the system and for all I know could be the reason for my own stellar results. This is just something to bear in mind.

I must also note that I never thought to implement their position sizing or stop loss idea. I will surely look into this in the next few days.

I would also urge anybody interested in what I’m doing to actually take some time and Google for Connors’ TPS. I have yet to find a single site where a total wipe out is detailed let alone a single site where the merits of Connors’ TPS ideas are not hailed. And that just has to be good enough for me if not for anyone else.

One interesting little factoid by the way: when stops were discussed with Connors he noted that not using stops for this is by design. His theory for no longer using stops was this: a robust system should be able to perform profitably over time without the implementation of stops i.e. with stops being implemented it is impossible to know whether any system is robust as a function of its design or rather as a pure function of stop placement.

Links:

http://easylanguagemastery.com/connors-rsi-update-for-2013/

http://easylanguagemastery.com/connors-rsi-update-2016/

Regards,

Dale.

103

Moving on…

Let the record show that in spite of my not being totally happy with where Gold closed last night I’ve indeed scaled-in further at the closing price of yesterday (not for any good reason other than the fact that right now it’s the only signal and therefore the only game in town) (still: it’s a valid trade nevertheless).

104

Let the record also show that I have placed a limit sell on Japan at 21341 (1-day signal). It’s trading below its 200-day SMA so in isolation it’s a valid signal (according to my adapted rules anyway).

I don’t know why but in just sitting monitoring the ticks I have a feeling, not based on anything is particular, that these markets are (currently anyway) struggling to move up. Suits me if they drop i.e. at least some valid signals to enter long at some point. Anyway. Just an opinion and nothing more.

105

Order for Japan cancelled i.e. even if a miracle were to happen in the next two hours and the order was executed there would not be enough time for it to be executed and move.

It’s at times like these when markets and this system can frustrate the life out of you. Every instrument that I monitor is pretty much stuck in no man’s land. Admittedly: I’m not used to sitting watching markets during the day so of course this adds to the frustration in copious amounts.

Only two possible signals on oil at the close tonight. Not even sure if it’s worth waiting i.e. will probably place orders in the morning. Will see.

Only good thing I suppose: when these market do wake up they will break in one direction or the other. From where I’m sitting at the moment I don’t really care in which direction that is i.e. some movement would be really nice either way.

106

I’ve not bothered posting on the thread for a day or two. It’s not been the greatest week for this trading system I’m afraid. Markets went nowhere until yesterday and then took off but no positions open at the time. In just looking at my post above though: I sure was right about them eventually moving one way or the other i.e. the moves yesterday were pretty nice especially oil. Anyway: I managed to get long most of the indices last night before the close and have just taken profit on those same positions (with the exception of my positions on the Hang Seng which are still open). These profits were as a result of my taking those infamous 1-day signals. And certainly nothing to write home about either. Realized profits as a percentage of the balance when I started trading again: 14.99% as things stand now. Am also long oil (both WTI and Brent) too but these are as advertised trades. Could not scale-in any further on Hong Kong as it closed as near as darn it to its close of yesterday. Oil: depending on what happens in the next two hours or less there MAY be a signal to scale-in further. And who knows: anything could happen in the next two hours or so i.e. any lower closes in any of the indices will generate valid as advertised signals. Will see what happens.

Anyways. That’s how it goes with this system. Slow but steady and reliable I guess. Certainly nothing to keep you riveted to your screen on a daily basis though. Not during weeks like this past week anyway. But then there are times, often enough, where you’re fully committed or all in on positions and the markets take off. Those are the days when it all seems worth it.

107

Hello.

For want of something better to do I guess I’ve been thinking about tracking the performance of this system along the lines of the way some of these other sites do automatically. I’d have to do this manually because my broker’s platform is a proprietary platform and is not supported by such sites. But I have some questions that try as I might I’m not coming up with answers that make sense.

Let’s assume a starting capital of 25K (currency is irrelevant here).

So I started trading again two or three weeks ago. And I’ve detailed by percentage gains. No problem thus far i.e. pretty easy to work out gains as a percentage of capital in the account to date. But here’s where it gets tricky:

What happens when you withdraw funds??? If there are not some few decent trades in the next week or two then I am going to be left with no choice but to withdraw some capital. So how is a withdrawal taken into account when building statistics??? In other words: if I generated graphs in Excel for this purpose then the moment I withdraw funds the equity figure is going to reduce by that amount obviously. But that is a withdrawal not draw down or a loss so should not reflect on the performance of the system itself. So how would one show this??? In addition: this could inflate gains artificially if such gains to date were based on the equity shown in the account after the withdrawal.

The other problem (and this has more to do with statistics in general I guess): let’s assume you have to withdraw funds from an account. Theoretically one would reduce trade sizes accordingly. But one may be in a situation where you could not reduce trade sizes any further i.e. you could be in a position that the minimum trade size at your broker is larger than what your theoretical maximum trade size should in relation to your capital. In effect and in this situation you’d be over trading the account. Let’s assume for one minute that the system is robust and that the higher risk would not result in a catastrophe. This would then surely result in improper statistics and performance results would it not??? This is a bit of a concern with ANY statistics where the % gain is shown as a % gain on capital for the period. Example: I’ve made (let’s round it up shall we i.e. call me a liar for %0.01) 15% return on my capital since I started again. But that gain was as a result of correct position sizing (without going into too much detail it turns out I’ve been pretty much on the mark with my position sizing according to some fancy calculations that I found as they pertained to this very trading system). But let’s say I wasn’t behaving myself and was totally over trading and going balls to the wall. That 15% could have been 50% or 60% for the last three weeks. Now that would sound and look fantastic. But if you really think about it: that’s absolutely no reflection on the performance of the trading system i.e. such stellar gains would only be as a result of taking far greater risks obviously at the risk of wiping out totally.

And conversely to the above: what when one adds capital to an account??? If I added capital to the account now (let’s say I doubled the initial capital right now). Then that 15% that I’m referring to above would drop to 7.5%. This again not being a fair reflection on the performance of the trading system itself.

So yeh. Those are my questions. Probably just pure math and quite possibly basic newbie questions. But I’ve never bothered with this type of thing before. Not ever. And I mean it’s fine to say that it’s the amount of money that I can withdraw at month end (or whenever) that counts to me. That’s true. But I just thought it’d be interesting to try track performance in real time (this as opposed to running tests on historical data which may be fine but does not take into account stuff like the above i.e. where the rubber meets the road type of thing).

Regards,

Dale.

P.S.

Just to be clear I’d be doing this in Excel (I suppose with graphs and whatever else) and then just keep uploading updates every week I suppose.

1 Like
108

Further to the above and in just thinking logically:

I’m seeing two different measures of performance here i.e. the trading system itself and my performance as a trader. And therein lies a potential pitfall when trying to measure performance of the trading system and something that needs to be clearly understood and stated with statistics. In other words (and to name but a few things):

I made some money shorting Australia this last week. That was not with this trading system. Not a big deal as it’s easy enough to just subtract the profit on that trade from the overall profits made with this system. Difference is in what’s being measured or rated i.e. with that included any statistics are going to be rating me as a trader and are no fair and accurate reflection of the performance of this trading system.

As was the case yesterday: I also took profits on open positions such positions not based on the as advertised version of this trading system. Furthermore: such profits would have been greater had I taken them earlier in the day or they could have been less as it turns out had I waited longer. Again: any statistics included would be rating my performance as a trader and not rating the trading system itself.

And now comes the real elephant in the room:

All the other statistics for the different variations of this trading system as published on the Internet differ from my experience with it. For one thing: there are one or two adaptations I’ve made that suit me e.g. these first day or one day signals. This explains at least one thing and that’s the sheer number of trades that I do with this i.e. they by far exceed the number of trades that the as advertised system would initiate for a given period of time. And there are one or two others (adaptations) which I’ve detailed on this thread already which definitely boost profitability but not of the as advertised system.

And now as I type this I find myself asking a question that’s never really crossed my mind:

All of these sites that allow for the publishing of results and performance: how can they possibly take into account what I’m talking about here. In other words: a particular trading system may have a straight line equity curve (growth) for a particular period. But at what POTENTIAL cost. In other words: it may just so happen that for a given period of time (let’s say for one year) the whole thing performs like clockwork. But there was always the potential for one or two trades to wipe the entire thing out. In other words: the stellar performance being shown may not be an accurate reflection and the results are only stellar because of excessive risk being taken and which may or may not be shown. And in the case of this particular trading system and its derivatives: because no stops are used no system can calculate potential risks on individual trades as they are initiated.

Hmmmnnn… Maybe I’m just going down a rabbit hole here i.e. attempting to calculate statistics for a trading system but that encompasses human factors or judgments that skew the said statistics.

And the above two posts was a hell of a lot of typing just to arrive at the above conclusion.

109

I think this really depends on who the statistics are aimed at. If it is only for your personal use then the definitions and structures are not so important and do not necessarily have to meet generally-accepted market standards - as long as you understand yourself what they mean.

If the withdrawals do not eat into your initial capital to the extent that it starts to reduce your position sizes then I think you only need a basic excel type spreadsheet that breaks down your various transactions into their various categories and treat them separately in the final analysis. i.e. initial capital plus gains/losses does not have to equal your current capital.

By way of example, here are two extracts from my own Excel workbook:

Capital development:

2019-05-25 results cum
2019-05-25 results cum496×624 23.5 KB

I’ve made up some figures here for illustration using your 25000 starting capital. So I have one column for the monthly end results and others for the monthly P/L% of capital and cumulative P/L% of capital (blue rings). These are unaffected by deposits and/or withdrawls.

Then comes deposits and withdrawals with separate cumulative figures and the last column is the current balance including P/L, deposits and withdrawals.

These various columns are then automatically transferred to another table which I call my equity analysis:

Equity Analysis:

2019-05-25 YTD

Here are separately listed initial capital/deposits/drawings/P/L YTD/ % YTD and current balance.

This is very basic but good enough for my purposes to track my progress. At the end of month it is easy to transfer the data to my official business accounts.

I do also keep separate data for various categories of trades i.e. Forex/Commodities/Indices and these could also be listed separately if desired in the summary table.

This is very basic and only good for personal use and can be reduced/expanded as required.

But if the aim is to build results data for future possible use in, for example, managed accounts, then I guess this is far from the right vehicle.

The issue does start to get more difficult, as you say, once you start to change the “initial capital” on which both position size and results analysis are based.

But I have written enough here for now, I am sure there are others here with much better input to this kind of issue. So this is just a prompter for more input from the experts! :slight_smile:

1 Like
110

Hello.

Hey thanks a lot for that. Much appreciated. Good “prompter” as you put it.

I guess when I got this bee in my bonnet last night the idea was to create statistics relating to the trading system itself. But after jotting down my notes above I don’t think it’s going to be possible unless I’m missing something. Any spreadsheet that I create and that takes into account what I started with and what I’m currently sitting with would be misleading at best to somebody who may be keen to see statistics for the trading system itself. And as I noted: in spite of my firmly believing myself (fooling myself) that I’m trading this thing purely mechanically in its original form that’s not actually true. This has become apparent as I’ve been detailing everything I’m doing on a daily basis on this thread i.e. have not spent five years keeping a journal or documenting those little things that I do in addition to what the as advertised trading system would have you do so when starting this thread these never really came to mind initially. In short: I’d be producing statistics of me as a trader and not for the trading system in its as advertised form which I guess is of use to no bugger. In other words: my performance is based on the as advertised system merely being a base (albeit a sound base thus far anyway). And even if I could document in detail every single little adaptation there’s still some interpretation involved. Fine example was Thursday night: in a crass attempt to not break as advertised rules I elected to not take two new long trades because the price in both cases was trading below the 200-day SMA. But under normal circumstances I would have taken those trades (and very sorry that I didn’t given that one was on the Nikkei and it shot up like a rocket yesterday and it’s the second best payer of all of the instruments that I trade). But then I say that: there will be times when I break this 200-day SMA filter rule and others when I don’t and this based purely on gut feel and I guess a certain amount of discretion developed over the last five years (but which I’ve always considered a very slippery slope when purporting to trade a system purely mechanically).

Oh well. If nothing else I’ve done a bit of brain exercise here for a change (certainly more productive than watching the S&P ticks) (or actually maybe not i.e. is it possible that after all these years somebody like myself could actually get a feel for something albeit that it plays out subconsciously but such feel as a result from staring at many charts and ticks over the last decade or more???).

Matter of fact and on a side note: I’ve always wondered what would happen if you put somebody in a room (maybe somebody serving a very very long jail sentence for example) and all they had in front of them day in and day out for year after year was a chart e.g. a 1 minute chart or a tick chart of the S&P. I have always wondered if said person after doing nothing else for such period of years other than seeing these movements would develop some type of ability that would allow them to sense what is to follow. Not talking about identifying chart patterns or candlesticks i.e. just the ebb and flow pattern (if there is one if you see what I’m saying).

NOT that I want to find out myself mind you!!! LOL!!!

Matter of fact and just reminded me of something else:

When I was fully committed in this business some years ago it was also one of my ideas to take trading to prisons. Again was (practically) laughed at (as was my idea to take trading to schools). Seems like much has gone to waste over the years to be honest. And I’ve run out of steam now i.e. no longer have the drive or the conviction that I used to have. Pretty sad. Sorry. Notes to self is all really. Need to go to the shops now.

1 Like
111

If a mechanical or rule-based method is “contaminated” by personal discretion then the statistics become meaningless as a measurement of the method but retain significance as a personal performance measure - which in itself is useful as a journal like many people do here.
It can help in building discipline and self-control as well as providing trading examples and issues for newbies.

It is just a case of clarifying the purpose! :slightly_smiling_face::upside_down_face:

1 Like
112

Love the way you manage to put in a few lines what takes me pages to do!!! LOL!!!

1 Like
113

Funny! That’s what i usually think about other people’s replies to me! :grinning:

1 Like
114

And then you “met” me!!! LOL!!!

Matter of fact: I think I do still hold the record for the longest post on here (actually found it the other day in the archive).

1 Like
115

One of the difficulties I’ve had is that I’m trading semi discretionary, so on any given day my performance depends on my mindset for the day. Tracking usual stats will not reflect this, I frequently have standout days when I am well rested, feel good and in tip top condition, and am able to follow my rules. Frequently in the past I’d compound profits daily and trade larger every day with better and better results, until I have a really bad day and blow up. That’s one thing I’m looking to measure now, some kind of metrics to determine my physical and mental well being versus daily performance.

1 Like
116

Maybe a plain old simple diary or journal would work??? Like first thing in the day before you start??? As for particular metrics e.g. amount of sleep the night before or how much to drink the night before could be valuable??? Not sure. Would be a pretty interesting read and statistics though.

P.S.

The above was actually typed last night but it was open on my notebook today i.e. obviously forgot to click post or send or whatever the option is.

117

The best days I have is when I make 50$ or 100$ the first hour or so trading, almost without any drawdown, and I’ve had breakfast or lunch, and 8 hours or more of sleep, and not worried about other things. These tend to be performance breakout days. But I am not sure, I’ve never really tracked these things.

Since I drink every night, beers consumed might actually be a metric I should look at hahaha! Imagine if I find a positive correlation between more beers consumed and more profit the next day LOL!!!

1 Like
121

TPS trades still open are WTI and Brent long as well as Hang Seng.

Am short as of today Swiss Market Index Future Expiry 21 June. This trade based on The Memory Of Price (double top and all time high on the Swiss Market Index).

All good.

122

Well what do you know. The Hang Seng finally getting a bid. About time.

123

ALMOST had to close out on the Hang Seng (for the absolute tiniest of a loss) but RSI(2) JUST missed the boat. So still long in the trade. Fact of the matter though: even if I was indeed forced to have closed out at the small loss tonight I would actually have made a profit on the trade overall because of loads of dividend payouts since having this long position open.

Still long Brent and WTI. Intraday: RSI(2) actually did reach over 70 so the profits were great. But price has retraced a little at this time and unless there’s a spurt of energy in the next hour or so well then tomorrow is another day as they say.

Took profit on my SMI Futures short. Really nice trade that was. Could have held onto it and let it run to its target but figured seeing as I’m now looking for signals to get long on any index that generates a signal it becomes counterproductive to be trading one index short and the rest long. Besides: not my core trading system (although The Memory Of Price has not let me down yet).

Not sure how I"m going to report results for the month (it’s actually only been three weeks by my rough calculation) as I’m probably going to be sitting with open positions (hopefully even some new ones). May reconcile at month end and report realized gains and showing open profit and loss separately. Cannot think of another way to do it.

Ah well. That’s me for the day. As I sit and type this there are 1-day signals forming on the S&P and Gold. But I dunno. Something is telling me to wait for as advertised signals. Unless there is a signal to go long the NASDAQ in which case I may change my mind. Long the NASDAQ is indeed a valid as advertised trade. I just wasn’t happy that the two consecutive closes were so close to each other. That and coupled with the fact that the US markets were closed yesterday.

And that’s it for now folks.