Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
Greg Reinacker, the Founder and CEO of Tradervue, will be monitoring this thread so that he may answer any questions that you post here relating to Tradervue's products and services.
Please keep in mind that some customer service/technical support issues are best handled through proper channels at Tradervue.
Greg Reinacker started Tradervue in 2011 to offer active traders a tool to help them keep a journal and improve their trading performance. Tradervue also has a social aspect which optionally allows traders to share their actual trades with others. Tradervue has detailed analytics to highlight strengths and weaknesses within your trading methodology and tools designed to improve on them. You can also find a highly regarded educational webinar that Greg presented on futures.io (formerly BMT) that covers the importance of maintaining a Trade Journal.
In addition to this thread, I will also be asking Greg Reinacker to stop by on occasion for a casual webinar where he can answer questions via audio while also sharing his screen to visually demonstrate any points as needed. The date/time of those sessions will be announced here in this thread. These sessions will be limited to questions only, there is no prepared presentation. After the session ends, the recording will be posted in this thread.
Feel free to ask any questions below and we'll do our best to get them answered.
The futures.io (formerly BMT) "AMA" (Ask Me Anything) series is by invitation only. It is part of a new program we are launching shortly called "Certified Trustworthy", something that has been months in the making. I will provide all the details of this new program as soon as it is ready for launch.
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
Hi Greg,
Let me ask the first question
I know you originally started Tradervue primarily for equity traders, and after you did your webinar on futures.io (formerly BMT) last year you added a lot of support for futures and forex, new bar types, and etc.
Since the majority of traders on futures.io (formerly BMT) are trading futures, can you tell us if you consider Tradervue to be "100%" compatible now with futures?
I believe it is indeed "100% compatible" with futures and forex - I'm not aware of anything that shouldn't work well.
The only thing that's not applicable is the liquidity reports - they are for equities only. One could make an argument that an analog exists for this in futures, but we don't have a way to calculate it just based on execution data.
Greg
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 3 users say Thank You to Tradervue for this post:
Greg I know you have a 'share trade' option, but do you have any reports that would show traders ranked on a ladder? For instance, ranked on a calendar month system?
Not at the moment. From the beginning, the idea of sharing trades was intended to be about sharing ideas and information about a trade, without sharing sensitive (to some) financial information. By default when you share a trade, we'll share your buy and sell points, along with your net position (e.g. long or short), but leave out actual size and P&L info. So we show enough that others can understand what happened in a trade, and leave out the fact that you perhaps traded 100 contracts and made $25,000.
Here's an example ( Tradervue | NEM Trade Detail) - we can see that he got short, and scaled out in three parts. We don't really care how much size he traded.
The idea was to create a community where people are talking about trades and the educational points therein, without bragging about their P&L and what a genius they are. There are other trading communities that focus on P&L, and it's a very different vibe there.
After lots of feedback about it, we eventually added the option to share P&L if you want to.
All that said, it's an interesting idea to "rank" traders, and a related idea is to have "verified" results where we know, in some way, that they haven't been tampered with. If anyone feels very strongly about any of this, I'd love to hear your thoughts...
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 3 users say Thank You to Tradervue for this post:
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
I don't care how much size someone trades, that is personal. But there is demand for a ranking system to demonstrate if someone is a good trader overall or not, ranked against their peers. Personally, I don't subscribe to the idea that someone's trading idea is only valid if they can prove they made money with it -- but most others do, especially newer traders. So something to consider perhaps for your service.
Maybe others can comment on the interest in a ranking system (without disclosing total P&L -- like perhaps listing it as "per contract")
We actually can accept Paypal...it's just not advertised on the web site as it's not fully integrated into our subscription system. It requires some manual work each time someone wants to subscribe, cancel, or change between subscription plans.
If you'd like to upgrade using Paypal, shoot us a note ( Tradervue | Contact Us) and we'll get it squared away!
Greg
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 3 users say Thank You to Tradervue for this post:
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
Greg,
I just thought I'd mention some other users journals on futures.io (formerly BMT) that contain good statistical information. I am not very good at Excel, so if something like Tradervue were to spit these out it would be a plus for me
Because I have more latitude in my decision-making now, I'm feeling like my trading isn't as "consistent" as it was before. I'm following my rules, but I'm also feeling unsure of myself because of the discretion …
An updated version of the VS 2010 project to sum cumulative profit of a strategy (or strategies) applied to a portfolio of instruments is attached ("PortfolioStrategyTradeSum.zip", including …
Attached is a series of charts that look at trades in relation to time:
trades by day of the week
time of day for each of my trades
time of day I took my first trade
time I let pass after closing my previous trade before entering the next
Trades …
There are countless others, but my brain is fried today and I have no energy to search for them.
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
Greg,
Do you provide any functionality for portfolio testing? Like if I want to group different classifications of trades (ie: methodologies, systems, time frames) together and then compress them into uniform report that shows me the total performance metrics of the combined class, but while also letting me still see individuals within the class (to spot poor performers, and also to use for analyzing correlation)?
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
My understanding of tags though is that you will see ONLY data for the tag you select. In my case what I need is to see data for a group of trades (tagged, whatever) but also see the total available data for the period.
The goal is to take say 5 different systems, and be able to view them as a whole while also being able to do correlative analysis using all the different metrics Tradervue supports. So it means I would need to see all the data, not just the tagged data, simultaneously.
Again, not an Excel wizard or I would have done this myself a long time ago. But basically in my head I envision (for each metric) multiple plots -- one plot for each methodology or "tag", and then a cumulative plot or totals.
If I'm understanding you correctly, you could indeed do this with tags. For example, suppose you had three systems:
System1
System2
System3
You could tag each of your trades with whichever system they belong to (or multiple systems if that's the case). Then, with the global trade filter, filter by
System1 OR System2 OR System3
That will show you only trades that belong to at least one of the systems mentioned. This will give you the total available data for the systems for that period. If you want to see ALL trades from the period, just skip that step - you can choose whether "all" means all trades, or all trades that are in one of the three systems.
To get a breakdown by system, you can use the tag breakdown report. This will show you something like this:
That's just total P&L and frequency, but it gives you the gist. If you want to _really_ dive in and compare your systems, you could use the "Compare" reports tab, and use
Group 1: System1 tag
Group 2: System3 tag
And then you could deep dive on a comparison between any two systems. You would then see something like this:
where you could flip through any of the report sub-groups and see your systems plotted against each other.
Does that answer your question?
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following user says Thank You to Tradervue for this post:
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
Yes but unfortunately not the answer I was looking for. I don't have any good portfolio analysis to show you as an example, which is why I've been searching for this for a while.
Your "compare" feature is not what I am after, I need overlayed results for more than two separate "tags" and specifically am looking for correlative analysis in all the various metrics computed. You have to think about it on a portfolio level.
As a simple example, the NQ and AAPL are correlated (first chart), but the SPY vs QQQ are almost 1.0 factor:
I am looking for analysis and correlation across net profit, trade frequency, timing, and all the usual metrics and distributions. The goal is to build a robust portfolio of methodologies that compliment each other from a risk/return standpoint, instead of trading highly correlated systems that simply increase risk.
Obviously instrument/product is a starting point here, but I've found trades are still highly correlated even in markets you would think are uncorrelated.
For example, here is GLD vs SPY:
Crude vs Gold:
Barclay IEI Bonds vs Gold:
What I am looking for is analysis that lets me analyze all my systems and their trading metrics. Not just a simple correlation chart based on price data of the instrument as I've posted above.
For some real great reading, check out @imPairsonator's Global Tactical Asset Allocation thread:
In this thread I'm going to document the development of an ETF-based "global tactical asset allocation" system. What are the goals of this system? It's meant to take the main role of allocating a long-term portfolio. Hopefully we will …
This is an interesting idea, Mike. What if it was just cumulative P&L for each system, plotted together on one chart, as well as being able to view correlation between any two? Would that cover what you're looking for, or if not, which other metrics do you think would be most useful?
This dovetails into something I've been thinking about, so there's a reasonable chance parts of this might appear in the product...
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following user says Thank You to Tradervue for this post:
I would want to see a view created where I can see these metrics on a per-system basis while also being able to view the total or average performance for all systems in the portfolio.
And when you view each individual metric, say MAE, MFE, avg time in winners, trades taken at 8AM -- whatever -- then I am looking to see both individual performance per-system as well as avg performance in portfolio, so I am looking for correlation here (or lack of).
Also picture a plot of trade executions by date time. When multiple trades are on simultaneously, I am looking at risk across these systems as a whole. Due to correlation, more positions will not necessarily result in more (likely) risk even if it does use more risk capital. The trick is naturally finding the right combination and thus the right kind of diversification, even with more positions.
I am not an expert in this subject. I have wanted to move this direction for years, but am very limited due to my poor math skills. I'm a visual person and a trial-by-error person. So I've been looking for a portfolio level tool that understands the importance of portfolio diversification and analysis, and then I wanted it to do the heavy lifting while I plug in my various systems.
Hopefully this is ok to post here...if not, Mike, feel free to flog me publicly.
Many futures.io (formerly BMT) folks have talked to me about using R to analyze their performance vs. risk, and asking whether this was possible in Tradervue. Up to now, the answer to this question has been "sort of..." - you could kind of fake it if you traded in fixed risk "buckets", but it certainly wasn't ideal.
It's much better now! A few days ago we added the ability to enter in your initial risk R value for a trade, and then any of the reports can be run showing average or aggregate performance in terms of R, instead of absolute $ return. And this works even if your R is different for different trades (maybe you risk more on trades where you're more confident, etc.).
I have seen that you have charts plotting the entries and also average mae and mfe for overall trade record , my questions is it possible to calculate mae and mfe for individual trade automatically or any other way for individual trades ? Thanks
The following user says Thank You to jinhar for this post:
Yep - we show MFE and MAE for every trade. When you open a trade, look just below the list of executions, right above the tabs for the different charts - there is a link that says "show trade stats". If you click that, you'll see the MFE/MAE data. There are a couple of screenshots of it on the blog here:
You can't annotate the auto-generated charts (at least not yet), but you can embed your own charts if you want to - quite a few folks do this. There is a help page about this at Tradervue | Adding Images to Trades
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 2 users say Thank You to Tradervue for this post:
Most of the time I don't talk here about the new features added to Tradervue (you can always find them on our blog), but this was another one of those things that was based on a ton of feedback from folks I met here.
For futures trades, you can now run any of the reports in Tradervue in terms of ticks, rather than in USD. This was a super popular request, and it wasn't until the risk reporting went in a few weeks ago that it dawned on me how to integrate tick reporting nicely.
Hey guys - this will almost certainly be of interest, since I know many of you are using NinjaTrader.
The folks over at Indicator Warehouse have built a plugin for NinjaTrader called Journal Lync, which will automatically import your trades during the trading day into Tradervue. You just install the plugin on your computer, set it up with your Tradervue account info, and then every trade you make will be auto-imported into Tradervue within a few seconds. Pretty cool...and it's even free!
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
Greg,
I would be interested in using your application for my backtesting analysis. In other words, I would import large subsets of trades (thousands) and then use your app to analyze them. I have hundreds of different backtests containing hundreds of thousands of trades that I would like to analyze.
What I would need is to know what your data formatting requirements are, so that I can have my strategies output (CSV) the necessary information for you to then import and analyze.
I would like to be able to save each import as a unique name or a way to group all of those trades together so I can easily run comparative reports from one backtest to another.
I also would like to utilize the tag system, having my strategy generate "tags" for each trade so I can group and analyze or filter within a particular backtest.
I searched around on your site but couldn't find info on how to format the data so that info can be imported easily.
Last, I know you use an external data provider for historical market info. How many years does this data go back? Some of my backtesting will go back many years and would like to know if this is a problem for some of the statistics like MAE/MFE.
You can tag all trades for a specific import. Or if you want to make sure these trades are kept completely separate (e.g. not merged with other trades from other imports), then import using account tags. You'd essentially treat each backtest as a specific "account", as referenced in this doc: Tradervue | Account Tags
At the moment, you can tag all trades in an import, or you can tag them in the user interface later, but there isn't a way to specify tags in the import data file. You could generate smaller files (so a file contains trades for one backtest that should have a specific set of tags) - to do that use an account tag for the backtest, and regular tags for the other tags. You could also do this with the import API, if you were inclined to write some code - docs at https://github.com/tradervue/api-docs
Our intraday data is available for up to 3 years back. I think the daily data is longer, but stats like MFE/MAE all use the intraday data for calculations.
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 4 users say Thank You to Tradervue for this post:
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
Thanks Greg. Using the API should be fine for me, but I will need to check it more thoroughly to know for sure.
If the API ends up not being the preferred method, it would be nice if you could allow tags to be imported on a per-trade basis with your generic import format. This would make what I am doing today extremely simple to adapt to your importer, whereas writing to the API is more complex because I would need to create a separate C# app to do it, basically reading my import file and then submitting to the API one trade at a time instead of just sending the file to your importer.
I guess I am just being lazy, what I want is a way to quickly test a few backtests without writing the API code If it works and I want to use it on an ongoing basis then of course the API method is preferred.
This has definitely come up before, and I've looked into implementing it. The complexity comes from the fact that tags are applied to logical trades, whereas a "line" in a CSV file is a single execution. If tags were allowed on each execution line in the file, we could merge them to apply to the trade the execution becomes part of. In the trivial case where you have an open and a close in a file, this is pretty straightforward. But in the more complex case, where say a new execution is being applied to an existing trade, that new tag would be applied to the old trade. And if for some reason the merge mode wasn't set the way you wanted, you could end up with some tags applied in a way you didn't want, and could potentially be difficult to undo.
All that said - I do like the idea...just haven't yet figured out the "right" way to implement it to avoid potential pitfalls...
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 2 users say Thank You to Tradervue for this post:
No immediate plans for such a beast. I had a long post about this typed here, but decided most of it probably wasn't interesting to anyone but me :-) So rather than bore you with all of that, basically we feel like we can provide more functionality and more advanced analytics for more people by having it available as an online service.
Greg
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 2 users say Thank You to Tradervue for this post:
It does break out scratch trades, defined as a trade that has a zero P&L. There isn't special handling for trades with a "small" loss at the moment - they will be considered to be a regular win or a loss.
Yes. Well, yes for my interpretation of the question, anyway.
Those three executions will all get grouped into one logical trade; stats like position MFE/MAE and such will be calculated to take into account the fact that you scaled in.
The individual execution detail is maintained as well - you can always see your individual buys and sells if you want to.
Hopefully that answers your question - if I totally missed the point, let me know!
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following user says Thank You to Tradervue for this post:
That particular report shows P&L based on the entry day of each trade.
And so you don't have to wait for me in the future , all of the reports have a Help link at the bottom of the page; in those help pages we try to be as specific as possible with respect to what exactly is being shown.
The Overview reports show realized P&L, as do the cumulative P&L and cumulative drawdown reports on the Detailed tab; most of the other reports show P&L allocated to the trade entry date.
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following user says Thank You to Tradervue for this post:
First, if you just want to view either long or short by themselves, you can use the filter bar to select the trades you want to see; from the "side" dropdown, select either long or short, then click "Filter". Once you do that, you can use any of the reports to view performance for just the selected trades.
Or, if you want to see long and short compared side-by-side on the same reports, then go to the Reports View, Compare tab, and select "Long vs Short" in the quick view menu at the top.
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 2 users say Thank You to Tradervue for this post:
awesome product, do you offer a desktop version? Looked on site but does not seem available. If No, Why? and any plans to introduce one? I ask cause not always available internet.
Thanks
Hey Greg, Have you considered a desktop only version?
No immediate plans for such a beast. I had a long post about this typed here, but decided most of it probably wasn't interesting to anyone but me :-) So rather than bore you with all of that, …
Basically, we can move faster, support more users, and have fewer complications with data licensing by offering Tradervue as a web service.
That said, some people have asked for specific features to be able to create an offline journal that they can print or save, and be able to review offline. We have done this - the trade details, the journal view, and the reports can all be printed (either to PDF or paper) for offline review...and you can also download trade data into Excel if you prefer to review that way.
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 3 users say Thank You to Tradervue for this post:
Hi @Tradervue greg yes I realized had posted before back in 2013, and your response was no "immediate plans", it's been a while. That's why I asked again.
But yes I see you will not. Thanks for the response.
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
I'd like to see a better embed option.
Here is a trade I shared. Unfortunately with the current embed, it's "poor quality" content.
I'd like to see this bigger (up to 620px wide), with a more granular lower time frame chart (ie 1-minute), and with the execution details.
Can you add a more detailed embed option? I realize following the link gives details, but I'd like more detail on the embed itself. Also please make the href open in a new target.
Max width is 464px - same max width that our widget allows.
The chart that's shown is always the most granular time frame chart you have configured for the trade, that will show all of the executions. The particular trade you showed above lasted for 12 hours - so not even the 5-min chart will show all of the executions. That's why the daily is shown.
If you added, say, a 15-min chart to the trade, that one would be used instead.
The executions are shown in the charts as red and green diamonds. In this particular example, though, they're all pretty much on top of each other on the daily.
We might be able to add another option to change the href target - we can't do this by default, as not everyone wants this, but if we add an option you could again perhaps add it to the embed script you use here. Let me take a look at this later this afternoon and see what I can do.
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 2 users say Thank You to Tradervue for this post:
I'll put this on the list to consider in the future. If we rolled out the custom columns to this view, that would cover it, but we haven't done that yet.
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
This one's a super popular request. The issue we ran into at first was that everyone who asked for it had a different idea about what VWAP timeframe they wanted (e.g. intraday, 2-day, 5-day, etc). It's unlikely we'll add of them, but rather just add the intraday version.
What we have on the to-do list right now is a VWAP that is calculated from the start of the trading day (regular session only for stocks), and it would be shown for intraday trades only. And for purposes of this calculation, a futures "day" would be the trading day from yesterday afternoon's open to today's afternoon close.
Also FYI, the big space between your avatar and the price chart is where an excerpt of the trade's notes are shown, if you add notes to the trade. Here's an example showing a trade with notes:
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 2 users say Thank You to Tradervue for this post:
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
Would it be possible to allow importing of the daily trade report that InteractiveBroker sends out?
I can forward a couple of examples to you via email if you need them. They are HTML, but extremely simple formatting.
I am using your bookmark javascript widget, but the problem is I have to login to IB every day to do that, which requires token authentication, generate the report, download, etc. And that report isn't available until midnight.
It would be far easier if I could take the email that IB sends immediately on the close, where they attach the daily log, and then import that directly.
Happy to take a look - forward over the email, and I can see what it looks like.
Even better, though, we've had some folks write some code to get trade data out of the IB desktop, and automatically import it into Tradervue using our API in near real time. If that's of interest, our API docs are here:
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
What does time 'adjusted' mean? These are all direct imports from IB, I've not adjusted anything. If I click 'detail', all the time stamps seem accurate to me.
"Adjusted" means it's a multi-day trade that was adjusted on this day (as opposed to opened or closed). Here's a blog post describing in detail how Tradervue handles multi-day trades and P&L realization:
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
Greg, I was wondering if you could come up with a way to show performance on a slider that would show "what-if" scenarios if a trade had been entered earlier or later, or if the exit had been earlier or later.
Say for example, if the last 100 trades had been exited 5 minutes, 30 minutes, or 1 hour later than they were. What would happen to performance?
I felt like @Tradervue was in a good position to attempt this, because you have the real-time market data to go along with the executions.
My goal in such an exercise would be to compare my actual performance against a hypothetical performance, to see if I am timing the market well.
We've talked with a few folks about things like this...two things come to mind.
First, for entries, it's extremely difficult to do this in a way that's actually useful without additional information. Presumably you have rules for when you enter and exit trades; e.g. you might exit when a particular stop is hit. Without knowing these rules, it's difficult to predict whether "entering 15 minutes earlier" is actually relevant to anything or not. If your (price-based) entry criteria was not met, then calculating a hypothetical P&L from that point isn't really actionable information.
However, for exits, we are actually working on some interesting things here. What we're looking at is, for your exit, how much did you leave on the table vs. the ideal exit? You can think of this in terms of either absolute P&L, or as an exit efficiency in %.
This too may or may not be relevant in aggregate; if you exited a momentum trade because the momentum stopped, and then it held around a price for an hour and then popped up 5 more points, well, you didn't make a mistake there - you were playing momentum, and correctly exited when the momentum stopped. For this trade, what happened afterwards is irrelevant.
However, for trades with a more discretionary exit (e.g. I think this is nearing resistance so I'm going to sell), we may be able to show you that you're not very good at identifying actual resistance levels, because you're consistently exiting too early on these trades and leaving money on the table.
Interestingly, the definition of exactly how it should be calculated is trickier than doing the calculation itself. For example, for trades where you're scaling in and out, we need to make intelligent decisions about what we try to guess. The fact that you sold half of your position before price ran up 5 points doesn't necessarily mean that first sale was a mistake, or that you left anything on the table, per se - maybe you were just managing risk. Similarly, if you sold, then it drops 5 points below your entry price, then jumps up 20 points, it's probably not useful to imagine how nice that 20 point gain would have been. I think we have it pretty nailed down now, at least for the first version...hopefully we can get something online in the coming weeks.
If anyone has additional ideas or feedback about what they'd like to see along these lines, I'd love to hear it!
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 3 users say Thank You to Tradervue for this post:
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
All good points.
In a dream scenario (I was thinking about this in my sleep last night), I would like a full sized chart on a daily basis that shows actual entry/exit positions, and then be able to slide a slider left/right on the x-axis that would show what would happen if exits were pushed/pulled by 'x' (minutes). This would let me visually see the impact on a granular basis.
I think the reporting mechanism has to be able to break it down in detail to be useful.
As for the exits you described, I tracked this for years in my personal Excel worksheet and I called it the "theoretical maximum" and later just called it "Efficiency". In other words, if I am 90% efficient on that trade, it means I captured 90% of what the maximum possible outcome was. But this isn't MFE. This included post-trade consideration. Which meant it was discretionary based on my rules.
I know it's a challenge, but I think any effort in this direction would be appreciated.
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
Would love to see a comparison against a benchmark like SP500. Comparison of buy/hold PnL vs my actual PnL, as well as some correlations on +/- days, comparison for drawdown, etc etc
In order to provide meaningful correlations against a benchmark (other than the +/- days), we need to know your overall portfolio performance. And in order to calculate that performance, we need to know your account balance, which we don't track at the moment for various reasons.
We do provide some reports (primarily useful for intraday trades) that break down your performance based on behavior of a market proxy (we have 7 of these in place). This will show you things like, for example, whether your performance is positive on days where the market closes up >2%. This would cover the +/- days that you mentioned. Here's a blog post about this:
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
Thanks. I ran the report using IB's portfolio tool, it wasn't that helpful anyway.
I would rather see you spend time developing the VWAP based entry/exit efficiency we previously discussed. BTW, another sleepless night had me thinking that it should be broken down in terms of % distance from VWAP (price entry/exit). This way you can group your trades in ranges, and see for instance where your best trades are positioned from (+1SD, +2SD, mid-line, etc).
Interesting idea to bring VWAP into the efficiency calculations - thanks, I'll give this some more thought.
Tangentially related, we do have a couple of ATR-related reports. The first (perf/dist by ATR) wouldn't be useful for you since you're primarily trading one instrument, but the second (showing performance based on your entry as a % of ATR) might be useful; both are available at Reports View / Detailed tab / Instrument group.
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following user says Thank You to Tradervue for this post:
In the meantime, you could do this with Excel fairly easily. In Tradervue, go to Reports View, Overview tab, Calendar group, and click "Download" near the bottom of the page. That will download the daily P&L to Excel. You could then add a column for cumulative P&L, and plot that with any curve-fitting you wanted.
Thanks!
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following user says Thank You to Tradervue for this post:
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
Thanks, I didn't see that as I am not Gold (although probably should be). I didn't do Gold because the risk analysis won't work for me, and including commissions is not that important to me.
If you already have a silver account, there isn't a trial for gold. That said, you can switch back and forth between silver and gold, and the charges will be pro-rated; for example, if you switch to gold to try it out for a day, and then decide to go back to silver, you'll get charged less than $1.
Yes - go to Reports View, Compare tab. For group 1, enter the range of open dates you want (e.g. 1/1 - 1/31), then for group 2, enter the other range (e.g. 2/1 - 2/28). You can use the same technique to compare any two groups of trades.
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following user says Thank You to Tradervue for this post:
I'd really like the ability to attach a template or my daily analysis and trade plans. Can you consider adding a way to put or attach one's Daily trading sheet/plan to the journal or?
The way I can think of offhand is to screenshot and import a screenshot of it. Then imbed the image. Basically using a word processor or similar template for my daily trading plans then screenshotting it then imbed it on tradervue.
On my trade plan I have my levels and hypothesis' , various other. Ultimately I may have 5 trades a day, and would like to have a link or attachment or a way to find what my hypothesis' were and which one came to fruition.
I'd also like to have ability to input custom parameters.
Can you be a bit more specific about what you'd like to be able to do here? Not just what you want to input, but what you want to be able to do with them...thanks!
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
I'd like to be able to track my own personal stats on different things, by allowing it to be customizable you can track what matters to the user.
Personally I'd like to track things like my ability to read the market state, my ability and ratio of my bias accuracy.
Say I read the market I'm trading is "in balance," and believe it's "transitioning to imbalance up."
I'd like to know how many times I'm accurately reading that. What my loss ratio is on days I get it wrong(flexibility).
also
I'd like to input certain things then have a gauge on them like 1-10. The 1-10 is like a gauge...seen at the doctors office for describing pain for non English speakers. Related - If using a spreadsheet you could use a 1-10 gauge.
Conviction of homework
Like my personal conviction of the trade on entry.
Psyche during trade.
etc.
Thanks for the thoughts. We've considered adding custom fields, so I'll add your comments to my notes on the topic. The tricky part isn't just adding fields you can type into - it's doing it in a way that can be meaningfully used for analytics and reporting.
In the meantime, consider using notes and tags for all of these things. If you use tags in particular, you'll be able to use them for filter and reporting. For example, if the market is "in balance", tag the trades you make those days with something like "balance", along with a tag for where you believe it's transitioning (e.g. "to-up"). Then, you could run a report for the combination by creating a tag filter for "balance AND to-up".
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 2 users say Thank You to Tradervue for this post:
Hi Greg, I was wondering, is there specific reason why TraderVue is limited to US equities?
Would it be possible to consider using custom symbols for those of us trading other markets.
It's purely an issue of market data. For each additional market we support, we need historical pricing data, both EOD and intraday. For some markets, this is prohibitively expensive. For example, for Australian ASX, the cost to license this data is something like 3x that of all the markets we currently support combined.
That said, we've added other markets in the past (for example, Eurex and BMF futures), and we base this sort of thing on how much demand we get for each market. So definitely let us know specifically which markets you'd like to see, and we'll get your vote counted!
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
Thanks Greg. Besides the US, I also trade the GCC markets (Dubai, Saudi, etc), so I doubt my vote will count as there must be very little interest in these markets
That said, I understand the need for market data to display the charts/trades, calculate MAE/MFE, etc, but there are also a lot of valuable individual trade statistics that TraderVue provides (win rate, trade expectancy, drawdown, etc) that are completely independent from market data.
Not trying to compare products here, but I've been doing some digging lately trying to find something that would analyze my trades, and while I found a a few tools that support custom instruments, the performance reports were no where near what TraderVue provides.
So my suggestion is that if custom instruments were supported, it would be a great way to leverage the analytic power of TraderVue and make it more versatile.
I understand that this may be a strategic decision on your part, but I just thought I'd share my suggestion.
The following user says Thank You to silacitrader for this post:
We just released this today...it's pretty cool. Starting today, Tradervue will calculate the “best exit” P&L for each of your eligible trades. That best exit is determined by “floating” the last exit executions up to their highest point (or lowest, in the case of a short); this exit point may be before or after your actual exit. By looking at this best exit value in comparison to your actual P&L in the trade, you can quickly see how efficient you were in terms of extracting P&L.
We also take risk into account - we won't show you a huge potential exit P&L, if that would have required an unrealistic drawdown to get there.
I wrote a blog post with a ton of detail and examples - rather than copy all that here, I'll just point you to the post:
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
@Tradervue, I am very excited with your new exit analysis including the "efficiency", this is how I tracked my trades for years when I used Excel.
One thing that is not yet clear to me is how this works when you scale out. Your article says it takes the "last trade" but what happens if you scale out over several hours? I assume that only the remaining size left on for the "last trade" is used?
Hi @Tradervue yes I know that I have asked in the past and you have responded and Thank you for that.
I guess what I was getting at all along is a solution for desktop only--I have no need to place my info onto a server somewhere, thus my request for a desktop version, yes I understand you do not offer such an animal. I respect that and your product is very robust and user friendly. With that said is there any possibility you would recommend a desktop solution for users that desire such? Realizing a conflict of interest but thought I would ask anyway!
Thanks in advance for any input on the subject.
You are correct, in that we only "play" with the _last_ scale-out. Specifically, we look at the last exit execution, and include any other executions within 5 seconds of that one (to allow for multiple fills that may be slightly delayed).
An early (unreleased) version of this analysis looked at all of the exits after the last entry-side execution. So for example, in this trade:
We originally would move all three of the sells up to the highest point at ~ 23.50, and use that as a best exit.
The result seemed to be statistically interesting, but it was almost an _entry_ efficiency, rather than an exit efficiency. It showed what the theoretical max P&L was. But it didn't seem to be actionable, in terms of using the data to improve performance. Perhaps your trading rules say you're going to sell 1/2 into the first pop, or perhaps sell 1/4 to cover risk as soon as possible. In these cases, you followed your plan - so "floating" those exits up to the maximum P&L doesn't give you anything useful.
However, for the _last_ scale-out, this one is either done as part of a plan, or it's discretionary.
Planned: perhaps you were making a momentum trade, and you sold when momentum dried up. You followed your plan, and executed correctly. In this case, Best Exit P&L means nothing, and can be safely ignored.
Discretionary: you sold because you thought you were done with the trade; perhaps you saw some sell signals. In this case, Best exit P&L is a good measure of whether you made a good choice.
This is all a long-winded way of saying yes, we are only looking at the last scale-out.
I certainly welcome any discussion!
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following user says Thank You to Tradervue for this post:
For MFE/MAE, it's not there because these calculations are made in the instrument's native currency, and may not all be the same. For example, ES trades have MFE calculated in USD, whereas FDAX trades have MFE in EUR. So, providing an average/total in the Trades View isn't possible.
That said - for USD-denominated instruments, you can go to the Reports View, Detailed tab, and the average position MFE/MAE are there in the stats table.
For exit efficiency - let me give that some thought. Like you said, some trades do not have this calculated (e.g. losing trades), so I'll have to think through whether having an average there makes sense in the general case. I can see how it would be useful - I just want to make sure we don't show something that someone might find misleading.
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
Would it be possible to change PnL and all other $ measures to % relative to account size?
Just came up in another thread & might really help people evolve in size as they can detach from the invcrease in absolutes more easily..
For equities, you can view P&L in % in the Trades View. However, that's showing a % return based on capital requirement for the trade - not account size.
In general, we don't know your account balance, so we can't calculate portfolio % returns. We've talked about adding this, and it's on the list to consider...the primary issue is most of the brokers/platforms we support don't include this information, so it would require manual entry and updating, which would be error-prone. And the account balance number we show would often be incorrect (until you updated it), which wouldn't be a great user experience.
Also on the idea of calculating an individual trade P&L relative to account size, it's not clear how useful this would be, in terms of actionable data to improve performance. If I make 5 ticks on 10 contracts, is that a better trade if I had $100K in my account, vs. if I had $110K in my account? I would argue that calculating returns in terms of risk (R) results in data that's more easily compared between trades.
That said - I can see how calculating overall portfolio performance in % terms is useful. But it seems more useful for portfolio performance, rather than _trade_ performance.
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following 2 users say Thank You to Tradervue for this post:
thanks for taking your time to reply. I already thought about the issue of where to get the account balance from.
My conclusion is that it should be relatively simple if the user enters an initial balance once and has the option to adjust for payments out of or into the account. All other changes would naturally be a result of trade result + commission.
Of course I have no idea how hard something like this would be to implement..
anyhow, thanks for considering!
Edit: as for the quality of a trade - the % in itself is not saying too much, but when looking at it along with things like the R relative, it can indeed help improving performance as you would see more clearly if you are trading too big or too small relative to your risk capital (ie. 50 ticks might be a good result for a 25.000$ account but not so much on a 250.000 account, unless you intent to risk only o.1% per trade.)
Oops - I believe I mis-spoke in the above reply. It is indeed possible to have a best-exit calculation be less than the position MFE.
In our exit calculations, we’re only moving the _last_ exit execution (and those within a 5-second window of that last one). The MFE may occur at an earlier point in the trade, and your trade may have backed off from the MFE before getting to those last exits. So, in some circumstances, it’s not possible to get all the way back up to the MFE just by moving the last-exit executions.
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following user says Thank You to Tradervue for this post:
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
In order for me to crunch data, can you provide the actual last scale out broken out as a column so it can be completed to the theoretical best scale out (aka best exit)
Right now, we calculate the best exit P&L, and compare that with the actual P&L. Let's look at two extremes:
Example 1. You have a trade with:
Actual P&L: $5000
Best Exit P&L: $5100
In this case, you extracted nearly all of the available P&L from the trade. Whether your last scale-out P&L was $100 or $3000 doesn't seem useful - in either case, there was only another $100 available for the taking.
Example 2. Your trade is:
Actual P&L: $5000
Best Exit P&L: $15000
In this case, we know that whatever your last scale-out P&L was, there was another $10000 available for the taking. And again - whether your last scale-out P&L was $100 or $3000 still doesn't seem terribly useful to know.
Now in general, I can see the draw in knowing something like "my last scale-outs are capturing 34% of their potential" - but while potentially interesting, I'm not sure I understand the value.
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
Site Administrator Swing Trader Data Scientist & DevOps
Manta, Ecuador
Experience: Advanced
Platform: Custom solution
Trading: Futures & Crypto
Posts: 50,028 since Jun 2009
Thanks: 32,487 given,
98,346
received
Maybe I am missing something, but right now in the CSV there is no way to determine what % of the 'actual pnl' the last scale was. But the 'best exit' is based solely on the last scale. So there is no way to really compare the two columns together.
In my situation, the last scale could be 20 contracts, it could be 1 contract, or anything in between. I'm just trying to find a more direct way to understand/utilize the 'best exit' column.
Maybe a short cut could be to show the last scale size. We already know the total trade size.
You're understanding correctly...we don't show you the size or P&L from the last scale.
I'll give this some more thought. But again, I don't think this calculation is meant to micro-analyze how good your last exit was; I mean, that's what it shows, but I don't think it's super useful to think of it that way. Rather, it's a measure of how well you exited the trade as a whole, given that your last exit was discretionary. That's why I think it's more useful to think of how good your last exit was in relation to the overall trade P&L.
That said, I do understand what you're saying, and I'll noodle on it a bit.
If you have any questions about Tradervue, please send me a Private Message or use the BMT "Ask Me Anything" thread.
The following user says Thank You to Tradervue for this post: