FXCM Release Extra Forex Trader Profitability Statistics
Just in case you hadn't noticed, today is October 18th 2010. Today is the day that all those new CFTC regulations come into force, and when a forex trader opens a new account brokers must now reveal to them quarterly account profitability statistics covering the previous year. FXCM have done just that, and here are their numbers:
| Quarterly Report | % Profitable | % Unprofitable | Total Non-Discretionary Accounts |
| Sep – Dec, 2009 | 26% | 74% | 22,371 |
| Jan – March, 2010 | 25% | 75% | 19,049 |
| April – June, 2010 | 23% | 77% | 17,771 |
| July – Sep, 2010 | 23% | 77% | 15,023 |
Comparing FXCM's profitability statistics with Interbank FX's that we published last week reveals firstly that FXCM's clients seem to be slightly less profitable than Interbank's, with an average of 24.25% against 29%. Secondly FXCM has slightly more active non-discretionary accounts, a quarterly average of 18,554 against IBFX's 17,525. Note that FXCM's figures include only accounts with FXCM LLC. I anticipate that FXCM's figures will increase far more than Interbank's for the final quarter of 2010, when lots of dollars that left the United States in search of a balmier forex trading climate on this side of the Atlantic are forcibly repatriated courtesy of the CFTC.
In addition to the legally required statistics, FXCM have also released to us additional information, breaking down profitability by trading account size:
| Equity Range | % Profitable |
| $0 – $999 | 27.89% |
| $1,000 – $4,999 | 40.52% |
| $5,000 – $9,999 | 42.36% |
| $10,000+ | 47.74% |
It's interesting, although not unexpected, that traders with larger accounts are more often profitable. However it's still true to say that in all categories most traders are unprofitable.
Note that we're not really comparing like with like here, since these figures are monthly averages. The "official" numbers instead cover individual quarters. Comparison between the two tables does provide additional evidence for the suggestion we made when we analysed Interbank FX's trader profitability statistics. FXCM's figures confirm that if you increase the time period you use when assessing profitability, the percentage of profitable traders decreases. Unfortunately this additional data doesn't include the number of traders in each equity range. Nonetheless the lowest percentage in the monthly figures is 27.89%, compared with the largest of 26% in the quarterly figures which certainly rams that point home. FXCM have assured us that as they crunch more numbers they will release additional statistical data to us, so hopefully in the not too distant future we'll be able to come to firmer conclusions about variability in profitability between different groups of traders.
However the unavoidable conclusion from all the numbers that have been crunched and then released so far is inescapable. Most retail forex traders lose money.
Filed under Regulation by 
Leave a Comment
Pings on FXCM Release Extra Forex Trader Profitability Statistics
Comments on FXCM Release Extra Forex Trader Profitability Statistics
Hi everyone,
I would say, the post is very educational. It would be even more interesting, how much of the traders are using automated trading and how much of them are profitable. Maybe, in the not so far future such very interesting information will be available to us also?
Hi Stan,
An excellent suggestion!
It would also be very illuminating to be able to compare like for like profitability percentages compiled over a range of timeframes. For example: 1 month, 3 months, 12 months, 2 years, 5 years, 10 years, 25 years.
Jim
Thanks for sharing this interesting information.
As the average is between 23% to 26%, and the breakdown figures are above the average, which accounts pull the average down? Micro accounts? These numbers still need an explanation…
@Yohay:
Hi Yohay,
The monthly numbers give a more detailed view of profitability by equity on a monthly timeframe. Comparing the quarterly average over the year will not necessarily equate to the monthly average over the year, and here is why.
Just because an account is unprofitable for the quarter, does not mean that it was unprofitable for all 3 months during that quarter. If the account is profitable for even 1 month (during the quarter) but was unprofitable overall for the entire quarter, then of course the monthly averages will be higher than the quarterly averages. The figures are more detailed than what was provided on a quarterly basis.
I hope you found this data insightful, and FXCM will be making more information available to give additional insight into profitability statistics based on accounts sizes, currency pairs traded and more.
-Jason
FXCM have now made the next snippet of additional information available. These do allow us to make like for like comparisons with the "official statistics", since they are based on quarterly percentage profitability, averaged over one year:
The equivalent number derived from the data in the first table above works out to be 24.25%. Once again these new numbers show that most traders are unprofitable, and that better capitalised traders are on average more profitable. Indeed over a 3 month period more than twice as many of the best capitalised traders are profitable compared to those with less than $1,000 in their "bank".
Although FXCM still haven't released the precise details, the two breakdown tables do indicate that most of FXCM LLC's "active self-directed" customers fall into that under $1,000 group. Note in particular their sharp drop in profitability from 27.89% over one month to 20.91% over three months.
Hi Yohay,
The breakdown figures by size of account are one month averages, over 12 months I assume. The headline "official statistics" are raw numbers for four different 3 month periods. Unfortunately the two different tables of information don't allow us to compare "like with like".
Comparing the two tables does however tell us that fewer traders are profitable when checked over a three month period than when checked over a one month period. That is what I would expect. The breakdown statistics do indeed suggest that "micro" traders are, on average, less profitable than "mini" traders. However strictly speaking the breakdown is by account balance rather than by account type.
What would be very interesting to see would be like for like figures over significantly longer periods of time. Again, I would expect the number of profitable traders to reduce as the period of measurement increases.
What would be most interesting of all is the percentage of traders that manage to stay in "the game" for several years and remain profitable. Is it the 5-10% that is frequently bandied about? Or not!!
A very interesting supplementary question, for us here at the Trading Gurus at least, was posed by Stan. How many successful long term forex traders have automated their trading, and how many still sit staring at several screens (SSSSS for short) for long periods?
Jim
Hello,
Apologies if the info is published somewhere else, but I am looking for the info regarding the account size statistics at the retail broker level, e.i. number of accounts breakdown by size. All tables above are citing 10K+ as the last size. Does that mean that accounts with 25K or 50K are rare with retail brokers? Or at least if educated guess is available what is generally considered to be the small, medium and big size account?
Thanks.
Hi John,
I did ask FXCM if they were able to reveal how many accounts there were in each equity range, but I have yet to receive an affirmative answer. Perhaps more information might be forthcoming once FXCM's impending IPO is out of the way?
Nonetheless FXCM's S-1 filing does impart some relevant information regarding your question:
Perhaps in such circumstances one should assume that the numbers of clients FXCM has with accounts in the $25K+ bracket are not statistically significant?
Jim
Well, there are some numbers in S-1… Customer equity $424,597K and 134,478 active accounts which amounts to $3,157 average balance. This strongly suggests that $25K+ bracket is most definitely irrelevant. It gets even more so when you take "tradeable accounts" instead of "active accounts" (whatever they use as criteria to distinguish those). There are 174,672 such accounts, which gives an average balance of $2,430. I would go further and say that it's obvious why they were reporting only 10K+ bracket as the highest because that looks to be the last statisticaly relevant range.
My investigation into this was due curiosity to see where do I fit with my account size and how much leverage I have (if any) when trying to get something resolved, mostly on the technical issues of API and such. I am fully automated and have an account with Dukascopy. While my account is way more than above numbers, I am not sure I do fall into high-end range of Duka accounts.
Another interesting number from S-1 is the institutional side of FXCM, which seems to be rather small, confirming that their retail is the core business. Revenue from institutional trading is listed as $20.7M versus $234.6M on the retail side.
As to your observation about the safety of funds, that is the one I personaly have at the very top of my list when selecting a broker, and the numbers from Tradestation are somewhat confirming the logic as well. They report 47,434 total brokerage accounts (no breakdown by type, but they say the vast majority are equity and futures accounts) with average balance of $70.7K for the equity accts and $21.8K for the futures accts. I did not find the source document myself, but I have seen a post which suggest that average size Universal Account at InteractiveBrokers is $125K.
However, I do not believe that safety of funds is prevailing reason for those numbers. I believe that retail FX is almost entirely made up of entry level, part-time traders, and professional/full-time and well funded traders are more exception than a rule (and by those I mean individual traders, not institutions and money managers). Regarding the profitability statistics from the original post, this would be the logical explanation why smaller size accounts are less profitable.
Lastly, I have just realized the huge difference in the number of accounts reported in the S-1 filing and your original post (174K/134K versus 18.5K). I hope I am not missing the point somewhere…
Thank you for providing the space for this topic.
Best regards,
Hi John,
Thanks for digging out and crunching all those extra numbers.
Regarding your final paragraph, I assume that the S-1 numbers include all the overseas accounts at the likes of FXCM UK and ODL. The figures in my original post include only accounts remaining in the U.S. after the rush to move offshore. I assume that when FXCM's Q4 2010 statistics are released the number of accounts will have swollen considerably thanks to the CFTC's insistence on U.S. dollars returning to their rightful home!
I don't know if you write your own software for automated trading with the Duke, but if so you might like to take a peek at the relevant section of our community forum.
Cheers,
Jim
It is interesting to see that most are unprofitable. I would think a platform that allows people to place positions on either side of the currency would create a situation where there would be equal loss to gain minus the early closed postions. As a broker, typically you are a bookie who takes your margin and has both sides covered to minimize your loss. Your margin is your take. Forex takes a margin and seems to take apx 26% of the losses (100%-24% =76% unprofitable, take the 76% and subtract the profitable 24% to get the profit taken by customer's closed positons. 28%) from its customers as well. Something does not seem kosher.