Welcome to NexusFi: the best trading community on the planet, with over 150,000 members Sign Up Now for Free
Genuine reviews from real traders, not fake reviews from stealth vendors
Quality education from leading professional traders
We are a friendly, helpful, and positive community
We do not tolerate rude behavior, trolling, or vendors advertising in posts
We are here to help, just let us know what you need
You'll need to register in order to view the content of the threads and start contributing to our community. It's free for basic access, or support us by becoming an Elite Member -- see if you qualify for a discount below.
-- Big Mike, Site Administrator
(If you already have an account, login at the top of the page)
Hello, I've been trying to search everywhere for a clear answer and I haven't found it, so I find it really hard to tame my expectations.
I've done a bunch of (manual) backtesting and forward testing in the last few months, 2500+ trades so far, and slowly I'm starting to find ways of trading that are much more robust.
I generally getting a 10%~ positive expectancy (~55% win rate with 1:1 reward/risk ratio, although I also do breakouts sometimes where I don't have a set TP) while doing about 6 trades per day where I risk ~$125. All of this is after accounting for B/A spread and commissions.
This means that I'm making ~$75 per day, per contract, and while risking 750$ to make that amount.
My question would be: is this realistic? Should I expect my edge to be a little smaller? If so, by how much?
And I know that different people get different results and it all depends on how you trade and such but, generally speaking, after going through a lot of trades and a lot of market conditions, would my results fall in the spectrum of what could be considered realistic for day trading?
This is a bit speculative obviously but a great topic.
But, if we're talking futures, I suspect a great day trading return would be 25% to 30% average return on risk. This would be equivalent to around $225 per day on $750 risk. At 3% risk per day, this translates to a theoretical maximum return of around 200%.
I think an active trader should shoot for a MINIMUM of 100% return on risk. However, it is very difficult to do this in practice. Generally speaking, an elite level day trader should be able to return anywhere from 50% to 100%. I consider any return in this level "successful". However, it obviously depends on how much you are risking. Returns at this level have been well documented. To go beyond this level, you would most likely need to special access or reduced trading cost and/or to risk a most of the account (over 50%).
If you want to get an idea from tracked/public accounts, you can look at Striker Securities, C2, or World Cup to see what the best public returns are. However, keep in mind that an independent trader should be able to beat these returns because these systems have much higher trading costs. But, they do give you a good baseline.
Now as for what you can expect.. here's the downer, there are very few good times for day trading over the past several years. Many day traders will make most of their profits over a few months where volatility is elevated.
We also need to address "shot taking". In order to understand shot taking, you must understand that if every bet you take is an independent bet with an edge then your edge will be the same no matter what your account size is. In other words, if I have only $500 and a trade cost me $500 and has an edge-- the edge is the same for me as for another trader who has $25,000. But, by the same token, it is very likely -- the most likely scenario is that I will blow out. You might question, how can this be? The reason this can be is that if I manage to hit a good streak, my return can be several times my actual account size/risk.
The way you can see this is imagine I have a system that requires 12k to trade historically and I put in 4k in my account. I have keep rest out. Now, I might get lucky and hit a good streak and get to 12k before losing the 4k, I could claim my return is like 3x. However, in reality we knew I might have to load more even to withstand normal fluctuations. Basically, if on average say I need 8k then I will blow out about 50% of the time.
As for your results, generally things will get worse. Over the long term, most CTA's has sharpe ratios at or below 1. This translates to roughly needing to risk 50% of an account to make 50% return.
I'd be weary of trusting anyone with telling me whats realistic or not. In all honesty the only real way to find out is by trying- thats when reality creeps in and the cracks become apparent. Backtesting and theorycrafting is one thing, but have you done enough testing on SIM to ensure this is something you are able to execute without hiccups?
I'll offer some of my initial thoughts but again be careful of trusting strangers on the net too much
- 1:1 R/R with $125 stop/T sounds scary to me(bit less so if you'd only be scalping few ticks on something thick like bonds) and would put a serious mind-block in most folks after a few consecutive bad days.. the math might add up in theory but you'd be leaning too hard on the win rate dependancy imo. Also a TP/target of $125 on CL (which you've listed in your profile) is a measly 12 ticks.. thats a fairly modest target for an instrument that loves to oscillate more than 12ticks gyrations, and usually puts in at least 80++ ticks daily range. Have you looked into bigger targets? Or simply letting winners run longer?
Also might be a good idea to take a look at something cheap like M6e or QM (CL mini) or perhaps even FX or just normal stocks to test out live without risking $750 a day. There are lots of instruments out there to trade which are far better suited for learning curves and ironing out kinks other than volatile outright futures- which is about as geared and expensive a lesson you can find.
"Hey! I've got 10 fresh apples to sell, any hungry buyers out here today?!" "Maybe! Just let me check my moving average and RSI!"
I was looking at Hedge Fund market wizards and here are some of the "Gain to pain" ratios ( = net returns / gross_losses, meaning it's the same as positive expectancy), most were using Global Macro I believe:
3.2, 4.1, 5.6, 4.2, 1.43, etc. Jack Schwager said that "For monthly data, roughly speaking, GPRs greater than 1.0 are good and those above 1.5 are very good. For daily data, the corresponding numbers would be approximately 0.17 and 0.25."
So it seems that @tpredictor was right in that 100%+ returns on risk are possible (I was surprised by that, hadn't realized this when I first read the book), and the 25%-30% is similar to Schwager's 17%-25%, but I'll maybe have to 'aim' significantly lower because I'm not doing huge swing trades on intraday charts, I'm doing something closer to scalping so commission alone probably reduces my profits a fair bit.
"Backtesting and theorycrafting is one thing, but have you done enough testing on SIM to ensure this is something you are able to execute without hiccups?"
I've only forward tested (SIM) to a very limited amount, basically just to ensure that I could execute the trades without getting overwhelmed, the returns were positive but it's way too small to know if they're similar to the backtests yet.
"1:1 R/R with $125 stop/T sounds scary to me(bit less so if you'd only be scalping few ticks on something thick like bonds) and would put a serious mind-block in most folks after a few consecutive bad days.. the math might add up in theory but you'd be leaning too hard on the win rate dependancy imo."
There's an upside to that, which is that if you're doing 6 trades per day across a couple of instruments (let's say 5, I've been able to do that on demo at least, on 15 min charts it's not that tiring), you're doing 30 trades per day, or about 600 per month. That means that your weekly/monthly results will (generally) be much closer to the positive expectancy and you can afford to stake a lot less per trade. Personally, I'm much better at dealing with "short term losses" rather than dealing with the long periods of doing nothing and waiting for the trades to 'play themselves out', although I'll be looking to do that in a systematic way.
"Also a TP/target of $125 on CL (which you've listed in your profile) is a measly 12 ticks.. thats a fairly modest target for an instrument that loves to oscillate more than 12ticks gyrations, and usually puts in at least 80++ ticks daily range. Have you looked into bigger targets? Or simply letting winners run longer?"
I did a backtest with 1100+ trades where I was trend following, on CL the average SL was almost 20 ticks, got about 20% positive expectancy... but I sometimes cheated ("Oh, I missed this, this is a clear signal, I wouldn't miss this if it was in real time") and I didn't note down when I did so, so the backtests are pretty close to worthless. That said, I've been testing a breakout system that's similar, but more selective, and my W% and edge are very similar. I'll definitely be careful when testing things live, thank you for the recommendation.
Spreading your 'eggs across multiple baskets' so to speak is a nice safety net in that you can expect an overall more stable performance on multiple instruments as opposed to only relying on one instrument to behave well towards your strategy. But there's a flipside to that upside, if you're not taking all your signals on all of those markets then you risk opting out on the winners and taking the losers. The minute you become selective in which signals to take, your statistics need to be viewed very differently. So the flipside becomes having to take your average 6 signals across 5 markets, resulting in 30 trades a day which in an(almost) worst case scenario will produce 30x $750 = $3750 dollar loss a day. If your strategy is sound and you're account can take 4k daily loss fairly easily then its a viable premium to consider for the general stability increase, but its a huge increase in your risk of ruin if you dont have the account size. Trading is a marathon and you need to be able to take the pain from the bad periods. Hope for the best, but always have a plan to survive the worst
I'll also leave you with a corky blanket statement, in that learning to trade is a long and costly road for the majority. Focus on doing all you can to minimize your tuition-fee. Many get into this too fast thinking they've found the solution, but ones odds of success straight out of the gates is minuscule.
In trading its better to be a pessimist proven wrong, rather than a broke optimist
"Hey! I've got 10 fresh apples to sell, any hungry buyers out here today?!" "Maybe! Just let me check my moving average and RSI!"
Then unless I'm missing something obvious ( quite normal!) I'm afraid as far as I can see that's a recipe for disaster unless your unqualified breakout discretionary trading can keep you afloat.
I get the point you're making but that wording will easily confuse folks. I'd say you're risking $125 to make $125 with a 55% win rate. The net effect over a large enough sample rate is that after the 55% winrate you expect to make an average of $12,5 per trade (even less after comms and slippage) which over 6 trades a day should statistically result in 12,5 x 6 = $75 daily average.
Basically: 100 trades
45 losses, 45 * 125 = -5625
55 wins, 55 * 125 = +6875
$1250 gains / 100 trades = $12,5 avg. theoretical gain per trade before comms, slippage, fatfingers, missed signals etc.
Throughout a day with 6 trades you are risking a worst case scenario of 6x125 = $750 max drawdown to make an expected average of $12,5 x 6= $75 a day.
But on a roll by roll basis you are risking 1 to gain 1.
To me these are not very favourable margins, and they come with a huge skew potential if/when your winrate dips (4 losses a day: -500 + 2 wins +250 = -250 net, meaning you need at least 2 wins the next day just to recoup previous day). But to each his own.
"Hey! I've got 10 fresh apples to sell, any hungry buyers out here today?!" "Maybe! Just let me check my moving average and RSI!"
